university of groningen genetic predisposition to ... · contains trophoblastic giant cells), yolk...

University of Groningen

Genetic predisposition to testicular cancerLutke Holzik, Martijn Frederik

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2007

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Lutke Holzik, M. F. (2007). Genetic predisposition to testicular cancer. s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 29-06-2019

https://www.rug.nl/research/portal/en/publications/genetic-predisposition-to-testicular-cancer(9ac5dc7f-5199-4d6d-8304-d8ebebc33fe0).html

CIP-gegevens Koninklijke Bilbiotheek, Den Haag

Lutke Holzik, M.F.

Genetic predisposition to testicular cancer

Proefschrift Groningen. – Met lit. opg. – Met samenvatting in het Nederlands.

Verschijningsvorm: Digitaal

ISBN: 978-90-367-3109-6

© Copyright 2007 M.F. Lutke Holzik

All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or

transmitted in any form or by any means, mechanically, by photocopying, recording or otherwise

without the written permission of the author.

Cover and lay-out: Annemiek van der Kleijn

Printed by: Gildeprint, Enschede

RIJKSUNIVERSITEIT GRONINGEN

Genetic Predisposition to Testicular Cancer

Proefschrift

ter verkrijging van het doctoraat in de

Medische Wetenschappen

aan de Rijksuniversiteit Groningen

op gezag van de

Rector Magnificus, dr. F. Zwarts,

in het openbaar te verdedigen op

maandag 1 oktober 2007

om 16:15 uur

door

Martijn Frederik Lutke Holzik

geboren op 25 februari 1974

te Enschede

Promotores Prof. dr. H.J. Hoekstra

Prof. dr. D.Th. Sleijfer

Copromotores Dr. R.H. Sijmons

Dr. J.E.H.M. Hoekstra-Weebers

Beoordelingscommissie Prof. dr. S. Horenblas

Prof. dr. J.W. Oosterhuis

Prof. dr. P.H.B. Willemse

Paranimfen Stephan Lutke Holzik

Deepu Daryanani

All studies in this thesis were designed and carried out at the department of Surgical Oncology,

in cooperation with the departments of -Medical Oncology, -Genetics, -Medical Biology, -Pathology

and Wenckebach Institute, University Medical Center Groningen, the Netherlands and the

Comprehensive Cancer Center North-Netherlands, Groningen, the Netherlands, the department

of Clinical Genetics, University of Antwerp, Belgium and the department of Surgery, Medisch

Spectrum Twente, Enschede, the Netherlands.

This research was financially supported by the Dutch Cancer Society (Koningin Wilhelmina Fonds)

grant RUG: 94-873 and RUG 99-2130 and “de Jan Kornelis de Cock” foundation.

Financial support for the publication of this thesis kindly provided by:

Medisch Spectrum Twente, Stichting Werkgroep Interne Oncologie, Integraal Kankercentrum

Noord-Nederland, KWF kankerbestrijding, Novartis, GlaxoSmithKline, Sanofi-Aventis, Roche,

Astra Zeneca, KCI Medical, Bioprof, Janssen-Cilag, Amgen, Tyco Healthcare, Stryker, Laprolan,

Johnson & Johnson Medical, Ethicon Endo-Surgery, Nycomed, Promega, Graduate School GUIDE

and MSD.

Aan mijn ouders,

Voor Marjolijn en Louise.

Contents

Chapter 1 General introduction, aim and outline of the thesis ............................................................................11

Chapter 2 Genetic predisposition to testicular germ-cell tumours..................................................................23

Lancet Oncololgy. 2004; 5:363-71

Chapter 3 Syndromic aspects of testicular carcinoma ...................................................................................................43

Cancer. 2003; 97:984-92

Chapter 4 Do the eastern and northern parts of The Netherlands

differ in testicular cancer? ...................................................................................................................................................57

Urology. 2001; 58:636-7(letter)

Chapter 5 Testicular carcinoma and HLA Class II genes...............................................................................................61

Cancer. 2002; 95:1857-63

Chapter 6 Absence of constitutional Y chromosome AZF deletions in

patients with testicular germ cell tumors ........................................................................................................75

Urology. 2005; 65:196-201

Chapter 7 Re-analysis of the Xq27-Xq28 region suggests a weak association

of an X-linked gene with sporadic testicular germ cell tumour without

cryptorchidism...................................................................................................................................................................................85

European Journal of Cancer. 2006; 42:1869-74

Chapter 8 Interest in and motivations regarding genetic testing for testicular

germ cell tumour susceptibility....................................................................................................................................97

Submitted

Chapter 9 Summary, discussion and future perspectives........................................................................................ 113

Chapter 10 Nederlandse samenvatting, conclusies en toekomstperspectieven.............................. 123

Reference list ................................................................................................................................................................................... 135

Addendum........................................................................................................................................................................................... 157

Dankwoord ......................................................................................................................................................................................... 163

List of publications................................................................................................................................................................... 169

Curriculum Vitae........................................................................................................................................................................... 175


1

11

Chapter 1

General introduction, aim and outline of the thesis

General introduction12


1

13

General introduction, aim and outline of the thesis

General introductionTesticular tumours can be divided into germ cell tumours, stromal tumours and other tumours

(e.g. malignant lymphomas). Tumours of paratesticular structures form a separate group. The

research in this thesis focused solely on the testicular germ cell tumours (TGCT) seminoma and

non-seminoma. TGCT are rare, but they are the most frequently occurring tumour in men aged

between 15 and 40 years. In the Netherlands, 536 men were diagnosed with TGCT in 2003, while

in 2004, 30 men died of this malignancy. Although the incidence of TGCT has increased sharply

in recent years, survival of patients with TGCT has improved enormously. Five year survival in

the nineteen seventies was about 65% compared to more than 90% at present (Figures 1 and

2).(1) Improved survival can chiefly be attributed to the cisplatin-based polychemotherapy that

was introduced in the nineteen eighties to treat patients with metastasized TGCT. In addition,

new strategies have been developed in the surgical approach to metastasized/non-metastasized

TGCT and alterations have been made to the radiotherapy technique and radiation dose for

seminoma.(2;3) The progress in diagnosis, treatment and the subsequent treatment outcomes

in patients with TGCT are the ultimate result of multidisciplinary team work. At the University

Medical Center Groningen (UMCG), this multidisciplinary approach was started at the end of the

nineteen seventies to provide every patient with tailored treatment. These accomplishments in

the treatment of TGCT have led to the present goal of further optimising the treatment for TGCT,

��

��

��

��

��

��

�

�

�

�

�

��

��

��

��

��

��

Figure 1: Incidence of TGCT



1

15

in which the research and treatment chiefly concentrate on reducing the toxic side-effects of

chemotherapy and radiotherapy. In patients with prognostically favourable factors (Table 2)(4),

the number of courses of chemotherapy can be reduced, whereas in patients with prognostically

unfavourable factors, more intensive chemotherapy is necessary to improve the chances

of survival. Nowadays the majority of TGCT patients can be cured by the multidisciplinary

treatment. Therefore, the number of TGCT survivors will continue to increase. In principle, these

TGCT survivors will be exposed to the long-term consequences of chemotherapy-related toxicity

(side-effects) cq the long-term side-effects of radiotherapy. TGCT survivors are mostly young men

who can be expected to have a long life ahead of them. This has meant that over the past few

years, scientific research has centred on studying the long-term effects of treatment(5-8) and the

quality of life of these TGCT survivors.(9) All this research has the ultimate aim of achieving further

improvement in the treatment and follow-up of TGCT patients.

The genetic background of TGCTCarcinoma in situ (CIS) (or intra-tubular germ cell neoplasia) is the precursor of TGCT and is

found in nearly all TGCT together with an invasive component. CIS cells originate from primordial

germ cells that “escaped” normal differentiation in utero. It is assumed that over the course of

time, CIS “develops” into an invasive TGCT, but the precise transformation of premalignant CIS

into a TGCT is not yet clear. It is suggested that the default pathway follows the development of

CIS into seminoma and that non-seminoma requires activation of pluripotency (reprogramming)

of a CIS or seminoma cell.(10;11) A theoretical model of TGCT development as part of testicular

dysgenesis, taking into account a range of reported TGCT risk factors, has been developed by

Skaekebaek et al(12) and is discussed in subsequent chapters. In recent years several studies

have looked into chromosomal abnormalities and more recently at gene mutations and gene

activity in TGCT to unravel the molecular pathways underlying these tumours. A detailed

overview of (non-inherited) genomic aberrations in TGCT was recently published by Von

Eyben.(13) Aneuploidy has been found in nearly all cases and triploidy is a common finding.

Seminoma have a mean hypertriploid DNA index and non-seminoma have a mean hypotriploid

DNA index (due to loss of chromosomal material during cancer progression).(11) When looking at

individual chromosomal regions, an isochromosome of the short arm of chromosome 12, i(12p),

(resulting in a duplication of the short arm of chromosome 12) is found in about 80% of TGCT.

The remainder have excess 12p genetic material in derivative chromosomes.(14) The exact relation

between these changes and TGCT is unclear but the absence of amplification of a section of

12p in intratubular germ cell neoplasia, suggests that this amplification may be related to

progression of the disease rather than initiation.(15) A recent gene expression profile study on

TGCT material identified differentiated expressed genes on 12p. Seventy-three genes on 12p were

significantly overexpressed, indicating that the p arm of chromosome 12 may play an important

role in TGCT tumorigenesis.(16) In addition to the genes located on 12p, a growing list of genes is

implicated in the various stages of TGCT development. In particular, TGCT has been shown to be

associated with a characteristic series of abnormalities in the retinoblastoma pathway including

upregulation of cyclin D2 and p27 and downregulation of RB1 and the cyclin-dependent kinase

inhibitors of p16, p18, p19 and p21.(13) A gain of activity of the KIT gene, a member of the tyrosine

kinase family, appears to play a role in the progression of CIS towards seminomas.(17) Recently,

the scope of genetic study of TGCT has been extended to include the role of naturally occurring

micro RNAs (miRNAs). Indeed, some of these miRNAs (miRNA-372 and 373) were shown to allow

tumorigenic growth. As they have also been observed to be expressed in human seminomas

and non-seminomas, but not in normal testicular tissue, it has been suggested that these miRNA

may represent a new class of oncogenes involved in TGCT.(18) Studies of hereditary genetic

changes and TGCT are widely discussed in the subsequent chapters.

HistologyTGCT can be divided into two important histological subtypes: seminoma and non-seminoma.

Pure seminoma occurs in about 50% of the cases and often (in 20%) contains trophoblastic

giant cells that can produce beta-human chorionic gonadotrophin (ß-HCG). Non-seminoma chiefly

comprises two or more cell types, for example embryonal carcinoma, choriocarcinoma (also

contains trophoblastic giant cells), yolk sac tumour, teratoma, whether or not in combination

with seminoma. Embryonal carcinoma and yolk sac tumour elements are associated with the

production of alpha-fetoprotein (AFP). Clinically, there are three important tumour markers in

the diagnosis and follow-up of TGCT: ß-HCG, alpha-fetoprotein (AFP) and lactate dehydrogenase

(LDH).(2)

��

��

��

��

�

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

�

��

��

��

��

��

��

��

��

��

��

Figure 2: Survival of TGCT

Figure 1 and 2 from: Signaleringsrapport kanker:

www.kwf.nl(1)



1

17

Clinical presentation and initial therapyPatients who present with a painless swelling of the testicle and increased levels of AFP, ß-HCG

and/or LDH have a TGCT until proven otherwise. Clinical presentation varies widely. TGCT can also

show extra-testicular clinical manifestation (without any testicular complaints). Retroperitoneal

lymph node metastases can cause (lower) back pain, while haemoptysis can be the result of

pulmonary metastases. Ultrasound examination of the testicle is useful to establish possible

testicular abnormalities. Radical inguinal orchidectomy with high ligation of the spermatic cord,

blood vessels and lymph vessels is the surgical treatment for patients who are suspected of

having a TGCT. Originally, the testicles descend embryologically via the retroperitoneal route

and inguinal canal into the scrotum. This often results in regional metastases from a TGCT that

first arise in the retroperitoneal lymph nodes. A right-sided TGCT tends to metastasize to the

inter-aorto-caval lymph nodes, whereas a left-sided TGCT tends to metastasize to the para-aortal

lymph nodes. At a higher, supradiaphragmatic level, metastases can spread via the thoracic

duct and result in mediastinal or supraclavicular metastases. Haematogenically, TGCT chiefly

metastasize to the lungs, later to the liver, the skeleton and the cerebrum.(3;19)

StagingWhen a patient has been diagnosed with a TGCT, the malignancy must be staged. This can be

done by means of tumour markers (AFP and ß-HCG, LDH) and spiral CT (computed tomography)

scanning of the lungs, the retroperitoneum and pelvis. On indication (anamnestic complaints

of the cerebrum and/or sharply elevated ß-HCG), CT scanning of the cerebrum is conducted.

The roles and/or additional value of magnetic resonance imaging (MRI) and positron emission

tomography (PET) are currently under investigation.(20) At the UMCG, patients with TGCT are staged

according to the Royal Marsden classification (Table 1). Patients with stage I have no radiological

or biochemical evidence of metastases. Patients with stages II to IV have metastasized disease.

These patients are subsequently classified according to the prognostic factors formulated by the

International Germ Cell Cancer Collaborative Group (IGCCCG) (Table 2) and a treatment plan is

drawn up on the basis of the subgroup classification (good / intermediate / poor).(4)

Table 1: Royal Marsden staging classification of testicular germ cell tumours

Stage Criteria

Stage I No evidence of metastases

Stage IM No clinical evidence of metastases, but persistent

elevation of serum tumour markers AFP and/or hCG

Stage II Infradiaphragmatic lymph node metastases

IIA Metastases < 2 cm in diameter

IIB Metastases 2-5 cm in diameter

IIC Metastases > 5 cm in diameter

Stage III Supradiaphragmatic lymph node metastases;

Status A, B, C as for stage II

Stage IV Extra lymphatic metastases

L1 <- 3 Lung metastases

L2 > 3 Lung metastases, all <- 2 cm in diameter

L3 > 3 Lung metastases, one or more > 2 cm in diameter

H+, Br+, Bo+ Liver, brain, or bone metastases

Table 2: IGCCCG prognostic classification for germ cell cancer (4)

Non-seminoma Seminoma

Good prognosis Testis / retroperitoneal primary Any primary site

and No non-pulmonary and No non-pulmonary

visceral metastases visceral metastases

and AFP<1000 ng/ml and and Normal AFP, any hCG,

hCG<1000 ng/ml and any LDH

LDH<1.5x N*

Intermediate prognosis Testis / retroperitoneal primary Any primary site

and No non-pulmonary and Non-pulmonary

visceral metastases visceral metastases

and and Normal AFP, any hCG,

1000 <- AFP <- 10.000 ng/ml or any LDH

1000 <- hCG <- 10.000 ng/ml or

1.5x N <- LDH <- 10x N

Poor prognosis Mediastinal primary No patients classified as

or Non-pulmonary visceral poor prognosis

metastases

or AFP>10.000 ng/ml or

hCG>10.000 ng/ml or

LDH>10xN

* N=normal range

TreatmentStage I disease (=non-metastasized disease)

About half of the patients with a non-seminoma present with stage I disease. Presently, the

treatment comprises radical orchidectomy with high ligation of the spermatic cord, blood vessels

and lymph vessels, followed by regular outpatient visits (wait and see policy) or modified

unilateral nerve-sparing retroperitoneal lymph node dissection. The UMCG has been applying

the wait and see policy to patients with stage I disease since 1982.(21) In the meantime, world-



1

19

Surgery after chemotherapy

After completion of the chemotherapy, patients with a metastasized tumour (TGCT) at the UMCG

undergo restaging with the aid of serum tumour marker analyses (as described above) and spiral

CT scanning of the lungs and retroperitoneum. When radiological investigation shows residual

disease after completion of the chemotherapy in patients with seminoma, surgical resection is not

performed, but instead the abnormality is followed radiologically. Generally, a new indication will

arise for chemotherapy or radiotherapy, which if necessary is combined with surgical resection of

the residual tumour. Presently, positron emission tomography (PET) scanning can be applied to

help identify viable cancer.(25) After patients with non-seminomatous TGCT have completed the

chemotherapy, there is no indication for surgical resection when they do not show any residual

disease, or the residual abnormality is smaller than 1 cm, the tumour markers have normalised

and mature teratoma is absent from the primary tumour. Follow-up is then conducted on the

basis of tumour markers. However, when residual (abdominal / pulmonary) disease is identified,

surgical resection must be performed on the residual retroperitoneal tumour mass (RRRTM), or

local resection in the case of residual pulmonary tumour tissue. When residual tumour persists

after completion of the chemotherapy and the tumour markers remain high, an individual

treatment plan must be drawn up: salvage chemotherapy or salvage surgery. In patients with

non-seminomatous TGCT, it is not possible to predict the histology of the residual tumour. After

surgical resection, histological examination shows that the mass consists of necrosis in 45% of

the cases, mature teratoma in 40%, viable tumour tissue in 10% and non-germ cell malignancies

in the remaining 5%.(26)

When the resected residual tumour only contains necrosis or mature teratoma, no further

treatment is necessary and the patient has an excellent prognosis. However, when viable

tumour tissue is present, the prognosis is less favourable and depending on factors such as

the initial prognostic classification (Table 2), the volume (percentage) of residual viable cancer

and the completeness of the resection, it may be necessary to administer adjuvant salvage

chemotherapy.(27)

The clinical significance of mature teratoma in the residual tumour is not yet completely clear

and it is impossible to predict the course that can be expected from mature teratoma left in

situ. In the literature, “growing teratoma” is a well-known phenomenon and it can lead to (very)

late tumour recurrence. In addition, there is a risk that mature teratoma will de-differentiate into

a non-germ cell malignancy (e.g. sarcoma). In such cases, the prognosis of the patient is far

less favourable. Radical surgery is the only curative treatment option for these non-germ cell

tumours.

When the primary TGCT contains elements of mature teratoma, there is a greater chance that the

residual tumour will also contain teratomatous elements. Therefore, at the UMCG, all patients

with elements of teratoma in the primary tumour undergo laparotomy and partial retroperitoneal

wide consensus has been reached about the treatment for stage I non-seminoma patients in

the low risk group (i.e. histological examination does not show any vascular invasion of the

tumour). These patients have 15% risk of TGCT relapse, thus the wait and see policy is justified

and comprises regular outpatient visits for 5 years after orchidectomy with physical examination,

tumour marker analysis and frequent radiological investigation. There is a great deal of

discussion about the current treatment policy for patients with stage I non-seminoma in the

high risk group (i.e. histo-pathological investigation shows vascular invasion). Unilateral nerve-

sparing retroperitoneal lymph node dissection (RPLND) or adjuvant chemotherapy leads to a

considerable reduction in the risk of relapse of about 5%. However, the disadvantages of RPLND

(loss of ejaculatory function) or chemotherapy (toxicity) must be taken into consideration. The

UMCG applies still the wait and see policy to these high risk patients. In the UK and many other

European countries, the preferred approach for high risk patients is often to administer two

courses of adjuvant chemotherapy, whereas in the USA it is the trend to perform nerve-sparing

unilateral RPLND.(2;3) Which treatment policy is the best for these so called ‘high-risk’ patients

is unknow.

About 75% of the patients with seminoma have stage I disease. The standard treatment after

orchidectomy is radiotherapy of the retroperitoneal para-aortal lymph nodes with a total dose of

20 Gy. In this way the relapse rate is reduced to 1-3%.(2) The advantage of para-aortal radiotherapy

(while excluding the ipsilateral lymph nodes) is less gastrointestinal and gonadal toxicity. As an

alternative for radiotherapy, adjuvant chemotherapy can be administered. The latter approach

results in the same relapse rates.(22) A wait and see policy alone would lead to relapse rates of

15% to 20% and is less suitable, because there are no sensitive tumour markers for seminoma

and the patient would therefore have to undergo frequent radiological investigation.(2;3)

Metastasized disease

Patients with stage IIa and IIb seminoma receive radiotherapy with a total dose of 30 Gy and

36 Gy, respectively. In contrast with stage I seminoma, the ipsilateral iliac lymph nodes are

also included in the treatment volume (dog-leg field) in patients with stage II seminoma. This

approach achieves 6-year relapse-free survival of 95% in stage IIa patients and 89% in stage

IIb patients. Alternative treatments for radiotherapy can be considered in stage IIb patients, for

example 3 courses of BEP (bleomycin, etoposide and cisplatin) or 4 courses of EP (etoposide

and cisplatin).(2;3)

Treatment for patients with stage IIc - IV metastasized seminoma and with stage II – IV

metastasized non-seminoma comprises chemotherapy in accordance with the prognostic factor

classification. In the good prognosis group, patients with non-seminoma / seminoma receive

chemotherapy in the form of 3 courses of BEP or 4 courses of EP. Patients in the intermediate

and poor prognosis groups are treated with 4 courses of BEP.(2;23;24)



1

21

dissection at the “original” tumour site after chemotherapy, even when there are no radiological

signs of residual disease.(26)

Aim and outline of this thesisOver the past few years, there has been growing interest in the genetic aspects of testicular

cancer for a variety of reasons. In general terms, there is a scientific need to unravel the

oncogenetic steps of all types of cancer in the hope that understanding the genetic changes will

lead to better (early) diagnosis and treatment options. Some of these genetic changes have a

hereditary nature, i.e. they can be passed on from one generation to the next and they must

therefore be distinguished from the non-hereditary, somatic mutations that occur in the course

of life. Studies on the hereditary predisposition of testicular cancer may lead to the identification

of men with a strongly increased risk of developing testicular cancer who might benefit from

preventive measures. Cure rates for testicular cancer are now high, so there is increasing interest

in fertility aspects and the possible consequences of passing on the predisposition of testicular

cancer to the offspring. Nowadays all patients with testicular cancer who undergo chemotherapy

are in principle offered the opportunity to freeze their semen (cryopreservation). If a patient

is less fertile or infertile after chemotherapy, it is possible to achieve pregnancy by means of

artificial insemination. In this way, there is a risk of transmitting the hereditary predisposition.

Another reason to study testicular cancer predisposition is that as a rule, finding hereditary

cancer-predisposing mutations in genes leads to increased insight into the development of non-

hereditary (so-called sporadic) tumours. In many cases, it has been found that genes involved

in hereditary tumours also play a role in non-hereditary types of cancer.

Now that the genomes of humans and other species are being systematically mapped and the

scope of molecular and statistical analyses in genetic studies is expanding, there has been rapid

progress in tumour genetic research. In contrast with studies on the hereditary aspects of breast

cancer, colorectal cancer and a series of relatively rare tumour syndromes (such as multiple

endocrine neoplasia (MEN1 and MEN2), Von Hippel Lindau, neuro-fibromatosis (NF1 and NF2)

and Li-Fraumeni’s syndrome), studies on the hereditary aspects of testicular cancer have been

unable to identify any germline mutations in genes that could be associated with a high risk of

developing testicular cancer.

The research described in this thesis aimed to contribute to knowledge on hereditary

predisposition to TGCT and focuses on the following aspects:

• The literature on the hereditary and syndromal aspects of TGCT

• Epidemiological aspects of TGCT

• The HLA and Xq27 regions on the genome that might harbour genes contributing to the

development of TGCT

• Occurrence of constitutional deletions of the Yq11 region in men with TGCT

• Interest of men in genetic testing for TGCT


2

23

Chapter 2

Genetic Predisposition to Testicular Germ-Cell Tumours

MF Lutke Holzik1, EA Rapley2, HJ Hoekstra1, DTh Sleijfer3, IM Nolte 4, RH Sijmons5

Departments of: 1Surgical oncology,3Medical oncology, 4Medical biology, 5Clinical genetics. University Medical Center Groningen, the Netherlands.2Institute of Cancer Research, Section of Cancer Genetics, London, United Kingdom

Lancet Oncology 2004; 5:363-371

Genetic Predisposition to Testicular Germ-Cell Tumours24 Genetic Predisposition to Testicular Cancer

2

25

Genetic Predisposition to Testicular Germ-Cell Tumours

IntroductionThe term testicular cancer encompasses a group of neoplasms that occur from childhood through

to old age. This review focuses on germ-cell tumours of adolescents and adults, specifically

seminomas and non-seminomas, which are thought to arise from carcinoma-in-situ.(28-30) We do

not discuss paediatric germ-cell tumours or spermatocytic seminomas which do not arise from

carcinoma-in-situ and probably have a different cause from germ cell tumours in adolescents

and adults.(28) We focus mainly on testicular germ-cell tumours (TGCT). However extragonadal

germ-cell tumours of adolescents and adults have a very similar cause and arise from carcinoma-

in-situ; they can therefore be thought of as a part of the same disorder.

TGCT is the most common type of malignant disorder in men aged 15 - 40 years. The

yearly incidence in this age group is about 7.5 per 100,000 people, but the incidence varies

substantially between countries. (19;28;31) TGCT can be classified histologically as seminoma (about

55%), which is most commonly diagnosed during the fourth decade of life, and non-seminoma

(about 45%), which is generally diagnosed in the third decade of life. Both subtypes probably

arise from preinvasive carcinoma-in-situ.(32;33) The strongest risk factors for TGCT are a family

history of the disorder(34;35), a previously diagnosed TGCT(36;37), and cryptorchism (undescended

testis)(38-41,81). In addition, patients with Klinefelter’s syndrome(42) or XY gonadal dysgenesis(43)

have a high risk of developing germ-cell tumours.(44) Other less strong risk factors documented

and confirmed in several studies include: testicular atrophy(45), inguinal hernia(34), infertility(46;47),

hydrocele(48), previously diagnosed extragonadal germ-cell tumour(44) and other disorders of male

sexual differentiation. Many of the risk factors for TGCT include disorders of male urogenital

differentiation. This feature has led to the term testicular dysgenesis syndrome (Figure 1), (12) and

both environmental and genetic factors are probably involved. In this review, we summarise the

current knowledge of genetic predisposition to TGCT.

Clues to the existence of genetic predisposition Family history

1 –3% of patients with TGCT report an affected first-degree relative, a proportion higher than

would be expected by chance (see Table 1).(35;48-57) The largest number of reported cases in

a family is five,(58) but most familial clustering concists of relative pairs such as two affected

brothers and to a smaller extent an affected father and son.(35;59) Figure 2 gives examples of

familial clustering in TGCT. Brothers of patients with TGCT have a relative risk of TGCT of 8-10,

and for father-son the relative risk is 4 – 6.(35;51;56) These relative risks for TGCT are higher than

for most other cancer types, which rarely exceed 4. The high relative risk for TGCT is difficult to

attribute only to a shared environmental component. To account for a relative risk greater than

3 without some form of genetic predisposition requires that the environmental risk factor or

factors are extremely potent. Khoury and colleagues(60) calculated that in this situation the risks


2

27

to exposed people are 50-100 times those of unexposed individuals. Such potent risk factors

have not yet been identified for TGCT.(61)

Racial differences and geographic clustering

Geographic clustering of TGCT(62) and racial differences in the incidence of this disease could

indicate a genetic component in the cause of the disease. The highest incidence is seen in white

people of northern European descent, whereas people of African or Asian descent seem to

have a universally low incidence.(28;59;61;63) The incidence of this disorder in African Americans is a

quarter of that among white Americans(61) , and is similar to that of native African populations;

thus the risk has not changed by much with migration to a new environment. Differences in

incidence persisting after migration argue in favour of genetic rather than exogenous risk factors.

This maintenance of risk contrasts sharply with the situation for cancers of breast, stomach,

colon and ovaries, for which the incidence in immigrant populations tends rapidly towards that

of the host population.(59;61;64;65)

Table 1: First-degree familial TGCT and estimates of relative risk in male first-degree

relatives of patients with TGCT 1985-2002

Author Year Study No. FTC / % FTC Sibs / RR

total Fath

group* -son

Tollerud (48) 1985 Multicentre hospital-based, 6/269 2.2 2/4 5.9

Retrospective

Kruse (52) 1987 Single-centre hospital 3/255 1.2 3/0 NC

based, Retrospective

Patel (54) 1990 Multicentre hospital-based, 5/500 1.0 1/4 NC

Retrospective

Forman (35) 1992 Multicentre hospital-based 12/794 1.5 8/4 9.8B

and national population


Westergaard (57) 1996 Data from Danish cancer 22/2261 1.0 10/12 12.3

registry, Retrospective /1.96A

Polednak (55) 1996 Population-based 12/1395 0,86 8/4 NC

Connecticut Tumor Registry

Heimdal (51) 1996 Multicentre hospital-based, 32/1159 2.8 Nc 7.6

Retrospective

Dieckmann (49) 1997 Multicentre hospital-based, 18/1692 1.1/1.7 9/9 3.1

Retrospective and 9/518 7/2 (retrospective

Prospective1 group)

Ondrus (53) 1997 Single-centre hospital 2/650 0.3 2/0 NC


Sonneveld (56) 1999 Single-centre hospital

based, Retrospective 17/686 2.5 11/6 8.5-12.7/1.7A

Dong et al.(50) 2001 Data from Swedish 62/4650 1.3% 38/24 8.3/3.9A

family-cancer database,

Retrospective

FTC = familial testicular cancer

NC = not calculated* = Number of familial TGCT cases compared with total group.

RR = Relative Risk for all male first-degree relativesA = RR for brothers / RR for fathersB = RR only for brothers1 = In a prospective multi centre study 18 of the 1692 patients had a first-degree relative with TGCT.

Also a selection of patients from the Berlin (Germany) hospital were investigated: 518 patients,

9 had a first degree relative with TGCT.

Figure 1: The testicular dysgenesis syndrome. The asterisk indicates the possibility that cryptorchidism

(testicular maldescent) acts as a causal risk factor. CIS: carcinoma in situ. Modified frame from:

Skakkebaek NE, Rajpert-De Meyts E, Main KM. Testicular dysgenesis syndrome: an increasingly

common developmental disorder with environmental aspects. Hum Reprod. 2001;16:972–978.

© European Society of Human Reproduction and Embryology.

Figure 1: The Testicular Dysgenesis Syndrome

Enviromental factors

Testicular dysgenesis

Hereditary disorders

and constitutional

chromosomal

anomalies and

somatic genetic

defects

Reduced semen

quality

CIS −> Testicular

Cancer

Testicular

maldescent

Hypospadias

Disturbed sertoli

cell function

Impaired germ cell

differentiation

Decreased leydig

cell function

Androgen

insufficiency

?*


2

29

Bilateral TGCT and risk of other tumours

A consistent risk factor for the development of TGCT is having a previous testicular tumour. For

patients who have had a first primary TGCT, the relative risk of developing a second is 25.(37;66)

The prevalence of bilateral TGCT ranges between 1.0% and 5.8% and studies have even shown

an increase.(67) Bilateral involvement of paired organs (breast, retina, and kidney) has proved to

be a clinical marker of hereditary cancer. Bilateral cases are commonly associated with a positive

family history for the same tumours.(56) Heimdal and co-workers(51) found that 2.8% of patients

with TGCT who did not have a family history had bilateral disease compared with 9.8% of those

with a positive family history. Bilateral disease in paired organs could also be the result of a very

early somatic mutation (in embryogenesis), and evidence suggests that a proportion of bilateral

TGCT could indeed be accounted for such a mechanism. Although mutations in the KIT (tyrosine

kinase receptor) gene are rare in TGCT(68) and mediastinal germ-cell tumours(69) , mutations of

this gene occur in a verry high proportion (95%) of tumours from patients with bilateral disease

compared with a smaller proportion (3%) of tumours from those with unilateral disease.(70) When

both tumours from bilateral cases could be assesed, the same mutation was present in both.

Together these results suggest that somatic KIT mutations occur early in embryogenesis, before

the primordial germ cells have divided and migrated to the gonads. Primordial germ cells with

KIT mutations are therefore distributed to both testes; hence KIT mutations are associated with

bilateral disease.(70) In patients with a family history of TGCT, the frequency of KIT mutations

in unilateral TGCT was similar to that detected previously by Looijenga and colleagues.(70)

However, in patients with bilateral disease and a family history of TGCT, only 28% of tumours

had mutations of KIT.(71) Although the reason for this difference is unclear, the evidence suggests

that bilateral disease in the context of familial TGCT has a different pathogenesis from sporadic

bilateral cases and that most familial bilateral cases are explained by the high risk conferred by

the underlying inherited genes.(71)

A genetic predisposition to TGCT does not necessarily have to be site specific; in analogy with

the situation with many proven hereditary tumour predispositions, patients with TGCT and

their relatives could have an increased risk of developing non-TGCT tumours. Dong and co-

workers(50) investigated this trend in 4650 patients with TGCT and with a mean follow-up of 11

years. They calculated a significantly increased standardised incidence ratio for various second

primary tumours after testicular seminomas: TGCT 11.6 (95% CI 7.0-18.1), colorectal cancer 1.9

(1.1-3.1), pancreatic cancer 3.8 (2.1-6.4), renal cancer 2.2 (1.1-4.0), bladder cancer 2.4 (1.5-3.7),

thyroid cancer 5.4 (1.4-14.1) and malignant lymphomas 2.5 (1.3-4.1). Some of the second primary

tumours could be associated with treatment, since seminomas (mostly treated by radiotherapy)

were associated with a higher frequency of second malignant disorders than non-seminomas

(mostly treated with chemotherapy). Furthermore, an analysis of the prevalence of tumours

in relatives also showed an increased risk of TGCT in male relatives of patients with TGCT

(standarised incidence ratio 8.3 for brothers and 3.8 for fathers and sons).(50) For all relatives

together there was no excess risk of other cancer types in first-degree relatives of patients with

Figure 2:

Examples of familial clustering in TGCT.

A. Brother pair, the most common type of pedigree.

B. Father-son pair, the next most common relative pair.

C. Cousin pair.

D. Uncle-nephew pair.

E and F. Rare TGCT pedigrees with many affected cases.


2

31

TGCT. However, mothers of patients with TGCT had significantly increased standarised incidence

ratios for lung cancer (1.9; 95%CI 1.4-2.6), non-endometrial uterine cancer (2.6; 1.2-4.7), soft-

tissue tumours (2.5; 1.1-4.9) and malignant melanoma (1.8; 1.1-2.6). Other studies(72) have shown

similar trends (although some not significant) of other cancers in mothers. However, all studies (57;73) show a high frequency of TGCT in relatives of patients with TGCT. These findings generally

suggest that relatives of patients with TGCT have a greatly increased risk of developing TGCT

but may not be at risk of developing other types of cancers.

Syndromic characteristics

Two constitutional chromosomal abnormalities are clearly associated with an increased risk

of TGCT. Patients with Klinefelter’s syndrome (47 XXY) have a very high risk of mediastinal,

(extragonadal) germ-cell tumours.(74) About 8% of patients with such tumours have Klinefelter’s

syndrome.(44) Patients with this syndrome rarely develop TGCT, probably because they do not

have germ cells in the testis from shortly after birth; however, about a third of patients with

extragonadal germ-cell tumours present with testicular carcinoma-in-situ and these patients have

a substantial risk of developing TGCT (see later).(44;75;76) Patients with XY gonadal dysgenesis

have a greatly increase risk of germ-cell tumours.(74;77) This increased risk is seen only in patients

with Y chromosome material; those with gonadal dysgenesis and XX or 45X karyotype do not

have an increased risk of TGCT.(51;76;77) Patients with Down’s syndrome (trisomy 21) might also be

at increased risk of TGCT but the numbers are too small for firm conclusions to be drawn. (78;79)

The presence of TGCT in a hereditary syndrome might be an indication of a hereditary

predisposition to TGCT. We have described the prevalence of TGCT in patients with various

hereditary disorders.(76) Owing to the rarity of most of these disorders and the scarce reports

of their occurrence in combination with TGCT, at present there is no statistical proof of an

association of TGCT with these disorders. An important clinnical issue is that in a proportion

of these disorders there is also a range of defects in urogenital differentiation, which suggests

that TGCT in these disorders is indeed a further complication of such a differentiation defect, as

postulated in the model of Skakkebaek and colleagues (figure 1).(12)

Cryptorchism and other disorders of testicular differentiation in patients and relatives

Cryptorchism and other testicular abnormalities such as atrophy, infertility, hydrocele and

inguinal hernia are risk factors for TGCT.(34;45;46;80;81) As regards family history, Forman and co-

workers(35) showed that the frequency of cryptorchism in patients with TGCT did not differ

between those with or without a positive family history. However, very few studies have looked

at the frequencies of these risk factors in male relatives of patients with TGCT. A small study

by Tollerud and colleagues(48) showed that cryptorchism occurred in a significantly greater

proportions of first-degree male relatives of patients with a family history of TGCT (17%) than

of relatives of patients with TGCT who did not have such a family history (5.3%) or of controls

(2.7%). Their study also showed that 50% of patients who had TGCT and a family history

reported a first-degree relative with inguinal hernia, compared with 10.3% of those without a

family history and with 12.7% of controls. The high prevalence of cryptorchism, inguinal hernias,

and hydrocele among men in these families suggests that an underlying alteration in urogenital

embryogenesis could be associated with the familial predisposition to testicular neoplasia.

Studies on Twins

Studies on twins with cancer have been used to address two general issues. First, are there any

carcinogenic effects of twinning? This question can be answered by comparison of the occurrence

of a cancer in twins and in singletons. The second issue, what the heritable components are to

that cancer, can be addressed by a proband-wise analysis ( proband = the affected individual

through whom a family with a genetic disorder is ascertained) of monozygotic twins compared

with dizygotic twins or siblings. If there is a heritable component the relative risk should be

higher for monozygotic than for dizygotic twins.

In a cohort of 14326 elderly twins aged 66 – 77 years, Braun and co-workers(82) noted that a

personal history of TGCT was reported by five (0.08%) of 5951 monozygotic twins and 11 (0.16%)

of 6992 dizygotic twins. Swerdlow and colleagues(83) als found a significantly higher risk of

TGCT in dizygotic than in monozygotic twins (odds ratio 1.5; 95% CI 1.1-2.2).(83) These finding

suggest that an environmental component was acting in utero to cause TGCT. To find a possible

genetic effect, twin studies must asses the risk of TGCT in the twin brothers of affected patients;

however, the numbers are too small in many studies for any conclusions to be drawn. Swerdlow

and colleagues(83) identified six pairs of concordant (both affected) twins. The risk of TGCT was

raised in twin brothers of patients with TGCT (relative risk 37.5; 95% CI 12.3 – 115.6) and was

greater in monozygotic (76.5) than dizygotic (35.7) twins, which would be expected if there

is a heritable part to TGCT. This relative risk is several times that found in non-twin brothers

(although the confidence limits are wide because of small numbers) but it does imply that the

genetic element for risk is far larger for TGCT than for most other cancers. Other twin studies

on TGCT have not had sufficient numbers of concordant twin pairs and have been unable to

determine zygosity so could not confirm this result.

Environmental components

The worldwide incidence of TGCT has more than doubled over the past 40 years. The increase

follows a birth-cohort effect, and a probable explanation is that factors in embryogenesis or

early life are contributing to the development of TGCT.(28;84;85) The rapid increase highlights

the importance of environmental factors in this disease, because the genetic composition of a

population simply cannot change in the course of one or two generations. Why this increase is

occuring is unclear, but one theory is that an increase in endogenous oestrogens is contributing

to increase in TGCT in addition to risk factors for the disorder such as the increase in incidence

of cryptorchism and the decrease in fertility and sperm quality.(86;87) Another hypothesis is that

abnormally high oestrogen concentrations in pregnancy predispose the developing gonad to


2

33

TGCT in adulthood. Direct investigation of this idea is difficult, and many studies have looked at

surrogate features that could reflect high oestrogen concentrations in pregnancy.(87) According to

this hypothesis, a high rate of TGCT in dizygotic twins (because uterine oestrogen concentration

during pregnancy are higher for dizygotic than for monozygotic twins), as well as older ages of

mothers and a higher rate of oestrogen related cancers (eg, breast cancer) in mothers and sisters

would be expected.(88) Overall published reports(87;88) provide conflicting results on the analysis

of these variables. These features are merely weak indicators of high oestrogens concentrations,

so the sample size and statistical power of many of these case-control studies might have

been too low to show any significant association. Although hormonal factors could be causally

involved in the development of TGCT, mothers of patients with TGCT do not have an increased

risk of oestrogen related cancer and the risks of breast cancers does not seem to be increased

in the sisters of these patients.(50;72) Even though many studies have showed similar results, this

hypothesis remains unconfirmed.

Several studies have confirmed a higher frequency of TGCT in dizygotic twins than in singletons

and monozygotic twins. This difference suggests that the concentration of circulating oestrogen,

which is significantly higher in pregnancies with dizygotic than in those with monozygotic twins

or singletons, is a contributory causative factor.(82;83;88) However, more research is needed into

the pathogenetic mechanisms that could cause the increase risk of TGCT in dizygotic twins.

Twinning itself is to a sustantial extent genetically determined, and coinherited genetic factors

could conceivably be contributing to this increase rather than circulating oestrogens.

Modelling studies

Two studies so far have tried to identify the mode of inheritance of TGCT susceptibility genes.

One was based on the frequency of bilateral TGCT,(89) and the other was a segregation analysis

on a group of Norwegian and Swedish families.(90) In a segregation analysis, a disease model

is sought for familial aggregation of a disease by fitting the patients and non-affected family

members in pedigrees to, for example, a dominant or recessive disease model. Heimdal and

colleagues(90) did such an analysis on all available patients with TGCT treated at a Norwegian

hospital and a Swedish hospital between 1981 and 1991 (n=978). For 30 patients, a first-degree

relative also had TGCT; there were no families with more than two affected members. The

investigators concluded that the familial clustering was best accounted for by a major gene for

TGCT with a recessive mode of inheritance. Their analysis took into account that the incidence

of TGCT has increased substantially over the past few years and that the treatment for TGCT

has improved greatly and led to better maintenance of fertility and longer survival. The study

showed that time trends in TGCT make little difference to the evidence for the recessive mode of

inheritance.(90) Under this recessive model, the estimated gene frequency was 3.8%. Thus 7.3%

(according to the Hardy-Weinberg equilibrium) of men in the population carry the mutant allele

and that 0.1% are homozygous. According to the calculations by Heimdal and colleagues, the

life-time risk of development of TGCT in homozygous men is 43%.(90)

Nicholson and Harland(89) aimed to define the incidence of genetically determined TGCT in

the general population. They analysed published data on the age of onset of TGCT and the

prevalence of bilateral TGCT in familial and unselected general cases according to the arguments

used by Knudson’s two-hit model for tumorigenesis in patients with a familial predisposition

to a certain cancer.(91) In the general population bilateral TGCT occurs much more frequently

than could be attributed to chance. On the basis of the comparison of the distribution of age

at diagnosis between patients with bilateral TGCT and familial cases (ie, those likely to be

genetically determined), the investigators assumed in their model that the increased risk was

due solely to genetic susceptibility. Thus patients with bilateral TGCT probably carry the same

susceptibility genotype as that causing familial TGCT. Nicholson and Harland(89) estimated that

about a third of the general patients with TGCT carry the susceptibility genotype and that the

penetrance is 0.45. Calculations with these values showed that a recessive disease model

showed a better fit with the observed risks for brothers (2.2%) and fathers (0.5%) of patients

with TGCT than a dominant disease model. The frequency of the susceptibility allele in the

recessive model was estimated to be 5%. Although this analysis, based on the simplest set of

assumptions, fitted the data reasonably well, Nicholson and Harland conceded that it was based

on data obtained under diverse conditions and could be subject to several unknown biases.

Therefore, the assumptions that underlie the model might be incorrect. They also conceded

that some unilateral tumours may be predisposed to a contralateral tumour by some as yet

unknown biological mechanism (indeed, such a mechanism was later identified as early somatic

KIT mutations, as iscussed above) or that there is more than one predisposition gene and that

the mode of inheritance is more complex.

X-linkage was not specifically addressed by the two studies mentioned. The higher relative risk

for brothers than for father-son pairs is compatible with a recessive mode of inheritance, but

since the early 1990s the incidence of TGCT has increased sharply, maintenance of fertility after

treatment for TGCT has improved greatly, and the introduction of cisplatin-based chemotherapy

has substantially lowered the number of deaths from TGCT.(92-94) These changes might lead to a

situation in which the relative risk for father-son is higher than was believed initially. A further

possibility is that because of shorter survival (ie, a lower chance that a TGCT patient will father

a child) and abolition of fertility in the past (before the 1990s), less transmission of TGCT from

father to son was seen. Although the study by Heimdal and colleagues(90) tried to take this

factor into account, the investigators admitted that to do so is very difficult and that more

complex assumptions are involved. The study by Nicholson and Harland(89) made allowance

for the time trends, but partly because only two studies have attempted to define the mode of

inheritance, there is a risk that the conclusion of inheritance according to a recessive model is

incorrect. A dominant form of inheritance should therefore not be totally excluded. Because a

proportion of bilateral cases are caused by early somatic mutations in KIT and because there is

clearly more than one TGCT susceptibility gene, susceptibility to TGCT is probably more complex

than suggested by either of these models.


2

35

Association studies and haplotype analyses

Association studies try to find statistical evidence of an association between polymorphic loci

on, or closely linked to, candidate genes (eg, genes involved in the metabolism of mutagenic

agents, known tumour-suppressor genes or oncogenes) and a particular phenotype (eg, TGCT)

by comparing frequencies between cases and controls. Because TGCT is a rare disease and

mapping of the causal genes is difficult, TGCT patients and families from founder populations

are very suitable people in whom to detect TGCT predisposition genes with the aid of

association or haplotype analyses.(95;96) Patients in founder populations are expected to share a

high number of mutations predisposing to TGCT from recent common ancestors.(62) They will also

share segments of DNA surrounding the disease mutations. Haplotypes consisting of conserved,

ancestral alleles at genetic markers covering the region of a disease mutation will therefore be

more frequent among patients than among controls.

Several association studies have been done on TGCT. In Table 2, an overview is given of these

studies. Much attention has been paid to HLA genes and TGCT. The HLA region is thought to be

interesting for such studies, because differences in immune response based on HLA variation

might have a role in the development of cancer and metastases. Hodgkin’s lymphoma, Kaposi’s

sarcoma, colorectal cancer and breast cancer are associated with this genomic region.(97;98) The

effects (based on HLA variation) of the immune response to carcinogenic factors such as viruses

that might be associated with TGCT, could contribute to the development of the disorder.(99;100)

Previous studies have shown that HLA factors might be associated with the development of

TGCT. In particular, consistent associations were found with the HLA class II antigens. No

association was found with the HLA A or C regions, and inconsistent associations were found

with the HLA B region(96) (Table 2). The method used in most studies was serotyping, which is

much less accurate and less efficient than the more recent genotyping studies. DNA typing is

far more sensitive and can identify larger numbers of alleles.(101) The first HLA genotyping study

by Özdemir and colleagues(102) in 55 Japanese patients, showed two associations: one HLA

susceptibility allele and one HLA protective allele (relative risk 3.26 and 0.26 respectively). A

much larger genotyping study on the HLA class II region of chromosome 6p21 in 151 patients

from the northern part of the Netherlands could not replicate this association.(96) The association

between TGCT and HLA class II alleles either does not exist or is much weaker than the earlier

report suggested.

The other association studies in Table 2 give an overview of the diverse associations that

researchers have attempted to show with TGCT. In many cases, only one study has investigated

a certain region. To date, no convincing associations has been shown between TGCT and a

genetic polymorphism.

Linkage analysis and genome-wide screens

Linkage analysis relies on the fact that during the formation of gametes through meiosis

chromosomal material is exchanged between homologous chromosomes. This recombination

of genes means that were once on the same homologue of a particular chromosome pair

become separated and those once on different homologues are brought together. The process

of recombination is achieved by crossing-over of the chromosomes during meiosis. The

probability that a cross-over will occur and two loci on the same chromosome will randomly

segregate increases with the distance between the genetic loci. For two genes close together

on a chromosome, a cross-over is unlikely to occur and the two loci will be inherited together.

Linkage analysis uses this feature to map genes by means of a series of polymorphic genetic

markers (microsatelite markers or single-nucleotide polymorphisms) along the chromosomes

to follow the inheritance pattern of marker loci and the disease locus through a family. If a

particular polymorphism or marker is inherited by a higher proportion of patients than would

be expected by chance, the locus is said to be linked to the disease and is in fact probably

very closely located to the disease locus on the DNA strand. Since the position of the marker

locus is known, the location of the unknown disease locus is then found. Linkage analysis in a

set of pedigrees with many cases of a cancer has identied genes associated with breast cancer,

colon cancer, familial melanoma, and others. For many of these cancers the pedigrees used had

many affected cases spread over several generations. In many cases, a single pedigree provides

sufficient numbers of affected cases to generate a statistically significant LOD score (logarithm

of odds, a statistic that indicates whether a locus is inherited by a higher proportion of patients

than expected by chance). The search for a TGCT susceptibility gene has been hampered by

a lack of these large multigenerational pedigrees. Most families identified are relative pairs,

generally siblings. Larger pedigrees with three, four or five affected cases have been reported

but these rarely extend beyond two generations.

The International Testicular Cancer Linkage Consortium (ITCLC) has the largest collection of

TGCT pedigrees with two or more cases of TGCT in a family.(103) An analysis by this group has

shown that no single autosomal gene accounts for all TGCT pedigrees.(104) The analysis of 160

TGCT pedigrees consisted of calculations under the best autosomal dominant and autosomal

recessive models given by the segregation analysis of Heimdal and colleaguesl(90), but allowing

for the possibility of genetic heterogeneity (ie, several genes are involved separately or in

interaction with each other). An X-linked component to susceptibility was not considered. The

analysis suggests that this set of pedigrees has sufficient power to detect two TGCT genes each

contributing to 50% of the set with a statistically significant LOD score greater than 3 under

both a dominant and a recessive model. The analysis also investigated the power to detect four

TGCT susceptibility genes each contributing to a quarter of the family set. In this case, significant

LOD scores would not be possible unless the number of families in the set approached 500

pedigrees. The last report by the ITCLC was on a total of 179 pedigrees and the failure to find

LOD scores greater than 3 in any autosomal locus suggests that no single locus can explain

occurrence of TGCT in at least 50% of the families. It also suggests that there are more than

two TGCT susceptibility genes.(103) The power to detect some or all of the susceptibility genes


2

37

Tabl

e 2:

Tes

ticul

ar C

ance

r As

soci

atio

n St

udie

s

Auth

or

Type

of st

udy

Gen

e As

soci

atio

n Co

nclu

sion

DeW

olf et

al.

In

vest

igat

ed g

enom

ic D

NA o

f 61

TGCT

HLA

A,

Incr

ease

d an

tige

n freq

uenc

y of

DW

7

Pos

sibl

e as

soci

atio

n19

79(1

06)

patien

ts for

52

HLA

A, B, C

B,C

,D

insu

bgro

up w

ith

tera

toca

rcin

oma

betw

een

HLA

DW

7

spec

ifici

ties

. HLA

D a

ssig

nmen

ts m

ade

ant

igen

s (p

<0.0

1)

antige

ns a

nd T

GCT

by

hom

ozyg

ous

typi

ng.

Carr e

t al

.

Ass

esed

20

patien

ts w

ith

HLA

In

crea

sed

antige

n freq

uenc

y of

AW

24

Pos

sibl

e as

soci

atio

n 19

79(1

07)

tera

toca

rcin

oma

antige

ns

in p

atie

nts

with

met

asta

tic

betw

een

HLA

AW

24

dise

ase

(p=

0.00

8)

and

met

asta

tic

TGCT

Maj

sky

et a

l. Te

sted

23

HLA

ant

igen

s of

A a

nd B

HLA

A

No

sign

ifica

nt res

ult

No

asso

ciat

ion

betw

een

1979

(108

) lo

ci In

62 p

atie

nts

with

test

icul

ar

and

B

TG

CT a

nd H

LA A

or B

ge

rmin

ativ

e tu

mou

rs a

nd 3

01 h

ealthy

an

tige

ns

an

tige

ns

unre

late

d su

bjec

ts.

Pol

lack

et al

. Ass

esed

exp

ress

ion

of H

LA a

ntig

ens

HLA

A,B

, Lo

w fre

quen

cy o

f DR3

in a

ll pa

tien

ts,

Caus

al a

nd p

rogn

ostic

1982

(109

) in

145

unr

elat

ed w

hite

pat

ient

s w

ith

C

and

DR

alth

ough

no

differ

ence

was

sig

nific

ant

impo

rtan

ce o

f ov

eral

l

TGCT

an

tige

ns

afte

r co

rrec

tion

for

num

ber of

de

crea

ses

in H

LA D

R3

an

tige

ns tes

ted

rem

ains

to

be d

eter

min

edOliv

er e

t al

. 11

4 TG

CT p

atie

nts

HLA

A, B,

Incr

ease

d DR5

in s

emin

oma

patien

ts,

Pos

sibl

e as

soci

atio

n19

86(1

10)

C

and

DR

(p<0

.04)

and

inc

reas

ed H

LA-D

R7

in

betw

een

HLA

DR5

and

antige

ns

stag

e IV

dis

ease

TGCT

(p<

0.05

) DR7

antige

ns a

nd T

GCT

Aig

inge

r et

1:

inv

estiga

ted

132

TGCT

pat

ient

s HLA

B13

1:

Inc

reas

ed fre

quen

cies

of HLA

B13

in

Pos

sibl

e as

soci

atio

nal

. 19

87(1

11)

2: c

ompl

eted

joi

nt c

alcu

lation

of

and

DR

TGCT

with

hae

mat

ogen

eous

met

asta

ses

betw

een

HLA

ant

igen

s

publ

ishe

d da

ta o

n HLA

ant

igen

s in

an

tige

ns

(p<0

·01)

, of

DR2

in T

GCT

witho

ut

and

TGCT

35

1 TG

CT p

atie

nts

m

etas

tase

s (p

<0·0

01), a

nd o

f DR1

in

se

min

oma

(p<0

·035

)

2: Inc

reas

ed fre

quen

cies

of DR1

(p

<0·0

25) an

d DR5

(p<0

·015

) in

sem

inom

a; inc

reas

ed fre

quen

cy o

f DR5

(p

<0·0

5) a

nd D

R7

(p<0

·05)

in

non-

se

min

oma

with

haem

atog

eneo

us

m

etas

tase

s Au

thor

Ty

pe o

f st

udy

Gen

e As

soci

atio

n Co

nclu

sion

Die

ckm

ann

Rev

iew

ed 1

19 fam

ilies

in

HLA

In

crea

sed

freq

uenc

y of

Hap

loty

pe s

haring

in

all (fam

ilial

) et

al.

1989

(112

) pu

blis

hed

repo

rts

with

an

tige

ns

HLA

ant

igen

s HLA

A3

TG

CT c

ases

tha

t is

mor

e th

an

ag

greg

atio

n of

TGCT

and

B7

(p<0

.02)

ex

pect

ed. Add

itio

nal ev

iden

ce for

theo

ry o

f ge

netic

influ

ence

in

the

ca

use

of

TGCT

Ryb

erg

et a

l. A

sses

sed

rest

rict

ion-

frag

men

t-

HRAS

Hig

her freq

uenc

y of

rar

e

Pat

ient

s w

ith

bila

tera

l ca

ncer

or

1993

(113

) le

ngth

pol

ymor

phis

m o

f th

e

al

lele

s in

TGCT

pat

ient

s

unila

tera

l ca

ncer

you

nger

tha

n 20

VNTR

reg

ion

flank

ing

the

HRAS

(23/

696)

tha

n in

con

trol

s;

year

s of

age

had

sig

nific

antly

gene

in

geno

mic

DNA fro

m 3

48

(p=

0·00

4). Hig

her

high

er inc

iden

ce o

f ra

re H

RAS

TG

CT p

atie

nts

and

343

heal

thy

freq

uenc

y of

rar

e al

lele

s al

lele

s th

an p

atie

nts

with

cont

rols

in b

ilate

ral TG

CT (p=

0·01

) un

ilate

ral ca

ncer

old

er tha

n 20

year

sDie

ckm

ann

In

vest

igat

ed b

lood

gro

ups

of

Lew

is a

ntig

en

Mor

e freq

uent

Lew

is

Ass

ocia

tion

s of

Le(

a-b-

) w

ith

TGCT

et

al.

1993

(114

) 57

7 pa

tien

ts tre

ated

for

TGCT

an

d HLA

an

tige

n Le

(a-b

-) in

patien

ts

and

of H

LA B

w41

with

sem

inom

a

and

cont

rols

; Le

wis

ant

igen

s of

B

w41

th

an in

cont

rols

(p=

0·04

6;

supp

ort co

nten

tion

tha

t ge

netic

14

3 pa

tien

ts tre

ated

for

TGCT

,

antige

ns

rela

tive

risk

2·38

). H

LA

fact

ors

are

invo

lved

in

the

caus

e

and

cont

rols

; an

d HLA

ant

igen

s

Bw

41 a

ssoc

iate

d w

ith

and

path

ogen

esis

of TG

CT.

of

215

pat

ient

s trea

ted

for TG

CT

se

min

oma

(p=0·

0001

;

HLA

-Bw

41 c

ould

be

used

as

risk

an

d co

ntro

ls

re

lativ

e ris

k 8·

2)

mar

ker fo

r se

min

oma

Hei

mda

l et

Com

pare

d TG

CT (n

= 4

42) an

d

WT1

(W

ilm’s

A1

alle

le o

f W

T1

Find

ings

mig

ht ind

icat

e an

al. 19

94(1

15)

cont

rol (n

= 3

84) po

pula

tion

tum

or); O

ne o

f po

lym

orph

ism

ass

ocia

ted

invo

lvem

ent of

the

WT1

bot

h in

fo

r th

e al

lele

fre

quen

cies

of tw

o t

he p

olym

orph

ism

s w

ith

bila

tera

l TG

CT (p=

0.03

) su

scep

tibi

lity

to T

GCT

and

in

po

lym

orph

ic loc

i lo

cate

d at

(W

T) w

as loc

ated

an

d di

ssem

inat

ed T

GCT

pr

ogre

ssio

n of

the

dis

ease

ch

rom

osom

e ba

nd 1

1p13

w

ithi

n an

d th

e

(p=

0.03

)

ot

her (D

11S3

25)

in c

lose

pro

xim

ity

to W

T1)


2

39

Cont

inue

Tab

le 2

: Te

stic

ular

Can

cer

Asso

ciat

ion

Stud

ies

Auth

or

Type

of st

udy

Gen

e As

soci

atio

n Co

nclu

sion

Olie

et al

. In

vest

igat

ed p

rese

nce

of N

RAS

NRAS

or K

RAS

Mut

atio

ns fou

nd in

five

sem

inom

a NRAS

or K

RAS

mut

atio

ns19

95(1

16)

and

KRAS

mut

atio

ns, in

cod

ons

(thr

ee in

NRAS

and

two

in K

RAS,

all

are

rare

and

app

aren

tly

12

, 13

, an

d 61

, in

sna

p-froz

en

codo

n 12

), a

nd in

one

non-

sem

inom

a no

t es

sent

ial fo

r in

itia

tion

sa

mpl

es o

f 10

0 pr

imar

y TG

CT,

(K

RAS,

cod

on 1

2)

or p

rogr

essi

on o

f TG

CT

cons

isting

of 40

sem

inom

a an

d

60 n

on-s

emin

oma

tum

ours

Hei

mda

l et

Ass

esse

d al

lelic

fre

quen

cies

of Oes

trog

en-

No

differ

ence

s in

alle

lic fre

quen

cies

Dat

a do

not

ind

icat

e th

atal

. 19

95(1

17)

thre

e po

lym

orph

ism

s w

ithi

n th

e re

cept

or g

ene

betw

een

canc

er p

atie

nts

and

cont

rols

va

riat

ion

in the

5’ en

d of

oe

stro

gen-

rece

ptor

gen

e in

(p=

0.08

) th

e oe

stro

gen-

rece

ptor

te

stic

ular

can

cer (n

=45

4) a

nd

gene

con

fers

sus

cept

ibili

ty

co

ntro

l po

pula

tion

s (n

=67

2)

to tes

ticu

lar ca

ncer

Har

ries

et

Ass

ocia

tion

stu

dy w

ith

GST

P1a

G

STP1b

alle

le

6.5%

of co

ntro

ls a

nd 1

8.7%

of TG

CT

The

GST

P1b

alle

le

al. 19

97(1

18)

and

GST

P1b

in

155

TGCT

(v

aria

nt o

f pa

tien

ts h

omoz

ygou

s fo

r th

e GST

P1b

as

soci

ated

with

test

icul

ar

patien

ts

glut

athi

one

al

lele

p=

0.00

2 ca

ncer

S-tran

sfer

ase

cDNAs

)

Özd

emir e

t H

igh-

reso

lution

gen

otyp

ing

HLA

gen

es

HLA

DRB1

susc

eptibi

lity

alle

le

Susc

eptibi

lity

and

al. 19

97(1

02)

stud

y in

HLA

cla

ss II re

gion

(rel

ativ

e risk

3.2

6) a

nd D

QB1

prot

ective

alle

le for

TGCT

on 5

5 pa

tien

ts w

ith

TGCT

prot

ectiv

e al

lele

(0.

26) fo

und

in (no

n)

iden

tifie

d

sem

inom

a pa

tien

ts

Rap

ley

et

Gen

otyp

ing

stud

y on

134

Xq

27 reg

ion

Link

age

of reg

ion

Xq27

in

patien

ts

Link

age

betw

een

Xq27

and

al. 20

00(1

05)

pedi

gree

s w

ith

TGCT

with

TGCT

and

at le

ast on

e fa

mily

TG

CT (pe

rhap

s al

so

mem

ber w

ith

bila

tera

l te

stic

ular

cr

ypto

rchi

sm)

ca

ncer

p=

0.0

34

So

nnev

eld

Hig

h re

solu

tion

gen

otyp

ing

HLA

gen

es

No

asso

ciat

ion

betw

een

HLA

cla

ss

No

asso

ciat

ion

betw

een

et a

l.

stud

y in

HLA

cla

ss II re

gion

on

II

gene

s an

d (n

on) se

min

oma

HLA

cla

ss II ge

nes

and

2002

(96)

15

1 TG

CT p

atie

nts

pa

tien

ts –

log 10

(p-v

alue

)<3

TGCT

Tum

our cl

assi

ficat

ion

/ ty

pe o

f st

udy

acco

rdin

g to

the

orig

inal

artic

le. TG

CT: te

stic

ular

ger

m c

ell tu

mou

r

for TGCT will depend greatly on the degree of genetic heterogeneity of this disease. Continued

identification and recruitment of families into this linkage study will be crucial to the efforts to

identify these genes.

The first evidence for a TGCT susceptibility gene was for the X chromosome.(105) The analysis on

99 pedigrees compatible with X linkage (no male-to-male transmission) yielded a heterogeneity

LOD score of 2.01 on chromosome Xq27. The data were subsequently stratified according to

the presence of at least one bilateral case, the presence of undescended testis, histology, and

age. Families with at least one case of bilateral disease showed strong evidence of linkage to

a locus on Xq27 (heterogeneity LOD score 4.76) and were more likely to show linkage to the X

chromosome than were families without a bilateral case. This score corresponds to a genome-

wide significance of p=0.034, equivalent to a LOD score of 3.78 in a genome-wide linkage search

of affected sibling pairs without subgrouping. This proposed susceptibility gene on chromosome

Xq27 was called TGCT1.(105) In Klinefelter’s syndrome, the increased risk of extragonadal germ-cell

tumours is associated with an extra copy of the X chromosome which further suggests X-linked

involvement in TGCT.(75;105) In addition, the subset analysis provided evidence that the TGCT1

might also predispose to cryptorchism. Linkage to this locus was found in 73% of families with

cryptorchism compared with 26% of families without cryptorchism (p=0.03). The results also

suggest that about a third of the excess familial TGCT risk to brothers is from TGCT1, with little

difference in the residual risks to brothers and sons after this locus has been accounted for. This

is the first cancer gene to be mapped in a genome wide search of predominantly sibling pairs

and was the third male cancer gene mapped to the X chromosome.(105)

Conventionally, such a result would be ratified in another set of families; however no other set

of TGCT pedigrees of similar size is known. These pedigrees are rare and collection of a similar-

sized set may take several years. The ITCLC reported preliminary results on an additional 25

pedigrees compatible with X linkage but the set was to small for any firm conclusions to be

drawn.(103) The gene in this region has not been identified.

TGCT1 is unlikely to be the only TGCT susceptibility gene and it could account for as few as

25% of all TGCT pedigrees. Several genome searches have now been done on the ITCLC set,

and suggestive evidence for linkage has been obtained for several autosomal regions including

3q27, 12q12-q13, 16p13, 18q22 – qter.(103) Many more pedigrees need to be assesed before any

of these regions can be conclusively identified as including a TGCT susceptibility gene.

DiscussionThere is evidence that TGCT susceptibility genes exist and are important in this disease.

This evidence includes increased TGCT risks associated with a positive family history, the higher

frequency of bilaterality in familial cases and the ethnic and racial differences that do not

change with migration, as well as the mathematical modelling of observed familial and non-


2

41

familial cases, and possible associations with known hereditary syndromes and constitutional

chromosomal anomalies; the latter commonly seen within the setting of defects in urogenital

differentiation.

Human TGCT susceptibility genes have not yet been identified. The putative gene mapped to

Xq27 is postulated to confer an increased risk of TGCT as well as cryptorchism. Completion of

the human gene map, further studies on animal models, the arrival of advanced gene arrays

(chips), genome-wide single- nucleotide polymorphism technology, and applied bioinformatics

are expected to facilitate further exploration of genetic predisposition to TGCT. One point of

interest is whether such predisposition can be linked to a genetic contribution from increased

intrauterine oestrogen concentrations and susceptibility to disruption of usual urogenital

differentiation or to an environmental factor that would account for the increasing incidence

of TGCT. Insight into genetic features of TGCT not only might contribute to the identification

of individuals at increased risk of developing the disorder, but also is likely to increase our

understanding of normal urogenital differentiation and non-hereditary TGCT. This knowledge will

contribute to improving the diagnosis and treatment of TGCT in the general population. From

a practical clinical perspective, identification of men with an increased risk of TGCT depends

on the presence of known risk factors, including the family history of cancer. Clinicians should

therefore record the family history of cancer and urogenital differentiation defects as part of their

routine clinical practice in patients with TGCT or urogenital differentiation defects. Brothers of

patients with TGCT should be informed about their risk of developing the disorder and should

be encouraged to examine their testicles regularly. The US National Cancer Institute and the

ITCLC are doing a genetic and causal multidisciplinary study of familial TGCT. (http://familial-

testicular-cancer.cancer.gov/). However, unlike some other increased risk situations determined

by positive family histories (eg, those of colorectal cancer), the effects of preventive measures,

including testis self-examination, in terms of tumour risk as well as psychosocial effects remain

to be investigated.

Search strategy and selection criteria

Articles and studies included in this review were identified by searches of titles, abstracts and

keywords of reports included in PubMed and through the list of references cited by the papers

found by those PubMed searches. Searches were done with combinations of the search terms:

” testicular”, “seminoma”, “non-seminoma”, ”familial”, “risk factors”, “genetic”, “tumo(u)r”,

“gene(s)”, “linkage” and “association”. Papers published before 1970 and those published in

languages other than English, French and German were excluded.


3

43

Chapter 3

Syndromic Aspects of Testicular Carcinoma

MF Lutke Holzik1, RH Sijmons2, DTh Sleijfer3, DJA Sonneveld1,

JEHM Hoekstra-Weebers4,5, J van Echten-Arends2, HJ Hoekstra1

Departments of: 1Surgical Oncology, 2Genetics, 3Medical Oncology, 4Wenckebach Institute, University Medical Center Groningen, the Netherlands.5Comprehensive Cancer Center North-Netherlands, Groningen, the Netherlands

Cancer 2003; 97:984-992

Syndromic aspects of testicular carcinoma44 Genetic Predisposition to Testicular Cancer

3

45

Syndromic Aspects of Testicular Carcinoma

IntroductionAlthough testicular carcinoma (seminoma and non-seminoma) is a rare disease, it is the most

common form of malignant disease in men between the ages of 20-40 years.(28;62) The exact

etiology of testicular carcinoma remains unknown; however, over the years, various risk

factors have been identified, including factors with an assumed or definite genetic basis (i.e.,

cryptorchidism, familial testicular carcinoma). Greater understanding of the molecular foundation

of hereditary tumor predisposition will not only facilitate the identification of men who have

an increased risk for testicular carcinoma but also will provide more insight into the origination

of nonhereditary forms of testicular carcinoma. There are various indications of a genetic

predisposition for testicular carcinoma.(61;105) Supporting arguments are the presence of familial

clustering and the racial differences in incidence of this tumor. Recently, DNA-linkage studies

have indicated the existence of a gene on the X chromosome that, when mutated in the germ

line, is associated with an increased risk for the development of testicular carcinoma. Another

argument is the presence of testicular carcinoma in patients with a hereditary abnormality or

with a constitutional chromosomal anomaly. Systematic research into the incidence of testicular

carcinoma in patients with these disorders is scarce in the literature.(119;120) The objectives of the

current study were to study disorders have been reported in combination with seminomatous

and non-seminomatous testicular carcinoma in the literature and to examine the extent to

which our current knowledge of the genetics and pathogenesis of these disorders contributes to

gaining a better understanding of the oncogenesis of testicular carcinoma.

Materials and MethodsA literature search was made for articles in the English language on seminomatous and/or

non-seminomamatous testicular carcinoma that cooccurred with hereditary disorders or

constitutional chromosomal anomalies. We also searched for articles on risk factors for testicular

carcinoma, because risk factors that have been established in the normal population might

also form part of a genetic condition and (partly) may explain testicular carcinoma in patients

with those conditions. The Literature sources were: Pubmed (www.ncbi.nlm.nih.gov/PubMed),

McKusick’s on-line catalogue of hereditary phenotypes (Online Mendelian Inheritance in Man,

www.ncbi.nlm.nih.gov/Omim) and Familial Cancer Database 1.2 (found at http://facd.uicc.org).(121)

The key words were (combinations of) testicular cancer, germ cell cancer ,seminoma, non-

seminoma, congenital anomalies, syndromes, hereditary ,inherited, mendelian , genetic and risk

factors. For an additional source, we used the references listed in the articles found using above-

described method. We excluded all articles which were not written in English and/or described

hereditary disorders or constitutional chromosomal anomalies with testicular carcinoma, other

than (non)seminoma.


3

47

Table 1: Hereditary Disorders and

Constitutional Chromosomal Anomalies

in Combination with Reported Testicular

Carcinoma (Alphabetical Order)

Name Syndrome a

Synonym b

OMIM # c

Androgen insensitivity syndrome XLR AR gene 1, 2, 3 X î (43;

Testicular feminization syndrome located at 152-160)

Androgen receptor deficiency Xq11-q12

Dihydrotestosterone receptor deficiency

# 300068

Adrenogenital syndrome AR CYP21 gene 1 (161;162)

Congenital adrenal hyperplasia (CAH1) located at

# 201910 6p21.3

Ataxia teleangiectasia AR ATM gene 4 11 (163)

Lois – Bar syndrome located

# 208900 at 11q22.3

Congenital total hemihypertrophy AR, (164)

Hemihyperplasia Spor,

Isolated hemihyperplasia impr

# 235000

Down’s syndrome 21 1 21 î (78;79;

Trisomie 21 165-168)

# 190685

Familial male-limited precocious AD LCGR gene 4 (169)

puberty (FMPP) located at

# 176410, 152790 2p21

Familial atypical multiple mole AD CDKN2A/p16, 12 î (170-

melanoma syndrome located at 173)

Familial dysplastic nevus syndrome 9p21, CDK4

FAMMM located at

# 155600, 155601 12q14, CMM1

located at

1p36

Hereditary persistence alpha-feto AD AFP located 4 (174-

protein (HPAFP) at 4q11-q13 177)

# 104150

Kallmann syndrome 1 XLR KAL1, located 1, 2, 4 X î (178)

# 308700 at Xp22.3

Mod

e of

In

her

itan

ce d

Gen

e /

Gen

e m

ap loc

us e

Ris

k fa

ctor

num

ber

co

rres

pond

ing

with

tab

le 2

f

Tum

orcy

togen

etic

ab

norm

ality

g

Lite

ratu

re

refe

renc

esî

î

Table 1 continued

Name Syndrome a

Synonym b

OMIM # c

Klinefelter syndrome XXY X, Y 2,4 X î (42;43;

Y 74;179-

185)

Li Fraumeni syndrome AD TP53 located (73;132)

# 151623, 114450 at 17p13.1,

CHK2 located

at 22q12.1

Marfan AD 15q21.1 (120;

# 154700 186)

Mixed gondal dysgenesis X,Y 1,2 X î (43;123;

45,X/46,XY Gonadal dysgenesis Y 187-191)

# 233420, 306100

Neurofibromatosis type I AD NF1, located (192)

Von Recklinghousen disease at 17q11.2

# 162200, 162220

Noonan’s syndrome AD PTPN11, 1,2 12 î (193;

Male turner syndrome, Pterygium located at 194)

Colli syndrome 12q24.1

# 163950

Persistent mullerian duct syndrome AR AMH located 1,2,3, 12 î (195-

# 261550 at 19p13. 4,5 200)

3-p13.2

AMHR2 gene

located at

12q13

Prader-Willi syndrome AD, cgd(15q)/UPD 1, 2, 4 (201;

Prader-Labhart-Willi syndrome impr (mat) located 202)

# 176270 at 15q11-q13

Proteus syndrome Spor, (203)

Encephalocraniocutaneous lipomatosis, AD?

ECCL

# 176920

Mod

e of

In

her

itan

ce d

Gen

e /

Gen

e m

ap loc

us e

Ris

k fa

ctor

num

ber

co

rres

pond

ing

with

tab

le 2

f

Tum

orcy

togen

etic

ab

norm

ality

g

Lite

ratu

re

refe

renc

es

î

î


3

49

ResultsOur literature search revealed 83 articles and a total of 25 hereditary disorders and constitutional

chromosomal anomalies in combination with testicular carcinoma (see Table 1). Table 2 presents

a list of recognized risk factors for testicular carcinoma that can occur as part of a hereditary

disorder or constitutional chromosomal anomaly (e.g., we excluded exposure to estrogens in

utero). Column 4 in Table 1 shows which risk factors for testicular carcinoma were present per

disorder in correspondence with the factors numbered in Table 2. Detection of statistically

significant correlations between the chromosomal / hereditary disorder and the testicular tumor

could not be performed due to the rarity of these disorders.

DiscussionThe literature search revealed 25 different hereditary disorders and constitutional chromosomal

anomalies that cooccurred with seminomatous or non-seminomatous testicular carcinoma.

These co-occurrences can be explained in two ways. They simply may be coincidental, or they

may result directly or indirectly, possibly interacting with other endogenous or exogenous risk

factors, from the constitutional genetic defect underlying the hereditary disorders / chromosomal

anomalies in question.

The majority of the hereditary conditions listed in Table 1 are extremely rare; only a few dozen to

100 patients with such a congenital disorder have been described. The combination of testicular

carcinoma and a hereditary condition often was described only in one or a few case reports, never as

the subject of clinical epidemiologic research into large groups of patients with a certain hereditary

abnormality. Moreover, many publications on hereditary disorders were limited to relatively young

patients, which means that complications in adulthood, such as testicular carcinoma, have received

little attention. This might have led to under-reporting of testicular tumors in patients with such

disorders. Conversely, there may be a publication bias in view of the remarkable nature of the

cooccurrence. These possible causes for bias and the small number of patients reported make it

difficult to provide statistical proof of a significant correlation between having a certain hereditary

Table 1 continued

Name Syndrome a

Synonym b

OMIM # c

Prune Belly syndrome Spor ?, 1,2 (204;

# 100100 AD?, 205)

AR?

Rubinstein-Taybi syndrome AD CREBBP (206)

Broad thumb-hallux syndrome gene located

# 180849 at 16p13.3

Russell-Silver syndrome Spor+ UPD(mat) 7 1,4,5,6 7 î (124)

Silver-Russell Dwarfism, Silver-Russell imprXL?, 17q23-q24

syndrome, SRS, Russel-Silver Dwarfism AD?,

# 180860, 312780 AR?

Supernumerary nipples, familial AD ? Multi- (207-

Polythelia, familial factorial? 209)

# 163700

Testicular germ cell tumour, familial XL Xq27 1,2,3 X î (105)

# 300228

Von Hippel-Lindau disease (VHL) AD VHL located (131)

# 193300 at 3p25-p26

X-linked ichthyosis XLR, STS located 1 X î (210;

Steroid sulfatase deficiency AD at Xp22.32 211)

# 308100 a Name: name of syndrome b Synonym = synonym of the syndrome c OMIM (#) number: McKusick’s on-line catalogue of hereditary phenotypes found at:

http://www3.ncbi.nlm.nih.gov/Omim/ d Mode of inheritance: AD = Autosomal dominant, AR = Autosomal recessive, XL = X-linked,

XLR = X-linked recessive, Spor = Sporadic, Impr = Imprinting, ? means that the mode of inheritance

is suggested in the literature, but is inconclusivee Gene/Gene map locus: name of the gene and locus, or only the locus of the gene if the gene has

only been mapped, but not cloned. UPD(mat) = uniparental (maternal) disomyf Risk factor number: risk factor corresponding with the number in Table 2g Tumorcytogenetic abnormalities: known (parts of) chromosomes that were identified as abnormal in

cytogenetic studies on (non)seminomatous tumours in general: = under-representation,

î = Over-representation

If a box is empty, the data are unknown

Mod

e of

In

her

itan

ce d

Gen

e /

Gen

e m

ap loc

us e

Ris

k fa

ctor

num

ber

co

rres

pond

ing

with

tab

le 2

f

Tum

orcy

togen

etic

ab

norm

ality

g

Lite

ratu

re

refe

renc

es

î

Table 2: Recognized Risk Factors for Testicular Carcinoma that May Be Part of Hereditary

Disorders and Constitutional Chromosomal Anomalies

Risk factor References

1. Cryptorchidism (34;212-217)

2. Subfertility (infertility) (34;46;218)

3. Inguinal hernia requiring surgery (34;39;80;214)

4. Hypogonadism (165)

5. Hypospadias (214)

6. Early age at puberty (34;218-220)


3

51

disorder and the development of testicular carcinoma, unless the risk of a testicular malignancy is

relatively high in certain hereditary disorders. The so-called intersex disorders, (and, in particular,

gonadal dysgenesis), are examples of such disorders. It is possible that 30% of individuals with

gonadal dysgenesis or mixed gonadal dysgenesis have an increased risk of developing gonadal

neoplasia. These patients have been known to develop a gonadoblastoma; this form of in situ germ

cell tumor has the ability to transform into an invasive germ cell tumor (e.g. seminoma). It appears

to be the presence of Y chromosome material in a dysgenetic gonad that predisposes to testicular

tumor development.(77;122;123)

Due to the fact that it is difficult to gain insight into the etiology of testicular carcinoma in patients

with a hereditary disorder through an epidemiologic-statistical approach, it is important to find other

ways to study the nature of these cooccurrences. A finding that is worthy of attention in this respect

is that many of the conditions listed in Table 1 also involve urogenital differentiation disorders,

several of which are known to be recognized risk factors for testicular carcinoma in the general

population (Table 2). An example is the Russel-Silver dwarf syndrome; over 40% of these patients

have cryptorchidism, hypospadia and an early onset of puberty.(124) Although there is no proof that

these recognized risk elements are direct causal factors or solely epidemiologic markers of an as

yet unknown causal factor, it is conceivable that, particularly in conditions that involve urogenital

differentiation disorders, testicular carcinoma develops as a further expression of such differentiation

disorders. Skakkebaek et al. developed a model that aims to explain the correlation between these

well-established epidemiologic risk factors, genetic factors and testicular carcinoma (Fig. 1).(12;125)

Those authors assume that the cause of testicular carcinoma lies in a condition they refer to as

testicular dysgenesis syndrome (TDS), which is postulated to be caused by a range of environmental

and/or genetic defects that disrupt the embryonal programming of gonadal development during fetal

life. In the Skakkebaek model, known risk factors, such as testicular maldescent (cryptorchidism),

infertility, and hypospadias, do not cause testicular carcinoma but, rather, result from TDS, as does

testicular carcinoma. The genes underlying the hereditary conditions listed in Table 1 may cause TDS

and subsequently may cause (indirectly) testicular carcinoma, together with a range of associated

urogenital anomalies. The Skakkebaek model takes into consideration a variety of urogenital

defects and their severity, including testicular carcinoma in the absence of congenital urogenital

anomalies. The type of genetic defect might influence the severity of TDS and, thus, the severity

and type of any associated urogenital anomalies. Currently, it is not clear whether cryptorchidism

results from TDS, as the model postulates or whether cryptorchidism (also) directly causes testicular

carcinoma. The decreased risk of testicular carcinoma after orchidopexia indicates a more direct role

in neoplastic development, although the data are conflicting and it remains to be seen whether

the Skakkebaek model is correct.(126-129)

From a theoretical point of view it cannot be excluded that in a number patients with testicular

carcinoma cases another pathway of tumor development has occurred, in which testicular

carcinoma develops in a normal differentiated testis, (i.e., in the absence of TDS, for instance

as a result of mutations in tumor suppressor and/or [proto] oncogenes). In the hereditary

disorders referred to as the human cancer syndromes, (130) these mutations are present in the

germline; therefore, these disorders are of special interest when looking at testicular carcinoma

predisposition. Testicular carcinoma has been reported in three of these hereditary disorders:

the Li-Fraumeni syndrome, neurofibromatosis type 1 (Recklinghausen disease) and Von Hippel-

Lindau disease.

Von Hippel-Lindau disease features (clear cell) renal cell carcinoma and epididymis cyst

adenomas as frequent urogenital anomalies.(131) However, only a single report has been published

on testicular carcinoma in Von Hippel-Lindau disease. Furthermore neurofibromatosis type 1 has

been described a number of times in combination with (bilateral) testicular carcinoma. Hartley

et al. and Heimdal et al. even suggested that testicular carcinoma may be a rare manifestation

in the Li-Fraumeni syndrome. (73;132)

Figure 1: The testicular dysgenesis syndrome. The asterisk indicates the possibility that cryptorchidism

(testicular maldescent) acts as a causal risk factor. CIS: carcinoma in situ. Modified frame from:

Skakkebaek NE, Rajpert-De Meyts E, Main KM. Testicular dysgenesis syndrome: an increasingly

common developmental disorder with environmental aspects. Hum Reprod. 2001;16:972–978.

© European Society of Human Reproduction and Embryology.

Figure 1: The Testicular Dysgenesis Syndrome

Enviromental factors

Testicular dysgenesis

Hereditary disorders

and constitutional

chromosomal

anomalies and

somatic genetic

defects

Reduced semen

quality

CIS −> Testicular

Cancer

Testicular

maldescent

Hypospadias

Disturbed sertoli

cell function

Impaired germ cell

differentiation

Decreased leydig

cell function

Androgen

insufficiency

?*


3

53

When attempting to unravel the molecular steps that lead to testicular carcinoma, the field of tumor

cytogenetics can be very helpful. Cytogenetic studies on testicular carcinoma have demonstrated an

increase in chromosome 12p in invasive testicular carcinoma. In addition, complex rearrangements

have been found with increases and decreases of specific chromosomal material: (parts of)

chromosomes 4, 5, 11, 13, 18 and Y are under-represented, whereas (parts of) chromosomes 7, 8,

12, 21 and X are over-represented.(133-136) The search for genes in these regions that are responsible

for testicular carcinoma, including 12p, is still in its early stages and it remains unknown which of

these genes also may carry germline mutations.(137) To date, linkage studies on these regions have

isolated only Xq27 as the locus for a gene (TGCT1) that might be responsible for familial clustering

of testicular carcinoma (105). It is not clear whether the HLA regions harbor a hereditary testicular

carcinoma gene: the data are controversial.(96;102;106;110-112;138)

Table 1 shows whether the genetic defects associated with the hereditary conditions lie in

chromosomal regions that appear to be of interest in cytogenetic studies on testicular carcinoma.

If there are correlations, then this might be another clue to a causal relationship between testicular

carcinoma and the genetic defect concerned. It is striking that, in the current literature review, we

found six hereditary conditions with underlying gene defects/regions that lie on the X chromosome

(Table 1); in addition, cryptorchidism has been described in association with these six hereditary

disorders. These findings, in combination with evidence concerning the Xq27 region suggest that

the X chromosome plays a role in the etiology of cryptorchidism resulting in testicular carcinoma

predisposition whether or not according to the model of testicular dysgenesis (Fig.1).

If, on the grounds of tumor cytogenetics or other considerations (e.g., the fact that a gene is

already known to play a role in other types of tumor), a gene appears to play a candidate role

in the oncogenesis of testicular carcinoma, then further molecular studies on tumors may help to

determine whether their role is more or less probable. Such research can involve searching for

somatic gene mutations, loss of heterozygosity (LOH) (loss of the normal allele in the tumor),

changes in methylation status, and gene expression with the aid of immunohistochemistry, or on a

large scale by means of gene expression arrays. Currently, such research has only been performed

on a very limited scale on genes that are responsible for the hereditary conditions listed in Table

1. Kume et al. recently described a patient with neurofibromatosis type 1 and testicular carcinoma

in whom no LOH of the NF1 gene could be demonstrated.(139) This makes it less probable that

NF1 mutation played a role in the pathogenesis of testicular carcinoma in this patient. Studies

concerning somatic mutations of the gene for Von Hippel-Lindau disease in patients with sporadic

gonadal tumors have not yet revealed any mutations.(140) Screening for the gene of the Li-Fraumeni

syndrome (TP53), in testicular tumors failed to demonstrate any pathogenic mutations, although

some missense mutations of unknown pathogenicity have been observed.(141) Gene expression

studies have suggested that another associated gene (CHK2) may play a role in testicular

carcinoma.(142) Somatic mutations of p16 (the gene mutated in the germline in a proportion of

families with familial dysplastic nevus syndrome) have been observed in testicular carcinoma.(143)

The insulin growth factor binding protein (IGFBP)1 gene may be involved in Russell-Silver

syndrome, and immunohistochemical studies have suggested a role of this protein in testicular

carcinoma development.(144;145) Mouse models with germ line defects in the above mentioned

genes also may have provided clues to associated tumors, although the spectrum of associated

tumors may differ significantly between mice and humans. No testicular tumors were found in

knock-out mice for TP53, NF1, p16, VHL and a range of other known tumor-suppressor genes.(146)

Based on data generated by research to date, we still cannot draw any definite conclusions

regarding the role of the above-discussed germline mutations in testicular oncogenesis.

Completion of the human gene map and the availability of advanced gene arrays and

bioinformatics undoubtedly greatly will facilitate further exploration of the role of hereditary

gene defects in testicular carcinoma. The first gene expression profiling studies on testicular

carcinoma that used large-scale gene arrays (chips) have been published recently and it can be

expected that more candidate genes will be found through studies like these.(147) It would be

interesting to include the genes that are associated with hereditary disorders mentioned in Table

1 in these expression arrays, because they may prove to be associated with testicular carcinoma.

The testicular dysgenesis model has implications for interpreting the results of these expression

studies. It is conceivable that some of the genes that are important in testicular dysgenesis and

(subsequent) neoplasia are expressed normally only in a narrow time window during early gonadal

development. Therefore, those genes will not be expressed in normal adult testicular tissue;

and, if they act as a step in tumor development through loss of action, then gene expression

studies comparing adult normal and testicular carcinoma tissue will not display any differences in

expression. Gene expression studies on normal gonads or dysgenic gonads during various stages of

development (including animal models) may suggest additional genes to be included in further

screening research for mutations.(148)

The objective of this review was to summarize the current knowledge about the hereditary

predisposition of testicular carcinoma. In a last remark about the possible psychological,

social, ethical and economic implications of the identification of men who are at increased

risk of hereditary carcinoma, we note that the literature on individuals with a family history

of malignancy shows that issues such as medicalization, stigmatization, coping with disease-

related worry and anxiety, greater sense of vulnerability, difficulty understanding statistical risks

and risk perception, aspects of decision making, changes in family dynamics and planning, and

difficulties with health insurance are related to the progress of genetic science.(149-151)

In summary, the identification of a hereditary predisposition for testicular carcinoma is likely to

contribute to our understanding of the development of the nonhereditary variety and help to

identify men with an increased risk of testicular carcinoma. We present an overview of hereditary

disorders and constitutional chromosomal anomalies that have been described in the literature

in combination with testicular carcinoma. Although, from an epidemiological point of view, there


3

55

seems to be only a direct or indirect correlation between testicular carcinoma in mixed gonadal

dysgenesis and Xq27-linked familial testicular carcinoma, the presence of urogenital differentiation

disorders and data from tumor cytogenetic research, combined with the knowledge on gene loci of

the discussed hereditary disorders, suggest that such a relation also may exist in other syndromes.

New techniques are rapidly becoming available that will enable us to complete the human gene

map and investigate the possible role of large numbers of candidate genes in the development of

testicular carcinoma.


4

57

Chapter 4

Do the Eastern and Northern Parts of The Netherlands Differ in Testicular Cancer?(letter to the Editor)

MF Lutke Holzik1, DJA Sonneveld1, HJ Hoekstra1, GJ te Meerman2, DTh Sleijfer3,

M Schaapveld4.

Departments of: 1Surgical oncology, 3Medical oncology.

University Medical Center Groningen, the Netherlands

Department of: 2Medical Genetics, University of Groningen, the Netherlands

Department of: 4Comprehensive Cancer Center North-Netherlands Groningen,

the Netherlands

Urology 2001; 58:636-637

Do the eastern and northern parts of The Netherlands differ in testicular cancer?58 Genetic Predisposition to Testicular Cancer

4

59

To the editor,

Spermon et al. (221) conclude in their article that in the Netherlands, brothers of patients with

testicular cancer have an increased risk of developing testicular cancer (TC). Recently, we also

studied familial TC in the northern part of the Netherlands and reviewed the medical records of

686 TC patients treated at the Groningen University Hospital.(56) These results compared with the

results of Spermon et al. are presented in Table I. In addition to familial TC, we studied the age-

adjusted incidence rates of TC in the Netherlands. Within a small country like the Netherlands,

there are geographic differences in incidence of TC present, with a statistically significant highest

incidence in the northern part of the Netherlands.(62) These results are in contrast to the findings

of Spermon et al., who mention that the incidence rates for cancer in the eastern parts of the

Netherlands are similar to other parts of the Netherlands. Shared environmental factors, as well

as genetic drift, might have resulted in the higher incidence of TC found in the stable founder

population in the northern part of the Netherlands.(62) The geographic clustering of TC in stable

founder populations in the northern Netherlands may lend support to a genetic susceptibility

to TC development. The lack of sufficient number of TC families with two or more affected

males has been a handicap to performing adequate linkage analysis studies.(222) Spermon et

al. recommended in their article the mapping of candidate genes for TC in TC families. We

think that not only TC families are suitable in the search for TC susceptibility genes. Our stable

founder population is very useful in tracing TC susceptibility genes because individuals in such

populations share a relatively high frequency of genes from common ancestors; that is, genes

that are identical by descent, which can be detected by association-based studies.(96)

Table 1: Familial testicular cancer: Nijmegen compared to Groningen

Item Nijmegen Groningen

Period 1986 - 1997 1977 - 1997

Patients 379 686

FTC patients 7 (1,8%) 17 (2,5%)

Seminomas 114 (30%) 153 (22%)

Nonseminomas 265 (70%) 540 (78%)

Bilateral TC en FTC 1 (14,3%) 1 (5,9%)

Prevalence of UDT in FTC’s* 0 or 1 (0 or 14,3%) 3 (17,6%)

Inguinal hernia in FTC ? 1 (5,9%)

RR father-son 0,96 1,75

RR for TC in brothers of TC patients 5,9 9 - 13

Key: TC= testicular cancer, FTC= familial testicular cancer, UDT= undescended testis, RR= relative risk.

* Spermon et al mention in their Results section that, in their population, patients with FTC had no

history of undescended testes, whereas in their Comment section (at page 751) they mention that

one of the FTC patients had an undescended testis.


5

61

Chapter 5Testicular Carcinoma and HLA Class II Genes

DJA Sonneveld1, MF Lutke Holzik1, IM Nolte2, DTh Sleijfer3,

WTA van der Graaf3, M Bruinenberg4, RH Sijmons5,

HJ Hoekstra1 and GJ te Meerman2

Departments of: 1Surgical oncology, 3Medical oncology, 5Genetics.

University Medical Center Groningen, the Netherlands.

Departments of: 2Medical genetics, 4Medical biology

University of Groningen, the Netherlands

Cancer 2002; 95:1857-1863

Testicular Carcinoma and HLA Class II Genes62 Genetic Predisposition to Testicular Cancer

5

63

Testicular Carcinoma and HLA Class II Genes

IntroductionTesticular germ cell tumors (TGCT) constitute the most common malignancy in men 20-40 years

of age. The etiology of TGCT is still poorly understood. In addition to possible environmental

predisposing factors, several observations point to a genetic susceptibility to the development

of TGCT.(19;28) First, familial and bilateral testicular carcinoma cases occur more frequently than

expected by chance. The relative risk (RR) for brothers of TGCT patients ranges from 3 to

13.(49;50;56) Second, a genetic susceptibility is reflected by an increased incidence of TGCT in

persons with certain rare malformations of the urogenital system, some of which have a definite

genetic component in the etiology.(61;123) Furthermore, higher rates of urogenital developmental

anomalies have been reported in families prone to TGCT.(48) Third, the age distribution of TGCT

may suggest a genetic origin of the disease. TGCT are usually diagnosed at a young age and

the incidence declines after the age of 50 years. The young age at onset of testicular neoplasms

indicates a role of important etiologic factors operating early in life, either in utero or shortly

after birth. In addition to in utero exposure to maternal estrogens or exposure to infectious

agents in early childhood, these early operating etiologic factors may also be genetic.(61;223-225)

Finally, racial differences in the incidence of TGCT may point to a genetic component in the

etiology of the disease. The highest incidence is observed in Caucasians of Northern European

descent. People of African descent have a universally low incidence of TGCT. In the United

States, the incidence of TGCT in African Americans is only one fourth of that observed in

Caucasians.(28;61;63) Based on findings in numerous clinical and epidemiological studies, a genetic

susceptibility to TGCT is very likely. Several candidate genes have been proposed to play a role

in the etiology of the disease.(61;105;212;226-229) Recently, a high-resolution genotyping study in 55

Japanese TGCT patients showed a histocompatibility antigen (HLA)-DRB1 susceptibility allele (RR

3.26) and an HLA -DQB1 candidate protective allele (RR 0.26) for TGCT.(102) The possible role

of the HLA system in TGCT development could result from effects of HLA variation on immune

response to carcinogenic factors, for example viruses that may be etiologically associated with

the development of TGCT.(230) The importance of the HLA system in regulating susceptibility to,

and tumor development in, a growing number of neoplastic conditions is becoming increasingly

clear. An impaired immune system, genetically or acquired, favors carcinogenic factors.(97)

Linkage studies in TGCT families could be performed to map candidate genes for TGCT.

Unfortunately, to date, the lack of a sufficient number of families with two or more affected men

with TGCT has been a handicap to perform linkage studies with enough power to find effects of

frequent alleles or genes with low marginal effects.(227) An alternative approach to linkage studies

in familial cases is to search for testicular carcinoma susceptibility genes among testicular

carcinoma patients in founder populations by means of association analysis including haplotype

methods. These approaches have more power than linkage analysis when high frequency alleles

are involved in the pathogenesis. However unlike linkage analysis, they require markers that

are very close to the causative genes. Patients in founder populations are expected to share


5

65

a relatively high number of alleles from recent common ancestors, which is also expected to

apply for mutations predisposing to testicular carcinoma. Therefore, founder populations are

particularly suitable for fi nding genes predisposing to the development of testicular neoplasms

through association-based methods. In a previous study(56) we showed the geographic clustering

of testicular carcinoma in the northern part of The Netherlands, which is indicative of the

importance of founder alleles. The current study, which includes testicular carcinoma patients

and their relatives from this founder population, is the fi rst extensive genotyping of the HLA

region on chromosome 6p21 in a large number of TGCT patients using both standard methods

and a new haplotype sharing method to examine the association between HLA class II genes

and TGCT.

Materials and MethodsPatients

A total of 151 TGCT patients treated at the University Medical Center Groningen (UMCG) in The

Netherlands during the period 1977-1998 were selected for the initial analysis. The majority

of these patients descended from three provinces (Groningen, Friesland and Drenthe) in the

northern part of The Netherlands, based on information collected about birthplace of the

patients’ great-grandparents. Patient characteristics are listed in Table 1. The difference between

the total number of nonseminomas (n=132, 87%) and pure seminomas (n=19, 13%) is due to

different referral patterns for these histologic subtypes. Histological diagnosis was established

for all patients by the Department of Pathology, UMCG. For 108 patients, DNA from both parents

was available for phase determination and for control (nontransmitted haplotype). For the

remaining 43 patients, DNA from children (older than 18 years) and spouses was available for

phase determination. In these cases, the haplotypes of the spouses were regarded as controls.

All participants gave their informed consent and the Ethical Committee of the UMCG approved

the study.

Genotyping

High molecular weight genomic DNA was extracted from peripheral lymphocytes from 20 mL

blood using standard protocols. After DNA extraction, a set of 15 polymorphic microsatellite

markers in the HLA region on 6p21 was genotyped in all patients and controls. Markers were

selected over a distance of approximately 8 cM. The marker order was determined by using

sequence data of the major histocompatibility complex (MHC).(231) Most of the 15 markers used in

the current study are located on this 3.6 megabase MHC sequence, particularly in the HLA class II

region. The position of the remaining markers on the immediate fl anking regions was determined

by using available marker information from data published in print or on the internet (see table

2 and fi gure 1 for marker information).(105;232;233) For each polymerase chain reaction (PCR): 0,25

units Taq DNA polymerase (Roche diagnostics, Mannheim, Germany) were used to amplify the

fragments. The reaction volume was 10 µl. Reaction mixtures contained 200 µM of each dNTP, 1.5

mM MgCl2 , 10 mM tris-HCL, 50 mM KCL and 0.25 µM of each primer (with one primer 5’ labeled

with a fl uorochrome 6-FAM, HEX or NED). Cycling was performed on a PTC-225 thermal cycler

(MJ research, Waltham, MA, USA). Amplifi cation consisted of an initial denaturation of 5 minutes

at 950C, 35 cycles of 30 seconds at 950C, 30 seconds at 550C and 1 minute at 720C. Post PCR

multiplexing was performed by combining 1-10 µl (based on signal strength) of PCR products.

Pooled fragments (2.3 µl)were mixed with 2.5 µl deionised formamide and 0.2 µl ET-400R size

standard (Amersham Pharmacia Biotech, Upsala, Sweden) and separated on a MegaBACE 1000

capillary sequencer (Amersham Pharmacia Biotech) according the manufacturer’s protocol.

Results were analyzed using genetic profi ler v1.1 (Amarsham Pharmacia Biotech).

Table 1: Patient Characteristics

No. of patients 151

Median age at diagnosis (range) 29.4 (15.9-63.0) yrs

Histologic type (%)

- Pure seminoma 19 (13)

- Nonseminomaa 132 (87)

Familial testicular carcinoma (%) 16 (11)

- Affected fi rst-degree relative 7 (5)

- Affected second-degree relative 9 (6)

Bilateral testicular carcinoma (%) 7 (5)

History of undescended Testis (%) 23 (15)a with or without a seminomatous component

Figure 1: Map of the HLA-region


5

67

Statistical methods

After genotyping the 15 markers in all participants (each marker meets the Hardy-Weinberg

criteria), the set of haplotypes present in patients and the set of nontransmitted haplotypes

present in parents or spouse control haplotypes was determined. Differences between these

two haplotype sets were analyzed using Haplotype Sharing Statistic (HSS), a new method

for quantitative analysis of haplotype similarity. The validity of this method is demonstrated

elsewhere by applying it extensively to simulated and empirical data.(232;234-236) HSS assumes that

mutations that predispose to TGCT will be present more often in patients than in controls. Some

of the patients will have inherited a possible predisposing mutation to TGCT development from

a common ancestor, especially in a founder population. Due to linkage disequilibrium, a small

haplotype surrounding this mutation will also be identical by descent among these patients.

The amount of identical DNA is dependent upon the number of recombinations that have taken

place on either side of the mutation since it occurred. This number is particularly influenced by

the number of meioses that have taken place and the recombination frequency of the region.

The number of meioses between patients, selected from a population on the condition that a

specific disease mutation is present on their haplotypes, is expected to be smaller than the

number of meioses between a random sample of controls from that population. Therefore, the

length of identical or shared DNA surrounding a predisposing mutation in patients is expected

to be larger than the length of shared DNA surrounding that locus in controls. The difference

in length of haplotype sharing surrounding a specific locus between patients and controls can

be used as an indication for involvement of this locus in susceptibility to the disease. This

difference is expected to be largest at the marker locus closest to the disease mutation. HSS

defines the haplotype sharing between pairs of haplotypes as the number of intervals between

loci that shows identical alleles on a row from a locus in both directions. This can be evaluated

for all pairs of haplotypes for all marker loci.

As a test of linkage disequilibrium, the statistical significance of the overlap of the observed

haplotypes is evaluated. For this test, the observed alleles are redistributed randomly over

their loci and the haplotype sharing in this randomized set is calculated. The mean haplotype

sharing observed in the data is compared to the mean haplotype sharing in the randomized data

in which there is linkage equilibrium among all marker loci. In addition to the new haplotype

sharing test, we also performed single and two locus association tests and a transmission

disequilibrium test (TDT). For the one and two locus association test, the frequencies of the

alleles and two locus haplotypes, respectively, are compared using a chi-square test, taking

only those alleles into account that have an expected frequency of at least one copy. The TDT

test evaluates the transmission distortion of each allele versus all other alleles, using the test

proposed by Spielman et al.(237) For each locus, the result given will be the maximum distortion

at that locus.

Tabl

e 2:

Mar

ker

Dat

a

Locu

s M

arke

r na

me

Posi

tion

(bp)

* Fo

rwar

d Pr

imer

(5´

->3´

) Re

vers

e Pr

imer

(5´

->3´

)

1 D6S

1560

te

lom

eric

CT

CCAGTC

CCCA

CTGC

CCCA

AGGCC

ACA

TAGC

2 DNRNGCA

35

5440

AGGAATC

TAGTG

CTCT

CTCC

CT

CTAGCA

AAAGGAAGAGCC

3 RIN

G3C

A

3753

87

TGCT

TATA

GGGAGACT

ACC

G

GATG

GGAAGTT

TCCA

GAGTG

4 D6S

2445

46

1158

AATA

TGATG

GAAGAAGTA

ATC

CAG

GGATT

ACA

GGTA

TAAGCC

ATT

G

5 TA

P1

4986

34

GCT

TTGATC

TCCC

CCCT

C GGACA

ATA

TTTT

GCT

CCTG

AGG

6 D6S

2444

60

1183

GAGCC

AAGAACC

CAGCA

TTC

GGAAGGATT

CTAAATA

GGGGAG

7 D6S

2443

63

1213

CC

ATA

CCAAAGTA

AAACC

CAG

GAGGATG

AAGGGAAATT

AGAG

8 G51

1525

64

8774

GGTA

AAATT

CCTG

ACT

GGCC

GACA

GCT

CTTC

TTAACC

TGC

9 D6S

1666

70

0043

CT

GAGTT

GGGCA

GCA

TTTG

ACC

CAGCA

TTTT

GGAGTT

G

10

D6S

273

1577

323

GCA

ACT

TTTC

TGTC

AATC

CA

ACC

AAACT

TCAAATT

TTCG

G

11

TNFa

17

2559

2 GCC

TCTA

GATT

TCATC

CAGCC

ACA

CC

TCTC

TCCC

CTGCA

ACA

CACA

12

C3-4

-13

2839

280

GCA

TGACA

CTATA

GTG

GCT

G

CATT

GCA

CTCC

AGTC

TGGGC

13

D6S

265

3165

149

ACG

TTCG

TACC

CATT

AACC

T ATC

GAGGTA

AACA

GCA

GAAA

14

D6S

478

3376

450

CCTC

CATA

ATT

GTG

TGAGCC

CC

AATC

TTCT

AACC

CAAGCA

15

D6S

258

cent

rom

eric

GCA

AATC

AAGAATG

TAATT

CCC

CTTC

CAATC

CATA

AGCA

TGG

* Po

sitio

n (b

ase

pair)

on

sequ

ence

as

publ

ishe

d by

the

San

ger Ce

ntre

(231

)

Mar

kers

use

d in

thi

s st

udy.

The

y co

ver fu

lly the

tw

o lo

ci fou

nd b

y Özd

emir

et a

l.(102

)


5

69

ResultsFigure 2 shows the test of linkage disequilibrium. In both patients and controls, there is

significant excess sharing (-log10 (p-value) >3) compared with random redistribution for all loci,

indicating strong linkage disequilibrium in the entire genotyped region. Haplotype sharing in

the presence of linkage disequilibrium indicates that similar haplotypes are likely to be identical

by descent, i.e., inherited from a common ancestor suggesting that HSS has the power to

detect differences between patients and controls. These differences in mean haplotype sharing

between patients and controls are plotted in Figure 3. No significant differences (-log10 (p-value)

<3) between patients and controls in haplotype sharing are observed over the entire region

between marker loci 1 and 15. In addition, single and two-locus association tests (Fig. 4), as well

as the transmission disequilibrium test (TDT) (Fig. 5), showed no significant association

(-log10 (p-value) <3) between TGCT and marker loci in the HLA region on 6p21. Subanalysis of

histological groups (i.e., seminoma, nonseminoma) demonstrated no significant differences

between patients and controls (data not shown). It must be noted, however, that the number

of seminoma patients is rather small due to the unique referral pattern of a cancer center. Our

results particularly pertain to nonseminomas.

Figure 3: HSS: difference in mean haplotype sharing between patients and controls

Figure 2: Linkage Disequilibrium

Deviation from multilocus linkage equilibrium (LE) in patients (solid line) and controls (dotted

line), evaluated by haplotype sharing. Multilocus LE is simulated by permutation of the alleles over

the haplotypes. The results are represented as the –log10 of the significance of the difference in

haplotype sharing between the observed and randomized haplotypes. Significance level is at -log10

(P value) greater than 3.

Figure 3: Difference between haplotype sharing between patient and control haplotypes as calculated

by a t test, expressed as –log10 of the significance. The standard deviation is calculated by repeated

sampling without replacement of 50% of the observed haplotypes.

Significance level is at –log10(P value) greater than 3.

Figure 4: Association of the markers with testicular germ cell tumors. Results of single-locus (solid

line) and two-locus (dotted line) association analysis. The distributions of alleles in patients and in

controls are compared using the chi-square test. The results are presented as –log10 of the significance.

Significance level is at –log10 (P value) greater than 3.6.

Figure 4: Association analysis


5

71

DiscussionIt is likely that immune response differences based on HLA variation may play a role in

carcinoma development and metastatic patterns. Several carcinomas have been reported to be

HLA associated (e.g., Hodgkin lymphoma, Kaposi sarcoma, colorectal carcinoma, and Burkitt

lymfoma.(97) HLA variation has also been suggested to play a role in TGCT development. This may

be caused by differences in effects of HLA variation on immune response to carcinogenic factors,

for example viruses that may be etiologically associated with the development of TGCT.(230) An

impaired immune system, genetically or acquired, favors carcinogenic factors. The theory that

TGCT may arise under conditions of reduced immune capacity is supported by the observation

that patients with immune deficiencies following renal transplantation have a two to fivefold

increased risk to develop a TGCT.(99) In addition, there is evidence that the incidence of TGCT in

patients with an acquired immunodeficiency syndrome (AIDS) is higher than in the general male

population.(238-241)

Recently, a high-resolution genotyping study comprising 55 Japanese TGCT patients showed a

HLA-DRB1 susceptibility allele (RR 3.26) and a HLA -DQB1 candidate protective allele (RR 0.26) for

TGCT.(102) This study suggested that one of the genetic factors involved in TGCT development may

be associated with HLA, in particularly the HLA class II region. In addition to Özdemir et al.,(102)

Oliver(110) found an association between DR5 and seminoma and an increase of DR7 in patients

with stage IV disease (extralymphatic metastasis). Aiginger et al.(111) pooled their data with those

from Oliver et al.’s study (total = 233 patients) and found a significant increased frequency of

the HLA antigenes DR1 en DR5 in seminoma patients. The current extensive genotyping study of

a large number of Dutch TGCT patients, however, fails to confirm the associations of HLA Class

II genes with susceptibility to TGCT. A previous study showed geographic clustering of testicular

carcinoma in the northern part of The Netherlands.(62) The majority of patients participating in

this study descend from this area. In addition to possible common environmental factors, this

population is likely to share a relatively high frequency of mutations in genes involved in TGCT

from recent common ancestors. However, for the HLA class II region analyzed in this study, no

difference between patients and controls is observed using both standard methods (association

and TDT analyses) and HSS, even though HSS has the power due to the presence of strong

linkage disequilibrium (Fig. 2). Strong linkgage disequilibrium suggests that only a few different

haplotypes are present in the data. Similar haplotypes are, therefore, likely to be identical by

descent, i.e., inherited from a common ancestor. Because this similarity will be centered around

the disease locus for patients and random over the entire region among controls, differences

due to disease mutation in this region would have been be revealed by HSS. The linkage

disequilibrium observed in this study is in agreement with the results of former studies that

have reported regions within the HLA region that show very little recombination (e.g., between

HLA-DR and DQ) and regions where recombination preferentially occurs.(242)

The study by Özdemir et al.(102) was the first HLA genotyping study of TGCT patients. The HLA

alleles for which associations were reported are also prevalent in the Dutch populations.(243)

However, Özdemir et al. genotyped only 55 patients whereas the current study genotyped almost

three times that number, resulting in more power, particularly for nonseminomas.(102) Moreover

with HSS, we have additional power from phase information to detect differences due to genetic

factors in the HLA region. Systematic analysis of patient and control haplotypes using a high

density genome screen can identify identity by descent within haplotypes of unrelated patients

in a founder population. This is because haplotypes that are similar for many consecutive

marker alleles are very likely to be inherited from common ancestors. The shared segment of

the haplotypes that is most significantly overrepresented in patients compared to controls is

likely to contain a predisposing gene. The previously suggested association between HLA and

TGCT need not result from a functional role of the HLA-system itself, but may also be due to

an effect from a separate gene that is closely linked with HLA loci. The analysis of haplotypes

where most genetic variation is associated with specific haplotypes enables conclusions about

all genes present on the haplotypes. Therefore, based on the results in the current study, a role

of HLA class II genes in the development of TGCT seems much more limited than previously

suggested by a few studies.

In previous studies by Oliver et al.(110) and Aiginger et al.,(111) the association between HLA and

TGCT was found by serotyping methods that are less precise and less efficient than genotyping

Figure 5: Transmission Disequilibrium Test

Results of the transmission disequilibrium test in 108 trios represented as –log10 of the significance

of the maximal transmission distortion at each locus. Significance level is at –log10 (P value) greater

than 3.2.


5

73

methods. DNA typing is more sensitive and identifies more alleles. The reason for this is that

the variation present in the HLA region is specific for the small number of founders. All variation

that contributes to disease can be identified with haplotype association methods. DNA based

methods are more accurate and allow higher definition HLA typing. High-resolution genotyping

is the method of choice because the polymorphism of the peptide binding domain of MHC class

II molecules is more precisely determined by genotypes than by serotypes.(101)

In conclusion, although there are proven associations between HLA and several malignancies

this is not the case for TGCT. The current genotyping did not confirm the previous reported

association between HLA class II genes and TGCT, despite a larger sample size, especially

nonseminomas. As the HLA alleles for which associations were reported are also prevalent in

the Dutch populations, these associations are likely nonexistent or much weaker than reported.

Further research focusing on other candidate loci should be performed to identify possible TGCT

susceptibility genes.


6

75

Chapter 6Absence of constitutional Y chromosome AZF deletions in patients with Testicular Germ Cell Tumors

MF Lutke Holzik1, K Storm2, RH Sijmons3, M D’Hollander2,

EGJM Arts4, ML Verstraaten1, DTh Sleijfer5, HJ Hoekstra1

Departments of : 1Surgical oncology, 3Genetics, 4Obstetrics and Gynecology, 5Medical oncology. University Medical Center Groningen, the Netherlands

Department of : 2Medical Genetics, University of Antwerp, Belgium

Urology 2005; 65:196-201

Absence of constitutional Y chromosome AZF deletions in patients with Testicular Germ Cell Tumors76 Genetic Predisposition to Testicular Cancer

6

77

Absence of constitutional Y chromosome AZF deletions in patients with Testicular Germ Cell Tumors

IntroductionAlthough testicular germ cell tumors (TGCT) constitute the most common malignancy in men

aged 15 to 40 years, their etiology is still poorly understood.(244) In the past few years a decrease

in fertility and an increase in TGCT has been reported.(46) This could suggest that fertility and

TGCT share a common etiological factor. A typical illustration is the East-West semen quality

gradient in the Nordic Baltic area and the incidence of TGCT. Finland and Estonia have only

one third of the TGCT incidence compared with Denmark and Norway, which is inversely related

to the lower sperm counts observed in Danish and Norwegian men compared with men from

Finland and Estonia.(245) A retrospective cohort study of more than 30,000 men from infertile

couples found an association between infertility and a subsequent risk of TGCT.(46) Men with

infertility were 1.6 times more likely to develop TGCT. The greatest risk of TGCT was in the first

2 years (standardized incidence ratio 1,8) after the first semen analysis. At 2 to 11 years after the

first semen analyses, the standardized incidence ratio was 1,5 to 1,6. This is a relatively constant

risk for TGCT, and impaired spermatogenesis may, therefore, have been present many years

before TGCT was diagnosed.(46) These results are in line with our previous results showing that

TGCT patients have already impaired spermatogenesis before orchiectomy was performed.(246)

In the model postulated by Skakkebaek et al.,(12) TGCT and poor semen quality are symptoms

of one underlying entity, the testicular dysgenesis syndrome (TDS). Endogenous, as well as

exogenous, risk factors, including inherited predisposing gene mutations may result in TDS. Past

research has traditionally focused on the range of possible exogenous risk factors for TDS and

TGCT, including prenatal exposure to maternal hormones.

Although TGCT susceptibility genes remain to be identified,(244) more light has recently been shed

on the genetics of infertility. Up to approximately 8% of infertility in the general male population

can be explained by the presence of constitutional deletions of part of the long arm of the Y

chromosome (Yq11), referred to as the azoospermia factor (AZF) region (subdivided into AZFa

to AZFd).(247-249) AZFc deletions are the most commonly found (60%). Although most of the AZF

deletions observed in infertile men are new (de novo) mutations, they have been inherited in

some cases from (apparently fertile) fathers, and 0.4% of fertile men in the general population

appear to carry an AZF deletion.(248)

Foresta et al. recently observed a particularly high percentage of 27.5% AZF (a-c) deletions

in patients with low sperm counts as well as unilateral cryptorchism.(250) Cryptorchism is an

acknowledged risk factor for TGCT and is one of the postulated other possible manifestations

of the TDS.(12;81;251) Taken together, these observations suggests that, at least from a theoretical

point of view, constitutional AZF deletions might be one of the genetic contributors to the


6

79

development of TDS and thereby of TGCT and other TDS manifestations. This possibility would

be in line with tumor cytogenetic studies that have demonstrated a nonconstitutional loss of

Y chromosome material in adult TGCT, as well as in their precursor carcinoma in situ, which

suggests that loss of Y chromosome material may indeed play a role in TGCT development.(252)

Bianchi et al.(253) have demonstrated that, noninherited mosaic AZF deletions can be observed in

tumor as well as nontumor, tissues from some patients with TGCT. Altogether, this might point

at a role for the loss of Y chromosome material, inculding AZF, in TGCT development. Given that

constitutional AZF deletions have been observed in fertile men in the general population and

TDS does not necessarily present with infertility, the possibility of constitutional AZF deletions

causing TDS and thereby TGCT in fertile men has not be ruled out. In the present study, we

investigated the frequency of Y chromosome deletions in the AZF region in a series of fertile, as

well as infertile, patients with TGCT.

Patients and MethodsPatients selection

A total of 112 patients with TGCT treated at the University Medical Center Groningen (UMCG) in

the Netherlands were randomly selected for initial analysis. Patient characteristics are listed in

Table 1. Familial TGCT was defi ned as more than one case in the family. Histologic diagnosis was

established in all patients by the Department of Pathology of the UMCG. Owing to the position

of our medical center (academic referral hospital for the northern part of The Netherlands) most

patients had undergone orchiectomy (before diagnosis) in the referring hospitals without prior

semen analysis and preservation. Therefore, in the current series of patients no data were

available on semen quality before orchiectomy; however, semen data were available for 25

patients after orchiectomy. We stratifi ed these 25 patients into four groups according to semen

concentration (Table 2). All participants gave their written informed consent and the ethical

committee of the UMCG approved the study.

Genotyping High-molecular-weight genomic DNA was extracted from peripheral blood lymphocytes according

to standard protocols.(254) After DNA extraction screening for AZF deletions was performed by

multiplex polymerase chain reaction (PCR) analysis using the Y Chromosome Deletion Detection

System, version 1.1 (Promega), and the addition of Multiplex Master Mix E (Promega). Version

1.1 has been extensively described by Aknin-Seifer et al.(255) and the addition of mix E has

improved accuracy. Currently, the system that includes mix E is known as the Y Chromosome

Deletion Detection System, version 2.0.(256) In the current study the system consisted of 24

primer pairs, of which 20 primer pairs are homologous to previously identifi ed and mapped

sequenced tag sites (STSs) within the AZF regions on the Y chromosome (locations provided in

Fig. 1 and Table 3). All the loci analyzed in this study have been recommended by the European

Quality Monitoring Network Group (EQMNG) for detection of Yq11 deletions associated with

male infertility.(256) Primers were combined into fi ve primer sets to use in fi ve parallel PCR

Table 1: Characteristics of the 112 TGCT patients

Type of Neoplasia Patients/total(%)

Patients with nonseminoma 94/112 (84%)

Patients with pure seminoma 18/112 (16%)

Other characteristics

Patients with bilateral TGCT 4/112 (3,5%)

Patients with cryptorchism 21/112 (18,8%)

Patients with familial TGCT 10/112 (9%)

TGCT = testicular germ cell tumor

Table 2: Distribution of semen concentration in 25 patients after orchiectomy

Sperm Concentration Patients (n)

Normozoospermia (>20 × 106/mL) 13*

Moderate oligozoospermia (5–20 × 106/mL) 5

Severe oligozoospermia (<5 × 106/mL) 2

Azoospermia 5** Normozoospermia and azoospermia groups both included 1 patient with cryptorchism.

Figure 1: Diagram of the Y chromosome with AZF regions, previously cloned genes

and pseudogenes, and STSs. Reprinted form Technical Manual No 248.(256) Used with

permission of Promega Corporationpermission of Promega Corporationpermission of Promega Corporation


6

81

amplifications (multiplex PCR A through E; Table 3). The slight modifications to the protocol(256)

provided by the manufacturer were: 500 ng DNA in a final volume of 25 µl multiplex Master Mix,

amplification in 35 cycles on a Perkin-Elmer GeneAMP® System 9700 thermal cycler (Applied

Biosystems), and annealing at 58°C for 1 minute 30 seconds. The control samples analyzed in

each multiplex PCR were a male genomic DNA control, a female genomic DNA control and a

blank (no-DNA) control. The separation and visualization of the PCR products were performed

by electrophoresis in 4% NuSieve 3:1 Plus agarose gels (Cambrex Bio Science), stained with

Ethidium Bromide (Fig. 2). The multiplex primer sets A through D contained a control primer pair

that amplifies a fragment of the X-linked SMCX locus.

Multiplex E contains a control primer pair that amplifies a unique region in both male and female

DNA (ZFX/ZFY). Both control primer pairs are internal controls for the amplification reaction and

the integrity of the genomic DNA sample. In addition multiplex E contains a primer pair that

amplifies a region of the SRY gene that is a control for the presence of the testis determining

factor on the short arm of the Y chromosome (Yp) and allows XX males (arising from Y to X

translocations) to be detected. The Y chromosome deletion detection system(256) is the standard

procedure in the laboratory of the Department of Medical Genetics, University of Antwerp, to

detect AZF deletions in men analyzed for infertility and subfertility. We previously found some

AZF deletions in our laboratory amongst infertile and subfertile men, who did not have a history

of TGCT (data not shown).

ResultsMicrodeletions analysis of the AZF region (Yq11) was successfully performed on genomic DNA of

112 patients with TGCT. Figure 2 includes representative examples of the electrophoresis gels,

showing amplification products for multiplex A-E. No PCR products detected in the blank (no

DNA) control. As expected the positive male control showed the appropriate number and sizes

of bands for each multiplex master mix. The positive female control only showed amplification

for the SMCX and ZFX loci. In the 112 patients with TGCT no deletions within the AZF region

were detected.

DiscussionA detailed analysis of microdeletions on the Y chromosome was performed in 112 Dutch patients

with TGCT by studying 24 STSs within the AZF regions on Yq11. These patients included bilateral

cases (n=4), cases with cryptorchism (n=21), cases with a positive family history (n=10) for TGCT

(Table 1) and patients with proven normal or low sperm counts (Table 2). No AZF deletions were

observed in any of the patients. The distribution of nonseminomatous TGCTs and seminomatous

TGCTs as shown in Table 1 is related to the referral pattern of TGCT’s to the UMCG. Our data

confirmed and extended the findings of another study recently published while our study was

in progress. Frydelund-Larsen et al.(257) screened 160 Danish TGCT patients for microdeletions

on chromosome Yq11. In 103 patients seven STSs spanning the three AZF regions (AFZa,

AZFb and AZFc) (plus SRY and ZFX/ZFY) were analyzed. In 57 patients, nine additional STSs

spanning AZFabc, (and TSPY on Yp) were studied. Four of the 16 STSs spanning AZFabc (sY84

in AZFa, sY134 in AZFb, and sY152 and sY254 in AZFc) were in common with those studied in

our patient group. No AZF deletions were observed in their study population. Because, in the

study by Foresta et al.,(250) AZF deletions were only found in a group of patients with a history

Table 3: Overview of the 24 STSs amplified

STS Locus PCR Fragment (bp) Multiplex PCR Position

sY81 DYS271 209 A distal to AZFa

sY86 DYS148 232 E AZFa

sY84 DYS273 177 E AZFa

sY182 KAL-Y 125 A proximal to AZFa

sY121 DYS212 190 C AZFb

SYPR3 SMCY 350 B AZFb

sY124 DYS215 109 D AZFb

sY127 DYS218 274 B AZFb

sY128 DYS219 228 C AZFb

sY130 DYS221 173 A AZFb

sY133 DYS223 177 D AZFb

sY134 DYS224 303 E AZFb

sY145 DYF51S1 143 C proximal to AZFc = AZFd

sY152 DYS236 285 D proximal to AZFc = AZFd

sY153 DYS237 139 D AZFd (nonpathogenic)

sY242 DAZ 233 B AZFc



sY254 DAZ 370 A AZFc

sY255 DAZ 126 C AZFc

sY157 DYS240 285 A distal to AZFc

sY14 SRY 400 E SRY gene

SMCX 83 A-D X chromosome (control)

ZFX/ZFY 496 E X/Y chromosome (control)

STSs = sequenced tag sites; PCR = polymerase chain reaction; AZF = azoospermia factor.

Overview of the 24 STSs amplified in five multiplex PCR amplifications

(A-E), including 20 STSs for detection of AZF deletions associated with male infertility,

one STS sY153 which seems to be polymorphic or in multiple copies and 3 control STSs

(SMCX, ZFX/ZFY and sY14).


6

83

of cryptorchism together with azoospermia or severe oligozoospermia, it is possible that AZF

deletions are only present in that minority of patients with TGCT who also have markedly

reduced fertility. In the current study, fertility status (Table 2) was known in 25 patients (22%)

and 5 of these had azoospermia after orchiectomy. Data on semen concentration before

orchiectomy were not available because the vast majority of patients were referred for treatment

after orchiectomy performed elsewhere. Furthermore, semen analyses after orchiectomy was

only offered to patients treated with adjuvant chemotherapy or radiotherapy with the intention

to father children in the near future. Frydelund-Larsen et al. (257) presented data on fertility for

70 of 160 of their patients before TGCT treatment. A total of 37 of these patients (23% of the

total group) had severe (n = 17; less than 5x106/ml) or very severe (n = 8; less than 0.2x106/ml)

oligozoospermia or azoospermia (n = 12). Although their study as well as the current study

did not detect any AZF deletions in patients with TGCT with reduced fertility, our study did not

have enough statistical power to exclude low percentages of AZF deletions in small subsets of

patients

In conclusion, the data suggest that a substantial contribution of constitutional large AZF

deletions to the development of TGCT, whether or not in the presence of reduced fertility,

cryptorchism, previous history of TGCT or positive family history, is unlikely. The present data

do not rule out the possibility of constitutional smaller deletions or other type of mutations

in genes mapped to the AZF region. These genes could therefore be the subject of additional

research. Given the complexity of urogenital differentiation and testicular tumor development,

only the ‘tip of the iceberg’ has been mapped with respect to genes involved in these processes

to date. As new tools for molecular study become available and mapping efforts advance, more

opportunities will undoubtedly arise to explore the molecular basis of the testicular dysgenesis

model and testicular tumor development and these explorations should include interactions with

environmental risk factors.

Note added in proof: A very recent study performed by Nathanson et al.(258) has revealed that

a small inherited or de novo deletion within the AZFc region, referred to as the gr/gr deletion

appears to be associated with an increased risk to develop TGCT. This gr/gr deletion on the Y

chromosome is not detected by the commonly used test for the larger AZF deletions, and was

recently found to be a risk factor for spermatogenic failure by Repping et al.(259)

Figure 2: Multiplexes A-E.

Figure 2A-2E. Electrophoresis gel (4% NuSieve 3:1 Plus agarose) showing amplifi cation products

for (A) multiplex PCR A, (B) multiplex PCR B, (C) multiplex PCR C, and (D) multiplex PCR D,

representing STSs within AZF regions and control STS (SMCX) and (E) multiplex PCR E showing

amplifi cation products for multiplex PCR E, representing STSs within AZF regions and control STSs

(ZFX/ZFY and sY14). Lanes 1 to 4 = patients with TGCT; L = 50 bp DNA Step Ladder; B = blank

(no-DNA control), M = normal male control; F = normal female control.

A B

C D

E


7

85

Chapter 7Re-analysis of the Xq27-Xq28 region suggests a weak association of an X-linked gene with sporadic Testicular Germ Cell Tumour without cryptorchidism.

MF Lutke Holzik1, HJ Hoekstra1, RH Sijmons2, DJA Sonneveld1, G van der Steege3,

DTh Sleijfer4, IM Nolte3

Departments of: 1Surgical Oncology, 2Genetics, 3Medical Biology, 4Medical Oncology,

University Medical Center Groningen, the Netherlands

European Journal of Cancer 2006; 42:1869-74

Re-analysis of the Xq27-Xq28 region suggests a weak association of an X-linked gene with sporadic Testicular

Germ Cell Tumour without cryptorchidism

86 Genetic Predisposition to Testicular Cancer

7

87

Re-analysis of the Xq27-Xq28 region suggests a weak association of an X-linked gene with sporadic Testicular Germ Cell Tumour without cryptorchidism.

IntroductionThe incidence of testicular germ cell tumour (TGCT) is still rising.(260) Although the aetiology of

TGCT is poorly understood, it is well known that male relatives (fathers/brothers/sons) of TGCT

patients have an increased risk of developing TGCT. Currently 1-3% of TGCT patients report an

affected relative.(261) Brothers of TGCT patients have an 8-10 fold increased risk of developing

TGCT and the relative risk (RR) to fathers and sons is approximately 4-6.(104;244) Because these RRs

associated with an affected first-degree relative are considerably higher than for other cancers,

which rarely exceed four, this observation most likely points to a genetic role in the aetiology of

TGCT.(244) Efforts have been made to identify TGCT predisposing genes. Although TGCT families

have been reported in the literature, multigenerational pedigrees with several affected cases are

rare and this limits the opportunities for linkage studies. In 1994, the International Testicular

Cancer Linkage Consortium (ITCLC) was formed with the aim to collect TGCT families from all

over the world and to perform genotyping studies. Recently, Rapley and colleagues (on behalf

of the ITCLC) presented evidence for a TGCT susceptibility gene on chromosome Xq27.(105) Their

genome-wide search for linkage in a set of 134 familial TGCT cases yielded a heterogeneity

LOD (hlod) score of 2.01 on chromosome Xq27 using all families (n=99) compatible with X-

linked inheritance and a hlod score of 4.7 on chromosome Xq27, if they included only families

with at least one bilateral TGCT case (n=15) (genome wide significance level p = 0.034). In

addition, 73% (n=14) of the familial TGCT cases with a history of cryptorchism (synonymous

with cryptorchidism), a well known risk factor for TGCT, were linked to locus Xq27. These results

provided evidence for a gene on chromosome Xq27 involved in TGCT susceptibility as well as

in cryptorchism. Two recombination’s, one between markers DXS8043 and DXS8028 on the

centromeric side and one between FRAXA.pcr2 and FMR1Di on the telomeric side, bounded

the identified TGCT1 locus, resulting in an interval of ~4 cM (~2.7 Mb). In this region, three

genes have been reported so far: FMR1, responsible for Fragile X syndrome, a single exon gene

Cxorf1 (expressed in the brain) and a tandemly duplicated gene LOC158813/158812.(262;263) As yet,

germline mutations in any gene in this region have yet to be identified as the cause of increased

risk to develop TGCT.

An alternative approach to linkage studies is searching for TGCT susceptibility genes among

unrelated TGCT cases in founder populations by means of association analyses on a dense

set of markers. These so-called linkage disequilibrium fine-mapping analyses are based on the

hypothesis that patients in founder populations inherited disease mutations from recent and

common ancestors. A previous study showed geographic clustering of TGCT in the northern part

of the Netherlands.(62;95) Another study presented the results of analyses of HLA microsatellite

markers and TGCT in this founder population.(96) The current study includes the previously used




7

89

study population, expanded with additional TGCT patients from the same founder population.

The aim of this study was to corroborate or refute the previously observed linkage between TGCT

and chromosome Xq27 using association analysis and the Haplotype Sharing Statistic (HSS).

Patients and MethodsPatients and controls

A total of 276 patients were randomly selected from all TGCT patients treated during the period

1977-2001 at the University Medical Center Groningen (UMCG), the Netherlands. The majority

of these patients descended from three provinces (Groningen, Friesland, and Drente) in the

Northern Netherlands, based on information collected about birthplace of the patient’s great-

grandparents. Histological diagnosis was established for all patients by the department of

Pathology of the UMCG.

Through the patients, family members (parents, children and spouse, brothers or sisters) were

asked to participate. For this study on association of the X-chromosome with TGCT, only unaffected

male first-degree family members were used and served as controls (n=169). Mothers of TGCT

patients, if available, were used to check for correct inheritance. Population characteristics are

presented in Table 1. TGCT cases were defined as familial cases when more than one TGCT case

was present in the family. TGCT cases with male-to-male transmission (e.g. father – son) of the

putative TGCT predisposition gene were excluded as these cases were obviously not compatible

with X-linked inheritance and would have obscured the test results. The difference between the

total number of non-seminomas (89%) and pure seminomas (11%) included is in particular due

to different referral patterns for these histological subtypes. Traditionally all patients diagnosed

with a non-seminoma within a defined area of the Comprehensive Cancer Centre in the northern

part of the Netherlands (CCNN) are referred to the UMCG for further management after having

been hemi-orchidectomised at the local hospital. In contrast, the majority of patients diagnosed

with a seminoma are referred to one of the three radiation facilities within the CCNN area

(including UMCG) for radiation treatment. All participants gave their informed consent and the

Medical Ethical Committee of the UMCG approved the study.

Genotyping

High molecular weight genomic DNA was extracted from peripheral lymphocytes from 20

ml EDTA blood using standard protocols.(254) After DNA extraction, a set of 16 polymorphic

microsatellite markers in the Xq27-Xq28 region was genotyped in all patients and their relatives

(see Table 2 for marker details, only markers that meet the quality criteria are shown, see section

Results). Microsatellite markers were selected over a distance of approximately 4.3 Mb in order

to cover the TGCT1 locus. Most markers were obtained from literature and the public databases.

To get an evenly distributed set, the markers starting with XTC0 were newly developed by

searching the downloaded sequences for putative dinucleotide repeats and amplify these loci

with primers selected with the online Primer3 software (http://frodo.wi.mit.edu/cgi-bin/primer3/

primer3_www.cgi). Relative marker positions were obtained from the contigs nt_011681.13 and

nt_019686, and NCBI Map Viewer build 34 version 3 was used to determine the distance between

both contigs. For each polymerase chain reaction (PCR), 0.25 units Taq DNA polymerase (Roche

diagnostics, Mannheim, Germany) was used to amplify the fragments. The reaction volume was

10 µl. Reaction mixtures contained 200 µM of each dNTP, 1.5 mM MgCl2 , 10 mM tris-HCL, 50 mM

KCL and 0.25 µM of each primer (with one primer 5’ labeled with a fluorochrome 6-FAM, HEX

or NED). Cycling was performed on a PTC-225 thermal cycler (MJ research, Waltham, MA, USA).

Amplification consisted of an initial denaturation of 5 min at 95ºC, 35 cycles of 30 sec at 95ºC,

30 sec at 55ºC and 1 min at 72ºC. Post PCR multiplexing was performed by combining 1-10 µl

(based on signal strength) of PCR products. 2.3 µl of the pooled fragments was mixed with 2.5 µl

deionised formamide and 0.2 µl ET-400R size standard (Amersham Pharmacia Biotech, Uppsala,

Sweden) and separated on a MegaBACE 1000 capillary sequencer (Amersham Pharmacia Biotech,

Uppsala, Sweden) according to the manufacturer’s protocol. Results were analyzed using Genetic

Profiler v1.1 (Amersham Pharmacia Biotech, Uppsala, Sweden).

Statistical methods

Only markers that fulfilled the following quality criteria were included in the statistical analysis:

failed genotyping (no results, alleles unknown) in less than 15% of the study population and

erroneously scoring of two alleles in less than 4% of the male sub-population (men of course

carry only one allele for each marker on their single X chromosome). In those few events where

genotyping of an included marker erroneously revealed two alleles in a male, the genotype for

that marker was set to “unknown”. All markers fulfilling the quality criteria were analyzed for

allelic association by a chi-square test using only those alleles with an expected count of at

least three. The data were also analyzed by the Haplotype Sharing Statistic (HSS).(264-267) HSS is a

linkage disequilibrium fine-mapping method that is based on the assumption that patients share

disease mutations inherited from recent common ancestors. It measures the sharing between

Table 1: Population characteristics

Number (%)

Patients 276

Non-seminoma 248 (89%)

Seminoma 28 (11%)

Bilateral TGCT 11 (4%)

Cryptorchism 43 (15%)

Familial TGCT 16 (6%)

Sporadic TGCT without cryptorchism 220 (79.7%)

Controlsa 169 a first degree, male family members




7

91

a pair of haplotypes at a marker locus as the number of consecutive marker loci carrying the

same alleles starting from the locus under analysis in both telomeric and centromeric direction.

HSS then hypothesises that, in genomic regions containing disease mutations, haplotypes of

patients show more sharing than haplotypes of controls. The association and HSS methods were

also applied to four subgroups of patients, namely to patients with bilateral TGCT, to patients

with familial TGCT, to patients with cryptorchism and to patients with sporadic TGCT without

cryptorchism (Table 1). It should be noted that the number of patients with bilateral TGCT and

the number of patients with familial TGCT were small.

Because multiple markers are analyzed, a multiple testing correction is required. This implicates

that a result is only regarded as significant when the p-value is smaller than 0.05 / 12 = 0.004,

where 12 is the number of markers that passed our quality criteria. A 99.57% confidence interval

corresponds to a 95% CI interval corrected for multiple testing. Power of the study was assessed

by standard statistical theory on normally distributed variables, assuming that the allele counts

follow a binomial distribution, which can be approximated by a normal distribution.

ResultsOf the 16 microsatellite markers that were genotyped, four did not meet our quality criteria (data

not shown). Hence, the analyses were performed on the 12 markers shown in Table2. Allelic

association analysis did not reveal any significant difference, i.e. p<0.004, between general TGCT

patients and controls at any marker (Figure 1). In addition, the HSS did not show a significant

result either (Figure 2). As the sample of Rapley and colleagues(105) consisted of familial TGCT

cases and the linkage evidence became stronger when selecting only cases with bilateral TGCT

or cases with cryptorchism, we also performed analyses on the subsets of patients. The results

are shown in Figures 1 and 2. No significant associations were observed with any of these

subgroups or any marker. We did however observe that for marker DXS1193 the major allele

was less frequent among cases (88.3%) than among controls (96.6%) and that difference was

smaller for cases with cryptorchism (94.7%) and those with familial TGCT (100%). Therefore,

we also analysed the subgroup of sporadic cases without cryptorchism. In this subgroup, both

allelic association and the HSS revealed evidence for a TGCT susceptibility locus (p=0.014 and

p=0.008, respectively), which remain not significant after multiple testing correction (Table

3). However, after the multiple testing correction, a suggestive relative risk was observed for

individuals carrying one of the minor allele at DXS1193 of 3.8 to develop TGCT and a significant

relative risk of 4.7 (99.57% CI: 1.1-19.6) to develop sporadic TGCT without cryptorchism,

compared with carriers of the major allele.

Tabl

e 2:

Cha

ract

eris

tics

of m

arke

rs u

sed

for

this

Xq2

7 as

soci

atio

n sc

reen

(on

ly m

arke

rs tha

t m

eet qu

ality

crite

ria

are

show

n)

Mar

ker

Rela

tive

Prim

ers

Fa

iled

Het

eroz

ygou

s

po

sitio

n (M

b)

Forw

ard

Reve

rsed

PC

R (%

) m

en (%

)

DXS

8043

0.

0000

0a AGTT

CTCA

GAAACA

TTTG

GTT

AGGC

AATT

ATT

GGCA

AAGAGTA

CAGGCA

G

6.5%

3.

50%

DXS

8028

0.

2147

2 a

TGATG

ACA

CTCG

GACT

GC

GAAATA

ATA

ATA

CTTG

CCTT

GCC

T 13

.7%

3.

05%

AFM

a113

zf5

0.50

108 a

AACA

CTGCA

CGATG

AGAA

AGCT

ATC

CTGATT

TTGGTA

CT

4.6%

2.

31%

DXS

8045

1.

4836

0 a

CAGGTA

AATC

TGAGAAATG

TTCT

GC

ACT

GCG

GTG

CTGACT

AGG

7.5%

2.

15%

DXS

1200

1.

7169

3 a

TACA

CACC

AAACA

ACA

GAGCC

T CT

AGGGGCA

CTTG

AAAACA

A

11.5

%

2.30

%

XTC0

08

2.03

909 a

TT

CTGTC

TCACA

AGCC

AGATA

A

CTGATC

CTCT

GACA

GCA

TATA

C 4.

0%

3.21

%

XTC0

221

2.57

815 a

TG

TATC

TGTG

CATG

TACC

TATC

AAGAAGTC

ATC

CACT

GAGTC

TA

2.2%

0.

45%

DXS

998

2.57

935 a

CA

GCA

ATT

TTTC

AAAGGC

AGATC

ATT

CATA

TAACC

TCAAAAGA

0.5%

2.

20%

Frax

ac1

2.95

775 a

GATC

TAATC

AACA

TCTA

TAGACT

TTATT

GATG

AGAGTC

ACT

TGAAGCT

GG

5.5%

0.

70%

DXS

1215

4.

0697

3 b

GGGCA

AAACA

TTAAACC

TCTC

GCC

CTCT

AAGTC

ATT

ACG

CT

4.4%

3.

87%

DXS

1193

2.

9580

6 b

AATT

CTGACT

CTGGGGC

TTATT

TTAAGGTG

AGTA

TGGTG

TGT

12.4

%

1.27

%

DXS

1113

4.

2867

9 b

GGGAGCT

TTAGAGATT

TTGGTA

AAC

ACC

TGTG

GAGGATA

GTA

GTC

TGACT

4.

6%

3.20

%

a lo

cate

d on

con

tig n

t_01

1681

.13

b lo

catio

n on

con

tig n

t_01

9686

, di

stan

ce b

etw

een

cont

igs

dete

rmin

ed u

sing

NCB

I M

ap V

iew

er b

uild

34

vers

ion

3




7

93

patients is higher than the increase in RR to fathers or sons of TGCT patients, which might

be explained by assuming an X-linked inheritance of a TGCT predisposing trait. In addition,

patients with Klinefelter syndrome (47, XXY constitutional karyotype) have a RR of 67 to develop

mediastinal germ cell tumours. Like TGCT, these tumours are thought to arise from carcinoma in

situ. The presence of an additional copy of the X-chromosome in Klinefelter syndrome suggests

a possible dose-effect of one or more genes on this chromosome, escaping X-inactivation,

on germ cell tumour development. Indeed, cytogenetic studies have revealed that generally

the X-chromosome is overrepresented in TGCT tumour DNA.(76;244;268) Ross and colleagues (269)

determined the sequence of over 99% of the gene-containing region of the X chromosome.

They predicted that nearly 10% of the 1098 genes on the X chromosome are in a class that is

upregulated in testicular and other cancers. Taken together, these observations suggest that

the X chromosome may well harbour TGCT predisposing genes. As yet, these genes and their

mutations remain to be identified.

Table 3: Statistical results for association analyses on marker DXS1193

patients sporadic,

Marker allele controls all Cryptorchism familial no cryptorchism

DXS1193 107 96.6% 88.3% 94.7% 100% 86.1%

105 0.0% 1.3% 0.0% 0.0% 1.7%

109 0.0% 3.9% 2.6% 0.0% 4.4%

110 2.7% 3.5% 2.6% 0.0% 3.9%

112 0.7% 2.6% 0.0% 0.0% 3.3%

114 0.0% 0.4% 0.0% 0.0% 0.6%

p-value 0.059 0.92 0.82 0.014

ORa 3.8 1.6 0.9 4.7

(99.57% CI) (0.9-15.9) (0.1-18.5) (0.0-61.8) (1.1-19.6)a OR (Odds Ratio) for combination of minor alleles Figure 1: Association analyses

Association analysis for alleles. The black squares represent the results at the markers. Lines between

the markers are drawn only for an easier interpretation of the results. The lines distinguish the

different (subgroup) analyses: a thick solid line for all patients, a thick dotted line for cases with

cryptorchism, a thin solid line for the familial case and a thin dotted line for the cases without a

family history of TGCT or cryptorchism. A p-value of < 0.004 is considered significant after multiple

testing correction.

DiscussionIn the current study, we did not find an association between Xq27 and familial TGCT,

cryptorchism or bilateral TGCT. We could therefore not confirm the results found by Rapley

and colleagues.(105) We should however notice that our subgroups of familial or bilateral cases

or cases with cryptorchism were small, which results in sufficient power only when a disease

mutation with a frequency of 5% (or higher) and the putative gene has a large effect (RR>8 for

cryptorchism; RR>20 for bilateral or familial cases). Hence, it cannot be excluded that an X-linked

gene with a smaller effect is involved in familial or bilateral TGCT or cryptorchism.

Interestingly, we did observe an association between the subset of TGCT cases without a family

history of TGCT or cryptorchism and specific alleles for the marker DXS1193 both by allelic

association analysis and by the HSS. The frequency of all minor alleles was increased among

these patients compared to controls: 13.9% versus 3.4%, respectively. The risk to develop

sporadic TGCT without a cryptorchism for an individual who has one of the minor alleles was

estimated to be 4.7 (99.57% CI: 1.1-19.6). This suggests that in our population one or more

low frequent mutations of an Xq27-linked gene contribute to TGCT development but not to

cryptorchism. Alternatively, particular genotypes in this region possibly protect the normal

population from developing TGCT. Further analyses on single nucleotide polymorphism (SNPs)

in candidate genes in this region should be performed to identify the causal gene and to unravel

the nature of its causality.

Several observations had been made that could be interpreted as suggestive of the existence

of a TGCT predisposing gene on the X chromosome. The increase in RR to brothers of TGCT




7

95

In a study by Rapley and colleagues a significant linkage was reported in 99 X-compatible

pedigrees, in particular in families with cases with bilateral TGCT.(105) Recently Crockford and

colleagues examined an additional 66 pedigrees with two or more cases of TGCT at Xq27. In

contrast to the previous findings, they found no evidence for linkage at this region in this new

set of pedigrees.(262) Moreover three candidate genes from the identified minimal region at

Xq27, FMR1, Cxorf1 and LOC58813/158812 were screened for small deletions, duplications and

missense/non-sense mutations and no pathogenic mutations were observed. Crockford and

colleagues also examined five genomic regions at four other chromosomes and for these regions

no significant results were observed either.(262) Their overall conclusion was that no single

major locus can account for the majority of the familial TGCT cases. They suggest that multiple

susceptibility loci with weak effects contribute to TGCT. Their conclusion is in line with our

results. The region surrounding marker DX1193 could contain one of these susceptibility loci.

Figure 2: HSS

The Haplotype Sharing Statistic. The black squares depict the results at the markers. The different

(subgroup) analyses are represented by different line styles: a thick solid line for all patients, a thin

solid line for the cases with cryptorchism and a thin dashed line for the cases without a family history

of TGCT or cryptorchism. The subgroup of bilateral cases was too small to perform a reliable haplotype

sharing analysis. A p-value of < 0.004 is considered significant after multiple testing correction.

In conclusion, we could not confirm the previously reported association of familial, bilateral

and cryptorchism-associated TGCT with Xq27, but we cannot exclude the presence of an X-

linked gene that slightly or moderately increases risk to develop these particular phenotypes.

Interestingly our data revealed an association between the subset of TGCT cases without a

family history of TGCT or cryptorchism and marker DXS1193. Our findings suggest that in our

population one but possibly more low frequent mutations of an Xq27-linked gene contribute to

TGCT development but not to cryptorchism. It will be interesting to see whether these results can

be confirmed in other populations. Until candidate genes from this region have been identified

and can be checked for mutations, variations and their functional roles, the question of causal

relation or statistical artefact remains unanswered.


8

97

Chapter 8Interest in and motivations regarding genetic testing for testicular germ cell tumour susceptibility

MF Lutke Holzik1, HJ Hoekstra1, RH Sijmons2, DTh Sleijfer3,

JEHM Hoekstra-Weebers4,5

Departments of: 1Surgical Oncology, 2Genetics, 3Medical Oncology, 4Wenckebach Institute, University Medical Center Groningen, the Netherlands.5Comprehensive Cancer Center North-Netherlands, Groningen, the Netherlands

Submitted

Interest in and motivations regarding genetic testing for testicular germ cell tumour susceptibility98 Genetic Predisposition to Testicular Cancer

8

99

Interest in and motivations regarding genetic testing for testicular germ cell tumour susceptibility

IntroductionTesticular Germ Cell Tumor (TGCT) is a rare disease. However, it is the most common malignancy

in males between 15 and 40 years of age. The incidence in the Netherlands is approximately

7.5 per 100.000 males and rises slightly each year.(1) Despite many investigations, the etiology

of TGCT remains elusive. In the past couple of years, however, several risk factors have been

identified, with a positive family history being one of the strongest predictors. Brothers of TGCT

patients have a 8-10 fold increased relative risk (RR) to develop TGCT, while the father-son RR is

4-6 times increased.(244) These findings suggest a genetic component in the etiology of TGCT and

it has been postulated that roughly one third of all TGCT cases occur in genetically predisposed

individuals.(89) However, the search for inherited gene mutations predisposing to the TGCT

development has not yet identified high penetrant mutations (i.e. mutations associated with a

high risk of developing TGCT). If such gene mutations would be identified then genetic testing

for hereditary TGCT would be facilitated. This test would either be diagnostic for affected men

suspected of having hereditary TGCT or pre-symptomatic in unaffected men at risk for carrying

the familial mutation. Similar DNA tests are already clinically used for a number of hereditary

tumor syndromes, including hereditary breast-ovarian cancer, several types of hereditary

colorectal cancer and multiple endocrine neoplasia (MEN) syndromes.(244;270)

A review study showed that the percentage of family members who decided to undergo genetic

testing for inherited cancer susceptibility varied between 27% and 80%. This percentage seemed

to depend on several factors including cultural differences (e.g. one study on hereditary breast

cancer showed that British female mutation carriers more often choose prophylactic surgery and

chemoprevention than those from France and Canada), differences in clinical facilities available

(e.g. laboratory facilities) and differences in study methodology (e.g. recruitment source and

response rates).(271;272) It has been reported that participants approached in the clinic are more

often willing to be tested than those recruited through databases and that motivations to

participate differ between these groups.(273)

The willingness to undergo genetic testing seems to depend on whether prevention of disease

is possible or wether the disease resulting from a possible gene mutation can be treated. If

this is not the case, interest in genetic testing is substantially lower, as can be shown by the

lack of interest in genetic testing for Huntington’s Disease and Cystic Fibrosis, compared to the

high patient interest testing for breast cancer and familial colon cancer.(271;274;275) The willingness

seems also to be dependent on the number of family members affected.(276) Factors preventing

people from undergoing DNA testing include an expected unfavorable test result and costs.(277;278)

Additionally, people with a positive family history for cancer frequently overestimate their own

risk of getting cancer themselves, which seem to be a reason to undergo genetic testing.(278)


8

101

Information about genetics, predisposition and possible pros and cons should be discussed

before proceeding with DNA testing. Actual knowledge about the different factors that affect the

choices people make for DNA testing is therefore of great importance.(271;279;280)

The majority of studies published focus on the interest of people in mutation analysis for

syndromes for which some of the genes involved have been identified (BRCA1 and BRCA2 for

breast/ovarian cancer, several mismatch repair genes for HNPCC, Lynch syndrome and CDKN2A

for cutaneous melanoma).(278;281-283) The respondents in these studies were mainly women or

older than TGCT patients, with the exception of those in studies that focused on melanoma.

The latter disease mainly affects young people as does TGCT. A difference with familial TGCT is

that families with the most common mutation for hereditary melanoma (in CDKN2A) also face

an increased risk for pancreatic cancer, a tumor with a poor prognosis and no proven early

detection benefit. Men with familial prostate cancer have also been studied. As in TGCT, no high

penetrant germline gene mutations have been identified. However, prostate cancer is common

in elderly men while TGCT is more frequent in young men.(284) A recent publication, the first on

the interest of men in a possible DNA test for hereditary TGCT, showed that the majority of the

participants (66%) expressed interest in this test.(285)

Many studies have examined the interest in genetic testing of patients with suspected or proven

hereditary cancer and their relatives. Less attention has been paid to the interest of men not

confronted with cancer and patients who developed TGCT but whose cancer seems unrelated

to a family history. Comparing these groups may reveal insight into possible differences in

interest in genetic testing. A distinct group of patients are those who become member of a

patient association. The main goal of patient associations is to provide and receive information

about the disease, give members the opportunity to come in contact with fellow patients, and

share experiences and information about TGCT. Members of a patient association may differ in

opinions/attitudes toward genetic testing from patients who do not decide to become members.

TGCT patient association members were reported to be younger, more recently treated, more

highly educated, to have more often experienced a recurrence and consequently received more

treatment, and they reported a worse quality of life compared to non-members.(9)

Anticipating the availability of a possible DNA test for TGCT, we conducted a study into the

interest in and motives for undergoing such a test. We postulated beforehand that more TGCT

patients, and in particular familial TGCT patients, would decide to undergo genetic testing and

that they would like to receive more information about the possibility of having a gene mutation

for cancer than people who had not suffered from cancer. Our thought was that patients with

familial TGCT would suspect they carry a gene mutation more than patients with a negative

family history of TGCT and that men who had not been confronted with cancer would least

suspect they carry a gene mutation. We explored motivations which may play a role in the

decision to undergo genetic testing and differences in motivations between TGCT patients and

men who had not suffered from cancer. Lastly, we expected a positive association between such

motivations for genetic testing and the interest of men to undergo genetic testing.

MethodsRespondents

Four groups of respondents were recruited for this study. Firstly, all familial TGCT patients

(N=44), i.e., patients with two or more confirmed cases of TGCT in the family, were selected from

the database including all TGCT patients/survivors (n=702) treated at the University Medical

Center Groningen between 1977 and 2003. This group is called “familial TGCT”. Secondly, a

similar number of non-familial TGCT patients were selected at random (“sporadic TGCT”) from

the same database. Thirdly, the Dutch TGCT patient association approached their 116 members.

This group is called “patient association”. Finally, the last group consisted of male patients

visiting the Emergency Department of Medisch Spectrum Twente, the Netherlands, following a

minor trauma (e.g. ankle distortion or a small wound). This group is called “controls”. Patients

were approached until the group’s size was comparable to the familial TGCT patient group.

Procedure

A letter with the objectives of the study and a brief explanation about TGCT and its hereditary

aspects, the questionnaire and a prepaid return envelope were sent to the men in groups 1

(familial TGCT) and 2 (sporadic TGCT) by the researchers. The Dutch patient association for TGCT

patients and survivors approached their members by sending the same letter with information,

the questionnaire and return envelope. The control group received the information and the

questionnaire in the Emergency waiting room. All participants gave written informed consent

and the study was approved by the Medical Ethics Committee of both hospitals, the University

Medical Center Groningen (UMCG) and Medisch Spectrum Twente.

Instruments

Data on the following socio-demographic variables were collected: age, educational level, marital

status, employment status, and number of children. Highest educational level completed was

measured on a seven-point scale: (1) primary school, (2) lower vocational, (3) lower secondary,

(4) middle secondary, (5) high secondary, (6) higher vocational degrees and (7) university.

Response categories for current marital status were: (1) married/cohabiting/LAT, (2) single,

(3) divorced, and (4) widower, and for current daily occupation: (1) paid job, (2) student, (3)

unemployed, (4) unable to work, and (5) retired. Also the prevalence/occurrence of other types

of cancer in the family was collected.

One question assessed respondents’ intention to undergo genetic testing for TGCT if this were

possible: The following answers could be given: (1) definitely not, (2) probably not, (3) probably

would, (4) definitely would. Participants were asked to indicate which result they would expect if

genetic testing were possible, using the following answer categories: (1) I have hereditary TGCT

(2) I probably have hereditary TGCT, (3) the chance/risk that I have hereditary TGCT is fifty/fifty,


8

103

(4) I probably do not have hereditary TGCT, (5) I do not have hereditary TGCT, (6) I do not have

any expectation about my chance/risk on having hereditary TGCT. Participants could indicate the

degree to which they wanted to have information about “their chance/risk on hereditary TGCT

and its consequences” on a scale ranging from 1 (not wanting to receive any information) to 10

(wanting to know as much as possible).

Fourteen self-constructed questions were used to gain insight into reasons/motivations on why

patients/controls would undergo genetic testing for TGCT/cancer (Table 2) and ten questions

were used to explore reasons why patients/controls would decide against undergoing genetic

testing (Table 3). Response categories ranged from: 1: does not play a role at all in my decision

to undergo genetic testing to 6: plays a very important role in my decision to undergo genetic

testing. Most of these questions were previously used in other studies on women with a genetic

predisposition for breast cancer.(271;278;286;287) The control group men responded to the same

questions but “TGCT” was replaced with the word “cancer”.

Statistical analyses

Statistical analyses were performed using SPSS 14.0. Descriptive statistics were used to

calculate means, frequencies and percentages. Chi square tests and ANOVAs were conducted to

examine differences between groups in socio-demographic variables. For employment status, a

dichotomous variable was created with the categories “not employed for wages” consisting of:

student, being unemployed, unable to work, or retired and “employed for wages” consisting

of paid job. For marital status, a dichotomous variable was created with the categories

“relationship” and “no relationship” consisting of single, divorced and widower.

Chi square tests and ANOVAs followed by post hoc Bonferroni tests were computed to examine

differences between groups in motivations for testing. Pearson’s correlations were calculated

to investigate relationships between variables. Correlation coefficients lower than .30 are

considered weak, those between .30 and .50 are moderately strong, and those higher than .50

are considered strong.(288) Results were considered statistically significant if the probability of

occurrence was <− 0.05.

Results

Of the 44 familial TGCT patients, 23 returned the questionnaire (response 52%); 27 of the 45

sporadic TGCT patients (60%), and 77 of the 116 members of the patient association (66%) did

so. All 42 men in the control group filled in the questionnaire. The study group totaled 169 men.

Mean age was 40.4 years (range 16.4– 69.2). An ANOVA showed a significant effect of group on

age. A Post-hoc Bonferroni test showed that familial TGCT patients were significantly older than

sporadic patients and controls. Mean educational level was relatively high and not significantly

different between the four groups. The vast majority of men had a paid job (86%) and were in

a relationship (79%), and 60% had children. The four groups did not significantly differ in these

variables. TGCT patients mentioned that other types of cancer than TGCT were significantly more

often prevalent in the family (X2 =8.43, p = 0.004) than controls did. (Table 1)

Table 1: Sample characteristics and comparison between groups

Variable Familial Sporadic Patient Controls Test result

TGCT TGCT association (n=42)

(n=23) (n=27) (n=77)

Age: mean (SD) 46.1 (9.9) 36.6 (7.9) 41.1 (8.5) 38.1 (11.9) F= 5.08A, p=.002

Highest education level: 4.5 (1.6) 4.6 (1.4) 5.1 (1.5) 4.7 (1.8) F= 1.1, p=.334

mean (SD)

Marital status X2= 1.45B, p=.69

• Relationship 20 20 61 31

• No relationship

- Single 2 4 10 7

- Divorced 1 3 6 3

- Widower 0 0 0 0

Daily occupation X2 = 3.64C , p=.30

• Employed for wages 18 26 65 36

• Not employed

- student 0 0 2 3

- unemployed 1 1 2 0

- unable to work 1 0 8 3

- retired 3 0 0 0

Children X2=6.95, p=.073

• yes 16 13 51 20

• no 6* 14 26 22

Family members X2=8.43D, p = .01

diagnosed with other

types of cancer?

• yes 16 20 47 25

• no 7 7 29* 17

Genetic testing X2=4.72D, p=.03

intention: n (%)

• Probably/definitely not 3(13) 5(19) 19(25) 16(38)

• Probably/definitely will 20(87) 22(81) 58(75) 26(62)

Perceived risk: n (%) X2=45.38E, p<.001

• (probably) yes 9(39) 2(7) 5(7) 1(2)

• chance/risk is fifty/fifty 7(30) 11(41) 22(29) 2(5)

• (probably) no 2(9) 8(30) 30(40) 26(62)

• no expectations 5(22) 6(22) 19(25) 13(31)

Information wish: 8.6 (1.5) 8.6 (1.7) 8.0 (2.3) 6.9 (2.3) F= 5.12F, p=.002

mean (SD)


8

105

A significantly greater percentage of the patients (79%) than of the controls (62%) would decide

to undergo genetic testing (X2= 4.72, p = 0.03). The percentage was highest in the familial

group (87%) but not significantly higher than in the other two patient groups (Table 1). Age was

not, but educational level was significantly but weakly related to interest in genetic testing (r =

-.16, p = .042). The lower the educational level was, the greater the interest in genetic testing.

Men in a relationship, with a paid job, and with children or family members with cancer other

than testicular cancer did not significantly differ from their counterparts in interest in genetic

testing.

Thirty-nine percent of the familial patients believed that they have a hereditary form of TGCT/

cancer, while 7% of the sporadic patients, 7% of the patient association group and 2% of the

controls did so. Only 9% of the familial patients believed that they would not have a hereditary

form of TGCT/cancer, in contrast to 62% of the controls, 40% of the patient association group and

30% of the sporadic patients. A Chi square test, excluding the category: “I have no expectations

about having hereditary TGCT”, revealed that the response pattern was significantly different

between groups (X2=44.16, p<.001) (Table 1). Of the 17 men who suspected that they would

(probably) have a hereditary form of TGCT, 16 (94%) would decide to undergo genetic testing.

Of the 42 men who believed they would have a fifty-fifty chance/risk, 37 (88%) would undergo

testing. 46 of the 66 men (70%) who believed they would (probably) not to have a hereditary

form would undergo testing; and of the 43 men who did not have any expectation about test

results, 27 (63%) would undergo genetic testing (X2=11.56, p=.009).

An ANOVA showed a significant effect of group on the wish for information about hereditary

TGCT/cancer and its consequences. A Bonferroni test showed that controls wanted to receive

significantly less information than patient association members (p = 0.04), sporadic patients (p

= 0.008) and familial patients (p = 0.01). The patients’ mean scores were high, varying from 8.0

to 8.6. The correlation coefficient between the wish for information and intention to undergo

genetic testing was significant (r = 0.54, p < 0.001), meaning that the more information men

Legend Table 1

A: post hoc Bonferroni test: familial TGCT patients significantly older than sporadic patients (p=.004)

and controls (p=.009)

B: X2 relationship versus no relationship

C: X2 paid job versus other

D: all patients versus controls

E: excluding response option “no expectation” from analysis

F: post hoc Bonferroni test: control group significantly lower mean score than familial patients

(p = 0.012 ), sporadic patients (p =0.008) and patient association (p = 0.037)

* information of one respondent is missing

Tabl

e 2:

Mot

ivat

ions

for

gen

etic

tes

ting

for

TGCT

com

parisi

on b

etw

een

grou

ps a

nd r

elat

ions

hip

with

tes

ting

inte

ntio

n

Item

To

tal

fam

ilial

sp

orad

ic

patie

nt

cont

rols

AN

OVA

Co

rrel

atio

nPos

sibl

e ra

nge:

1=

no rol

e–6=

very

gr

oup

patie

nts

patie

nts

asso

-

with

test

ing

impo

rtan

t ro

le ran

king

by

mea

n

ciat

ion

in

tent

ion

scor

e of

tot

al g

roup

m

ean

mea

n m

ean

mea

n m

ean

(SD)

(SD)

(SD)

(SD)

(SD)

F p

r1)

To

supp

ort sc

ient

ific

rese

arch

4.

5 (1

.3)

4.9

(1.1

) 4.

9 (1

.3)

4.5

(1.2

) 3.

9 (1

.6)

4.2

A)

.007

.4

0***

2) T

o ga

in m

ore

certai

nty

abou

t risk

4.

4 (1

.7)

5.0

(1.1

) 4.

6 (2

.0)

4.6

(1.6

) 3.

7 (1

.6)

4.6

B)

.004

.4

9***

of

get

ting

TGCT

/can

cer

3) T

o be

mor

e ce

rtai

n ab

out m

y ch

ildre

n’s

risk

3.

9 (2

.0)

4.1

(2.1

) 4.

0 (2

.0)

4.1

(2.1

) 3.

5 (2

.0)

0.7

.57

.48**

*

4) T

o ge

t re

gula

r m

edic

al c

heck

-ups

3.

9 (1

.7)

4.4

(1.2

) 4.

7 (1

.5)

3.8

(1.5

) 3.

2 (1

.8)

6.2

C)

.001

.4

2***

5) T

o ta

ke p

reve

ntiv

e m

easu

res

3.6

(1.7

) 3.

6 (1

.6)

4.0

(1.9

) 3.

3 (1

.7)

4.0

(1.8

) 1.

6 .2

0 .4

5***

6) B

eing

afrai

d fo

r ca

ncer

rec

urre

nce

3.6

(1.8

) 3.

9 (1

.5)

4.0

(1.6

) 3.

6 (1

.8)

3.1

(1.9

) 2.

0 .1

2 .4

1***

7) A

t m

y pa

rtne

r’s/

child

ren’

s in

sist

ence

3.

4 (1

.8)

3.3

(1.9

) 3.

7 (1

.5)

3.2

(1.8

) 3.

7 (1

.8)

1.0

.39

.32**

*

8) T

o de

crea

se m

y fe

ar o

f ca

ncer

3.

3 (1

.8)

3.3

(1.8

) 3.

9 (1

.4)

3.2

(1.8

) 3.

0 (1

.9)

1.3

.28

.38**

*

9) B

ecau

se the

re is

canc

er in

my

fam

ily

3.1

(1.8

) 4.

3 (1

.3)

3.5

(1.9

) 2.

8 (1

.8)

2.9

(1.7

) 5.

1 D)

.002

.3

1***

10) Ano

ther

fam

ily m

embe

r w

ants

gen

etic

res

earc

h 3.

0 (1

.7)

3.1

(1.8

) 2.

9 (1

.6)

2.8

(1.7

) 3.

5 (1

.9)

1.5

.22

.23**

11)

At a

doct

or’s

ins

iste

nce

2.9

(1.6

) 2.

7 (1

.6)

3.1

(1.5

) 2.

6 (1

.5)

3.4

(1.8

) 2.

3 .0

76

.17*

12) At m

y pa

rent

s’/s

iblin

g(s)

’s ins

iste

nce

2.7

(1.6

) 2.

4 (1

.6)

2.9

(1.5

) 2.

5 (1

.5)

3.0

(1.6

) 1.

4 .2

6 .1

8*

13) To

hel

p de

cide

whe

ther

to

have

chi

ldre

n 2.

7 (1

.9)

2.1

(1.7

) 2.

9 (2

.0)

2.6

(1.9

) 3.

0 (2

.0)

1.3

.27

.21**

14) Gen

eral

fut

ure

plan

ning

(pa

rtne

r, job

) 2.

6 (1

.8)

2.4

(1.7

) 2.

6 (1

.6)

2.6

(1.7

) 2.

8 (1

.9)

0.3

.86

.30**

*

* =

p < 0.

05, **

= p

< 0

.01,

***

= p

< 0

.001

Post

Hoc

Bon

ferron

i te

sts:

A) c

ontrol

gro

ups’

mea

n si

gnifi

cant

ly low

er tha

n th

at o

f sp

orad

ic (p=

0.02

) an

d fa

mili

al p

atie

nts

(p=

0.03

)

B) co

ntro

l gr

oups

’ m

ean

sign

ifica

ntly

low

er tha

n th

at o

f pa

tient

ass

ocia

tion

(p=

0.02

) an

d fa

mili

al p

atie

nts

(p=

0.00

8)

C) c

ontrol

gro

ups’

mea

n si

gnifi

cant

ly low

er tha

n th

at o

f fa

mili

al (p=

0.02

) an

d sp

orad

ic p

atie

nts

(p=

0.00

1)

D) co

ntro

l gr

oups

’ (p

=0.

01) an

d pa

tient

ass

ocia

tion

grou

ps’ m

ean

(p=

0.00

2) s

igni

fican

tly low

er tha

n th

at o

f fa

mili

al p

atie

nts


8

107

would like to receive regarding their chance of having hereditary TGCT, the more likely they were

to decide to undergo genetic testing (Table 1).

Based on the mean scores of the total group, the motivations: “to support scientific research”,

“to gain more certainty on the risk of getting TGCT/cancer, “to get regular medical check-ups”

and “to be more certain about my children’s risk” played a relatively important role. The

motivations: “at the insistence of parents/siblings”, “to help decide whether to have children”

and “general future planning” played a relatively small role (Table 2). An ANOVA showed a

significant effect of group on four items. Post-hoc Bonferroni tests showed that: “to gain more

certainty about getting TGCT/cancer” and “because there is cancer in my family” played a

significantly less important role for the control group than for the familial patients (p = 0.008

and p=0.01) and for the patient association group (p = 0.02 en p = 0.002). ”To get regular

medical check-ups” and “to support scientific research” played a significantly less important role

for the controls than for the familial (p=0.02 and p = 0.03 respectively) and sporadic patients

(p=0.001 and p=0.02 respectively).

Correlational analyses of these 14 motivations regarding genetic testing resulted in 10 significant

and moderately strong relationships. The remaining 4 relationships were significant but weak

(Table 2).

Barriers for undergoing genetic testing for TGCT are shown in Table 3. The mean scores appeared

to be relatively low (varying from 1.3 to 2.9). Based on the mean scores of the total group, the

barriers: “possible trouble with insurance company” and “I do not want to undergo surgery”

played a relatively greater role while “it will have implications for my choosing a partner” and

“I do not have faith in such a test” played a smaller role. An ANOVA showed a significant effect

of group on the barrier “I do not want to undergo surgery”. A post-hoc Bonferroni test showed

that this barrier played a significantly less important role for the controls than for the patient

association group (p = 0.005). Correlational analyses between these ten questions and intention

on genetic testing showed two significant but weak correlations. The greater the intention to

undergo genetic testing, the less of a role the barriers “medical check-ups” and “time” played.

Discussion

The first goal of this study was to investigate whether patients treated for TGCT were more

interested in and if they had different motivations for genetic testing for TGCT (if such a test was

available) than people who had not had TGCT. As hypothesized, we found that a significantly

larger percentage of men who had been treated for TGCT would undergo genetic testing for TGCT

(79 %), compared to men (62 %) who had not suffered from cancer. A difference in intention

to undergo genetic testing was also found in a study comparing breast cancer patients with

healthy controls.(289) In that study, the first group was almost six times more likely to express

interest in the genetic test than women without cancer. A second important finding from that

Tabl

e 3:

Mot

ivat

ions

aga

inst

gen

etic

tes

ting

(bar

rier

s) for

TGCT

, co

mpa

risi

on b

etw

een

grou

ps a

nd r

elat

ions

hip

with

tes

ting

inte

ntio

n

Item

To

tal

fam

ilial

sp

orad

ic

patie

nt

cont

rols

AN

OVA

Co

rrel

atio

n

Pos

sibl

e ra

nge:

1=

no rol

e–6=

very

gr

oup

patie

nts

patie

nts

asso

-

with

test

ing

impo

rtan

t ro

le ran

king

by

mea

n

ciat

ion

in

tent

ion

scor

e of

tot

al g

roup

m

ean

mea

n m

ean

mea

n m

ean

(SD)

(SD)

(SD)

(SD)

(SD)

F p

r

1) P

ossi

ble

prob

lem

s w

ith

insu

ranc

e co

mpa

nies

2.

6 (1

.7)

2.6

(1.8

) 2.

5 (1

.6)

2.9

(1.7

) 2.

0 (1

.6)

2.5

.06

-.04

2) I

do n

ot w

ant to

und

ergo

sur

gery

2.

4 (1

.6)

2.7

(1.6

) 2.

3 (1

.6)

2.8

(1.7

) 1.

8 (1

.1)

4.0A

.0

08

-.10

3) I

am a

frai

d fo

r ca

ncer

2.

2 (1

.3)

1.9

(1.0

) 2.

1 (1

.1)

2.3

(1.4

) 2.

2 (1

.3)

0.5

.71

.01

4) I

do n

ot w

ant re

gula

r m

edic

al c

heck

-ups

2.

0 (1

.3)

1.8

(1.0

) 1.

7 (1

.0)

2.2

(1.4

) 1.

7 (1

.1)

2.3

.08

-.22

**

5) It

cou

ld h

ave

impl

icat

ions

for

my

job

1.9

(1.3

) 1.

4 (.7)

2.

0 (1

.5)

2.1

(1.4

) 1.

8 (1

.2)

2.0

.11

-.01

6) It

cou

ld a

ffec

t m

y de

cisi

on to

have

chi

ldre

n 1.

9 (1

.5)

1.6

(1.2

) 1.

9 (1

.6)

1.8

(1.4

) 2.

1 (1

.7)

0.7

.58

.09

7) It

cos

ts m

oney

1.

8 (1

.2)

2.0

(1.3

) 1.

7 (1

.0)

1.9

(1.2

) 1.

6 (1

.2)

0.9

.42

.02

8) It

tak

es tim

e 1.

7 (1

.1)

1.7

(0.8

) 1.

4 (0

.6)

1.9

(1.2

) 1.

5 (0

.9)

2.6

.06

-.16

*

9) I

do n

ot h

ave

faith

in s

uch

a te

st

1.5

(1.0

) 1.

3 (0

.6)

1.4

(0.8

) 1.

6 (0

.9)

1.6

(1.2

) 0.

7 .5

5 -.04

10) It w

ill h

ave

impl

icat

ions

for

my

choi

ce

1.5

(1.1

) 1.

4 (1

.1)

1.3

(0.5

) 1.

6 (1

.1)

1.5

(1.2

) 0.

7 .5

4 -.01

of

a p

artn

er

* =

p < 0.

05, *

* =

p < 0

.01

A: P

ost Hoc

Bon

ferron

i te

st: co

ntro

l gr

oup

sign

ifica

ntly

low

er m

ean

than

pat

ient

ass

ocia

tion

(p =

0.0

05)


8

109

study is the revelation that the intention to be tested may not automatically translate into

actually undergoing testing. This observation was also reported by others.(278;290) In contrast to

our expectations, we did not find that significantly more familial TGCT patients would decide to

undergo testing than non-familial TGCT patients. Our finding is not in line with a study focusing

on cutaneous melanoma that reported that interest in genetic testing is greater in patients with

one or more family member with cutaneous melanoma than in patients who do not have family

members with cutaneous melanoma.(282)

It is striking that a large percentage of the male controls (62%) would request genetic testing

for cancer if such a test was available. Anticipated regret could be an important motive for

requesting a genetic test. Anticipated regret means that patients want to prevent feeling regret

in the future because of possibly making a wrong decision at this moment.(291) An even higher

percentage of healthy controls expressed interest in genetic testing in studies on prostate cancer.

One study reported that 92% of (healthy) men who visited a prostate screening clinic expressed

an interest in genetic testing for prostate cancer and 89% indicated that they would undergo

testing if available.(275) This interest of healthy men in an as yet non-existing test was not related

to their having a positive or negative family history of prostate cancer. In contrast, another study

showed that only 27% of healthy men indicated that they would undergo testing for a prostate-

specific antigene (PSA).(292) Approximately 45% of these men knew of the existence of this test

and 15% had already undergone a PSA test. The large contrast in these two studies’ findings

may be due to the fact that a large number of the men in the study of Watson et al.(292) were

well-informed about the test. A review study showed that the more information men have about

the limitations and/or the risk of PSA testing, the less likely they are to decide to be tested.(293)

In the current study, it was found that the lower the educational level, the greater the interest in

genetic testing. It might be that lower educational level is associated with less knowledge about

heredity which may result in a more positive attitude towards testing.

Our expectation that more familial TGCT patients would expect a positive test result than other

TGCT patients and than controls is supported. However, only 39% of the familial TGCT patients

expect a positive test result, which is surprising given that these men have a positive family

history. An explanation could be that although these men were informed in the clinic and the

brief information accompanying the request to participate in the study of the possibility of a

genetic defect, they were also informed that a positive family history is not proof of heredity.

Patients with a negative family history of TGCT are not very likely to have a hereditary form. It

is, therefore, remarkable that 48% of the sporadic TGCT patients assume that they may have a

genetic predisposition for cancer or that their chance is 50%. It is possible that people are not

able to think about risk in “chances”. Many people view their odds of becoming sick as 50-50:

you either get it or you don’t. This is referred to as binary thinking. Our results show that almost

all of the patients who think that they carry a genetic predisposition for TGCT or that their chance

is fifty-fifty, intend to undergo genetic testing.

In line with our expectations, this study showed that men who were not treated for cancer require

less information about their chance/risk on hereditary TGCT and its consequences than men who

were confronted with cancer and its consequent treatment. In contrast to our expectations,

familial patients do not differ on wish for genetic information from sporadic patients and patient

association members. The mean scores on information desire are high, which indicates that

patients are very interested in receiving information on genetic predisposition and available

options. Therefore, it is very important that both patients and controls are given adequate

information about the (possible) hereditary aspects of cancer and the implications of positive

and negative DNA test results, before genetic testing is offered, to prevent misconceptions about

possible genetic testing results. This information may result in less interest for genetic testing,

as was shown in the study from Watson et al.(292) but for TGCT this is not clear yet.

Relatively more important reasons for undergoing genetic testing were: “to support scientific

research”, “to gain more certainty about getting TGCT/cancer”, “to be more certain about my

children’s risk”, “to get regular medical check-ups” and “to take preventive measures”. This

is in line with earlier research in women with breast cancer.(271;278;286) Men in the current study

identified “at the insistence of a doctor”, “at the insistence of parents/siblings”, “to help decide

whether to have children” and “general future planning” as motivations that played a smaller

role which is conform a Dutch study on individuals at increased risk for breast/ovarian cancer

and colon cancer.(294)

Motives to undergo genetic testing differ between patients and controls. For patients, the

motivation: “to support scientific research” plays a significantly more important role than for

controls. It seems that patients have more an altruistic desire to aid genetic and oncology

research than controls. “To get regular medical check-ups” is another motivation that played

a more important role for patients than for controls. Patients diagnosed with TGCT can also

develop a malignancy in the second testicle as women with breast cancer in the other breast.

Medical check-ups may help TGCT patients feel more secure.

This study also examined barriers for undergoing genetic testing. It is remarkable that the mean

scores are low (under median score) suggesting that these motivations play a relatively small

role. Possible problems with insurance companies seemed to be the most important barrier

for the intention to undergo testing. Possible discrimination by insurance companies has been

identified as a concern by others as well.(285;295) The least important barriers were implications for

their choosing a partner and faith in the test. Time and money seem to also play a minor role.

In The Netherlands, the cost of genetic testing is covered by insurance companies, which may

explain why cost was not important in the decision-making process.

All the motivations for genetic testing were significantly related to the intention to undergo

genetic testing (Table 2). The motivations for testing that seemed to play a more important role


8

111

were more highly related to the intention to undergo testing (moderately strong relationships)

than the ones that played a less important role. Only two barriers to undergo genetic testing

are related to the intention to undergo testing (Table 3). “It takes time” and “I do not want

regular medical check-ups” are weakly correlated, meaning that these motivations probably do

not prevent men from genetic testing.

Our study had some limitations which have to be taken into account when interpreting the

results. Firstly, the choice to undergo genetic testing for TGCT was hypothetical. Men do not

have any information about or experience with a genetic test for TGCT. They do not have friends,

family members, or neighbors who have been tested for TGCT with whom they could exchange

information. In addition, it is unknown if/which preventive measures can be taken. This lack

of information/experience may influence the findings, as compared to a genetic test for breast

cancer, which is actually available and fairly common. Secondly, low numbers may have resulted

in a lack of power. The low numbers are the result of the rarity of TGCT. However, 127 TGCT

patients returned the questionnaire. This is the largest number of TGCT patients to have ever

been studied regarding their interest in and motivations for genetic testing, as far as the authors

know. This is a strength of the study. Additionally, the diversity of the studied groups and the

comparison with a healthy control group give this study strength.

When genetic testing for TGCT becomes possible, studies should closely look at the medical,

psychological, ethical and legal value of positive and negative DNA test results as well as the

outcome of monitoring. It will be interesting to examine the actual number of men who would

undergo a genetic test. Additionally, future studies should focus on psychosocial support in

genetic counseling, something the current study did not focus on. Questions that should be

examined include: is it possible to characterize participants/patients who could benefit from

psychosocial support, what kind of support would they require, and who should offer support?

Based on the answers to these questions and with help from this and other studies’ findings,

protocols could be developed regarding patient education and counseling. These protocols could

be used by a multidisciplinary team of doctors, nurses, psychologists and genetic counselors.

ConclusionThis is the second study to investigate the willingness, motivations and expectations of men

treated for TGCT and of men not confronted with cancer regarding genetic testing for TGCT.

Those more inclined to undergo genetic testing are men with lower education; men who have

cancer; men who suspect they have a genetic mutation or those who think the chance is fifty-

fifty; those who want more information about the possibility of carrying a gene mutation; and

men for whom a number of motivations play a greater role. Important motivations in deciding

to undergo genetic testing are: “to support scientific research”, “to gain more certainty on the

risk of getting TGCT/cancer”, “to be more certain about my children’s risk” and “to get medical

check-ups”. Possible insurance problems and not wanting to undergo surgery are relatively

important barriers.


9

113

Chapter 9

Summary, discussion and future perspectives

Summary, discussion and future perspectives114 Genetic Predisposition to Testicular Cancer

9

115

Summary

Insight into hereditary aspects of Testicular Germ Cell Tumours (TGCT) may lead to the

identification of individuals at increased risk for developing TGCT, increase our understanding

of the mutation pathways that lead to sporadic TGCT, and is expected to contribute to

improvement of TGCT diagnosis (e.g. genetic screening for men at risk) and treatment (e.g. gene

therapy). The current thesis focuses on genetic predisposition for TGCT

In the general introduction on TGCT described in Chapter 1, epidemiological, therapeutical and

clinical aspects of TGCT are discussed. The literature on the genetic aspects of TGCT is reviewed

in Chapter 2. Several observations have suggested that TGCT susceptibility genes exist and

are important in this disease. These include increased TGCT risks associated with a positive

family history, the increased incidence of bilaterality in familial cases and the ethnic and racial

differences that do not change with migration, statistical analysis of observed familial and

nonfamilial cases (e.g. segregation analyses) and possible associations with known hereditary

syndromes and constitutional chromosomal anomalies. In Chapter 3, the literature on these

syndromes and chromosomal anomalies is reviewed. Twenty five of these disorders have

been reported in patients who also developed seminomatous or nonseminomatous testicular

carcinoma. Although these malignancies were too rare to enable the detection of statistically

significant correlations between the chromosomal / hereditary disorder and the testicular cancer,

it was striking that many of the patients had other urogenital abnormalities in addition to their

TGCT. Urogenital abnormalities are caused by disrupted urogenital differentiation and the same

disturbance may have lead to testicular dysgenesis and / or testicular carcinoma. This theory

of shared genetic and environmental risk factors in testicular tissue differentiation has been

postulated by Skakkebaek and colleagues.(12) Tumour cytogenetics has also provided hints that

at least some of the genes involved in the hereditary disorders in question may be causally

related to TGCT as these genes have been mapped to regions that appear to be involved in the

development of sporadic TGCT. Molecular studies on candidate genes will be required to provide

definite answers.

In Chapter 4, a difference is described in the incidence of TGCT between the northern and

eastern part of the Netherlands. Within the Netherlands, which is a relatively small country,

geographic differences in incidence of TGCT exist, with a statistically significant highest incidence

in the northern part of the Netherlands. This geographic clustering of TGCT may be caused by

a stable founder population that likely share a relatively high frequency of genes from common

ancestors, including genes possibly related to TGCT. This founder population is suitable for

searching TGCT susceptibility genes

In search of these candidate genes, a study was undertaken to study the association between

histocompatibility antigens (HLA), in particular class II genes (DQB1, DRB1) and TGCT. This study


9

117

described in Chapter 5 used genotyping of microsatellite markers in order to confirm or refute

previously published associations. In 151 patients, along with parents or spouses, the HLA-

region (particularly class II) on chromosome 6p21 was genotyped for a set of 15 closely linked

microsatellite markers. In both patients and controls, strong linkage disequilibrium was observed

in the genotyped region indicating that similar haplotypes are likely to be identical by descent.

However, association analysis as well as the transmission disequilibrium test did not show

significant results. An alternative approach to linkage studies is searching for TGCT susceptibility

genes among unrelated TGCT cases in founder populations by means of association analyses on

a dense set of markers. These so-called linkage disequilibrium fine-mapping analyses are based

on the hypothesis that patients in founder populations inherited disease mutations from recent

and common ancestors. This analysis is called haplotype analysis (Haplotype Sharing Statistic,

HSS). Haplotype analysis did not show differences in haplotype sharing between patients and

controls. Therefore, this study did not confirm the previously reported association between HLA

class II genes and TGCT.

Another genomic region thought to harbour TGCT associated gene mutations was the Y

chromosome. Reduced fertility is associated with TGCT and reduced fertility and TGCT might

share genetic risk factors according to the testicular dysgenesis hypothesis (designed by

Skakkebaek et al).(12) Up to 8% of infertility and reduced fertility in the general male population

can be explained by the presence of constitutional deletions of part of the long arm of the Y

chromosome (Yq11), referred to as the azoospermia factor (AZF) region. In Chapter 6, a study

is presented that investigated the frequency of azoospermia factor (AZF) deletions in Dutch

patients with TGCT. In 112 patients with TGCT, screening for constitutional deletions in the AZF

region was performed by multiplex polymerase chain reaction (PCR) analysis in DNA extracted

from peripheral blood lymphocytes. A set of 24 primer pairs of which 20 primer pairs are

homologous to previously identified and mapped sequenced tag sites (STSs) within the AZF

region were used. No deletions in the Yq11 region were detected in any of the 112 patients. The

conclusion of this study is that large Y chromosome microdeletions in the AZF region are not a

major contributor to the development of TGCT and TGCT-associated reduced fertility.

Global efforts have been made to identify TGCT predisposing genes. Although TGCT families have

been reported in the literature, multigenerational pedigrees with several affected cases are rare

and this limits the power of linkage studies. In 1994, the International Testicular Cancer Linkage

Consortium (ITCLC) was formed with the aim to collect TGCT families from all over the world

and to perform genotyping studies. The ITCLC performed genotyping studies and their results

provided evidence for the location of a gene involved in TGCT and cryptorchism susceptibility as

the consortium found a HLOD score for such a gene of 2.01 on chromosome Xq27. In Chapter

7 a genotyping study aimed at this Xq27 region is presented. We used the Haplotype Sharing

Statistic (HSS), which was also used in the HLA screen as described in chapter 5. In 276 patients

and 169 unaffected first-degree male relatives, 12 microsatellite markers covering the candidate

region were genotyped and used to study possible association of TGCT with Xq27 both by

single locus association analysis as well as the HSS. We observed a suggestive association

between the subset of TGCT cases without a family history of TGCT or cryptorchism and marker

DXS1193 both by allelic association (p=0.014) and the HSS (p=0.09), both were not significant

after multiple testing correction. However, when taken all minor alleles at this marker together,

this marker remained significantly different between this subset of TGCT patients (13.9% had

minor allele) and controls (3.4%). An individual without cryptorchism carrying a minor allele

had a 4.7-fold risk to develop sporadic TGCT (99.57% CI 1.1-19.6) compared with carriers of the

major allele. This study could not confirm the previously observed linkage of TGCT to Xq27,

which was particularly observed in families with bilateral cases and cases with cryptorchism, but

we did observe an association between the subset of TGCT cases without a family history of

TGCT or cryptorchism and marker DXS1193. Multiple minor alleles at this locus had an increased

frequency among patients. This suggests that in our population several mutations of an Xq27-

linked gene have a moderate to large effect on TGCT development but not on cryptorchism.

Although there is currently no genetic test available for testing genetic predisposition to TGCT,

this may very well change in the future. In Chapter 8 a study was performed to examine interest

in and motivations for TGCT susceptibility testing if such a test would be available today,

by comparing the opinions of TGCT patients with controls. Four groups of respondents were

recruited. From the database consisting of all TGCT patients/survivors (n=702) treated at the

University Medical Center Groningen between 1977 and 2003, all TGCT patients with an affected

first or second degree relative (familial TGCT) and a randomly selected group of TGCT patients

without a family history (sporadic TGCT) were selected. The third group consisted of TGCT

patients recruited by the Dutch TGCT patients/survivors association (patient association group)

and the fourth of age-matched male patients visiting the hospital’s emergency room for a minor

trauma (controls). A letter with information about the objectives of the study and a questionnaire

was given to all participants. The results: 23/44 familial TGCT patients (response 52%), 27/45

randomly selected sporadic TGCT patients (60%), 77/116 patient association members (66%) and

42 control group men (100%) completed the questionnaire. More patients (79%) than controls

(62%) would undergo genetic testing (X2=4.72, p=0.03). More familial patients (40%) expected

that they would have a genetic mutation than sporadic patients (7.4%), patient association

members (6.6%), and controls (2.4%) (X2=45.4, p<.001). A significant effect of group was found

on 4 of 14 motivations for TGCT genetic testing: “To support scientific research” (F=4.1, p=.007),

“to gain more certainty about risk of getting TGCT/cancer” (F=4.6, p=.004), “to get regular

medical check-ups” (F=6.2, p=.001) and “because there is cancer in my family” (F=5.1, p=0.02).

These motivations played a smaller role for controls than for patients. “Possible insurance

problems” and “I do not want to be operated” are the most important reasons why men would

restrain from genetic testing. We concluded that more patients than controls would choose to

undergo genetic testing if such a test were available. Familial TGCT patients significantly more

often than controls assume that they carry a genetic predisposition. Some motivations play a


9

119

greater role for patients than for controls in their decision on genetic testing. The greater role a

motivation played the more inclined respondents were to undergo genetic testing

Discussion and future perspectivesThe incidence of TGCT has doubled over the last 40 years. This increase is unlikely to be caused

by genetic predisposition alone and environmental factors most likely play an important role

as well. Although TGCT is an infrequent disease and it is highly curable, it is still important

to unravel the environmental risk factors as well as genetic predisposition for TGCT. Early

identification of men with increased risk for inherited TGCT might lead to early detection and

improved treatment outcome. As mentioned in chapter 1, disseminated TGCT can be classified in

three prognostic groups (good, intermediate and poor prognosis). Survival of TGCT is dependent

on the prognostic group, with 5 years survival rates of 94% (good), 83% (intermediate) and

71% (poor) respectively.(296) In the current literature there are no data on the clinical behaviour

(or prognosis) of TGCT in patients with familial TGCT. Clinical observation does also not point to

a difference in survival. Presently, there is no reason to expect that familial TGCT has a different

outcome than sporadic TGCT. Thus identifying TGCT in genetically predisposed men at an earlier

stage, (i.e. in stage I or in the group with a good prognosis), is expected to improve survival.

Early detection could theoretically be reached by ultrasound of the testes, palpation (routine

testicular self examination) and/or by measuring tumour markers. Presently, however, there is

no proof that men who routinely examine their testes are more likely to detect earlier stage

tumours or improve their prospects for survival.(297) From a therapeutical point of view, treatment

of TGCT may be further optimized using new insights in the oncogenetic pathways involved and

possibly reduce the toxic side-effects of chemotherapy and radiotherapy. Identifying patients

at an earlier stage of dissimenation with favourable prognostic factors, the number of courses

of chemotherapy / radiotherapy can be reduced, resulting in less toxic side effects. Besides

this, the unravelling of the genetic pathway of TGCT may result in the opportunity to develop

gene therapy. There are some trials (in other types of cancer) with promising results. The

use of genetically modified autologous tumour cells appears to be a promising approach for

cancer therapy. Further studies are required to determine whether promising effects on immune

activation will result in an actual clinical benefit for patients.(298)

In chapter 6 we discussed the study of microdeletions in the AZF region of the Y chromosome in

TGCT patients. The AZF deletions are well known deletions of the Y chromosome and account for

8% of the male infertility. This region was studied in TGCT patients because they often present

with abnormal semen characteristics (reduced fertility). We did not find an association between

AZF deletions and TGCT in our study. However, recently a novel Y-chromosome 1.6-MB deletion

was described, named: gr/gr. This deletion was associated with spermatogenic failure as well.

The gr/gr deletion is much smaller than the deleted AZF region studied previously and removes

only part of the AZFc region, including copies of DAZ and a copy of CDY1, as well as other

transcription units (see Chapter 6). Nathason et al recently described an association between

these gr/gr deletions and TGCT. Familial TGCT patients had a threefold increased risk having

these gr/gr deletions.(258) This study focusing on a small part of the AZF region elucidated a small

deletion associated with moderately increased risk for TGCT. Therefore the Y chromosome, in

particular the AZFc region, seems still relevant for further research on genetic predisposition of

TGCT.

Although circumstantial evidence points to the existence of TGCT predisposition, no germ line

gene mutations have yet been identified that confer a high risk to develop TGCT. The question

arises whether such mutations exist at all or whether genetic predisposition for TGCT (as the

gr/gr deletions did) will turn out to consist of a range of mutations, each with only relatively

weak effects. One barrier to resolving this issue is the fact that large families with a strong family

history of TGCT are rare. The international testicular cancer linkage consortium (ITCLC) was

established to collect large number of TGCT families to perform linkage studies. In a previous

study, the ITCLC presented evidence for linkage of TGCT with Xq27 (in a study that was based on

99 pedigrees).(105) However, in their last study, in which 179 pedigrees were studied genome wide,

no regions showed a significant HLOD score of 2 or higher. The genomic regions: 2p23, 3p12,

3q26, 12p13-q21, 18q21-q23 and Xq27 (the Xq27 region was examined in a further 66 pedigrees

compared to their previous study) had a HLOD score between 1 and 2, which is not considered

significant, but neither provides evidence against a TGCT gene in this region.(262) Since the ITCLC

has the largest series of familial TGCT patients (459 pedigrees), and could not identify genomic

regions associated with high TGCT risk, the ITCLC suggests that the susceptibility to TGCT is

determined through multiple loci as opposed to a single locus. Possibly, very strong inherited

TGCT predisposition does not exist. Furthermore, linkage analysis in contrast to association

studies is not a powerful tool for gene detection when frequent genetic variants play a role.(299)

With 459 pedigrees, of which the ITCLC examined only 237, because the rest was considered

not well sampled, only relative risk of 5 or larger for carriers of a genetic variant (frequency

of 15%) can be detected. To detect a relative risk of 2, which is realistic in complex human

diseases, one would need to study at least 2000 pedigrees to have 80% power to detect such

a genetic variant. On the other hand, it cannot be excluded that (part of) the familial clusters

can be attributed to different, rare, strongly predisposing gene mutations present in subsets of

families, that, because of their low frequency in the population, may very well escape detection

in such relatively small linkage studies. From a theoretical point of view, it may be interesting

to perform linkage studies in a subset of familial TGCT patients with urogenital malformations

(like hypospadia, cryptorchism, bilateral disease etc), because in this subgroup of patients

(identical phenotypes), it is very well possible that they share a causative gene which results in

urogenital malformations as well as TGCT (Skakkebaek model). The same approach has been

used for Xq27: In the first study, significant results in a subgroup of familial patients with TGCT:

cryptorchism and/or one case of bilateral disease in the family(12;105)


9

121

Recently the genetic defect in mouse strain 129.MOLF-Chr19 chromosome substitution strain,

known to develop TGCTs at a high frequency (70-80%), was reported.(300) A germ line mutation

in Dead End Gene (DND1) was found to cause this high tumour risk (and some testicular and

spermatogenic abnormalities).(258) Very little is known about the human homologue of DND1

and it is therefore worthwhile to study its possible association with human TGCT. Therefore the

University Medical Center Groningen (UMCG) is currently screening the gene for mutations in a

series of familial and sporadic TGCT patients and controls.

The collection of more families with TGCT will facilitate genetic studies. Some hospitals

systematically collect DNA from their cancer patients in a prospective way in order to expand

the material for genetic studies. The deCODE initiative (www.decode.com) takes this approach

a large step further by linking genealogical data with disease status and DNA markers in the

Icelandic population. Such approaches are to be encouraged from a pure scientific point of

view; however, privacy and other ethical and legal issues involved in these approaches need to

be addressed very thoroughly. International collaboration will be facilitated if DNA and patient

data collecting will be performed according to (to be developed) international standards. The

HapMap project, a partnership of scientists and funding agencies all over the world to develop

a public resource that will help researchers find genes associated with human disease, is a good

example (www.hapmap.org).

Not only study populations are expanding through collaborative efforts, statistical techniques

and insight in cancer biology and its resulting molecular tools evolve as well. Recently, the scope

of genetic study of TGCT has been extended to include the role of naturally occurring micro

RNAs (miRNAs). Normally miRNAs function as regulators of genetic pathways by manipulating

translational regulation. Some of these miRNAs (miRNA-372 and 373) were shown to allow

tumorigenic growth. As they have also been observed to be expressed in human seminomas and

non-seminomas, but not in normal testicular tissue, it has been suggested that these miRNAs

may represent a new class of oncogenes involved in TGCT development. (18) Time will tell whether

miRNAs play a role in (testicular) cancer predisposition

Should future studies identify clinically important TGCT predisposing mutations, then testing for

these mutations might be welcomed by a substantial subset of TGCT patients and families as

has been the case for many families with common hereditary tumour syndromes. Research on

early detection techniques for TGCT in high risk men will be necessary. Although many studies

have looked into the psychosocial aspects and requirements for genetic testing programmes in

these syndromes, very little is known with respect to testing in the setting of TGCT. Research

on early detection techniques for TGCT in high risk men will be necessary. As in any other new

genetic testing programme, testing would need to be performed in a research setting first, where

medical as well as psychosocial issues would need to be carefully monitored.


10

123

Chapter 10Nederlandse samenvatting, conclusies en toekomstperspectieven

Nederlandse samenvatting, conclusies en toekomstperspectieven124 Genetic Predisposition to Testicular Cancer

10

125

Nederlandse samenvatting, conclusies en toekomstperspectieven

SamenvattingKanker ontstaat als gevolg van een opeenstapeling van beschadigingen (mutaties) van stukjes

erfelijke aanleg (genen). Deze beschadigingen kunnen zowel erfelijk als niet-erfelijk zijn. Bij

de meeste gevallen van kanker spelen individuele erfelijke genbeschadigingen geen sterke

rol bij het ontstaan van de gezwellen. Als echter erfelijke beschadigingen wel sterk bijdragen

aan het ontstaan van kanker dan spreken we van erfelijk kanker. Ouders met de aanleg voor

erfelijke kanker kunnen die aanleg aan hun kinderen doorgeven. Als deze aanleg bij iemand

aangetoond kan worden nog voordat kanker is ontstaan dan kunnen (afhankelijk van de soort

kanker) soms preventieve maatregelen worden getroffen. Van veel soorten kanker zijn inmiddels

erfelijke vormen beschreven en raken steeds meer details bekend over de aard en de rol van

de betrokken genmutaties. De genen die van belang zijn bij het ontstaan van erfelijke vormen

van kanker blijken vaak ook een rol te spelen bij het ontstaan van niet-erfelijke vormen van

kanker. In dat laatste geval gaat het dan niet om erfelijke maar om niet-erfelijke beschadigingen

van die genen. Dit proefschrift richt zich op het zoeken naar erfelijke factoren die de kans op

zaadbalkanker vergroten. Inzicht in de erfelijke aspecten van zaadbalkanker ( testis carcinoom

(TC)) zou kunnen leiden tot het identificeren van mannen met een erfelijk verhoogd risico op het

krijgen van TC (diagnostiek, evt. genetisch screenen) en tot het vergroten van de kennis over

genmutaties die leiden tot niet-erfelijk (sporadisch) TC.

In een algemene introductie worden in Hoofdstuk 1 epidemiologische, therapeutische en

klinische aspecten van TC beschreven. In Hoofdstuk 2 wordt een overzicht gegeven van de

huidige literatuur over genetische aspecten van TC. Verscheidene observaties suggereren dat

er een genetische predispositie voor TC bestaat en dat de betrokken genen belangrijk zijn bij

het ontstaan c.q. het beloop van deze ziekte. Deze observaties zijn: 1) verhoogd risico op het

krijgen van TC bij familiair voorkomen van TC, 2) verhoogde incidentie van dubbelzijdig TC in

families met TC, 3) de etnische en raciale verschillen in optreden van TC, die niet veranderen

na migratie, 4) resultaten van statistische analyses van familiare en niet-familiaire patiënten

(bijvoorbeeld segregatieanalyse, dit is een statistische techniek die het patroon van het

voorkomen van een aandoening in families vergelijkt met bekende overervingsmodellen) en 5)

mogelijke associatie van TC met bekende erfelijke syndromen en constitutionele chromosomale

afwijkingen. In Hoofdstuk 3 wordt de literatuur over de associatie tussen TC en deze syndromen

en chromosomale afwijkingen samengevat. Vijfentwintig syndromale afwijkingen worden

beschreven waarbij de betreffende patiënten tevens TC hebben ontwikkeld. Aangezien het om

zeer zeldzame combinaties gaat van TC met erfelijke syndromen en constitutionele chromosomale

afwijkingen konden hierbij geen statistische relaties worden aangetoond. Opvallend was dat

veel patiënten naast TC tevens urogenitale afwijkingen hadden (stoornis in de aanleg van

blaas, genitalia etc., bijvoorbeeld cryptorchisme, dit is niet-ingedaalde testikel). Urogenitale


10

127

afwijkingen kunnen worden veroorzaakt door een gestoorde urogenitale differentiatie (stoornis

in celuitgroei van bijvoorbeeld organen) en dit zou tevens kunnen leiden tot testiculaire

dysgenesie (niet juiste aanleg) en / of TC. Deze theorie van gedeelde risicofactoren, d.w.z.

genetische en omgevingsrisicofactoren die een rol spelen in de differentiatie van testisweefsel is

beschreven door Skakkebaek et al.(12) De tumorcytogenetische studies (chromosomenonderzoek

van gezwellen) suggereren dat bepaalde genen, betrokken bij de betreffende erfelijke afwijking,

ook gerelateerd zouden kunnen zijn aan TC, aangezien deze genen zich in chromosoomregio’s

bevinden die mogelijk betrokken zijn bij het ontstaan van (sporadisch) TC. Moleculaire studies

naar kandidaat-genen zijn nodig om hier definitieve antwoorden op te geven.

In Hoofdstuk 4 wordt een verschil beschreven in incidentie van TC bij patiënten woonachtig in

Noord- en Oost-Nederland. In een klein land als Nederland bestaan dus geografische verschillen in

incidentie van TC, met een statistisch significante hoger incidentie in het Noorden van Nederland.

Deze geografische clustering van TC zou kunnen worden veroorzaakt door zogenaamde stabiele

founderpopulaties in Noord-Nederland (= “bron populaties” of oorspronkelijke populatie).

Mensen uit een dergelijke founderpopulatie delen waarschijnlijk relatief veel genen van een

klein aantal gemeenschappelijke voorouders, en mogelijk dus ook genen gerelateerd aan TC.

Onderzoek naar TC predisponerende genen in dergelijke founderpopulaties is vanuit theoretisch

oogpunt dan ook aantrekkelijk.

Om kandidaat-genen te onderzoeken werd een studie verricht naar de associatie tussen histo-

compatibiliteitsantigenen (HLA, Humane Leukocyten Antigenen), in het bijzonder klasse II genen

(DQB1, DRB1), en TC. Deze studie, beschreven in Hoofdstuk 5, is een genotyperingsstudie met

microsatellietmerkers (merkers voor korte stukjes DNA) om een eerdere gepubliceerde associatie

te bevestigen dan wel te verwerpen. In het bloed van 151 patiënten samen met de ouders of

eventueel de echtgenote en kinderen vond genotypering plaats van de HLA-regio (klasse II) op

chromosoom 6p21 met behulp van 15 aan elkaar gelinkte microsatellietmerkers. Bij zowel de

patiënten als de controlepersonen was er sprake van sterk linkage disequilibrium (allelen op

verschillende loci erven niet onafhankelijk van elkaar over) in de betreffende regio, hetgeen

betekent dat haplotypes (combinatie van allelen voor meerdere merkers op 1 chromosoom)

gelijk zijn door gemeenschappelijke afkomst (afstamming). Een alternatieve methode voor

koppelingsstudies (linkage-analyse) is het zoeken naar TC predisponerende genen in niet-

verwante TC patiënten in een founderpopulatie met behulp van associatieanalyse op een

set dicht bij elkaar liggende merkers. Deze linkage disequilibrium fine-mapping analyses zijn

gebaseerd op de hypothese dat patiënten in founderpopulaties ziektemutaties hebben geërfd

van gemeenschappelijke voorouders. Deze analyse wordt haplotype-analyse genoemd. Eén zo’n

methode is de Haplotype Sharing Statistic (HSS). In de HLA genotyperingsstudie liet haplotype-

analyse geen verschil zien tussen patiënten en controles, net als klassieke associatieanalyse

en de transmissie disequilibrium test. Deze studie kon daarom de eerder genoemde associatie

tussen HLA klasse II genen en TC niet bevestigen.

Het Y-chromosoom is een andere regio van het genoom waar TC gerelateerde genmutaties zich

zouden kunnen bevinden. Afgenomen fertiliteit is geassocieerd met TC en afgenomen fertiliteit

en TC zouden dus gemeenschappelijke onderliggende genetische risicofactoren kunnen hebben

volgens het eerder genoemde testiculaire dysgenesiemodel (ontworpen door Skakkebaek(12)).

In de algemene populatie kan 8% van de infertiliteit en afgenomen fertiliteit worden verklaard

door de aanwezigheid van constitutionele deleties van een deel van de lange arm van het Y

chromosoom (Yq11), ook wel de azoospermie (AZF) regio genoemd. In Hoofdstuk 6 wordt een

studie beschreven die de prevalentie analyseert van AZF-deleties bij Nederlandse patiënten met

TC. Bij 112 patiënten met TC werd DNA uit bloed geanalyseerd op constitutionele deleties in

de AZF-regio met behulp van multiplex polymerase ketting reacties. Er werd gebruik gemaakt

van 24 primerparen, waarvan 20 primerparen homoloog waren aan eerder geïdentificeerde en

gemapte sequence tag sites (STS) binnen de AZF-regio. Er werden bij deze 112 patiënten geen

deleties gevonden in de Yq11 regio. De conclusie van deze studie is dat (grote) Y-chromosoom

microdeleties in de AZF-regio niet geassocieerd lijken met TC en met TC geassocieerde

afgenomen fertiliteit.

Wereldwijd wordt een zoektocht naar TC predisponerende genen verricht. Ondanks dat er TC

families in de literatuur worden beschreven, zijn families met meerdere aangedane familieleden

schaars en dit belemmert linkage-analyse. In 1994 is het International Testicular Cancer Linkage

Consortium (ITCLC) opgericht met het doel TC families over de gehele wereld te verzamelen en

hierbij genotyperingsstudies te verrichten. Het ITCLC heeft deze studies verricht en de resultaten

toonden aanwijzingen voor een gen gelokaliseerd in de regio Xq27, dat geassocieerd leek te

zijn met TC en cryptorchisme. In Hoofdstuk 7 wordt een genotyperingsstudie gepresenteerd in

onze eigen TC families gericht op deze regio Xq27. Wederom werd HSS gebruikt zoals eerder

beschreven bij de HLA studie (Hoofdstuk 5). Bij 276 patiënten en 169 niet aangedane, mannelijke

eerstelijns verwanten, werd een genotyperingsstudie verricht met 12 microsatellietmerkers

(die de gehele kandidaat-regio op Xq27 dekken) om een mogelijke associatie van TC met de

betreffende Xq27 regio te analyseren(single locus associatieanalyse en HSS). Een suggestieve

associatie werd gevonden tussen een subgroep TC patiënten zonder aangedane familieleden

of zonder cryptorchisme en merker DXS1193, zowel bij allelische associatie (p=0.014) en HSS

(p=0.09), die beide niet significant waren na correctie voor multipele testen. Echter wanneer alle

laagfrequente allelen ter plaatse van deze merker werden samengevoegd bleek deze merker wel

significant verschillend te zijn tussen deze subgroep TC patiënten (13.9% had een laagfrequent

allel) en de controles (3.4%). Een individu zonder cryptorchisme die drager is van een

laagfrequent allel, had een 4.7 keer verhoogd risico om (sporadisch) TC te ontwikkelen (99.57%

betrouwbaarheidsgebied 1.1-19.6) vergeleken met dragers van het frequente allel. Deze studie

kon de eerder gevonden associatie tussen de Xq27 regio en TC patiënten met dubbelzijdig TC

en TC patiënten met cryptorchisme niet bevestigen. We vonden dus echter wel een associatie

tussen een subgroep TC patiënten zonder aangedane familieleden of zonder cryptorchisme en

marker DXS1193. Dit resultaat suggereert dat in onze populatie mutaties van een aan Xq27


10

129

gelinkt gen een middelmatig tot groot effect hebben op het ontwikkelen van TC maar niet op

het ontwikkelen van cryptorchisme.

In de toekomst zou een genetische test beschikbaar kunnen komen om genetische predispositie

voor TC te testen. In Hoofdstuk 8 wordt een studie gepresenteerd die de interesse in en motivatie

voor genetisch testen onderzoekt van TC patiënten en controles. Vier groepen respondenten

werden benaderd. Uit alle TC patiënten (n=702), behandeld in het Universitair Medisch Centrum

Groningen tussen 1977 en 2003, werden alle TC patiënten geselecteerd met een aangedaan

eerste of tweede graads familielid (familiair TC; n=44) en een even grote, willekeurige groep TC

patiënten zonder aangedane familieleden (sporadisch). De derde groep bestond uit TC patiënten

van de Nederlandse TC patiëntenvereniging (stichting De Kernzaak) en de vierde groep bestond

uit een aan leeftijd aangepaste controlegroep van mannen die de spoedopvang (EHBO) van

een ziekenhuis bezochten vanwege een klein trauma (verzwikte enkel etc.). De mannen kregen

een informatiebrief over het onderzoek en een vragenlijst. De resultaten toonden aan dat 23/44

familiaire TC patiënten (respons 52%), 27/45 sporadische TC patiënten (60%), 77/116 patiënten

van de patiëntenvereniging (66%) en alle 42 controlemannen de vragenlijst hadden ingevuld.

Meer patiënten (79%) dan controles (62%) zouden overgaan tot genetisch testen (X2=4.72,

p=0.03). Meer familiaire patiënten (40%) dan sporadische TC patiënten (7.4%), patiënten van

de patiëntenvereniging (6.6%) en controles(2.4%) (X2=45.4, p<0.001) veronderstelden dat zij

een genetische predispositie hadden voor TC. Een significant verschil tussen de groepen werd

gevonden bij 4 van de 14 motivaties die een rol kunnen spelen bij mensen in hun beslissing

tot genetisch testen. “Om wetenschappelijk onderzoek verder te helpen (F=4.1, p=.007), “Om

meer zekerheid te verkrijgen over mijn kans op TC/kanker” (F=4.6, p=.004), “Om regelmatig

door een arts te worden gecontroleerd” (F=6.2, p=.001) en “Omdat er kanker voorkomt in

de familie” (F=5.1, p=0.02) spelen een kleinere rol voor de controles dan voor de patiënten.

“Mogelijke problemen met de verzekeringsmaatschappij” en “Ik heb geen zin in operaties” zijn

de voornaamste redenen waarom mannen zouden afzien van genetisch testen. Geconcludeerd

kan worden dat significant meer patiënten dan controle mannen zouden overgaan tot genetisch

testen wanneer dit in de toekomst mogelijk zou zijn. Familiaire TC patiënten veronderstellen

(zoals verwacht) significant vaker dat zij drager zijn van een genetische predispositie. Bepaalde

motivaties spelen een belangrijkere rol voor patiënten dan voor controles. Hoe belangrijker de

motivatie, des te meer geneigd zijn de respondenten om over te gaan tot genetisch testen.

Conclusies en toekomstperspectievenDe incidentie van TC is de afgelopen 40 jaren verdubbeld en neemt nog steeds toe. Deze

toename komt waarschijnlijk niet alleen door een genetische predispositie; omgevings factoren

lijken ook een belangrijke rol te spelen. Ondanks dat TC een zeldzame en goed te genezen ziekte

is, is het toch belangrijk de risico verhogende omgevingsfactoren en de genetische predispositie

voor TC op te helderen. Vroege herkenning van mannen met een verhoogd risico op erfelijke

TC, zou kunnen leiden tot vroege detectie en hierdoor tot een beter behandelbare ziekte en

een betere overleving. Zoals vermeld in Hoofdstuk 1 kan uitgezaaide TC worden onderverdeeld

in drie prognostische groepen (goed, middelmatig en slecht). Overleving is afhankelijk van

de prognostische groep, waarbij de 5 jaars overleving als volgt is: goede prognose: 94%,

middelmatige prognose 83% en de slechte prognose 71%.(296) In de huidge literatuur zijn

geen data bekend over het klinische gedrag cq prognose van specifiek familiair TC. Klinische

observatie toont in ieder geval geen verschil in overleving. Derhalve is er geen aanwijzing om

aan te nemen dat familiair TC zich klinisch anders zou gedragen dan sporadisch TC. Wanneer

bij een man een genetische predispositie voor TC kan worden vastgesteld en hierdoor TC in

een vroeg ziektestadium kan worden opgespoord (stadium 1 of goede prognose groep), dan

zal daarom zoals verwacht ook de overleving van deze patiënt verbeteren. Vroege detectie

van TC zou theoretisch kunnen worden bereikt door palpatie van de testikel (routinematig

zelfonderzoek), echografie van de testikels, en/of door het meten van tumormerkstoffen (zie

Hoofdstuk 1). Naar dit onderwerp is slechts zeer beperkt onderzoek verricht. Tot op heden is er

geen bewijs dat bijvoorbeeld zelfonderzoek van de testikels zal leiden tot het vaststellen van TC

in een vroeg stadium en hiermee tot een betere prognose voor de patiënt.(297)

Theoretisch gezien zou de behandeling van TC verder kunnen worden geoptimaliseerd met behulp

van nieuwe inzichten in de oncogenetische aspecten van TC en mogelijk door het reduceren van

de toxische bijwerkingen van chemotherapie en radiotherapie. Daarnaast zou het ontrafelen van

de genetische aspecten van TC kunnen leiden tot het ontwikkelen van gentherapie. Bij andere

vormen van kanker zijn hiermee al bemoedigende resultaten bereikt. Het gebruik van genetisch

gemodificeerde autologe tumorcellen lijkt een veel belovende behandelingsstrategie te worden.

Verder onderzoek zal moeten bepalen of het veelbelovende effect van immuunactivatie zal

leiden tot een klinisch voordeel voor de patiënt.(298)

In Hoofdstuk 6 is het onderzoek besproken naar microdeleties in de AZF regio op het Y-

chromosoom bij TC patiënten. De AZF deleties zijn bekende deleties van het Y-chromosoom en

zijn verantwoordelijk voor ongeveer 8% van de mannelijke infertiliteit (afgenomen fertiliteit).

Deze regio is bestudeerd in TC patiënten aangezien deze mannen ook vaak minder fertiel zijn

(abnormale semen karakteristieken). Een associatie tussen de AZF-deleties en TC werd niet

gevonden in onze studie. Recent is een nieuwe deletie op het Y-chromosoom beschreven (1.6-

Mb), genoemd: gr/gr. Deze deletie is ook geassocieerd met verminderde fertiliteit. De gr/gr

deletie is veel kleiner dan de microdeleties die bestudeerd werden in de AZF regio en door deze

deletie is slechts een klein gedeelte van de AZFc-regio verwijderd, onder andere kopieën van

DAZ en een kopie van CDY1, evenals andere transcriptie-units (zie Hoofdstuk 6). Nathason et al.

hebben recent een associatie beschreven tussen deze gr/gr deletie en TC. Patiënten met familiair

TC hebben een 3x verhoogd voorkomen van de gr/gr deletie.(258) Deze studie, gericht op een erg

klein gedeelte van de AZF regio toonde een deletie geassocieerd met een gemiddeld verhoogd

risico op TC. Hiermee is het Y-chromosoom, in het bijzonder de AZFc-regio, nog steeds relevant

voor verder onderzoek naar de genetische predispositie van TC.


10

131

Hoewel er indirect bewijs bestaat voor een genetische predispositie voor TC, is de kiemlijn-

genmutatie die een hoge kans op het ontstaan van TC veroorzaakt nog niet gevonden. De

vraag doet zich voor of zo’n genmutatie wel bestaat of dat de genetische predispositie voor

TC (zoals de gr/gr deletie) bestaat uit een serie mutaties met elk slechts een zwak effect.

Een van de beperkingen van de mogelijkheden om deze theorie te staven is het feit dat er

slechts zeer weinig families zijn met meerdere familieleden met TC. Het International Testicular

Cancer Linkage Consortium (ITCLC) is opgericht om grote aantallen TC families te verzamelen

en hiermee linkage-studies te verrichten. In een recente studie toonde het ITCLC bewijs voor

linkage met chromosoom Xq27 (gebaseerd op 99 families).(105) Echter in de laatste studie door

het ITCLC, waarbij 179 families met merkers over het gehele genoom werden bestudeerd, werd

geen betekenisvolle regio op het genoom gevonden. De regio’s: 2p23, 3p12, 3q26, 12p13-q21,

18q21-q23 en Xq27 (Xq27 regio werd bij nog eens 66 families onderzocht) hadden een HLOD

score tussen de 1 en 2 (niet significant) en vormen dus geen bewijs voor een genoomregio

geassocieerd met TC. Het is overigens ook geen bewijs dat er in deze regio geen (zeldzaam)

predisponerend gen ligt voor TC.(262) Aangezien het ITCLC over het grootste aantal familiaire

TC patiënten beschikt (459 families) en zij geen genomische regio kunnen aantonen die

geassocieerd is met een verhoogd risico op TC, suggereert de ITCLC dat de predispositie voor

TC wordt bepaald door multipele loci (polygenen) in plaatst van één enkele locus (monogeen).

Het ITLC concludeert met andere woorden dat er waarschijnlijk geen sterke erfelijk monogeen

overervende vorm van TC bestaat. Linkage studies zijn minder krachtig in het detecteren van

een mutant gen dan associatie-studies wanneer frequente genetische varianten een rol spelen

(linkage studie heeft dan minder power en er is dus een veel grotere sample size nodig).(299)

Met 459 families, waarvan de ITCLC er slechts 237 analyseerde (de rest was niet goed in kaart

gebracht) kan slechts een relatief risico van 5 of hoger voor dragers van een genetische variant

(frequentie 15%) worden gedetecteerd. Om een relatief risico van 2 te detecteren (hetgeen

realistisch is in complexe humane ziekten) zijn er tenminste 2000 families nodig om met een

power van 80% een genetische variant te kunnen detecteren. Het kan echter niet worden

uitgesloten dat (gedeelten van) families met TC het gevolg zijn van verschillende, zeldzame,

sterk predisponerende genmutaties, die, vanwege de zeldzaamheid in de populatie, niet worden

gedetecteerd in de relatief kleine linkage-studies. Theoretisch gezien zou het interessant zijn

om linkage-studies te verrichten bij een subgroep van familiaire TC patiënten met urogenitale

afwijkingen (als hypospadie, cryptorchisme, bilateraal TC etc.), aangezien het in deze subgroep

(identieke fenotypen) goed mogelijk is dat zij genen delen die leiden tot urogenitale afwijkingen

en tot TC (Skakkebaek model). Dezelfde benadering is destijds toegepast voor Xq27: In de

eerste studie(105) werden significante resultaten gevonden voor een mogelijk gen in regio Xq27

in een subgroep van familiaire patiënten met TC, deze patiënten hadden allemaal minimaal 1

familielid met dubbelzijdig TC. Daarnaast leek Xq27 tevens geassocieerd met cryptorchisme. (12;105) Het probleem met subgroepanalyses is dat, wanneer je een groot aantal verschillende

subgroepen gaat analyseren, de kans dat er bij toeval één een significant verschil laat zien

aanzienlijk toeneemt. Hiervoor moet statistisch gecorrigeerd worden (multipele-testen-correctie),

wat tot gevolg heeft dat een verschil tussen subgroepen groter moet zijn om significant te zijn

naarmate meer subgroepen worden geanalyseerd. Het is dus belangrijk om vooraf te bepalen

welke subgroepen biologisch gezien relevant zijn om te testen. Welke subgroepen je wilt

analyseren, zal moeten afhangen van de functie van het gen dat je onderzoekt. In het geval

van een analyse van het gehele chromosoom is dat niet mogelijk en zul je dus voor een groot

aantal testen moeten corrigeren wat het onderscheidende vermogen van de analyse niet ten

goede komt.

Recent is het genetische defect in een muizenstam 129.MOLF-Chr19 chromosoom substitutiestam,

die bekend staat om de hoge frequentie (70-80%) van TC ontwikkeling, gerapporteerd.(300) Een

kiemlijn mutatie in het Dead End Gene (DND1) bleek verantwoordelijk voor dit hoge risico op

TC (en ook op semenafwijkingen).(258) Er is erg weinig bekend over het humane homoloog van

DND1 en dit maakt het de moeite waard om een mogelijke associatie tussen DND1 en humaan TC

te bestuderen. In het Universitair Medisch Centrum Groningen (UMCG) wordt momenteel in een

serie familiaire en sporadische TC patiënten en controles het DND1 gen gescreend op mutaties.

Het verzamelen van meer TC families zal genetisch onderzoek naar TC vergemakkelijken. Er

zijn ziekenhuizen die systematisch DNA verzamelen (prospectief) om zo materiaal in handen

te krijgen voor genetische studies. Het deCODE initiatief (www.decode.com) gaat nog een stap

verder en verbindt in de populatie op IJsland de genealogische data met ziekte-status en DNA

merkers. Wetenschappelijk gezien moeten deze projecten zeer sterk worden aangemoedigd,

echter privacy, ethische en wettelijke aspecten moeten hierbij niet uit het oog worden verloren.

Internationale samenwerking zal worden vergemakkelijkt wanneer de verzameling van DNA en

patiëntengegevens volgens (nog te ontwikkelen) internationale standaarden wordt verricht. Het

HapMap project, (een samenwerkingsverband van wetenschappers over de gehele wereld dat

een openbare verzameling gegevens beschikbaar maakt, die behulpzaam zijn bij het vinden van

genen geassocieerd met menselijke ziekten), is een goed voorbeeld (www.hapmap.org).

Niet alleen studiepopulaties zullen zich door samenwerking uitbreiden maar ook de statistische

technieken, de inzichten in kankerbiologie en de hieruit voortvloeiende moleculaire technieken

zullen zich verder ontwikkelen. Recent is het genetisch onderzoek naar TC uitgebreid in de

richting van de rol van het, in het genoom natuurlijk voorkomende, micro RNA (miRNA). Normaal

gesproken functioneert miRNA als een regulator van genetische wegen door de translatieregulatie

te manipuleren. Enkele van deze miRNAs (miRNA-372 and 373) blijken oncogeen te zijn. Deze

miRNAs zijn ook al aangetoond in humane seminomen en non-seminomen, maar niet in het

normaal testisweefsel. Dit wekt de suggestie dat deze miRNAs een nieuw soort oncogenen zijn

die betrokken lijken te zijn bij TC ontwikkeling.(18) De tijd zal leren of miRNAs een rol spelen in

de predispositie voor TC.


10

133

Wanneer in de toekomst klinisch relevante TC predisponerende mutaties worden geïdentificeerd,

dan zal waarschijnlijk een groot aantal TC patiënten (en families) zich hierop laten testen,

zoals thans ook het geval is voor families met bekende erfelijke tumorsyndromen (bijvoorbeeld

erfelijke borstkanker en erfelijke darmkanker). Alhoewel veel studies de psychosociale aspecten

van genetische testen (en naar de benodigdheden / voorzorgsmaatregelen hiervoor) bij de al

bekende tumorsyndromen hebben onderzocht, is hierover erg weinig bekend bij TC. Onderzoek

naar vroege opsporingstechnieken bij mannen met een hoog risico op TC blijft nodig. Zoals bij elk

ander nieuw genetisch testprogramma, moet het genetisch testen eerst in een onderzoekssetting

plaatsvinden zodat medische en psychologische kwesties goed bestudeerd kunnen worden.

Genetic Predisposition to Testicular Cancer 135

Reference List

Reference List136 Genetic Predisposition to Testicular Cancer 137

Reference List

(1) Cancer in the Netherlands,trends,prognoses and implications for call of health. 2004

www.kwfkankerbestrijding.nl/content/documents/SCK_rapport_Kanker_in_Nederland.pdf

(2) Albers P, Albrecht W, Algaba F, Bokemeyer C, Cohn-Cedermark G, Horwich A et al.

Guidelines on testicular cancer. Eur Urol 2005; 48(6):885-894.

(3) Schmoll HJ, Souchon R, Krege S, Albers P, Beyer J, Kollmannsberger C et al. European

consensus on diagnosis and treatment of germ cell cancer: a report of the European

Germ Cell Cancer Consensus Group (EGCCCG). Ann Oncol 2004; 15(9):1377-1399.

(4) International Germ Cell Consensus Classification: a prognostic factor-based staging

system for metastatic germ cell cancers. International Germ Cell Cancer Collaborative

Group. J Clin Oncol 1997; 15(2):594-603.

(5) Abouassaly R, Klein EA, Raghavan D. Complications of surgery and chemotherapy for

testicular cancer. Urol Oncol 2005; 23(6):447-455.

(6) Meinardi MT, Gietema JA, van der Graaf WT, van Veldhuisen DJ, Runne MA, Sluiter WJ et

al. Cardiovascular morbidity in long-term survivors of metastatic testicular cancer. J Clin

Oncol 2000; 18(8):1725-1732.

(7) Nuver J, Smit AJ, Wolffenbuttel BH, Sluiter WJ, Hoekstra HJ, Sleijfer DT et al.

The metabolic syndrome and disturbances in hormone levels in long-term survivors of

disseminated testicular cancer. J Clin Oncol 2005; 23(16):3718-3725.

(8) Nuver J, Smit AJ, Sleijfer DT, van Gessel AI, van Roon AM, van der Meer J et al. Left

ventricular and cardiac autonomic function in survivors of testicular cancer. Eur J Clin

Invest 2005; 35(2):99-103.

(9) Fleer J. Quality of life of testicular cancer survivors.

Thesis: http://irs.ub.rug.nl/ppn/291569412. University of Groningen, The Netherlands 2006.

(10) Hoei-Hansen CE, Rajpert-De Meyts E, Daugaard G, Skakkebaek NE. Carcinoma in situ

testis, the progenitor of testicular germ cell tumours: a clinical review. Ann Oncol 2005;

16(6):863-868.

(11) Oosterhuis JW, Looijenga LH. Testicular germ-cell tumours in a broader perspective. Nat

Rev Cancer 2005; 5(3):210-222.

(12) Skakkebaek NE, Rajpert-De Meyts E, Main KM. Testicular dysgenesis syndrome: an

increasingly common developmental disorder with environmental aspects. Hum Reprod

2001; 16(5):972-978.

(13) von Eyben FE. Chromosomes, genes, and development of testicular germ cell tumors.

Cancer Genet Cytogenet 2004; 151(2):93-138.

(14) Reuter VE. Origins and molecular biology of testicular germ cell tumors. Mod Pathol

2005; 18 Suppl 2:S51-S60.

(15) Dearnaley D, Huddart R, Horwich A. Regular review: Managing testicular cancer. BMJ

2001; 322(7302):1583-1588.


(16) Korkola JE, Houldsworth J, Chadalavada RS, Olshen AB, Dobrzynski D, Reuter VE et al.

Down-regulation of stem cell genes, including those in a 200-kb gene cluster at

12p13.31, is associated with in vivo differentiation of human male germ cell tumors.

Cancer Res 2006; 66(2):820-827.

(17) McIntyre A, Summersgill B, Grygalewicz B, Gillis AJ, Stoop J, van Gurp RJ et al.

Amplification and overexpression of the KIT gene is associated with progression in the

seminoma subtype of testicular germ cell tumors of adolescents and adults. Cancer Res

2005; 65(18):8085-8089.

(18) Voorhoeve PM, le Sage C, Schrier M, Gillis AJ, Stoop H, Nagel R et al. A genetic screen

implicates miRNA-372 and miRNA-373 as oncogenes in testicular germ cell tumors. Cell

2006; 124(6):1169-1181.

(19) van Basten JP, Schraffordt Koops H, Sleijfer DT, Pras E, van Driel MF, Hoekstra HJ.

Current concepts about testicular cancer. Eur J Surg Oncol 1997; 23(4):354-360.

(20) Hain SF, O’Doherty MJ, Timothy AR, Leslie MD, Partridge SE, Huddart RA.

Fluorodeoxyglucose PET in the initial staging of germ cell tumours. Eur J Nucl Med

2000; 27(5):590-594.

(21) Gels ME, Hoekstra HJ, Sleijfer DT, Marrink J, De Bruijn HW, Molenaar WM et al. Detection

of recurrence in patients with clinical stage I nonseminomatous testicular germ cell

tumors and consequences for further follow-up: a single-center 10-year experience. J Clin

Oncol 1995; 13(5):1188-1194.

(22) Oliver RT, Mason MD, Mead GM, von der Maase H, Rustin GJ, Joffe JK et al. Radiotherapy

versus single-dose carboplatin in adjuvant treatment of stage I seminoma:

a randomised trial. Lancet 2005; 366(9482):293-300.

(23) de Wit R, Stoter G. Treatment of metastatic testicular carcinoma according to prognosis;

new development. Ned Tijdschr Geneeskd 2001; 145(25):1194-1199.

(24) de Wit R, Roberts JT, Wilkinson PM, de Mulder PH, Mead GM, Fossa SD et al.

Equivalence of three or four cycles of bleomycin, etoposide, and cisplatin chemotherapy

and of a 3- or 5-day schedule in good-prognosis germ cell cancer: a randomized study

of the European Organization for Research and Treatment of Cancer Genitourinary Tract

Cancer Cooperative Group and the Medical Research Council. J Clin Oncol 2001; 19(6):

1629-1640.

(25) De Santis M, Becherer A, Bokemeyer C, Stoiber F, Oechsle K, Sellner F et al. 2-18

fluoro-deoxy-D-glucose positron emission tomography is a reliable predictor for viable

tumor in postchemotherapy seminoma: an update of the prospective multicentric

SEMPET trial. J Clin Oncol 2004; 22(6):1034-1039.

(26) Lutke Holzik MF, Hoekstra HJ, Mulder NH, Suurmeijer AJ, Sleijfer DT, Gietema JA.

Non-germ cell malignancy in residual or recurrent mass after chemotherapy for

nonseminomatous testicular germ cell tumor. Ann Surg Oncol 2003; 10(2):131-135.

(27) Fizazi K, Tjulandin S, Salvioni R, Germa-Lluch JR, Bouzy J, Ragan D et al. Viable

malignant cells after primary chemotherapy for disseminated nonseminomatous

germ cell tumors: prognostic factors and role of postsurgery chemotherapy-results from

an international study group. J Clin Oncol 2001; 19(10):2647-2657.

(28) Bosl GJ, Motzer RJ. Testicular germ-cell cancer. N Engl J Med 1997; 337(4):242-253.

(29) Giwercman A, Muller J, Skakkebaek NE. Prevalence of carcinoma in situ and other

histopathological abnormalities in testes from 399 men who died suddenly and

unexpectedly. J Urol 1991; 145(1):77-80.

(30) Jacobsen GK, Henriksen OB, von der Maase H. Carcinoma in situ of testicular tissue

adjacent to malignant germ-cell tumors: a study of 105 cases. Cancer 1981; 47(11):

2660-2662.

(31) Boyle P. Testicular cancer: the challenge for cancer control. Lancet Oncol 2004; 5(1):56-61.

(32) Skakkebaek NE. Possible carcinoma-in-situ of the testis. Lancet 1972; 2(7776):516-517.

(33) Jones RH, Vasey PA. Part I: testicular cancer--management of early disease. Lancet

Oncol 2003; 4(12):730-737.

(34) Aetiology of testicular cancer: association with congenital abnormalities, age at puberty,

infertility, and exercise. United Kingdom Testicular Cancer Study Group. BMJ 1994;

308(6941):1393-1399.

(35) Forman D, Oliver RT, Brett AR, Marsh SG, Moses JH, Bodmer JG et al. Familial testicular

cancer: a report of the UK family register, estimation of risk and an HLA class 1 sib-pair

analysis. Br J Cancer 1992; 65(2):255-262.

(36) Osterlind A, Berthelsen JG, Abildgaard N, Hansen SO, Hjalgrim H, Johansen B et al. Risk

of bilateral testicular germ cell cancer in Denmark: 1960-1984. J Natl Cancer Inst 1991;

83(19):1391-1395.

(37) Wanderas EH, Fossa SD, Tretli S. Risk of a second germ cell cancer after treatment of a

primary germ cell cancer in 2201 Norwegian male patients. Eur J Cancer 1997; 33(2):

244-252.

(38) Brown LM, Pottern LM, Hoover RN. Testicular cancer in young men: the search for

causes of the epidemic increase in the United States. J Epidemiol Community Health

1987; 41(4):349-354.

(39) Swerdlow AJ, Huttly SR, Smith PG. Testicular cancer and antecedent diseases. Br J

Cancer 1987; 55(1):97-103.

(40) Swerdlow AJ, Huttly SR, Smith PG. Prenatal and familial associations of testicular

cancer. Br J Cancer 1987; 55(5):571-577.

(41) Swerdlow AJ, Higgins CD, Pike MC. Risk of testicular cancer in cohort of boys with

cryptorchidism. BMJ 1997; 314(7093):1507-1511.

(42) Hasle H, Mellemgaard A, Nielsen J, Hansen J. Cancer incidence in men with Klinefelter

syndrome. Br J Cancer 1995; 71(2):416-420.

(43) Verp MS, Simpson JL. Abnormal sexual differentiation and neoplasia. Cancer Genet

Cytogenet 1987; 25(2):191-218.

(44) Fossa SD, Aass N, Heilo A, Daugaard G, Skakkebaek E, Stenwig AE et al. Testicular

carcinoma in situ in patients with extragonadal germ-cell tumours: the clinical role of

pretreatment biopsy. Ann Oncol 2003; 14(9):1412-1418.


(45) Harland SJ, Cook PA, Fossa SD, Horwich A, Mead GM, Parkinson MC et al. Intratubular

germ cell neoplasia of the contralateral testis in testicular cancer: defining a high risk

group. J Urol 1998; 160(4):1353-1357.

(46) Jacobsen R, Bostofte E, Engholm G, Hansen J, Olsen JH, Skakkebaek NE et al. Risk

of testicular cancer in men with abnormal semen characteristics: cohort study. BMJ

2000; 321(7264):789-792.

(47) Moller H, Skakkebaek NE. Risk of testicular cancer in subfertile men: case-control study.

BMJ 1999; 318(7183):559-562.

(48) Tollerud DJ, Blattner WA, Fraser MC, Brown LM, Pottern L, Shapiro E et al. Familial

testicular cancer and urogenital developmental anomalies. Cancer 1985; 55(8):

1849-1854.

(49) Dieckmann KP, Pichlmeier U. The prevalence of familial testicular cancer: an analysis of

two patient populations and a review of the literature. Cancer 1997; 80(10):1954-1960.

(50) Dong C, Lonnstedt I, Hemminki K. Familial testicular cancer and second primary cancers

in testicular cancer patients by histological type. Eur J Cancer 2001; 37(15):1878-1885.

(51) Heimdal K, Olsson H, Tretli S, Flodgren P, Borresen AL, Fossa SD. Familial testicular

cancer in Norway and southern Sweden. Br J Cancer 1996; 73(7):964-969.

(52) Kruse C. Bilateral and familial occurrence of germ cell tumors of the testis. Urologe A

1987; 26(2):61-62.

(53) Ondrus D, Kuba D, Chrenova S, Matoska J. Familial testicular cancer and developmental

anomalies. Neoplasma 1997; 44(1):59-61.

(54) Patel SR, Kvols LK, Richardson RL. Familial testicular cancer: report of six cases and

review of the literature. Mayo Clin Proc 1990; 65(6):804-808.

(55) Polednak AP. Familial testicular cancer in a population-based cancer registry. Urol Int

1996; 56(4):238-240.

(56) Sonneveld DJ, Sleijfer DT, Schraffordt Koops H, Sijmons RH, van der Graaf WT, Sluiter

WJ et al. Familial testicular cancer in a single-centre population. Eur J Cancer 1999;

35(9):1368-1373.

(57) Westergaard T, Olsen JH, Frisch M, Kroman N, Nielsen JW, Melbye M. Cancer risk in

fathers and brothers of testicular cancer patients in Denmark. A population-based study.

Int J Cancer 1996; 66(5):627-631.

(58) Goss PE, Bulbul MA. Familial testicular cancer in five members of a cancer-prone

kindred. Cancer 1990; 66(9):2044-2046.

(59) Forman D, Gallagher R, Moller H, Swerdlow TJ. Aetiology and epidemiology of testicular

cancer: report of consensus group. Prog Clin Biol Res 1990; 357:245-253.

(60) Khoury MJ, Beaty TH, Liang KY. Can familial aggregation of disease be explained by

familial aggregation of environmental risk factors? Am J Epidemiol 1988; 127(3):674-683.

(61) Heimdal K, Fossa SD. Genetic factors in malignant germ-cell tumors. World J Urol 1994;

12(4):178-181.

(62) Sonneveld DJ, Schaapveld M, Sleijfer DT, te Meerman GJ, van der Graaf WT, Sijmons

RH et al. Geographic clustering of testicular cancer incidence in the northern part of The

Netherlands. Br J Cancer 1999; 81(7):1262-1267.

(63) Senturia YD. The epidemiology of testicular cancer. Br J Urol 1987; 60(4):285-291.

(64) Parkin DM, Iscovich J. Risk of cancer in migrants and their descendants in Israel: II.

Carcinomas and germ-cell tumours. Int J Cancer 1997; 70(6):654-660.

(65) Geddes M, Buiatti E. Review of studies of cancer in Italian migrants. IARC Sci Publ 1993;

123:24-30.

(66) Osterlind A, Berthelsen JG, Abildgaard N, Hansen SO, Jensen H, Johansen B et al.

Incidence of bilateral testicular germ cell cancer in Denmark, 1960-84: preliminary

findings. Int J Androl 1987; 10(1):203-208.

(67) Sonneveld DJ, Schraffordt Koops H, Sleijfer DT, Hoekstra HJ. Bilateral testicular germ

cell tumours in patients with initial stage I disease: prevalence and prognosis--a single

centre’s 30 years’ experience. Eur J Cancer 1998; 34(9):1363-1367.

(68) Tian Q, Frierson HF, Jr., Krystal GW, Moskaluk CA. Activating c-kit gene mutations in

human germ cell tumors. Am J Pathol 1999; 154(6):1643-1647.

(69) Przygodzki RM, Hubbs AE, Zhao FQ, O’Leary TJ. Primary mediastinal seminomas:

evidence of single and multiple KIT mutations. Lab Invest 2002; 82(10):1369-1375.

(70) Looijenga LH, de Leeuw H, van Oorschot M, van Gurp RJ, Stoop H, Gillis AJ et al. Stem

cell factor receptor (c-KIT) codon 816 mutations predict development of bilateral

testicular germ-cell tumors. Cancer Res 2003; 63(22):7674-7678.

(71) Rapley EA, Hockley S, Warren W, Johnson L, Huddart R, Crockford G et al. Somatic

mutations of KIT in familial testicular germ cell tumours. Br J Cancer 2004; 90(12):

2397-2401.

(72) Kroman N, Frisch M, Olsen JH, Westergaard T, Melbye M. Oestrogen-related cancer risk

in mothers of testicular-cancer patients. Int J Cancer 1996; 66(4):438-440.

(73) Heimdal K, Olsson H, Tretli S, Flodgren P, Borresen AL, Fossa SD. Risk of cancer in

relatives of testicular cancer patients. Br J Cancer 1996; 73(7):970-973.

(74) Nichols CR, Heerema NA, Palmer C, Loehrer PJ, Sr., Williams SD, Einhorn LH. Klinefelter’s

syndrome associated with mediastinal germ cell neoplasms. J Clin Oncol 1987; 5(8):

1290-1294.

(75) Houldsworth J. Genetics and biology of male germ cell tumors. Chest Surg Clin N Am

2002; 12(4):629-643.

(76) Lutke Holzik MF, Sijmons RH, Sleijfer DT, Sonneveld DJ, Hoekstra-Weebers JE, Echten-

Arends J et al. Syndromic aspects of testicular carcinoma. Cancer 2003; 97(4):984-992.

(77) Levin HS. Tumors of the testis in intersex syndromes. Urol Clin North Am 2000; 27(3):

543-51.

(78) Hasle H, Clemmensen IH, Mikkelsen M. Risks of leukaemia and solid tumours in

individuals with Down’s syndrome. Lancet 2000; 355(9199):165-169.


(79) Satge D, Sasco AJ, Cure H, Leduc B, Sommelet D, Vekemans MJ. An excess of testicular

germ cell tumors in Down’s syndrome: three case reports and a review of the literature.

Cancer 1997; 80(5):929-935.

(80) Gallagher RP, Huchcroft S, Phillips N, Hill GB, Coldman AJ, Coppin C et al. Physical

activity, medical history, and risk of testicular cancer (Alberta and British Columbia,

Canada). Cancer Causes Control 1995; 6(5):398-406.

(81) Pettersson A, Richiardi L, Nordenskjold A, Kaijser M, Akre O. Age at surgery for

undescended testis and risk of testicular cancer. N Engl J Med 2007; 356(18):1835-1841.

(82) Braun MM, Caporaso NE, Page WF, Hoover RN. Prevalence of a history of testicular

cancer in a cohort of elderly twins. Acta Genet Med Gemellol (Roma ) 1995; 44(3-4):

189-192.

(83) Swerdlow AJ, De Stavola BL, Swanwick MA, Maconochie NE. Risks of breast and

testicular cancers in young adult twins in England and Wales: evidence on prenatal and

genetic aetiology. Lancet 1997; 350(9093):1723-1728.

(84) Adami HO, Bergstrom R, Mohner M, Zatonski W, Storm H, Ekbom A et al. Testicular

cancer in nine northern European countries. Int J Cancer 1994; 59(1):33-38.

(85) Bergstrom R, Adami HO, Mohner M, Zatonski W, Storm H, Ekbom A et al. Increase in

testicular cancer incidence in six European countries: a birth cohort phenomenon. J Natl

Cancer Inst 1996; 88(11):727-733.

(86) Giwercman A, Carlsen E, Keiding N, Skakkebaek NE. Evidence for increasing incidence of

abnormalities of the human testis: a review. Environ Health Perspect 1993; 101 Suppl 2:

65-71.

(87) Sharpe RM. The ‘oestrogen hypothesis’- where do we stand now? Int J Androl 2003; 26:

2-15.

(88) Dieckmann KP, Endsin G, Pichlmeier U. How valid is the prenatal estrogen excess

hypothesis of testicular germ cell cancer? A case control study on hormone-related

factors. Eur Urol 2001; 40:677-683.

(89) Nicholson PW, Harland SJ. Inheritance and testicular cancer. Br J Cancer 1995; 71(2):

421-426.

(90) Heimdal K, Olsson H, Tretli S, Fossa SD, Borresen AL, Bishop DT. A segregation analysis

of testicular cancer based on Norwegian and Swedish families. Br J Cancer 1997; 75(7):

1084-1087.

(91) Knudson AG, Jr. Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad

Sci U S A 1971; 68(4):820-823.

(92) Moller H. Trends in sex-ratio, testicular cancer and male reproductive hazards: are they

connected? APMIS 1998; 106(1):232-238.

(93) Petersen PM, Skakkebaek NE, Giwercman A. Gonadal function in men with testicular

cancer: biological and clinical aspects. APMIS 1998; 106(1):24-34.

(94) Sonneveld DJ, Hoekstra HJ, van der Graaf WT, Sluiter WJ, Mulder NH, Willemse PH et al.

Improved long term survival of patients with metastatic nonseminomatous testicular

germ cell carcinoma in relation to prognostic classification systems during the cisplatin

era. Cancer 2001; 91(7):1304-1315.

(95) Lutke Holzik MF, Sonneveld DJ, Hoekstra HJ, te Meerman GJ, Sleijfer DT, Schaapveld M.

Do the eastern and northern parts of The Netherlands differ in testicular cancer? Urology

2001; 58(4):636-637.

(96) Sonneveld DJ, Lutke Holzik MF, Nolte IM, Sleijfer DT, van der Graaf WT, Bruinenberg M

et al. Testicular carcinoma and HLA Class II genes. Cancer 2002; 95(9):1857-1863.

(97) Bateman AC, Howell WM. Human leukocyte antigens and cancer: is it in our genes? J

Pathol 1999; 188(3):231-236.

(98) de Jong MM, Nolte IM, de Vries EG, Schaapveld M, Kleibeuker JH, Oosterom E et al.

The HLA class III subregion is responsible for an increased breast cancer risk. Hum Mol

Genet 2003; 12(18):2311-2319.

(99) Birkeland SA, Storm HH, Lamm LU, Barlow L, Blohme I, Forsberg B et al. Cancer risk

after renal transplantation in the Nordic countries, 1964- 1986. Int J Cancer 1995; 60(2):

183-189.

(100) Spano JP, Atlan D, Breau JL, Farge D. AIDS and non-AIDS-related malignancies: a

new vexing challenge in HIV-positive patients. Part II. Cervical and anal squamous

epithelial lesions, lung cancer, testicular germ cell cancers, and skin cancers. Eur J Intern

Med 2002; 13(4):227-232.

(101) Corzo D, Salazar M, Granja CB, Yunis EJ. Advances in HLA genetics. Exp Clin

Immunogenet 1995; 12(3):156-170.

(102) Özdemir E, Kakehi Y, Mishina M, Ogawa O, Okada Y, Özdemir D et al. High-resolution

HLA-DRB1 and DQB1 genotyping in Japanese patients with testicular germ cell

carcinoma. Br J Cancer 1997; 76(10):1348-1352.

(103) Rapley EA, Crockford GP, Easton D, Stratton MR, Bishop DT. Localisation of

susceptibility genes for familial testicular germ cell tumour. APMIS 2003; 111:128-135.

(104) Rapley EA, Crockford GP, Easton D, Stratton MR, Bishop DT. The Genetics of Testicular

Germ Cell Tumours in proceedings of: Germ Cell Tumours V. Harden, P, Joffe,

JK and Jones, W G Eds 2002;1-22.

(105) Rapley EA, Crockford GP, Teare D, Biggs P, Seal S, Barfoot R et al. Localization to Xq27

of a susceptibility gene for testicular germ-cell tumours. Nat Genet 2000; 24(2):197-200.

(106) DeWolf WC, Lange PH, Einarson ME, Yunis EJ. HLA and testicular cancer. Nature 1979;

277(5693):216-217.

(107) Carr BI, Bach FA. Possible association between HLA-AW24 and metastatic testicular

germ-cell tumours. Lancet 1979; 1(8130):1346-1347.

(108) Majsky A, Abrahamova J, Korinkova P, Bek V. HLA system and testicular germinative

tumours. Oncology 1979; 36(5):228-231.

(109) Pollack MS, Vugrin D, Hennessy W, Herr HW, Dupont B, Whitmore WF, Jr. HLA antigens

in patients with germ cell cancers of the testis. Cancer Res 1982; 42(6):2470-2473.


(110) Oliver RT, Stephenson CA, Parkinson MC, Forman D, Atkinson A, Bodmer J et al. Germ

cell tumours of the testicle as a model of MHC influence on human malignancy. Lancet

1986; 1(8496):1506.

(111) Aiginger P, Kuzmits R, Kratzig C, Schwarz HP, Zielinski CC, Kuhbock J et al. HLA antigens

and germ-cell tumours. Lancet 1987; 1(8527):276-277.

(112) Dieckmann KP, von Keyserlingk HJ. HLA typing in familial testicular cancer. Eur Urol

1989; 16(5):361-365.

(113) Ryberg D, Heimdal K, Fossa SD, Borresen AL, Haugen A. Rare Ha-ras1 alleles and

predisposition to testicular cancer. Int J Cancer 1993; 53(6):938-940.

(114) Dieckmann KP, Klan R, Bunte S. HLA antigens, Lewis antigens, and blood groups in

patients with testicular germ-cell tumors. Oncology 1993; 50(4):252-258.

(115) Heimdal KR, Lothe RA, Fossa SD, Borresen AL. Association studies of a polymorphism in

the Wilms’ tumor 1 locus in Norwegian patients with testicular cancer. Int J Cancer 1994;

58(4):523-526.

(116) Olie RA, Looijenga LH, Boerrigter L, Top B, Rodenhuis S, Langeveld A et al. N- and

KRAS mutations in primary testicular germ cell tumors: incidence and possible biological

implications. Genes Chromosomes Cancer 1995; 12(2):110-116.

(117) Heimdal K, Andersen TI, Skrede M, Fossa SD, Berg K, Borresen AL. Association studies

of estrogen receptor polymorphisms in a Norwegian testicular cancer population. Cancer

Epidemiol Biomarkers Prev 1995; 4(2):123-126.

(118) Harries LW, Stubbins MJ, Forman D, Howard GC, Wolf CR. Identification of genetic

polymorphisms at the glutathione S- transferase Pi locus and association with

susceptibility to bladder, testicular and prostate cancer. Carcinogenesis 1997; 18(4):

641-644.

(119) Sakashita S, Koyanagi T, Tsuji I, Arikado K, Matsuno T. Congenital anomalies in children

with testicular germ cell tumor. J Urol 1980; 124(6):889-891.

(120) Dexeus FH, Logothetis CJ, Chong C, Sella A, Ogden S. Genetic abnormalities in men with

germ cell tumors. J Urol 1988; 140(1):80-84.

(121) Sijmons R H., Burger G.T N. Familial cancer database: a clinical aide-mémoire. Familial

Cancer 2001; 1(1):51-55.

(122) Rescorla FJ, Breitfeld PP. Pediatric germ cell tumors. Curr Probl Cancer 1999; 23(6):

257-303.

(123) Savage MO, Lowe DG. Gonadal neoplasia and abnormal sexual differentiation.

Clin Endocrinol (Oxf) 1990; 32(4):519-533.

(124) Weiss GR, Garnick MB. Testicular cancer in a Russell-Silver dwarf. J Urol 1981; 126(6):

836-837.

(125) Boisen KA, Main KM, Rajpert-De Meyts E, Skakkebaek NE. Are male reproductive

disorders a common entity? The testicular dysgenesis syndrome. Ann N Y Acad Sci 2001;

948:90-99.

(126) Cortes D, Visfeldt J, Moller H, Thorup J. Testicular neoplasia in cryptorchid boys at

primary surgery: case series. BMJ 1999; 319(7214):888-889.

(127) Depue RH, Pike MC, Henderson BE. Cryptorchidism and testicular cancer. J Natl Cancer

Inst 1986; 77(3):830-833.

(128) Pike MC, Chilvers C, Peckham MJ. Effect of age at orchidopexy on risk of testicular

cancer. Lancet 1986; 1(8492):1246-1248.

(129) Pottern LM, Brown LM, Hoover RN, Javadpour N, O’Connell KJ, Stutzman RE et al.

Testicular cancer risk among young men: role of cryptorchidism and inguinal hernia. J

Natl Cancer Inst 1985; 74(2):377-381.

(130) Fearon ER. Human cancer syndromes: clues to the origin and nature of cancer. Science

1997; 278(5340):1043-1050.

(131) Cendron M, Wein AJ, Schwartz SS, Murtagh F, Livolsi VA, Tomaszewski JE. Germ cell

tumor of testis in a patient with von Hippel-Lindau disease. Urology 1991; 37(1):69-71.

(132) Hartley AL, Birch JM, Kelsey AM, Marsden HB, Harris M, Teare MD. Are germ cell tumors

part of the Li-Fraumeni cancer family syndrome? Cancer Genet Cytogenet 1989; 42(2):

221-226.

(133) de Jong B, van Echten J, Looijenga LH, Geurts vK, Oosterhuis JW. Cytogenetics of the

progression of adult testicular germ cell tumors. Cancer Genet Cytogenet 1997; 95(1):

88-95.

(134) Oosterhuis JW, Looijenga LH, van Echten J, de Jong B. Chromosomal constitution and

developmental potential of human germ cell tumors and teratomas. Cancer Genet

Cytogenet 1997; 95(1):96-102.

(135) Mostert M, Rosenberg C, Stoop H, Schuyer M, Timmer A, Oosterhuis W et al.

Comparative genomic and in situ hybridization of germ cell tumors of the infantile

testis. Lab Invest 2000; 80(7):1055-1064.

(136) Summersgill B, Osin P, Lu YJ, Huddart R, Shipley J. Chromosomal imbalances associated

with carcinoma in situ and associated testicular germ cell tumours of adolescents and

adults. Br J Cancer 2001; 85(2):213-220.

(137) Qiao D, Zeeman AM, Deng W, Looijenga LH, Lin H. Molecular characterization of hiwi,

a human member of the piwi gene family whose overexpression is correlated to

seminomas. Oncogene 2002; 21(25):3988-3999.

(138) Kratzik C, Aiginger P, Kuzmits R, Spona J, Kirnbauer M, Seiser A et al. HLA-antigen

distribution in seminoma, HCG-positive seminoma and nonseminomatous tumours of

the testis. Urol Res 1989; 17(6):377-380.

(139) Kume H, Tachikawa T, Teramoto S, Isurugi K, Kitamura T. Bilateral testicular tumour in

neurofibromatosis type 1. Lancet 2001; 357(9253):395-396.

(140) Foster K, Osborne RJ, Huddart RA, Affara NA, Ferguson-Smith MA, Maher ER. Molecular

genetic analysis of the von Hippel-Lindau disease (VHL) tumour suppressor gene in

gonadal tumours. Eur J Cancer 1995; 31A(13-14):2392-2395.


(141) Ye DW, Zheng J, Qian SX, Ma Y, Zheng X, Li D et al. p53 gene mutations in Chinese

human testicular seminoma. J Urol 1993; 150(3):884-886.

(142) Bartkova J, Falck J, Rajpert-De Meyts E, Skakkebaek NE, Lukas J, Bartek J. Chk2 tumour

suppressor protein in human spermatogenesis and testicular germ-cell tumours.

Oncogene 2001; 20(41):5897-5902.

(143) Heidenreich A, Gaddipati JP, Moul JW, Srivastava S. Molecular analysis of P16(Ink4)/

CDKN2 and P15(INK4B)/MTS2 genes in primary human testicular germ cell tumors. J Urol

1998; 159(5):1725-1730.

(144) Drescher B, Lauke H, Hartmann M, Davidoff MS, Zumkeller W. Immunohistochemical

pattern of insulin-like growth factor (IGF) I, IGF II and IGF binding proteins 1 to 6 in

carcinoma in situ of the testis. Mol Pathol 1997; 50(6):298-303.

(145) Drescher B, Zumkeller W, Lauke H, Hartmann M, Davidoff MS. Insulin-like growth

factor-binding protein (IGFBP)-5 in human testicular tubules. Adv Exp Med Biol 1997;

424:153-154.

(146) Ghebranious N, Donehower LA. Mouse models in tumor suppression. Oncogene 1998;

17(25):3385-3400.

(147) Skotheim RI, Monni O, Mousses S, Fossa SD, Kallioniemi OP, Lothe RA et al.

New insights into testicular germ cell tumorigenesis from gene expression profiling.

Cancer Res 2002; 62(8):2359-2364.

(148) Jiang LI, Nadeau JH. 129/Sv mice--a model system for studying germ cell biology and

testicular cancer. Mamm Genome 2001; 12(2):89-94.

(149) Kash K, Dabney M, Holland JC, Osborne MP, Miller D. Familial cancer and genetics:

Psychosocial and ethical aspects. In: L.Baider, CL Cooper, A.Kaplan De-Nour (Eds.).

Cancer and the Family. Chichester, Engeland: John Wiley., 1997: 389-403.

(150) Melzer D, Zimmern R. Genetics and medicalisation. BMJ 2002; 324(7342):863-864.

(151) Press NA, Yasui Y, Reynolds S, Durfy SJ, Burke W. Women’s interest in genetic testing

for breast cancer susceptibility may be based on unrealistic expectations. Am J Med

Genet 2001; 99(2):99-110.

(152) Sakai N, Yamada T, Asao T, Baba M, Yoshida M, Murayama T. Bilateral testicular tumors

in androgen insensitivity syndrome. Int J Urol 2000; 7(10):390-392.

(153) Collins GM, Kim DU, Logrono R, Rickert RR, Zablow A, Breen JL. Pure seminoma arising

in androgen insensitivity syndrome (testicular feminization syndrome): a case report and

review of the literature. Mod Pathol 1993; 6(1):89-93.

(154) Kocak M, Yalvac S, Pata O, Turan H, Haberal A. A seminoma case which occurred in a

patient with familial testicular feminization syndrome. Acta Obstet Gynecol Scand 2000;

79(10):890-891.

(155) Muller J, Skakkebaek NE. Testicular carcinoma in situ in children with the androgen

insensitivity (testicular feminisation) syndrome. Br Med J (Clin Res Ed) 1984; 288(6428):

1419-1420.

(156) Cassio A, Cacciari E, D’Errico A, Balsamo A, Grigioni FW, Pascucci MG et al. Incidence of

intratubular germ cell neoplasia in androgen insensitivity syndrome. Acta Endocrinol

(Copenh) 1990; 123(4):416-422.

(157) Rutgers JL, Scully RE. The androgen insensitivity syndrome (testicular feminization):

a clinicopathologic study of 43 cases. Int J Gynecol Pathol 1991; 10(2):126-144.

(158) Kaisary AV, Williams G. A case of bilateral intra-abdominal testes and seminomas in a

previously unrecognised hermaphrodite. Br J Urol 1982; 54(3):323.

(159) Rutgers JL, Scully RE. Pathology of the testis in intersex syndromes. Semin Diagn Pathol

1987; 4(4):275-291.

(160) Hellwinkel OJ, Bassler J, Hiort O. Transcription of androgen receptor and

5alpha-reductase II in genital fibroblasts from patients with androgen insensitivity

syndrome. J Steroid Biochem Mol Biol 2000; 75(4-5):213-218.

(161) Adesokan A, Adegboyega PA, Cowan DF, Kocurek J, Neal DE, Jr. Testicular “tumor” of

the adrenogenital syndrome: a case report of an unusual association with myelolipoma

and seminoma in cryptorchidism. Cancer 1997; 80(11):2120-2127.

(162) Rutgers JL, Young RH, Scully RE. The testicular “tumor” of the adrenogenital syndrome.

A report of six cases and review of the literature on testicular masses in patients with

adrenocortical disorders. Am J Surg Pathol 1988; 12(7):503-513.

(163) Phillips HA, Howard GC. Testicular seminoma in a patient with ataxia-telangiectasia.

Clin Oncol (R Coll Radiol ) 1999; 11(1):63-64.

(164) Parra G, Seery W, Buchbinder M, Cole AT. Congenital total hemihypertrophy and

carcinoma of undescended testicle: a case report. J Urol 1977; 118(2):343-344.

(165) Dieckmann KP, Rube C, Henke RP. Association of Down’s syndrome and testicular

cancer. J Urol 1997; 157(5):1701-1704.

(166) Yang Q, Rasmussen SA, Friedman JM. Mortality associated with Down’s syndrome in the

USA from 1983 to 1997: a population-based study. Lancet 2002; 359(9311):1019-1025.

(167) Jacobsen GK, Henriques UV. A fetal testis with intratubular germ cell neoplasia (ITGCN).

Mod Pathol 1992; 5(5):547-549.

(168) Satge D, Jacobsen GK, Cessot F, Raffi F, Vekemans M. A fetus with Down syndrome and

intratubular germ cell neoplasia. Pediatr Pathol Lab Med 1996; 16(1):107-112.

(169) Martin MM, Wu SM, Martin AL, Rennert OM, Chan WY. Testicular seminoma in a patient

with a constitutively activating mutation of the luteinizing hormone/chorionic

gonadotropin receptor. Eur J Endocrinol 1998; 139(1):101-106.

(170) Avril MF, Chompret A, Verne-Fourment L, Terrier-Lacombe MJ, Spatz A, Fizazi K et al.

Association between germ cell tumours, large numbers of naevi, atypical naevi and

melanoma. Melanoma Res 2001; 11(2):117-122.

(171) Hemminki A. The molecular basis and clinical aspects of Peutz-Jeghers syndrome.

Cell Mol Life Sci 1999; 55(5):735-750.

(172) Raghavan D, Zalcberg JR, Grygiel JJ, Teriana N, Cox KM, McCarthy W et al. Multiple

atypical nevi: a cutaneous marker of germ cell tumors. J Clin Oncol 1994; 12(11):2284-2287.


(173) Sigg C, Pelloni F. Dysplastic nevi and germ cell tumors of the testis--a possible

further tumor in the spectrum of associated malignancies in dysplastic nevus syndrome.

Dermatologica 1988; 176(3):109-110.

(174) Flechon A, Droz JP. Hereditary persistence of alpha-fetoprotein in testis disease. J Urol

2001; 165(6 Pt 1):2004.

(175) Schefer H, Mattmann S, Joss RA. Hereditary persistence of alpha-fetoprotein. Case report

and review of the literature. Ann Oncol 1998; 9(6):667-672.

(176) Staples J. Alphafetoprotein, cancer, and benign conditions. Lancet 1986; 2(8518):1277.

(177) Cochran PK, Chauvenet AR, Hart PS, de Graaf SS, Cushing B, Kroovand L et al.

Hereditary persistence of alpha-fetoprotein in a child with testicular germ cell tumor.

Med Pediatr Oncol 1999; 32(6):436-437.

(178) Albers DD, Males JL. Seminoma in hypogonadotropic hypogonadism associated with

anosmia (Kallmann’s syndrome). J Urol 1981; 126(1):57-58.

(179) Bebb GG, Grannis FW, Jr., Paz IB, Slovak ML, Chilcote R. Mediastinal germ cell tumor in

a child with precocious puberty and Klinefelter syndrome. Ann Thorac Surg 1998; 66(2):

547-548.

(180) Carroll PR, Morse MJ, Koduru PP, Chaganti RS. Testicular germ cell tumor in patient with

Klinefelter syndrome. Urology 1988; 31(1):72-74.

(181) Isurugi K, Imao S, Hirose K, Aoki H. Seminoma in Klinefelter’s syndrome with 47, XXY,

15s+ karyotype. Cancer 1977; 39(5):2041-2047.

(182) Kurabayashi A, Furihata M, Matsumoto M, Sonobe H, Ohtsuki Y, Aki M et al. Primary

intrapelvic seminoma in Klinefelter’s syndrome. Pathol Int 2001; 51(8):624-628.

(183) Muller J, Skakkebaek NE, Ratcliffe SG. Quantified testicular histology in boys with sex

chromosome abnormalities. Int J Androl 1995; 18(2):57-62.

(184) Schimke RN, Madigan CM, Silver BJ, Fabian CJ, Stephens RL. Choriocarcinoma,

thyrotoxicosis, and the Klinefelter syndrome. Cancer Genet Cytogenet 1983; 9(1):1-7.

(185) Sogge MR, McDonald SD, Cofold PB. The malignant potential of the dysgenetic germ

cell in Klinefelter’s syndrome. Am J Med 1979; 66(3):515-518.

(186) Epenetos AA, Collis CH. Seminoma in Marfan’s syndrome. Postgrad Med J 1979;

55(648):762-764.

(187) Beheshti M, Hardy BE, Mancer K, McLorie G, Churchill BM. Neoplastic potential in

patients with disorders of sexual differentiation. Urology 1987; 29(4):404-407.

(188) Donahoe PK, Crawford JD, Hendren WH. Mixed gonadal dysgenesis, pathogensis, and

management. J Pediatr Surg 1979; 14(3):287-300.

(189) Gourlay WA, Johnson HW, Pantzar JT, McGillivray B, Crawford R, Nielsen WR. Gonadal

tumors in disorders of sexual differentiation. Urology 1994; 43(4):537-540.

(190) Kulkarni JN, Kamat MR, Borges AM. Bilateral synchronous tumors in testes in

unrecognized mixed gonadal dysgenesis: a case report and review of literature. J Urol

1990; 143(2):362-364.

(191) Takazawa R, Kawakami S, Kageyama Y, Kihara K, Oshima H. Mixed gonadal dysgenesis

detected after finding a large seminoma mimicking a uterine inguinal hernia. BJU Int

2001; 87(9):903.

(192) Groot-Loonen JJ, Voute PA, de Kraker J. Testicular tumor concomitant with von

Recklinghausen’s disease. Med Pediatr Oncol 1988; 16(2):116-117.

(193) Aggarwal A, Krishnan J, Kwart A, Perry D. Noonan’s syndrome and seminoma of

undescended testicle. South Med J 2001; 94(4):432-434.

(194) Tartaglia M, Mehler EL, Goldberg R, Zampino G, Brunner HG, Kremer H et al. Mutations

in PTPN11, encoding the protein tyrosine phosphatase SHP-2, cause Noonan syndrome.

Nat Genet 2001; 29(4):465-468.

(195) Berkmen F. Persistent mullerian duct syndrome with or without transverse testicular

ectopia and testis tumours. Br J Urol 1997; 79(1):122-126.

(196) Duenas A, Saldivar C, Castillero C, Flores G, Martinez P, Jimenez M. A case of bilateral

seminoma in the setting of persistent mullerian duct syndrome. Rev Invest Clin 2001;

53(2):193-196.

(197) Kulkarni JN, Bhansali MS, Tongaonkar HB, Kamat MR, Borges AM. Carcinoma in the third

testis in a case of polyorchidism and persistent mullerian structure syndrome. Eur Urol

1992; 22(2):174-176.

(198) Melman A, Leiter E, Perez JM, Driscoll D, Palmer C. The influence of neonatal orchiopexy

upon the testis in persistent Mullerian duct syndrome. J Urol 1981; 125(6):856-858.

(199) Wu HC, Chen JH, Lu HF, Shen WC. Persistent mullerian duct syndrome wtih seminoma:

CT findings. AJR Am J Roentgenol 2000; 174(1):102-104.

(200) Eastham JA, McEvoy K, Sullivan R, Chandrasoma P. A case of simultaneous bilateral

nonseminomatous testicular tumors in persistent mullerian duct syndrome. J Urol 1992;

148(2 Pt 1):407-408.

(201) Robinson AC, Jones WG. Prader Willi syndrome and testicular tumour. Clin Oncol (R Coll

Radiol ) 1990; 2(2):117.

(202) Uehling D. Cryptorchidism in the Prader-Willi syndrome. J Urol 1980; 124(1):103-104.

(203) Hornstein L, Bove KE, Towbin RB. Linear nevi, hemihypertrophy, connective tissue

hamartomas, and unusual neoplasms in children. J Pediatr 1987; 110(3):404-408.

(204) Parra RO, Cummings JM, Palmer DC. Testicular seminoma in a long-term survivor of the

prune belly syndrome. Eur Urol 1991; 19(1):79-80.

(205) Massad CA, Cohen MB, Kogan BA, Beckstead JH. Morphology and histochemistry of

infant testes in the prune belly syndrome. J Urol 1991; 146(6):1598-1600.

(206) Miller RW, Rubinstein JH. Tumors in Rubinstein-Taybi syndrome. Am J Med Genet 1995;

56(1):112-115.

(207) Cohen PR, Kurzrock R. Miscellaneous genodermatoses: Beckwith-Wiedemann syndrome,

Birt-Hogg- Dube syndrome, familial atypical multiple mole melanoma syndrome,

hereditary tylosis, incontinentia pigmenti, and supernumerary nipples. Dermatol Clin

1995; 13(1):211-229.


(208) Goedert JJ, McKeen EA, Javadpour N, Ozols RF, Pottern LM, Fraumeni JF, Jr. Polythelia

and testicular cancer. Ann Intern Med 1984; 101(5):646-647.

(209) Mehes K. Association of supernumerary nipples with other anomalies. J Pediatr 1979;

95(2):274-275.

(210) Lykkesfeldt G, Bennett P, Lykkesfeldt AE, Micic S, Rorth M, Skakkebaek NE et al. Testis

cancer. Ichthyosis constitutes a significant risk factor. Cancer 1991; 67(3):730-734.

(211) Lykkesfeldt G, Hoyer H, Lykkesfeldt AE, Skakkebaek NE. Steroid sulphatase deficiency

associated with testis cancer. Lancet 1983; 2(8365-66):1456.

(212) Chaganti RS, Houldsworth J. Genetics and biology of adult human male germ cell

tumors. Cancer Res 2000; 60(6):1475-1482.

(213) Hutson JM, Hasthorpe S, Heyns CF. Anatomical and functional aspects of testicular

descent and cryptorchidism. Endocr Rev 1997; 18(2):259-280.

(214) Prener A, Engholm G, Jensen OM. Genital anomalies and risk for testicular cancer in

Danish men. Epidemiology 1996; 7(1):14-19.

(215) Strader CH, Weiss NS, Daling JR, Karagas MR, McKnight B. Cryptorchism, orchiopexy,

and the risk of testicular cancer. Am J Epidemiol 1988; 127(5):1013-1018.

(216) Swerdlow AJ, De Stavola BL, Swanwick MA, Mangtani P, Maconochie NE. Risk factors for

testicular cancer: a case-control study in twins. Br J Cancer 1999; 80(7):1098-1102.

(217) Ulbright TM. Testis risk and prognostic factors. The pathologist’s perspective. Urol Clin

North Am 1999; 26(3):611-626.

(218) Swerdlow AJ, Huttly SR, Smith PG. Testis cancer: post-natal hormonal factors, sexual

behaviour and fertility. Int J Cancer 1989; 43(4):549-553.

(219) Moss AR, Osmond D, Bacchetti P, Torti FM, Gurgin V. Hormonal risk factors in testicular

cancer. A case-control study. Am J Epidemiol 1986; 124(1):39-52.

(220) Wilkinson TJ, Colls BM. Testicular cancer and age at puberty. BMJ 1994; 309(6959):955.

(221) Spermon JR, Witjes JA, Nap M, Kiemeney LA. Cancer incidence in relatives of patients

with testicular cancer in the eastern part of The Netherlands. Urology 2001; 57(4):747-752.

(222) Harland SJ. Conundrum of the hereditary component of testicular cancer. Lancet 2000;

356(9240):1455-1456.

(223) Akre O, Ekbom A, Hsieh CC, Trichopoulos D, Adami HO. Testicular nonseminoma and

seminoma in relation to perinatal characteristics. J Natl Cancer Inst 1996; 88(13):883-889.

(224) Mills PK, Newell GR, Johnson DE. Familial patterns of testicular cancer. Urology 1984;

24(1):1-7.

(225) Newell GR, Mills PK, Johnson DE. Epidemiologic comparison of cancer of the testis and

Hodgkin’s disease among young males. Cancer 1984; 54(6):1117-1123.

(226) Candidate regions for testicular cancer susceptibility genes. The International Testicular

Cancer Linkage Consortium. APMIS 1998; 106(1):64-70.

(227) Leahy MG, Tonks S, Moses JH, Brett AR, Huddart R, Forman D et al. Candidate regions

for a testicular cancer susceptibility gene. Hum Mol Genet 1995; 4(9):1551-1555.

(228) Murty VV, Bosl GJ, Houldsworth J, Meyers M, Mukherjee AB, Reuter V et al. Allelic loss

and somatic differentiation in human male germ cell tumors. Oncogene 1994; 9(8):

2245-2251.

(229) Skotheim RI, Kraggerud SM, Fossa SD, Stenwig AE, Gedde-Dahl JT, Danielsen HE et al.

Familial/bilateral and sporadic testicular germ cell tumors show frequent genetic

changes at loci with suggestive linkage evidence. Neoplasia 2001; 3(3):196-203.

(230) Bodmer WF. The HLA system: structure and function. J Clin Pathol 1987; 40(9):948-958.

(231) Complete sequence and gene map of a human major histocompatibility complex. The

MHC sequencing consortium. Nature 1999; 401(6756):921-923.

(232) Boon M, Nolte IM, Bruinenberg M, Spijker GT, Terpstra P, Raelson J et al. Mapping of a

susceptibility gene for multiple sclerosis to the 51 kb interval between G511525 and

D6S1666 using a new method of haplotype sharing analysis. Neurogenetics 2001; 3(4):

221-230.

(233) Foissac A, Crouau-Roy B, Faure S, Thomsen M, Cambon-Thomsen A. Microsatellites in

the HLA region: on overview. Tissue Antigens 1997; 49(3 Pt 1):197-214.

(234) Beckmann L, Fischer C, Deck KG, Nolte IM, te MG, Chang-Claude J. Exploring haplotype

sharing methods in general and isolated populations to detect gene(s) of a complex

genetic trait. Genet Epidemiol 2001; 21 Suppl 1:S554-S559.

(235) Dorum A, Moller P, Kamsteeg EJ, Scheffer H, Burton M, Heimdal KR et al. A BRCA1

founder mutation, identified with haplotype analysis, allowing genotype/phenotype

determination and predictive testing. Eur J Cancer 1997; 33(14):2390-2392.

(236) Levinson DF, Kirby A, Slepner S, Nolte I, Spijker GT, te MG. Simulation studies of

detection of a complex disease in a partially isolated population. Am J Med Genet 2001;

105(1):65-70.

(237) Spielman RS, McGinnis RE, Ewens WJ. Transmission test for linkage disequilibrium: the

insulin gene region and insulin-dependent diabetes mellitus (IDDM). Am J Hum Genet

1993; 52(3):506-516.

(238) Bernardi D, Salvioni R, Vaccher E, Repetto L, Piersantelli N, Marini B et al. Testicular

germ cell tumors and human immunodeficiency virus infection: a report of 26 cases.

Italian Cooperative Group on AIDS and Tumors. J Clin Oncol 1995; 13(11):2705-2711.

(239) Calista D. Five cases of melanoma in HIV positive patients. Eur J Dermatol 2001; 11(5):

446-449.

(240) Hentrich MU, Brack NG, Schmid P, Schuster T, Clemm C, Hartenstein RC. Testicular germ

cell tumors in patients with human immunodeficiency virus infection. Cancer 1996;

77(10):2109-2116.

(241) Timmerman JM, Northfelt DW, Small EJ. Malignant germ cell tumors in men infected with

the human immunodeficiency virus: natural history and results of therapy. J Clin Oncol

1995; 13(6):1391-1397.

(242) Carrington M. Recombination within the human MHC. Immunol Rev 1999; 167:245-256.


(243) Roychoudhury AK, Nei M. Human polymorphic genes world distribution. New York:

Oxford university press 1988;234-239.

(244) Lutke Holzik MF, Rapley EA, Hoekstra HJ, Sleijfer DT, Nolte IM, Sijmons RH. Genetic

predisposition to testicular germ-cell tumours. Lancet Oncol 2004; 5(6):363-371.

(245) Jorgensen N, Carlsen E, Nermoen I, Punab M, Suominen J, Andersen AG et al.

East-West gradient in semen quality in the Nordic-Baltic area: a study of men from the

general population in Denmark, Norway, Estonia and Finland. Hum Reprod 2002; 17(8):

2199-2208.

(246) Nijman JM, Schraffordt Koops H, Kremer J, Willemse PH, Sleijfer DT, Oldhoff J. Fertility

and hormonal function in patients with a nonseminomatous tumor of the testis.

Arch Androl 1985; 14(2-3):239-246.

(247) Kent-First M, Muallem A, Shultz J, Pryor J, Roberts K, Nolten W et al. Defining regions

of the Y-chromosome responsible for male infertility and identification of a fourth AZF

region (AZFd) by Y-chromosome microdeletion detection. Mol Reprod Dev 1999; 53(1):

27-41.

(248) Layman LC. Human gene mutations causing infertility. J Med Genet 2002; 39(3):153-161.

(249) Vogt PH, Edelmann A, Kirsch S, Henegariu O, Hirschmann P, Kiesewetter F et al. Human

Y chromosome azoospermia factors (AZF) mapped to different subregions in Yq11.

Hum Mol Genet 1996; 5(7):933-943.

(250) Foresta C, Moro E, Garolla A, Onisto M, Ferlin A. Y chromosome microdeletions in

cryptorchidism and idiopathic infertility. J Clin Endocrinol Metab 1999; 84(10):3660-3665.

(251) Herrinton LJ, Zhao W, Husson G. Management of cryptorchism and risk of testicular

cancer. Am J Epidemiol 2003; 157(7):602-605.

(252) Echten van J, Jong de B. Cytogenetics of testicular germ cell tumors of adults. The

Cancer Journal 1998; 11(5):242-246.

(253) Bianchi NO, Richard SM, Peltomaki P, Bianchi MS. Mosaic AZF deletions and

susceptibility to testicular tumors. Mutat Res 2002; 503(1-2):51-62.

(254) Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from

human nucleated cells. Nucleic Acids Res 1988; 16(3):1215-1216.

(255) Aknin-Seifer IE, Touraine RL, Lejeune H, Laurent JL, Lauras B, Levy R. A simple, low cost

and non-invasive method for screening Y-chromosome microdeletions in infertile men.

Hum Reprod 2003; 18(2):257-261.

(256) Y chromosome deletion detection system version 2.0. technical manual no 248 2003;

(Promega Corporation USA).

(257) Frydelund-Larsen L, Vogt PH, Leffers H, Schadwinkel A, Daugaard G, Skakkebaek NE et

al. No AZF deletion in 160 patients with testicular germ cell neoplasia. Mol Hum Reprod

2003; 9(9):517-521.

(258) Nathanson KL, Kanetsky PA, Hawes R, Vaughn DJ, Letrero R, Tucker K et al. The Y

eletion gr/gr and susceptibility to testicular germ cell tumor. Am J Hum Genet 2005;

77(6):1034-1043.

(259) Repping S, Skaletsky H, Brown L, van Daalen SK, Korver CM, Pyntikova T et al.

Polymorphism for a 1.6-Mb deletion of the human Y chromosome persists through

balance between recurrent mutation and haploid selection. Nat Genet 2003; 35(3):

247-251.

(260) Richiardi L, Bellocco R, Adami HO, Torrang A, Barlow L, Hakulinen T et al. Testicular

cancer incidence in eight northern European countries: secular and recent trends. Cancer

Epidemiol Biomarkers Prev 2004; 13(12):2157-2166.

(261) Hemminki K, Li X. Familial risk in testicular cancer as a clue to a heritable and

environmental aetiology. Br J Cancer 2004; 90(9):1765-1770.

(262) Crockford GP, Linger R, Hockley S, Dudakia D, Johnson L, Huddart R et al. Genome-wide

linkage screen for testicular germ cell tumour susceptibility loci. Hum Mol Genet 2006;

15(3):443-451.

(263) Redolfi E, Montagna C, Mumm S, Affer M, Susani L, Reinbold R et al. Identification of

CXorf1, a novel intronless gene in Xq27.3, expressed in human hippocampus. DNA Cell

Biol 1998; 17(12):1009-1016.

(264) Nolte IM, Spijker G, Boon M, et al. Haplotype Sharing Statistic: fine-mapping of disease

gene loci by comparing patients and controls for the length of haplotype sharing. In:

Nolte I.M. Statistics and population genetics of haplotype sharing as a tool for

fine-mapping of disease gene loci. Groningen, PhD thesis, 2003, 67-85.

(265) Te Meerman GJ, Van der Meulen MA, Sandkuijl LA. Perspectives of identity by descent

(IBD) mapping in founder populations. Clin Exp Allergy 1995; 25 Suppl 2:97-102.

(266) Van der Meulen MA, Te Meerman GJ. Association and haplotype sharing due to Identity

by Descent, with an application to genetic mapping. In: Pawlowitzki IH, Edwards JH,

Thompson E (eds), editors. Genetic mapping of disease genes. London: Academic Press,

1997: 115-135.

(267) Van der Meulen MA, te Meerman GJ. Haplotype sharing analysis in affected individuals

from nuclear families with at least one affected offspring. Genet Epidemiol 1997; 14(6):

915-920.

(268) Kawakami T, Okamoto K, Sugihara H, Hattori T, Reeve AE, Ogawa O et al. The roles

of supernumerical X chromosomes and XIST expression in testicular germ cell tumors.

J Urol 2003; 169(4):1546-1552.

(269) Ross MT, Grafham DV, Coffey AJ, Scherer S, McLay K, Muzny D et al. The DNA sequence

of the human X chromosome. Nature 2005; 434(7031):325-337.

(270) Honrado E, Benitez J, Palacios J. The molecular pathology of hereditary breast cancer:

genetic testing and therapeutic implications. Mod Pathol 2005; 18(10):1305-1320.

(271) Bleiker EM, Hahn DE, Aaronson NK. Psychosocial issues in cancer genetics-current status

and future directions. Acta Oncol 2003; 42(4):276-286.

(272) Julian-Reynier CM, Bouchard LJ, Evans DG, Eisinger FA, Foulkes WD, Kerr B et al.

Women’s attitudes toward preventive strategies for hereditary breast or ovarian

carcinoma differ from one country to another: differences among English, French, and

Canadian women. Cancer 2001; 92(4):959-968.


(273) Geller G, Doksum T, Bernhardt BA, Metz SA. Participation in breast cancer susceptibility

testing protocols: influence of recruitment source, altruism, and family involvement on

women’s decisions. Cancer Epidemiol Biomarkers Prev 1999; 8(4 Pt 2):377-383.

(274) Craufurd D, Dodge A, Kerzin-Storrar L, Harris R. Uptake of presymptomatic predictive

testing for Huntington’s disease. Lancet 1989; 2(8663):603-605.

(275) Miesfeldt S, Jones SM, Cohn W, Lippert M, Haden K, Turner BL et al. Men’s attitudes

regarding genetic testing for hereditary prostate cancer risk. Urology 2000; 55(1):46-50.

(276) Bottorff JL, Ratner PA, Balneaves LG, Richardson CG, McCullum M, Hack T et al.

Women’s interest in genetic testing for breast cancer risk: the influence of

sociodemographics and knowledge. Cancer Epidemiol Biomarkers Prev 2002; 11(1):89-95.

(277) de Wert G. Ethics of predictive DNA-testing for hereditary breast and ovarian cancer.

Patient Educ Couns 1998; 35(1):43-52.

(278) Pasacreta JV. Psychosocial issues associated with genetic testing for breast and ovarian

cancer risk: an integrative review. Cancer Invest 2003; 21(4):588-623.

(279) Bluman LG, Rimer BK, Regan SK, Lancaster J, Clark S, Borstelmann N et al. Attitudes,

knowledge, risk perceptions and decision-making among women with breast and/or

ovarian cancer considering testing for BRCA1 and BRCA2 and their spouses.

Psychooncology 2003; 12(5):410-427.

(280) Shiloh S, Ilan S. To test or not to test? Moderators of the relationship between risk

perceptions and interest in predictive genetic testing. J Behav Med 2005; 28(5):467-479.

(281) Guillem JG, Wood WC, Moley JF, Berchuck A, Karlan BY, Mutch DG et al. ASCO/SSO

review of current role of risk-reducing surgery in common hereditary cancer syndromes.

J Clin Oncol 2006; 24(28):4642-4660.

(282) Kasparian NA, Meiser B, Butow PN, Soames Job RF, Mann GJ. Anticipated uptake

of genetic testing for familial melanoma in an Australian sample: An exploratory study.

Psychooncology 2007; 16(1):69-78.

(283) Riedijk SR, de Snoo FA, van Dijk S, Bergman W, van Haeringen A, Silberg S et al.

Hereditary melanoma and predictive genetic testing: why not? Psychooncology 2005;

14(9):738-745.

(284) Cormier L, Valeri A, Azzouzi R, Fournier G, Cussenot O, Berthon P et al. Worry and

attitude of men in at-risk families for prostate cancer about genetic susceptibility and

genetic testing. Prostate 2002; 51(4):276-285.

(285) Peters JA, Vadaparampil ST, Kramer J, Moser RP, Court LJ, Loud J et al. Familial testicular

cancer: interest in genetic testing among high-risk family members. Genet Med 2006;

8(12):760-770.

(286) Bruno M, Tommasi S, Stea B, Quaranta M, Schittulli F, Mastropasqua A et al. Awareness

of breast cancer genetics and interest in predictive genetic testing: a survey of a

southern Italian population. Ann Oncol 2004; 15 Suppl 1:I48-I54.

(287) Jager M, Oosterwijk B, Middel B, Hoekstra-Weebers JEHM. Motivatie voor genetische

counseling en angst voor kanker bij een verhoogd familair risico op erfelijke borst- en

eierstokkanker.(motivations regarding genetic counseling and fear for cancer in women

having an increased risk for familial breast and ovarian cancer). Gedrag en Gezondheid

2004; 32:18-27.

(288) Cohen J. Statistical power analysis for the behavioral sciences. Erlbaum, Hillsdale,

Nj 1988.

(289) Cappelli M, Surh L, Humphreys L, Verma S, Logan D, Hunter A et al. Psychological and

social determinants of women’s decisions to undergo genetic counseling and testing for

breast cancer. Clin Genet 1999; 55(6):419-430.

(290) Lerman C, Narod S, Schulman K, Hughes C, Gomez-Caminero A, Bonney G et al. BRCA1

testing in families with hereditary breast-ovarian cancer. A prospective study of patient

decision making and outcomes. JAMA 1996; 275(24):1885-1892.

(291) Tijmstra TJ. The imperative character of medical technology and the significance of

‘anticipated decision regret’. Ned Tijdschr Geneeskd 1987; 131(26):1128-1131.

(292) Watson E, Hewitson P, Brett J, Bukach C, Evans R, Edwards A et al. Informed decision

making and prostate specific antigen (PSA) testing for prostate cancer: a randomised

controlled trial exploring the impact of a brief patient decision aid on men’s knowledge,

attitudes and intention to be tested. Patient Educ Couns 2006; 63(3):367-379.

(293) Evans R, Edwards A, Brett J, Bradburn M, Watson E, Austoker J et al. Reduction in

uptake of PSA tests following decision aids: systematic review of current aids and their

evaluations. Patient Educ Couns 2005; 58(1):13-26.

(294) Bleiker EM, Aaronson NK, Menko FH, Hahn DE, van Asperen CJ, Rutgers EJ et al. Genetic

counseling for hereditary cancer: a pilot study on experiences of patients and family

members. Patient Educ Couns 1997; 32(1-2):107-116.

(295) Harel A, Abuelo D, Kazura A. Adolescents and genetic testing: what do they think about

it? J Adolesc Health 2003; 33(6):489-494.

(296) van Dijk MR, Steyerberg EW, Habbema JD. Survival of non-seminomatous germ cell

cancer patients according to the IGCC classification: An update based on meta-analysis.

Eur J Cancer 2006; 42(7):820-826.

(297) Austoker J. Screening for ovarian, prostatic, and testicular cancers. BMJ 1994;

309(6950):315-320.

(298) Moiseyenko VM, Danilov AO, Baldueva IA, Danilova AB, Tyukavina NV, Larin SS et al.

Phase I/II trial of gene therapy with autologous tumor cells modified with tag7/

PGRP-S gene in patients with disseminated solid tumors: miscellaneous tumors.

Ann Oncol 2005; 16(1):162-168.

(299) Risch N, Merikangas K. The future of genetic studies of complex human diseases.

Science 1996; 273(5281):1516-1517.

(300) Youngren KK, Nadeau JH, Matin A. Testicular cancer susceptibility in the 129.MOLF-Chr19

mouse strain: additive effects, gene interactions and epigenetic modifications.

Hum Mol Genet 2003; 12(4):389-398.


Addendum

Addendum158 Genetic Predisposition to Testicular Cancer 159

Genetische termen

DNA

Deoxyribonucleic acid ofwel DNA is een dubbele helix van nucleotiden (ook wel basen

genoemd), die alle genetische informatie van een cel bevat. Er bestaan vier nucleotiden die

worden aangeduid met A, C, G en T.

Chromosoom

Een chromosoom is een compact DNA-molecuul in associatie met eiwitten. Elk mens heeft 46

chromosomen, om precies te zijn, 23 chromosomenparen. Van elk paar is er één geërfd van de

vader en één van de moeder.

Genoom

Het genoom is de verzameling van chromosomen van een organisme.

Merker

Een merker is een locatie op het genoom (d.i. locus) die varieert tussen mensen. Alle mensen

hebben voor ~99,9% hetzelfde DNA. Verschillen tussen mensen worden bepaald door ~0,1%

van het DNA. Aangezien het genoom bestaat uit zo’n 3,5×109 baseparen (bp), betekent dit nog

steeds ongeveer 3,5×106 verschillen. Wij werken in ons laboratorium hoofdzakelijk met twee

verschillende soorten merkers: microsatellietmerkers en single nucleotide polymorfismes (SNPs).

Een microsatelliet is een polymorf locus met een DNA-volgorde bestaande uit een variabel aantal

van oligonucleotide (di-, tri-, tetra- of pentanucleotide) eenheden. Een SNP is een locus van één

nucleoide dat kan variëren. (voor een betekenis van een polymorfisme, zie mutatie)

Allel

Een allel is een specifieke vorm van een merker. Bijvoorbeeld, voor een microsatelliet-merker

duidt men het allel aan met het aantal herhaalde oligonucleotide-eenheden.

Haplotype

Een haplotype is een combinatie van allelen voor meerdere merkers op één chromosoom.

Genotype

Het genotype is de genetische samenstelling van allelen op één of meerdere loci voor beide

chromosomen uit een chromosomenpaar.

Addendum - Verklarende woordenlijst

Addendum160 Genetic Predisposition to Testicular Cancer 161

Fenotype

Het fenotype is de uiterlijke verschijning van een ziekte of een specifiek kenmerk, d.w.z. de

fysieke expressie van genen / allelen). Dit in tegenstelling tot het genotype.

Fase van haplotypes

Fase voor haplotypes afleiden wil zeggen dat bepaald wordt welke allelen bij elkaar op hetzelfde

chromosoom liggen. Omdat iedereen twee allelen per merker heeft en het laboratorium niet kan

zeggen welke allelen voor naburige merkers bij elkaar horen op een chromosoom, zijn er voor

twee merkers twee mogelijke combinaties. Door te kijken welke allelen in de ouders aanwezig

zijn, kan worden bepaald welke allelen bij elkaar op hetzelfde chromosoom liggen.

Heterozygositeit of informativiteit

Personen die op een merkerlocus twee dezelfde allelen hebben, worden homozygoot genoemd.

Als een persoon twee verschillende allelen heeft, dan is hij heterozygoot. De heterozygositeit

van een merker geeft aan welk percentage van de mensen heterozygoot is.

Mutatie

Een mutatie is een verandering in het DNA die bijvoorbeeld een ziekte of afwijking tot gevolg

heeft. Als zo’n verandering geen aanwijsbare invloed heeft op het fenotype, dan wordt hij

‘neutraal’ genoemd en heet hij een polymorfisme of (genetische) variant.

Recombinatie

Een recombinatie beschrijft het verschijnsel dat, tijdens het ontstaan van geslachtscellen (d.i. een

meiose), de twee chromosomen van een ouder stukken met elkaar uitwisselen zodat allelen, die

bij de ouder nog op twee veschillende chromosomen aanwezig waren, bij het kind op hetzelfde

chromosoom terechtkomen. Per meiose komt een recombinatie redelijk vaak voor. De lengte van

een gebied, dat 1% kans heeft op een recombinatie, wordt centiMorgan (cM) genoemd.

Linkage disequilibrium

Linkage disequilibrium wil zeggen dat allelen op verschillende loci niet onafhankelijk van elkaar

overerven. Als gevolg daarvan kunnen ze statistisch met elkaar gecorreleerd zijn. Dit ontstaat

doordat er geen of weinig recombinaties optreden tussen loci die dicht bij elkaar op het

chromosoom liggen.

Penetrantie

De penetrantie van een mutatie is de kans dat iemand die deze mutatie draagt daadwerkelijk

ziek wordt.

Fenokopie

Met een fenokopie wordt iemand bedoeld die niet de benodigde mutatie(s) draagt, maar die

toch de ziekte of een kenmerk heeft.

Statistische termen

Type I fout

De type I fout geeft bij het testen van een nul-hypothese tegen een alternatieve hypothese de

kans weer dat de nul-hypothese onterecht wordt verworpen.

Onderscheidingsvermogen of power

Het onderscheidingsvermogen ofwel power geeft de kans weer dat terecht de nul-hypothese

wordt verworpen.

Associatieanalyse

Associatieanalyse vergelijkt voor één of meerdere merkers de frequenties van de verschillende

allelen, genotypes of haplotypes tussen onafhankelijke patiënten en controlepersonen. Als een

merker het ziektegen is of in linkage disequilibrium met het ziektegen is, wordt verwacht dat het

ook geassocieerde allel, genotype of haplotype in een verhoogde frequentie aanwezig is in een

steekproef van patiënten vergeleken met een steekproef van controlepersonen. Het kan echter

ook zijn dat de aanwezigheid van een bepaald allel beschermt tegen het krijgen van een ziekte.

Zo’n beschermend of protectief allel zal dan een lagere frequentie hebben in een steekproef van

patiënten in vergelijking tot een steekproef van controlepersonen.

Transmission/disequilibrium test

De transmissie/disequilibrium test bekijkt trio’s bestaande uit een patiënt met beide ouders en

analyseert of ouders een bepaald allel van een merker vaker doorgeven aan hun aangedane kind

dan andere allelen. Voor een merker die niet gelinkt (d.w.z. gekoppeld) is met de te onderzoeken

ziekte, wordt verwacht dat beide allelen, aanwezig in de ouder van de patiënt, even veel kans

(d.i. 50%) hebben om doorgegeven te worden. Beide allelen moet wel verschillend zijn, d.w.z.

de ouder moet heterozygoot zijn.

Hardy-Weinberg equilibrium test

Het Hardy-Weinberg evenwicht geeft weer hoeveel homo- en heterozygoten verwacht worden

in de populatie. Bijvoorbeeld, als een allel een frequentie heeft van 10% in de populatie,

dan verwacht je 10%×10% =1% homozygote personen voor dit allel en 2×10%×90%=18%

heterozygote personen in de populatie.

In ons laboratorium wordt voor elke merker getest of het Hardy-Weinberg evenwicht aanwezig

is. Een afwijking van het evenwicht duidt meestal op een systematische fout in het scoren van

de genotypes. Zoals wij de test gebruiken, is het dus een test om de kwaliteit van de geleverde

genotype-data te checken.

Overgenomen uit: Statistics and population genetics of haplotype sharing as a tool for fine-

mapping of disease gene loci. Proefschrift Dr. I.M. Nolte, Groningen 2003.


Dankwoord

Dankwoord164 Genetic Predisposition to Testicular Cancer 165

Dankwoord

Dit onderzoek was niet mogelijk geweest zonder de bereidheid van alle patiënten en controle

personen om DNA-materiaal af te staan. Ik wil mijn oprechte waardering hiervoor uitspreken.

Het onderzoek dat in dit proefschrift wordt beschreven werd uitgevoerd bij de afdelingen

Chirurgische Oncologie van het Universitair Medisch Centrum Groningen, in samenwerking met

de afdeling Genetica, Medische Oncologie, Medische Biologie, Pathologie, Wenckebach Instituut

en het Integraal Kankercentrum Noord-Nederland(IKN). Velen hebben mij in staat gesteld mijn

promotie-onderzoek tot een eind te brengen. Iedereen die hieraan heeft bijgedragen wil ik

hartelijk bedanken voor zijn of haar inzet. Natuurlijk zal ik een aantal mensen in het bijzonder

noemen.

Graag wil ik de promotores Prof. dr. H.J. Hoekstra en Prof. dr. D.Th. Sleijfer en de copromotores

Dr. R.H. Sijmons en Dr. J.E.H.M. Hoekstra-Weebers bedanken voor het in mij gestelde vertrouwen,

de steun en begeleiding.

Prof. dr. H.J. Hoekstra, Beste Harald, drijvende kracht achter het zaadbalkanker onderzoek.

Jij bood mij de kans om promotie-onderzoek te verrichten. Je enthousiasme, begeleiding en

laagdrempelige mogelijkheden voor overleg zijn van grote waarde. De inzet waarmee jij op

persoonlijke wijze (vele) onderzoekers begeleid is erg motiverend en lovenswaardig. Bovendien

beschik jij over een vermogen om grote vaart in de wetenschap te houden waardoor de

onderzoeksmolen blijft draaien en mijn enthousiasme alleen maar groter werd. Harald ik heb de

afgelopen jaren veel van je geleerd en ik ben je hiervoor zeer dankbaar.

Prof. dr. D.Th. Sleijfer, Beste Dirk, veel heb ik geleerd van je gestructureerde en kritische

werkwijze. Manuscripten kwamen vaak zeer snel gecorrigeerd retour met zeer verhelderend

commentaar. Bovendien tipte je mij geregeld als er weer een “interessant” stuk was gepubliceerd

over zaadbalkanker. Je kritische blik is van belangrijke waarde geweest voor de totstandkoming

van mijn proefschrift. Ik wil je hiervoor bedanken, maar ook voor de bemoedigende woorden

die af en toe nodig waren.

Dr. R.H. Sijmons, Beste Rolf, jij hebt me kennis laten maken met de “ins en outs” van de

Genetica. Het enthousiasme waarmee jij je kennis overbrengt in heldere begrijpelijke taal is

bewonderenswaardig. Voor overleg was je altijd bereikbaar en veel tijd hebben we samen

doorgebracht (in jullie noodgebouw waarin je veel te lang was gehuisvest) bij de totstandkoming

van de manuscripten en het kritisch analyseren van gepubliceerde artikelen. Wanneer ik iets

“definitiefs” inleverde ging jij er nog even “met het stof kammetje door”, wat resulteerde in

fraaie zinnen met mooie nuanceringen. Je verhelderende benaderingswijze en sturende inbreng

heeft grote invloed gehad op de totstandkoming van dit proefschrift. Ik ben je voor dit alles

zeer dankbaar.

Dankwoord166 Genetic Predisposition to Testicular Cancer 167

Dr. J.E.H.M. Hoekstra-Weebers, Beste Josette, Het was voor mij natuurlijk ideaal om een

“echtpaar” als onderzoeksbegeleiders te hebben. Met één telefoontje naar jullie huis, had ik

jullie beiden aan de telefoon en kon ik “dingen” regelen of vragen, 24 uur per dag, 7 dagen per

week. Jij hebt me veel geleerd over “kwaliteit van leven” en “psychosociaal” onderzoek en het

ontwerpen en verwerken van vragenlijsten. In het beoordelen van mijn laatste manuscript was

je kritisch en het heeft me veel moeite gekost om aan je eisen te voldoen maar je bleef geduldig

en nam ruim de tijd voor heldere uitleg. Psychosociaal onderzoek was volledig nieuw voor mij

en ik ben blij dat ik hier ervaring mee heb opgedaan. Ik heb hier veel van geleerd en ik wil je

hiervoor hartelijk bedanken.

De leden van de beoordelingscommissie, Prof. dr. S. Horenblas, Prof. dr. J.W. Oosterhuis en

Prof. dr. P.H.B. Willemse wil ik bedanken voor de beoordeling van het manuscript.

Daanaast ben ik Dr. I.M. Nolte veel dank verschuldigd voor de hulp bij de vele statistiek. Beste

Ilja, genetische statistiek is een vak appart en ik ben je zeer dankbaar voor je vele hulp hierbij.

Je hebt veel tijd genomen om mij uitleg te geven en je hebt de manuscripten kritisch beoordeeld

en de statistiek geweldig verzorgd.

Het laboratorium onderzoek heb ik kunnen uitvoeren met hulp van medewerkers op het lab van

de afdeling Medische Biologie van het Universitair Medisch Centrum Groningen. In het bijzonder

wil ik hier bedanken Dr. ir. G.J. te Meerman, Marcel Bruinenberg en Gerrit van der Steege.

“Mijn onderzoeksgroep / kamergenoten / collega-onderzoekers”: Imi Veldman, Joke Fleer,

Eric Sonneveld, Rudi Komdeur, Marrit Tuinman en David Cobben. Ik ben blij dat ik jullie heb

leren kennen. We hebben veel tijd doorgebracht in onze kamer / op de gang / tijdens de koffie.

Ik heb veel van jullie geleerd (en jullie hadden meer discipline dan ik en dat benijd ik). Gelukkig

was er ook tijd voor ontspanning en “social talk” veel dank hiervoor.

Mijn paranimfen: Stephan Lutke Holzik, mijn broertje. Vaak vroeg je wanneer ik nou eens klaar

zou zijn met mijn onderzoek. Welnu, dat stadium is bereikt. Ik ben blij en trots dat je de taak

van paranimf op je wilt nemen. Deepu Daryanani, collega en vriend, jij weet ook wat het is om

onderzoek en opleiding te combineren. Als collega’s hebben we veel contact en we zijn goede

vrienden geworden, veel dank dat je mijn paranimf wilt zijn.

Mijn ouders, pap en mam, jullie stonden garant voor een onbezorgde jeugd en jullie hebben mij

alle kansen gegeven om mijzelf te ontwikkelen tot wie ik nu ben. Jullie onvoorwaardelijke steun

en interesse zijn onmisbaar, daarom draag ik dit boekje ook aan jullie op.

Lieve Marjolijn, ik vind het echt geweldig dat jij mij de ruimte hebt gegeven om in Groningen

te promoveren terwijl jij daar de nodige kilometers voor hebt moeten reizen. Jij hebt mij altijd

weten aan te moedigen en je stimuleert me om van elke situatie iets te leren en iets positiefs

mee te nemen. Bedankt voor je steun en liefde en de allergrootste vreugde die je ons hebt

gegeven met Louise. Lieve Louise, je bent zo vrolijk en nog zo klein, jij laat zien waar het in het

leven om gaat. Marjolijn en Louise dit is ook jullie boekje!


List of Publications

List of Publications170 Genetic Predisposition to Testicular Cancer 171

List of publications

Lutke Holzik MF, Hoekstra HJ, Sijmons RH, Sleijfer DTh, Fleer J, Hoekstra-Weebers JEHM.

Interest in and motivations regarding genetic testing for testicular germ cell tumour

susceptibility, Submitted 2007

Lutke Holzik MF, Hoekstra HJ, Sijmons RH, Sonneveld DJ, van der Steege G, Sleijfer DTh, Nolte IM.

Re-analysis of the Xq27-Xq28 region suggests a weak association of an X-linked gene with

sporadic testicular germ cell tumour without cryptorchidism.

Eur J Cancer. 2006 Aug; 42(12):1869-74.

Nuver J, Lutke Holzik MF, van Zweeden M, Hoekstra HJ, Meijer C, Suurmeijer AJ, Groen HJ,

Hofstra RM, Sluiter WJ, Groen H, Sleijfer DTh, Gietema JA.

Genetic variation in the bleomycin hydrolase gene and bleomycin-induced pulmonary toxicity

in germ cell cancer patients.

Pharmacogenet Genomics. 2005 Jun; 15(6):399-405.

Lutke Holzik MF, Storm K, Sijmons RH, D’hollander M, Arts EG, Verstraaten ML, Sleijfer DTh,

Hoekstra HJ.

Absence of constitutional Y chromosome AZF deletions in patients with testicular germ cell

tumors.

Urology. 2005 Jan; 65(1):196-201

Lutke Holzik MF, Rapley EA, Hoekstra HJ, Sleijfer DTh, Nolte IM, Sijmons RH.

Genetic predisposition to testicular germ-cell tumours.

Lancet Oncol. 2004 Jun; 5(6):363-71.

Lutke Holzik MF

A Non-Germ Cell Malignancy in a Recurrent Retroperitoneal Tumor Mass After Combined

Treatment for a Nonseminomatous Testicular GermCell Tumor

The American Journal of Urology Review. 2003 Nov/Dec; 1(6):2-5

Lutke Holzik MF, Hoekstra HJ, Mulder NH, Suurmeijer AJ, Sleijfer DTh, Gietema JA.

Non-germ cell malignancy in residual or recurrent mass after chemotherapy for

nonseminomatous testicular germ cell tumor.

Ann Surg Oncol. 2003 Mar; 10(2):131-5.

List of Publications172 Genetic Predisposition to Testicular Cancer 173

Lutke Holzik MF, Sijmons RH, Sleijfer DTh, Sonneveld DJ, Hoekstra-Weebers JEHM,

van Echten-Arends J, Hoekstra HJ.

Syndromic aspects of testicular carcinoma.

Cancer. 2003 Feb 15; 97(4):984-92.

Fleer J, Hoekstra HJ, Sleijfer DTh, Lutke Holzik MF, Hoekstra-Weebers JEHM.

Postmortem diagnosis of testicular cancer.

Lancet. 2002 Nov 9; 360(9344):1511-2 (letter)

Sonneveld DJ, Lutke Holzik MF,Nolte IM, Sleijfer DTh, van der Graaf WT,

Bruinenberg M, Sijmons RH, Hoekstra HJ, Te Meerman GJ.

Testicular carcinoma and HLA Class II genes.

Cancer. 2002 Nov 1; 95(9):1857-63.

Lutke Holzik MF, Sonneveld DJ, Hoekstra HJ, te Meerman GJ, Sleijfer DTh,

Schaapveld M.

Do the eastern and northern parts of The Netherlands differ in testicular cancer? (letter)

Urology. 2001 Oct; 58(4):636-7

Lutke Holzik MF, Mastboom, WJB

Galsteenileus. Nederlands tijdschrift voor geneeskunde (studenten editie). 2000; (3):48-50


Curriculum Vitae

Curriculum Vitae176 Genetic Predisposition to Testicular Cancer 177

Curriculum Vitae

Martijn Frederik Lutke Holzik werd op 25 februari 1974 te Enschede geboren. Hij bezocht het

Jacobus College in Enschede waar hij in 1991 zijn HAVO diploma en in 1993 zijn VWO diploma

behaalde. Aansluitend studeerde hij geneeskunde aan de Rijksuniversiteit Groningen. Na het

doctoraal examen in 1998 vonden de co-schappen plaats in het Medisch Spectrum Twente

te Enschede. In februari 2000 behaalde hij (cum laude) zijn artsen titel. Aansluitend werd hij

arts-assistent chirurgie in het Sophia ziekenhuis locatie Weezenlanden te Zwolle. Medio 2000

werd hij voor de opleiding chirurgie aangenomen. Voorafgaand aan zijn opleiding chirurgie

startte hij vanaf eind 2000 als arts-onderzoeker bij de afdeling chirurgische oncologie in het

Academisch Ziekenhuis Groningen (thans Universitair Medisch Centrum Groningen). Gedurende

ruim anderhalf jaar werd de basis voor dit proefschrift gelegd. Eind 2002 begon hij de opleiding

tot chirurg in het Universitair Medisch Centrum Groningen (Prof. Dr. H.J. ten Duis) en in 2004

werd dit vervolgd in het Medisch Spectrum Twente te Enschede (Dr. W.J.B. Mastboom) alwaar hij

zijn opleiding ook zal afronden. Martijn is getrouwd met Marjolijn en hun dochter Louise werd

14 januari 2007 geboren.

university of groningen genetic predisposition to ... · contains trophoblastic giant cells), yolk...

Documents