epidemiological, clinical and genetic aspects of ...jultika.oulu.fi/files/isbn951425418x.pdf ·...

EPIDEMIOLOGICAL, CLINICAL AND GENETIC ASPECTS OF NEUROFIBROMATOSES IN NORTHERN FINLAND

MINNAPÖYHÖNEN

Department of Clinical Genetics

OULU 1999

OULUN YLIOPISTO, OULU 1999

EPIDEMIOLOGICAL, CLINICAL AND GENETIC ASPECTS OF NEUROFIBROMATOSES IN NORTHERN FINLAND

MINNA PÖYHÖNEN

Academic Dissertation to be presented with the assent of the Faculty of Medicine, University of Oulu, for public discussion in Auditorium 6 of the University Hospital of Oulu, on November 12th, 1999, at 12 noon.

Copyright © 1999Oulu University Library, 1999

OULU UNIVERSITY LIBRARYOULU 1999

ALSO AVAILABLE IN PRINTED FORMAT

Manuscript received 11 October 1999Accepted 15 October 1999

Communicated by Docent Jaakko Ignatius Professor Juha Peltonen

ISBN 951-42-5418-X(URL: http://herkules.oulu.fi/isbn951425418X/)

ISBN 951-42-5417-1ISSN 0355-3221 (URL: http://herkules.oulu.fi/issn03553221/)

4

To neurofibromatoses patientsand their families

6

Pöyhönen, Minna, Epidemiological, Clinical and Genetic Aspects of Neurofibromatoses inNorthern FinlandDepartment of Clinical Genetics, University of Oulu and Oulu University Hospital, FIN-90220 Oulu,Finland and Department of Medical Genetics, Väestöliitto, The Family Federation of Finland,P.O.Box 849, FIN-00101 Helsinki, Finland1999Oulu, Finland(Manuscript received 11 October 1999)

Abstract

A population-based study to investigate the epidemiological, genetic and clinical features ofneurofibromatoses (NF) in Northern Finland was carried out between 1989-1996. The area concernedwas that served by Oulu University Hospital, with a total population of 733 037.

A total of 197 patients with neurofibromatosis type 1 (NF1), five with neurofibromatosis type 2(NF2) and eight with segmental neurofibromatosis (NF5) fulfilling the diagnostic criteria wereidentified among several hundred patients examined on account of a possible NF diagnosis. The 197NF1 patients came from 119 families. 77 of these cases were sporadic, 117 familial, and three weremothers of children suffering from NF1 who were themselves diagnosed as having segmental NF. Themale/female ratio was 0.93 (95 males and 105 females). The geographical distribution of the patientsroughly corresponded to that of the general population in the area. The overall prevalence of NF1 was23/100 000, with a peak prevalence of 34/100 000 in the age group 10 to 19 years. The overall birthincidence of NF1 was estimated to be 27/100 000, with the highest figure, 37/100 000, recorded in thesix–year period 1990-1995. The mean age at the time of diagnosis was 20 ± 16 years in the wholepopulation and 6 ± 4 years in the children born in the 1980’s.

A new mutation was suspected in 49% of the NF1 cases (96/197), and a mutation rate of 4.37 ±0.72 x 10-5 was obtained for the period 1960-1995. The relative fitness of the NF1 patients was 0.48,being reduced more in the affected males (0.24) than in the females (0.72). The mean maternal andpaternal ages of the sporadic patients were 30 ± 6 and 33 ± 6 years, respectively, which is significantlyhigher than in the general population. Two cases with a deletion of the NF1 gene were identified, oneencompassing the loci from EVI-20 to INT-38 and the other the INT-27 locus, representing 3% of the66 cases analysed. In seven familial cases the parental origin of the new mutation could be verified andlinkage studies showed that the oldest affected individual in the family had inherited the mutation fromthe father in 6/7 cases.

In one family seven members in three generations were affected with a rare spinalneurofibromatosis, and a linkage to the NF1 gene was shown. Of these seven patients, four areincluded among the 197 studied here while the other three lived outside the area.

The diagnostic features of the 164 NF1 patients aged from three months to 73 years who wereexamined clinically included café au lait spots (CLS) in 96% of cases, freckles in 87%, neurofibromasin 69%, plexiform neurofibromas in 20%, Lisch nodules in 70%, optic glioma (asymptomatic) in 20%and pseudarthrosis in 3%. 56% of the cases had an affected first degree relative. A plexiformneurofibroma was diagnosed in 33 individuals and this became a malignant peripheral nerve sheathtumour (MPNST) during the seven years of monitoring in 15% of cases (5/33). Hyperintense T

2lesions in a MRI scan of the brain were found in 94 % of the children under the age of six years whohad had such a scan (n=17) and in 84% of those under 16 years (n=50).

Symptoms related to NF1 which needed medical intervention, rehabilitation or follow-up werediagnosed in about 2/3 of the cases, and in 38% of cases such medical problems of this kind had beentreated before NF1 was actually diagnosed. All these findings emphasise the need for amultidisciplinary approach to the follow-up of neurofibromatoses.

Keywords: NF1, population-based, prevalence, incidence

8

Acknowledgements

This research was carried out during the years 1989-1999 at the Department of ClinicalGenetics , Oulu University Hospital and University of Oulu, and at the Department ofMedical Genetics, Väestöliitto, the Family Federation of Finland.

I wish to express my sincere gratitude to the head of the Department of Clinical Genetics,Professor Jaakko Leisti, M.D., Ph.D., my supervisor, for introducing me to the fascinatingfield of clinical genetics. I also owe my gratitude to him for his support during these years.

I would like to express my deep gratitude to the head of Department of Medical Genetics,Docent Helena Kääriäinen, M.D., for providing me with excellent working facilities in herdepartment. She has always had the energy and time for discussions.

I wish to acknowledge most sincerely Docent Jaakko Ignatius, M.D. and Professor JuhaPeltonen, M.D., Ph.D., for their constructive, valuable and supportive criticism and manyinspiring discussions during the preparation of this thesis.

I am deeply grateful to my co-author Docent Riitta Herva, M.D., for her never-failinggenuine and supportive interest in this work. Her comments have been most valuable. It hasbeen a pleasure to work with her.

I extend my warmest gratitude to my co-authors Docent Eila Mustonen, M.D., EmeritusProfessor Erkki Heikkinen, M.D., Docent Soili Kytölä, Ph.D., Dr. Eeva-Liisa Leisti, Dr. UlpuSeppänen, M.D., Dr. Pentti Kinnunen, M.D., and Dr. Solja Niemelä, for their fruitful co-operation during this work.

I am grateful to the staff of the Department of Clinical Genetics for their kind help in thecourse of this study, especially to genetic nurse Riitta Mattlar and head nurse Liisa Ukkola,M.Sc., who between them arranged the examinations and collected the blood samples fromthe patients and their families and have been taking care of the families for many years. I wishto thank Dr. Aki Mustonen, M.D., Dr. Leila Pajunen, M.D., Ph.D., Dr. Outi Vierimaa, Dr. TuijaLöppönen, M.D., Dr. Sirpa Kivirkko, M.D., Ph.D., Dr. Pia Huusko, M.D., Docent MarkettaKähkönen, Ph.D., Ms. Hanna Kokkonen, M.Sc., Ms. Marja-Leena Väisänen, Ph.D., Ms.Ritva Haataja, M.Sc., Ms. Kyllikki Haapala, M.Sc., Docent Robert Winqvist, Ph.D., Ms. VirpiLaunonen, Ph.D. and Mr. Arto Mannermaa, Ph.D., for creating a friendly atmosphere overthese years.

I wish to thank all my colleagues in Northern Finland who have helped me in collectingthe series of patients. Without their support this work would not have been possible.

I sincerely thank all the patients and their families who have participated in the study.

9

They have been my best motivation during this work.I wish to express my particular gratitude to Dr. Susan Huson, M.D., Department of

Clinical Genetics, Oxford Radcliffe Hospital, Professor Birgitta Rembeck, M.D., Departmentof Psychiatry, Sahlgrenska University Hospital, Gothenburg, Dr. Markku Sainio, Departmentof Occupational Medicine, Finnish Institute of Occupational Health, Helsinki, and geneticnurse Patricia Birch, Department of Medical Genetics, University of British Columbia,Vancouver, for their encouragement, many supportive discussions and friendship duringthis work.

I warmly thank the staff of the Department of Medical Genetics, Väestöliitto, for theirkind understanding towards this work. Especially, I would like to thank Dr. Mirja Somer, M.D.for her encouragement and always positive attitude towards this work.

I am grateful to Mr. Malcolm Hicks, M.A., who has skilfully revised the language of thismanuscript and original papers numbers I, II, III and VI, Mr. Anthony Heape, M.Sc., forrevising the language of original paper number IV, and Ms. Sirkka-Liisa Leinonen, Lic.Phil.,for revising the language of original paper number V.

I wish to thank all my friends, especially colleagues Dr. Kati Juva, M.D., Professor HannuKalimo, M.D., Ph.D. and Professor Matti Viitanen, M.D., for their positive support duringthese years and Ms. Leena Ruohola for her valuable help with technical problems.

Most of all I would warmly like to thank my family for their unfailing support,understanding and caring during this research.

This work has received financial support from the Rinnekoti Foundation, the Alma andK.A. Snellman Foundation, and Oulu University Hospital.

Helsinki, November 1999

Minna Pöyhönen

10

Abbreviations

CLS café au lait spotsCNS central nervous systemCPT congenital pseudarthrosis of the tibiaCT computed tomographyDNA deoxyribonucleic acidERM Ezrin-Radixin-Moesin proteinEVI2A ecotropic viral integration site 2AEVI2B ecotropic viral integration site 2BGAP GTPase-activating proteinGTP guanosine triphosphateJCML juvenile chronic myeloid leukaemiaKb kilobaseMIM Mendelian Inheritance in ManMPNST malignant peripheral nerve sheath tumourmRNA messenger ribonucleic acidMRI magnetic resonance imagingNF neurofibromatosisNF1 neurofibromatosis type 1NF2 neurofibromatosis type 2NF5 segmental neurofibromatosisOMGP oligodendrocyte-myelin glycoproteinPCR polymerase chain reactionRNA ribonucleic acidTBR translocation breakpoint region

12

List of original papers

The thesis is based on the following articles, referred to in the text by their Roman numerals:

I Poyhonen* M, Kytölä S & Leisti J (1999) Epidemiology of neurofibromatosistype 1 (NF1) in Northern Finland. J Med Genet (in press).

II Poyhonen* M A clinical assessment of neurofibromatosis type 1 (NF1) andsegmental NF (NF5) in Northern Finland. (Submitted)

III Poyhonen* M, Leisti E-L, Kytölä S & Leisti J (1997) Hereditary spinalneurofibromatosis: a rare form of NF1? J Med Genet 34: 184-187.

IV Poyhonen* Minna, Niemela Solja & Herva Riitta (1997) Risk of Malignancy andDeath in Neurofibromatosis. Arch Pathol Lab Med 121: 139-143.

V Mustonen Eila, Poyhonen* Minna & Leisti Eeva-Liisa (1997) Neuro-ophthalmological findings in neurofibromatosis. Clinical and neuroradiologicalstudy of 125 patients. Neuro-ophthalmol 17: 117-126.

VI Heikkinen Erkki S, Poyhonen* Minna H, Kinnunen Pentti K & Seppänen Ulpu I(1999) Congenital pseudarthrosis of the tibia. Treatment and outcome at skeletalmaturity in 10 children. Acta Orthop Scand 70: 275-282.

Some unpublished results will also be presented.

* Due to the different orthography of Finnish and English the author’s name in all theoriginal publications is written differently from that in the thesis.

13

Contents

AbstractAcknowledgementsAbbreviationsList of original papersContents1. Introduction ................................................................................................................... 152. Review of the literature ................................................................................................. 17

2.1. History of neurofibromatoses .............................................................................. 172.2. Current classification of neurofibromatoses ...................................................... 182.3. Neurofibromatosis type 1 ...................................................................................... 19

2.3.1. Diagnostic criteria ............................................................................... 192.3.2. Clinical manifestations ........................................................................ 202.3.3. Epidemiology and genetics ................................................................ 282.3.4. Pathophysiology ................................................................................. 31

2.4. Neurofibromatosis type 5 (Segmental type) ...................................................... 322.5. Neurofibromatosis type 2 ...................................................................................... 33

2.5.1. Diagnostic criteria ............................................................................... 332.5.2. Clinical manifestations ........................................................................ 342.5.3. Epidemiology and genetics ................................................................ 35

3. Aims of the research ..................................................................................................... 364. Patients and methods .................................................................................................... 37

4.1. Collection of patients ............................................................................................. 374.2. Diagnostic criteria and clinical examination ....................................................... 384.3. Laboratory examinations ....................................................................................... 394.4. Epidemiology, genetics and statistics ................................................................. 394.5. DNA studies ........................................................................................................... 40

5. Results ............................................................................................................................ 425.1. Neurofibromatosis type 1 ..................................................................................... 42

5.1.1. Population genetic findings of neurofibromatosis type 1 (I) ....... 425.1.2. Clinical findings regarding neurofibromatosis type 1 (II- VI) ....... 44

5.2. Neurofibromatosis type 5 (Segmental type) (II, IV, V) ...................................... 495.3. Neurofibromatosis type 2 (IV, V) ......................................................................... 49

14

6. Discussion ...................................................................................................................... 516.1.Population genetics of neurofibromatosis type 1 (I) ......................................... 516.2. Clinical findings in neurofibromatosis type 1 (II-VI) ......................................... 546.3. Neurofibromatosis type 2 (IV-V) .......................................................................... 576.4. Genetic counselling, care and follow-up of neurofibromatoses (II) ............... 57

7. Summary and conclusions ........................................................................................... 598. References ...................................................................................................................... 61

Appendix

15

1. Introduction

Phakomatoses or hamartomatoses are a group of hereditary neurocutaneous syndromesincluding neurofibromatoses, tuberous sclerosis and von Hippel-Lindau disease. They areinherited in an autosomal dominant fashion and each shows both genetic and clinicalheterogeneity. An important feature of all of them is the occurrence of benign tumours suchas angiofibromas, neurofibromas and hemangioblastomas, all of which have a risk of turningpremalignant. The genes involved belong to the class of tumour suppressor genes.

The first reports of neurofibromatoses were published 200 to 300 hundred years ago, butit was not until 1987 that a consensus was achieved over the diagnostic criteria forneurofibromatosis type 1 (NF1) and type 2 (NF2). The identification of the NF1 and NF2genes at 17q11.2 and 22q12.2, respectively, and characterisation of their mutations andproteins in the 1990’s further pointed the way to our present understanding of thesedisorders, but in spite of such achievements, the diagnosis of NF1 and NF2 is still basedmainly on clinical features.

NF1, the most common of the inherited neurocutaneous diseases, has been identifiedworldwide in different countries and ethnic groups. It is caused by a mutation in the NF1tumour suppressor gene, and is clinically characterised mainly by the presence of café au laitspots (CLS) and neurofibromas. Its estimated birth incidence is about 1/2500 and itsprevalence about 1/4000, with about two million affected individuals worldwide, 200 000 inEurope and more than one thousand in Finland. NF1 is a classical example of an autosomaldominantly inherited disorder showing wide phenotypic variability even in the same family.About 50% of cases are due to new mutations (Riccardi 1992, Huson 1994).

NF2 is a rarer type of NF with symptoms mainly characterised by bilateral vestibularschwannomas leading to vertigo, loss of balance and deafness. It is caused by a mutation inthe NF2 tumour suppressor gene. Its estimated birth incidence is about 1/33 000 andprevalence about 1/210 000, and thus there are less than fifty affected individuals in Finland(Evans et al. 1992a).

While the major diagnostic criteria for NF1 and NF2 have been agreed upon and themolecular basis of the disease is known, their pathogenesis and the mechanisms related totheir pleiotropy and clinical variation still remain mostly unknown. Also, much remains to belearned about the progression of neurofibromatoses, how to follow up the patients and howto deal with the medical problems.

In spite of actively ongoing research into NF, there are still only a few population-based

16

studies that investigate its clinical variation and genetic aspects, and only a small numberthat concentrate on the inheritance of the disease and its clinical picture based on asystematic study of the first degree relatives. Furthermore the possible role of moleculargenetic analysis in assisting normal clinical diagnosis has seldom been studied. Eventhough neurofibromatoses seem to be similar worldwide, Finnish NF families and theirpossible special clinical features and forms have not previously been studiedsystematically.

17

2. Review of the literature

2.1. History of neurofibromatoses

The first case report of NF1 is probably a description of a man suffering from anelephantiasis neuromatosa and short stature, as reported posthumously by Aldrovandi in1642. The first case report published in English was in 1768 by Akenside, who discovered aman with multiple small wens, or cysts, having a tendency to grow again on the site fromwhich they had been removed, a condition which was also seen in his father. Another 18th

century report was the case report of Johann Gottfried Rheinhard, also known as the “WartMan“ who had multiple fibrous tumours of the skin and whose “Case History ofExtraordinarily Unsightly Skin“ was published in 1793 by Ludvig and Tilesius (reviewed byMulvihill 1990a).

Neurofibromatosis type 1, also known as von Recklinghausen disease, is named afterFriederich Daniel von Recklinghausen (1833-1910), professor of pathology, who published amonograph “Über die multiplen Fibrome der Haut und ihre Beziehung zu den multiplenNeuromen“ (On Multiple Neurofibromas of the Skin and their Relationship to MultipleNeuromas) in Berlin in 1882. He was the first to use the term neurofibroma and to recognisethat skin and nerve tumours represented a mingling of both neural elements and connectivetissue cells (Crump 1981).

The most famous patient claimed to have had NF and often quoted in the literature is the“Elephant Man“, Joseph Merrick (1862-1890), who was thought by Weber in 1909 to havehad an incomplete and anomalous form of von Recklinghausen’s disease. Actually it seemscertain that he did not have NF1, but probably Proteus syndrome, a rare and highly variablehamartomatous disorder characterised mainly by asymmetrical overgrowth, exostoses andlarge soft-tissue and vascular tumours (Clark 1994, Seward 1994).

The familial occurrence of NF1 was probably first described by Akenside in 1768 andlater in 1847 by Virchow, who reported NF in multiple family members. It was not until 1918that Preiser and Davenport established the autosomal dominant inheritance in this disorder(Mulvihill 1988).

In contrast to the highly visible cutaneous manifestations of NF1, the clinicalmanifestations of NF2 are more subtle and were characterised in detail later. A unilateralvestibular schwannoma (acoustic neurinoma) was first reported by Eduard Sandifort in

18

1777, while Wishart reported in 1822 the first known detailed descriptions of bilateralvestibular schwannomas and of meningioma. Eighty years later, in 1903, Henneberg andKoch published a description of the intracranial variant of NF, considered to be the firstreport of NF2. The first known report describing the familial occurrence of bilateralvestibular schwannomas was published by Feiling and Ward in 1920, and ten years laterGardner and Frazier published their famous NF2 family with 217 individuals in fivegenerations and 38 cases of bilateral vestibular schwannomas (Gardner & Frazier 1930, Ahnet al. 1996).

As far as Finland is concerned, mainly case reports of NF have been published(Granroth 1946, Oravisto 1949, Katila & Granroth 1955, Karjalainen 1964).

Before the 1980’s NF1 and NF2 were usually considered variants of the same disorder.The Consensus Development Conference held at the National Institute of Health (NIH) in1987 was the first attempt to clarify the various clinical types of NF. The clinicalmanifestations associated with classical von Recklinghausen’s disease were differentiatedfrom those of the predominantly intracranial subtype, and the NF types 1 (NF1) and 2 (NF2)were defined (Stumph et al. 1988). In 1990 a second NIH conference was held to furtherevaluate the clinical aspects of NF. As a result of these meetings, the present clinicaldiagnostic criteria for NF1 and NF2 were proposed. This effort has greatly improved theaccuracy of clinical diagnosis. The molecular cloning of NF1 and NF2 genes and thedemonstrating of these in 1990’s finally defined NF1 and NF2 as separate disorders(Cawthon et al. 1990, Wallace et al. 1990, Rouleau et al. 1993, Trofatter et al. 1993).

2.2. Current classification of neurofibromatoses

Other types of neurofibromatoses have been described besides NF1 and NF2, but theirexistence, clinical features and genetics are less clear. These include the rare segmental NF(NF5) and other even less common types designated as NF types 3-4 and 6-7, and NF-NOS(not otherwise specified) by Riccardi (1992; see Table 1). They are mainly characterised byatypical distribution of café au lait spots and/or neurofibromas. Additional rareneurofibromatosis types proposed by Riccardi are the gastrointestinal type, multipleschwannomatosis and the familial brain tumour type.

Another classification, proposed by Viskochil and Carey in 1994, and further evaluatedin 1999 is based on clinical and molecular genetic aspects of NF (Table 2) (Viskochil & Carey1994, Carey & Viskochil 1999).

19

2.3. Neurofibromatosis type 1

2.3.1. Diagnostic criteria

Certain clinical findings occur in NF1 patients so often that their presence, especially incombination, can be used as diagnostic criteria. They have proved useful for research

MIM

20

purposes (Table 4) and have been accepted as current diagnostic criteria in consensusmeetings in 1987 and further evaluated in 1997 (Table 3) (Stumph et al. 1988, Gutmann et al.1997).

Table 3. Diagnostic criteria for neurofibromatosis type 1 (NF1) (Gutmann et al. 1997).The NF1 patient should fulfil 2 or more of the following criteria :

2.3.2. Clinical manifestations

NF1 is a neurocutaneous disorder which shows wide interfamilial and intrafamilial clinicalvariations due to pleiotropy and variants in the expression of the NF1 gene. Many of themanifestations are mild, but some affected individuals have significant medical problems.

The cutaneous manifestations are usually clear and easy to recognise without anyspecial instruments, and special experience of NF1 is usually needed only in atypical cases(Riccardi 1992).

Café au lait spots (CLS) are brown macules on the skin with a colour of coffee with milkand they are a hallmark of NF1. The history of CLS related to NF1 dates back to 1896, whentwo French articles were published about the importance of pigmentary changes as aprimary aspect of the disease (Chauffard 1896, Marie & Bernard 1896). Crowe & Schull (1953)pointed out the diagnostic importance of CLS in NF1 and concluded that six or more maculeswith ≥ 1.5 cm could be regarded as pathognomonic for the disorder. The diameter is usuallya few centimetres, but can vary from 0.5 to 10 cm. They are typically ovoid in shape withsharp, well-defined smooth borders with the surrounding normal skin (Figure 1a). It is notclear whether they are already present at birth, but in any case they tend to increase innumber and become clearer during the first year of life. CLS may be seen all over the body,with the possible exception of the scalp, eyebrows, palms of the hands and soles of the feet.The histological finding of a macule consists of focal hyperpigmentation in the basal layer ofan otherwise normal epidermis, this hyperpigmentation being due to an increased amount ofmelanin in both the melanocytes and keratinocytes. Both the number of melanocytes andtheir dopa oxidase activity are increased, and “giant“ melanosomes are also present in thesecells (von Deimling & Krone 1997).

This change is not specific to NF, however. Crowe & Schull (1953) studied theprevalence of CLS among 6856 institutionalised patients without cutaneous manifestations

No. Criteria

1. Six or more café au lait macules- diameter ≥ 1.5 cm in postpubertal individuals- ≥ 0.5 cm in prepubertal individuals

2. Two or more neurofibromas of any type or one plexiform neurofibroma3. Multiple freckles in the axillary area or groin4. Optic glioma5. Two or more Lisch nodules (iris hamartomas)6. A distinct osseous lesion such as sphenoid dysplasia or thinning of the bone cortex with or

without pseudarthrosis7. A first degree relative (parent, sibling or offspring) who meets the above criteria for NF1

21

Figure 1a. Café au lait spots. 1b. Freckles.

Figure 2a. Neurofibromas 2b. Plexiform neurofibroma

22

of NF and found that about 10% had from one to five macules, but all of them less than six.In another study of skin findings in 601 healthy adult patients the prevalence of CLS was14%, although there were always less than four present (Kopf et al. 1985). The frequency ofCLS as a diagnostic criterion for NF1 varies from 79% to 100% between populations, havinga tendency to decrease in adulthood (Table 4).

Freckles were first observed to be typical of NF1 only about 35 years ago, in 1964, byCrowe, who pointed out that patients with NF often show freckling on parts of the body notexposed to sunshine, such as the axillary and inguinal areas. Freckles are defined as a clusteror clusters of hyperpigmented macules of diameter <5 mm (usually 2-3 mm) and with similarcolour and histological findings as for café au lait spots (Figure 1b). The typical places forfreckling are axillary and other intertriginous and skin fold areas, such as the inguinal region,upper eyelids, base of the neck, under the chin, and the submammary region in women. Smallbrown macules in these regions are considered to be highly suggestive of NF1. The frecklesare not usually observed at birth, but they appear early and are reported to be present in 81%of children by six years of age (Obringer et al. 1989). The overall frequency of freckles in theaxillary or inguinal region in NF1 is reported to be between 55% and 86% (Table 4). Frecklesfound in the axillary or groin area are of great help, especially when making an earlydiagnosis of NF1 in young children presenting with several CLS (Riccardi 1992, Huson1994).

Neurofibromas are hamartomas composed mainly of those cell types forming the fine-ordered supportive compartments of peripheral nerves, but organised in a haphazard way.More specifically, the normally highly well-organised perineurial tube surrounding theSchwann cell-axon units is disrupted, so that escaping Schwann cells, endoneurialfibroblasts and perineurial cells form the majority of cells in these tumours. In addition,neurofibromas may contain solitary muscle cells and clusters of epithelial cells (Peltonen etal. 1984, Peltonen et al. 1988, Jaakkola et al. 1990).

A cutaneous (dermal) neurofibroma is typically a small, soft nodule, pink or purple incolour, which may become larger and pedunculated with time (Figure 2a). A subcutaneous(nodular) tumour is more firm and ovoid in shape. Von Recklinghausen was the first to namethese tumours neurofibromas, but actually Smith had already postulated in 1849 that theyarise from the fibrous connective tissue of small nerves (Mulvihill 1990a). In NF1 theneurofibromas are most often dermal, but they may also occur in deeper peripheral nerves orin the spinal canal. Pulst et al. (1991) reported a family with symmetrical spinalneurofibromas and CLS in several generations and showing linkage with the NF1 gene,suggesting that the family manifested a rare form of NF1. A biopsy of a suspectedneurofibroma is useful for a specific diagnosis and helps to differentiate it from other softtissue tumours, although a solitary neurofibroma is not pathognomonic for NF1 (Harkin &Reed 1969, Peltonen et al. 1988, von Deimling et al. 1995).

The neurofibromas involved in NF1 do not usually appear before late childhood oradolescence, and their reported frequencies vary from 40% in a study of 150 children under18 years (Cnossen et al. 1998a) to 100% in a study of 72 adults (Samuelsson 1981) (Table 4).Nearly all adult patients have dermal neurofibromas. They may become numerous, causecosmetic problems and troublesome itching, or lead to neurological problems such as pain,weakness and autonomic dysfunction, but they have not been reported to transform intomalignant tumours (Riccardi 1992, Wiestler & Radner 1994).

24

Plexiform neurofibromas consist of the same elements as dermal neurofibromas but areassociated with nerve plexuses and major nerve trunks and their branches. They mayinvolve any nerve, including the spinal nerve roots and cranial nerves. The tumour retains afascicular organisation and may be associated with soft tissue hyperplasia.Characteristically it has the appearance of a soft tissue mass which may extend from thedeep parts of the body to the skin, and it may be accompanied by a diffusehyperpigmentation and hypertrichosis of the overlying skin (Figure 2b). It can vary in sizefrom a few cm to large areas of the body, causing hypertrophy and gross disfigurement.Plexiform neurofibroma is congenital and almost exclusively associated with NF1, with afrequency of 15% to 40% (Table 4). There is an age-dependent risk of a benign plexiformneurofibroma transforming into a malignant peripheral nerve sheath tumour (MPNST) (Aloi& Massobrio 1989, Wanebo et al. 1993, Fisher et al. 1997, von Deimling & Krone 1997, Korf1999).

Other cutaneous findings have also been reported to be present in NF1 patients morefrequently than in the general population. Hyperpigmentation, both generalised andlocalised or patchy, giving a dirty skin appearance, is often present in NF1 patients,distinguishing them from their healthy relatives. The pathogenesis is unknown. It has beenspeculated that this is one of the manifestations of the NF1 gene itself, because in a few rarecases where a parent of an individual affected with NF1 has had a segmental NF, his/her skinfindings has also shown similar hyperpigmentation but in a segmental region. Thehypopigmented macules observed in NF1 are similar to those seen in other genodermatoses.Pseudoatrophic macules are areas of skin that appear to be atrophic, with a very pale colour,a thin epidermis and a lack of skin hairs. Skin angiomas, or Campbell de Morgan spots, area few mm in diameter, cherry-coloured, slightly raised spots mostly found on the trunk,abdomen and thighs, which increase with age. Juvenile xanthogranulomas may also beassociated with NF1, but also with juvenile chronic myeloid leukaemia (JCML). They areyellowish, dome-shaped nodules consisting of histiocytes and multinucleated giant cells.None of these other skin manifestations are pathognomonic for NF1 (Westerhof & Kondrad1982, Viskochil & Carey 1994, Zvulunov et al. 1995).

Lisch nodules are harmless, elevated, dome-shaped, smooth, yellowish to brown irishamartomas composed of melanocytes (Figure 1 in paper V). The first description of a NFpatient with true Lisch nodules was probably that presented by Snell in 1900. Confirmationof characteristic iris nodules with NF was published by Sakurai in Japan in 1935, and Lischdescribed details of similar findings in three of his patients two years later, in 1937, pointingout that the nodules were pathognomonic for NF and suggesting that they are frequentlyfound in NF patients (Sakurai 1935, Lisch 1937). They vary in size (diameter ≤ 2 mm), areusually present in both eyes, and increase with age, starting to develop from earlychildhood. Their frequency varies with age, from about 50% in childhood to over 95% inadolescence. They have been observed in 52% to 87% of individuals in differentpopulations (Table 4) (Lewis & Riccardi 1981, Huson et al. 1987, Ragge et al. 1993a).

To distinguish Lisch nodules from iris nevi, a slit lamp examination by anophthalmologist familiar with NF1 is recommended. Lisch nodules are highly characteristicof NF1 and Watson syndrome (a rare autosomal dominant syndrome characterised by CLS,mental retardation and pulmonary stenosis), but are rare in the general population or in otherdiseases (Lubs et al. 1991). They have been observed four times in association with anisolated case of NF2 (Charles et al. 1989, Garretto et al. 1989, Michels et al. 1989, Kaye et al.

25

1992), twice with segmental NF (Weleber & Zonana 1983, Rubenstein et al. 1983) and oncewith Cushing’s disease (Bouzas et al. 1993).

Histologically, an optic glioma is a pilocytic astrocytoma of grade I occurring in theoptic nerve, and is identical to other pilocytic astrocytomas occurring elsewhere in thecentral nervous system. Von Michel reported a patient with optic glioma in 1873, but it wasonly in 1940 that Davis established optic glioma as a part of NF (Mulvihill 1990a). Listernickreported that 19% of children with NF1 who had had a cranial and orbital CT or MRI scan(33/176) were diagnosed as having an optic pathway tumour. Only three of them showed anyprogression of the disease, however (Listernick et al. 1994). The prevalence of opticpathway gliomas among NF1 patients in other studies varies from about 1% to 15%. This ismainly due to the different protocols used, the lower figures including only symptomaticcases while the higher ones also featured asymptomatic ones identified in neuroimagingscreening (Table 4). The age of presentation with optic pathway gliomas is usually less thanseven years, with a peak incidence between four and six years, and the tumour seldomappears at an older age (Listernick et al. 1995, Listernick et al. 1999). A recent studyperformed in Japan showed a 12.5% incidence of NF1 in cases of optical pathway gliomaamong 26 084 children with cancer. Thus the majority of optic gliomas are attributable tocauses other than NF1 (Lewis et al. 1984, Matsui et al. 1993, von Deimling et al. 1995).

Other ophthalmological findings fairly common in NF1 are choroidal lesions such asnevi and hamartomas, which are small, flat or minimally elevated, yellow-white or brown,avascular lesions, observed with a frequency of 51% in the series of Lewis and Riccardi(1981). These findings are not specific to NF1, and choroidal hamartomas have beenreported in NF2, for example.

Skeletal changes are common in NF1. Adrian, reviewing the literature of NF in 1901,found scoliosis, skeletal anomalies and atrophy or hypertrophy of individual bones to bementioned frequently. Pseudarthrosis was probably mentioned for the first time inconnection with NF in 1872, when Fremmert described a patient with molluscum fibrosumand pseudarthrosis following a fracture (Holt & Wright 1948, Crawford & Bagamery 1986).

Congenital bowing and thinning of the long bones appears in children with NF1 eitherwith or without pseudarthrosis. Generally, pseudarthrosis means a false joint which mayresult from a fracture within a long bone that does not heal normally. In NF1, however, itrepresents a congenital defect in long bone formation, even though it may not beappreciated clinically or radiographically until weeks, months or even years after birth. Byfar the most common site is the tibia. It is estimated that congenital pseudarthrosis of thetibia (CPT) turns out to be associated with NF1 in about a half of cases. The overallincidence of the pseudarthrosis has been estimated in a Danish study from the 1970’s to beabout 1/190 000 live births (Andersen 1978), and its frequencies in NF1 from 1% to 4%between populations (Table 4). There is one case report of two brothers having bothpseudarthrosis of the tibia and NF1 (their mother had NF1 without pseudarthrosis), but in allother reports pseudarthrosis has been sporadic (Wellwood et al. 1971). Histopathologicallyaffected bones show non-specific changes (Boyd 1982, Stevenson et al. 1999).

The prevalence of scoliosis was estimated to be about 10% in a series of 220 NF1patients (Akbarnia et al. 1992). There are two types of lateral spinal curvature in NF1, anidiopathic one and a dystrophic one. The latter is rare and involves five or less vertebrae,causing a very sharp, angular curve and ending up in a more difficult form of scoliosis that

26

is almost exclusively observed in patients with NF1. This requires early detection, and oftentreatment with spinal fusion (MacEwen 1990).

The different types of other skeletal abnormalities include erosive defects due to thepresence of neurofibromas contiguous to bone, disorders of growth, including both under-development and over-development, intra-osseous cystic lesions, and congenitalanomalies of the skeleton. An example of the last group is sphenoid bone wing dysplasia,which is a rare finding but is said to be specific to NF1 (Holt & Wright 1948).

Other typical, although less specific, clinical features observed are short stature,macrocephaly, dysmorphic features and high intensity T

2 lesions in MRI.

Short stature is a common finding in NF1 that was first recognised by Riccardi in 1981. Ina Welsh study (Huson et al. 1988) height was ≤ the 3rd percentile in about 32% of thepatients, with no difference between the sexes. The corresponding figure in a study of 200Australian patients was 27% (North 1993). Growth curves in infancy show that growthappears to be delayed even at this stage. In the National Neurofibromatosis Foundation’sdatabase study of 2374 NF1 patients in different countries, comparison of their heights withage and sex-specific standards showed both sexes to be shorter than normal, the meanaverage height for children being – 0.59 SD (Szudek et al. 1998). The cause of short staturein NF1 is unknown. The explanation has only occasionally been an endocrinologicaldisturbance such as growth hormone deficiency or abnormalities of the pituitary gland, norhave any structural skeletal defects been found (Riccardi 1992, Huson 1994, Zacharin 1997).

Macrocephaly, as defined by a head circumference at or above the 97th percentile, isanother common feature of NF1, occurring in more than 40% of patients in populationstudies. The onset is postnatal and usually becomes evident in childhood. MRI studies ofNF1 cases with macrocephaly have shown that it is due to enlargement of the whole brainand not to hydrocephalus (Riccardi 1981, Hughes 1994, Cnossen et al. 1998b, Szudek et al.1998).

There are no common facial dysmorphic features associated with the NF1 phenotype,although a small group of patients have displayed an asymmetric face with hypertelorismand a broad nasal bridge. An incomplete Noonan phenotype has been seen in a few patientswith NF1, and in a few families it has been associated with a mutation in the NF1 gene(Allanson et al. 1985, Tassabehji et al. 1993, Colley et al. 1996, Bahuau et al. 1998). Inaddition, a clinical overlap has been described between patients with NF1 and Watsonsyndrome, so that these may be allelic disorders (Allanson et al. 1991).

The reported incidence of the high signal intensity lesions in T2 -weighted MR images

of the brain in NF1 patients varies from about 50% to up to 79 %, being highest in children.Itoh et al. (1994) reported these lesions in 93% of patients under 15 years of age and in 29%over 31 years. These T

2 lesions have been previously called as “hamartomas“,

“heterotopias“, “unidentified bright objects“ (UBO), or “unidentified neurofibromatosisobjects“. They have the appearance of well-circumscribed, non-enhancing hyperintenselesions without any mass effect or surrounding oedema, and are not visible on a CT scan.The most frequent places where they are observed are in the basal ganglia, thalamus,cerebellum and brainstem regions. Although these lesions are distinctive and associatedwith NF1, there is not yet enough information about their specificity to this disorder, nor anunderstanding of their pathological basis (North 1997).

A number of studies have been made of the connection between T2 lesions and learning

difficulties or low IQ scores in children with NF1. Some suggest a possible correlation

27

between these lesions in the thalamus and neuropsychological disturbances such as lowerscores for the IQ, memory, motor, distractibility or attention domains (Moore et al. 1996,North 1997), while others have failed to show any such connection (Dunn & Roos 1989,Duffner et al. 1989, Ferner et al. 1993).

The incidence of mental retardation (IQ < 70) in NF1 patients has been estimated to be3-8%, a slightly higher figure than in the general population (1-2%). Developmental delay isnot progressive and it is not correlated with the frequency of other features of the disorder.Instead, learning disabilities are observed in 30% to 60% of affected individuals and theyshow visuospatial integration deficits and an increased incidence of school performanceproblems (especially in reading and spelling). There is no characteristic profile of learningdisability in children with NF1. Macrocephaly is not especially associated with cognitivedeficits (Riccardi 1981, Legius et al. 1994, Moore et al. 1994, North 1997).

There is an increased risk of both primary and secondary malignancies in NF1 (Jensenet al. 1998). In a long-term follow-up of 212 NF patients in Denmark spanning 42 years,Sorensen et al. (1986) reported a relative risk of 4.0 (95 % confidence limits 2.8 to 5.6) ofdeveloping a malignant neoplasm or a benign central nervous system tumour in probandswith NF1 relative to the general population. In addition, about one fifth of the NF1 patientswith a primary malignancy developed a secondary malignancy, as compared with only a fewpercent in the general population. Correspondingly, a twelve-year follow-up study inGothenburg, Sweden, by Zöller et al. (1995) reported 22 deaths among their 70 adult NF1patients and found a malignancy in 55% of them (12). Matsui et al. (1993) studied 26 084children in Japan with cancer and observed the incidence of NF1 to be 6.45 times higher thanexpected (0.21%). The incidence of NF1 was 31.4% among the children who had a malignantperipheral nerve sheath tumour (MPNST), 0.9% among those with nervous system gliomasand 1.36% among those who had rhabdomyosarcoma.

Malignant tumours especially known to be associated with NF1 are MPNSTs,astrocytomas, rhabdomyosarcomas, pheochromocytomas, carcinoid tumours andchildhood leukaemia. In addition, the risk of hemispheric, brain stem and cerebellarastrocytomas and glioblastomas is increased in both NF1 and NF2 (Ilgren et al. 1985,Babtiste et al. 1989, von Deimling et al. 1995).

Malignant peripheral nerve sheath tumour (MPNST), previously known asneurofibrosarcoma, neurogenic sarcoma, malignant neurilemmoma, or malignantschwannoma, is a tumour that arises from the Schwann cells or peripheral fibroblasts. Itsincidence in the general population is 0.001%, and in about half of the cases it is connectedwith NF1. The incidence rate of MPNST among NF1 patients varies from 2% to 29% indifferent reports. Plexiform neurofibromas have a lifetime minimum risk of malignanttransformation of 5% (Ducatman et al. 1986, Riccardi & Powell 1989, Wanebo et al. 1993).

Fuller and Williams (1991), reviewing the gastrointestinal manifestations of NF1,described NF1 patients with duodenal carcinoid as having a higher risk of developing apheochromocytoma. This tumour type has been described in NF1, but its frequency is low,approximately 1% (Fuller & Williams 1991). The literature includes a few case reports of thecoexistence of Wilms’ tumour and NF1 (Walden et al. 1977, Stay & Vawter 1977), althoughno clear causal relationship has been established (Matsui et al. 1993, Mulvihill 1994).

Juvenile chronic myeloid leukaemia (JCML) has been reported to be associated withNF1. Morier et al. (1990) suggested that NF1 patients with juvenile xanthogranulomas havea higher risk of JCML, but others have failed to identify this (Gutmann et al. 1996).

28

NF1 is a slowly progressive disorder, with new hamartomas appearing in the course oftime. It occasionally manifests itself as a congenital plexiform neurofibroma or apseudarthrosis of the tibia in newborn infants, while in those affected in later childhood thepresence of optic glioma, learning disabilities and an intellectual handicap can play animportant role. Neurofibromas, scoliosis and psychological problems are the mainsymptoms in adolescence, and vascular hypertension and malignancy are lifelongproblems. Most individuals with NF lead a normal life with only cosmetic problems, whileothers develop medical problems (Gutmann et al. 1997).

Both Scandinavian long-term cohort studies of NF1 suggest a reduction in the life spanof NF1 individuals. In the Danish study the 39-year survival curves for patients with NF1(probands, male/female, and affected relatives in separate curves) showed a clear reductionin survival as compared with the expected curves for the general population, especially infemale NF1 probands (Sorensen et al. 1986). In the Swedish 12-year follow-up study the riskratio for death among NF1 patients relative to the normal population was 4.3 (95 %confidence limits 2.7 to 6.5) and the mean age at death was 61.6 (29.1 – 81.9) years, about 15years lower than in the general population (Zöller et al. 1995). The lower than expectedprevalence of NF1 in older age groups also suggests that patients die prematurely(Samuelsson 1981, Huson et al. 1988, Fuller et al. 1989, Clementi et al. 1990).

Pregnancy causes an increase in the size and number of cutaneous neurofibromas in anaffected mother. Pre-eclampsia, pre-term delivery, intrauterine growth restriction,pregnancy-induced high blood pressure, stillbirth, miscarriage and perinatal mortality arealso reported to be more common. On the other hand, a study of 247 pregnancies in 105 NF1women showed the numbers of obstetric complications to be similar to those in the generalpopulation with the exception of caesarean section, which was about 15% more frequent(Dugoff & Sujansky 1996). These results suggest that there is a small group of women withNF1 who are at risk because of manifestations of NF1 and might thus require closemonitoring at a special obstetric centre (Edwards et al. 1983, Sharma et al. 1991, Weissmanet al. 1993, Hagymásy et al. 1998).

Prenatal diagnosis of NF1 has seldom been performed, possible because the result willonly tell the family of the gene carrier status of the fetus but not of either the severity of thedisorder or its prognosis (Upadhyaya et al. 1992, Benjamin et al. 1993, Lazaro et al. 1995a,Ponder et al. 1998).

2.3.3. Epidemiology and genetics

NF1 occurs worldwide and in different ethnic groups. Its estimated prevalence in studiesperformed in the 1970’s or later has varied from 1/960 to 1/7800, and its estimated birthincidence from 1/2558 to 1/4292 (Table 5). The penetrance of NF1 is very high, and only a fewpossible non-penetrant cases have been reported (Carey et al. 1979, Spence et al. 1983,Spence et al. 1986). The mutation rate in the population-based studies has been calculatedto be between 3.1-6.5 x 10-5 per gamete per generation, one of the highest among the humandisease genes. This is partly explained by the large size of the gene, the alternatively splicedtranscript and the possibility that there may be sequences within the gene that are highly

29

susceptible to mutations (Crowe et al. 1956, Sergeyev et al. 1975, Samuelsson 1981, Husonet al. 1989a, Fuller et al. 1989, Clementi et al. 1990, Garty et al. 1994, Shen et al. 1996).

The majority of NF1 mutations occur on the patient’s paternally derived chromosome 17,but gross gene deletions are predominantly maternally derived (Upadhyaya et al. 1998).Some authors have postulated an older age for fathers of sporadic NF1 cases (Sergeyev1975, Riccardi et al. 1984, Takano et al. 1992, Bunin et al. 1997), while others have not foundany paternal age effect (Borberg 1951, Samuelsson 1981, Huson et al. 1989a). Riccardi et al.(1984) concluded that men aged 35 years and older have an increased risk of fathering a childwith a new mutation for NF1 (odds ratio 2.04; 95% confidence interval 1.25 to 3.33). However,two thirds of the new mutation individuals in their population had fathers younger than 35years (Riccardi 1992). In a recent case-control study of 89 families, Bunin et al. (1997) foundthat the fathers of the patients with NF1 were 1.5 years older than those of the controlsubjects at the birth of the child (p=0.07).

Crowe, Samuelsson and Huson all reported a reduction of between 0.8 and 0.47 ingenetic fitness, or the ability to procreate. There was a significant difference in geneticfitness between the married men (0.6) and the women (0.9) in the series of Crowe et al. Thisreduction in fitness was partly attributed to biological factors related to premature death andpartly to non-biological factors such as selection against parenting in affected subjects,especially males (Crowe et al. 1956, Samuelsson 1981, Huson et al. 1989a).

The NF1 gene locus was first mapped to the centromeric region of chromosome 17 in1987 as a result of an analysis of 15 families with several affected members who all fulfilledstrict clinical criteria for NF1 (Barker et al. 1987). In 1989 two NF1 patients with balancedchromosomal translocations involving the long arm of chromosome 17 were identified,allowing analysis of the genetic region by physical mapping techniques (Fountain et al.1989). The first translocation involved chromosomes 1p34.3 and 17q11.2, and the second17q11.2 and 22q11.2 (Schmidt et al. 1987, Ledbetter et al. 1989). Screening of various cDNAlibraries with probes derived from genomic clones mapping between the two translocationbreakpoints led to the identification of four candidate genes, EVI2A (Cawthon et al. 1990),EVI2B (Cawthon et al. 1991), OMGP (Viskochil et al. 1991) and TBR (Viskochil et al. 1990).The two groups showed independently that the fourth candidate was the NF1 gene(Cawthon et al. 1990, Wallace et al. 1990).

The NF1 gene spans a region of approximately 350 kb of genomic DNA and is dividedinto 60 known exons. Three of these are alternatively spliced as in-frame insertion exons.The gene encodes a 11-13 kb mRNA sequence. The three previous candidate genes (EVI2A,EVI2B, OMGP) are embedded within one large intron and are transcribed in the oppositedirection to the NF1 gene (Figure 3) (Viskochil 1998).

More than 240 mutations have been reported in the NF1 gene, showing that most of thefamilies have their own mutations. These vary from small or large deletions to insertions orpoint mutations, and usually lead to a shorter protein product, as has been observed in othertumour suppressor genes. Some of the reported mutations have been recurrent, although notrue “hotspots“ have so far been detected. One of the most frequent mutations has been anonsense mutation in exon number 31, which has been reported in at least 14 independentcases (Lazaro et al. 1995b, Blanquet et al. 1995, Korf 1998).

31

Figure 3. The NF1 gene (Ragge 1993b).

The only constant genotype-phenotype correlation has been found in patients havingdeletions of the entire NF1 gene and variable amounts of flanking DNA associated with adistinctive phenotype. The characteristic features in these cases have been severedevelopmental delay, learning disability, dysmorphic features, early onset and a largenumber of neurofibromas (Kayes et al. 1992, Wu et al. 1995, Cnossen et al. 1997, Wu et al.1997, Upadhyaya et al. 1998).

2.3.4. Pathophysiology

The fact that neurofibromas may undergo malignant transformation can be explained by thetumour suppressor gene “two-hit“ hypothesis originally described by Knudson in 1971 forretinoblastoma (Knudson 1971). According to this model the affected individual has inheri-ted one mutant copy of the NF1 gene from his parent or developed one as a result of a spon-taneous mutation. The second mutation is a somatic one and occurs in the remaining wildtype copy of the NF1 gene. As a result both copies are non-functional and NF1 gene proteinexpression is missed. The loss of this protein expression has been observed in both malig-nant tumours (MPNST, pheochromocytoma, juvenile myeloid leukemia) (Legius et al. 1993,Xu et al. 1992, Shannon et al. 1994) and benign ones (neurofibroma) (Colman et al. 1995,

32

Sawada et al. 1996, Serra et al. 1997). Similar findings regarding the tumour suppressor genemechanism have been reported in NF2, tuberous sclerosis and von Hippel-Lindau’s disease(Skuse & Ludlow 1995).

The protein encoded by the NF1 gene, neurofibromin, shows homology to the GTPaseactivating family of proteins (GAPs) in mammals and to the IRA1 and IRA2 proteins in yeast,which are all known to stimulate ras-GTPase activity. The NF1 gene product may function asa negative growth regulator in growth-arrested cells, allowing for the onset of cellulardifferentiation, and loss of NF1 expression would be predicted to result in aberrant cellulardifferentiation (Figure 4) (Seizinger 1993, Gutmann 1997).

Inhibition of p21RAS activity(e.g. in the mitogenic signalling pathway)

Neurofibromin Other functions not related to p21RAS

(e.g. interaction with microtubuli)

Downstream effector(e.g. in the mitogenic p21RAS activity signalling pathway)

Fig. 4. Possible cellular function(s) of neurofibromin (Seizinger 1993).

Alternative splicing generates different isoforms of NF1 mRNAs, the best characterisedto date being types I and II (Gutmann et al. 1995). Hirvonen et al. (1999) studied alternativeNF1 gene products in peripheral nerves at different developmental stages and found thatthe neurofibromin immunosignal in the peripheral nerves was increased in Schwann cells,perineural cells and axons during the second and third trimesters of gestation.

The major clinical features and tumours of NF1 (CLS, freckles, neurofibromas, plexiformneurofibromas, pheochromocytoma and MPNST) arise from tissue of neural crest origin,considering NF1 as a neurocristopathy. Not all of the pathogenesis can be explained byneurocristopathy, however, as some other typical findings such as pseudarthrosis andscoliosis, optic glioma and learning disabilities arise from different tissues (Mulvihill et al.1990b, Huson 1994, Carey & Viskochil 1999).

Transgenic animals have shown some important findings regarding neurofibromin intheir developmental biology. Homozygous knock-out NF1 mice show embryonic lethality by14.5 days of gestation because the normal structure of the heart fails to develop. In additionthey show hyperplasia of neural crest derived sympathetic ganglia. Heterozygous NF1 micelack the typical skin manifestations, but show similarities to human NF1 in tumorigenesisand learning difficulties (Brannan et al. 1994, Jacks et al. 1994, Silva et al. 1997).

2.4. Neurofibromatosis type 5 (Segmental type)

NF5, or the segmental type of NF is thought to represent a mosaic form of NF1 in which themainly cutaneous NF1 features are limited to a unilateral segmental area of the body, not

33

usually crossing the midline. No other features of NF are normally found, and familialoccurrence has been reported in only seven families (Rubenstein et al. 1983, Boltshauser etal. 1989, Riccardi 1992, Moss & Green 1994, Huson & Ruggieri 1996). In these cases of aparent with segmental NF and a child with classical NF1 the situation has been explained byassuming that the affected parent has gonadal mosaicism. The first definition of thislocalised form of the disease was published in 1931 by Gammel, and it was Crowe et al. (1956)25 years later, who published further cases and started to use the term sectorial NF. Later on,Riccardi described segmental neurofibromatosis in 1982 as entailing CLS and/orneurofibromas in a single, unilateral segment of the body, without crossing the median lineand having no familial history or systemic involvement (Riccardi 1982). Five years later, Rothet al. subdivided the segmental NF into four subtypes, described as the true segmental type(Riccardi’s NF5), a localised type with deep involvement, a hereditary type and a bilateraltype (Roth et al. 1987). At present about 150 segmental NF cases have been described in theliterature (Riccardi 1992, Viskochil & Carey 1994, Huson 1998).


Neurofibromatosis type 2 is characterised by bilateral vestibular schwannomas, which arenot seen in NF1. Its birth incidence is about one tenth of that of NF1.

2.5.1. Diagnostic criteria

The diagnostic criteria for NF2 were established by the NIH Consensus DevelopmentConference in 1987 and defined further in 1990 and 1997. The main features are bilateralvestibular schwannomas or a family history of NF2 (Table 6) (Stumpf et al. 1988, Mulvihill etal. 1990b, Gutmann et al. 1997).

Table 6. Diagnostic criteria for neurofibromatosis type 2 (NF2) (Gutmann et al. 1997).

No. Criteria

I. Individuals with the following clinical features have NF2:-Bilateral vestibular schwannomas or-A family history of NF2 [first degree relative (parent, sibling or offspring)] plus either1. unilateral vestibular schwannoma < 30 years or2. any two of the following criteria: meningioma, glioma, neurofibroma, schwannoma orjuvenile posterior subcapsular lenticular opacities/juvenile cortical cataract.

II. Individuals with the following clinical features should be evaluated for NF2 (probable NF2):-Unilateral vestibular schwannoma < 30 years plus at least one of the following:meningioma, glioma, neurofibroma, schwannoma or juvenile posterior subcapsular lenticularopacities/juvenile cortical cataract. or-Multiple meningiomas (2 or more) plus unilateral vestibular schwannoma or one of thefollowing: meningioma, glioma, neurofibroma, schwannoma or juvenile posterior subcapsularlenticular opacities/juvenile cortical cataracts.

34

2.5.2. Clinical manifestations

Two clinical subtypes of NF2 have been described. The severe type, also known as theWishart type, first described by Wishart in 1820, is characterised mainly by early onset, fastprogression and a predisposition to spinal tumours and meningiomas. The mild type, alsoknown as the Gardner type (Gardner & Frazier 1930), is characterised mainly by late onset, amore benign course, slow progression and a low incidence of spinal tumours andmeningiomas. Both types also involve a predisposition to vestibular schwannomas. Theclinical manifestations of NF2 are usually similar among members of a family, but variabilityin expressivity is manifest between families (Evans et al. 1992a, Parry et al. 1994, Ahn et al.1996).

The most common feature is bilateral vestibular schwannoma, the early symptoms ofwhich include tinnitus, balance dysfunction and sensorineural hearing loss. The mean ageat onset was 27 years in the series of Evans et al. (1992b), showing that the symptoms of NF2start about 20 years earlier than those of sporadic unilateral vestibular schwannomas. Thesetumours are histologically benign. Eldridge found no malignant vestibular schwannomasafter reviewing 134 cases in the literature (Eldridge 1981, Short et al. 1994).

Other common tumours in NF2 include meningioma, glioma, neurofibroma andschwannoma. Histologically, the meningiomas and gliomas are low-grade neoplasms ofmeningothelial cells and glial cells. These tumours are most often central. Dysplastic andmalformative central nervous system lesions are common features of both NF1 and NF2.About half of all NF2 patients have meningiomas. Several meningiomas are usually found,and the age at onset is earlier than in sporadic cases (Evans et al. 1992b, Parry et al. 1994,Sainio et al. 1994, Louis et al. 1995).

Spinal tumours are diagnosed in about two-thirds of all NF2 patients. These tumours areusually schwannomas, but meningiomas and glial cell tumours have also been reported.Peripheral nerve tumours are reported in about 70% of cases, and are histologically primarilyschwannomas. A NF2 plaque is a useful diagnostic hallmark, being reported in about half ofthe cases. It is a small (usually diameter ≤ 2 cm) cutaneous lesion described as a pigmented,well-circumscribed, slightly raised, coarse and sometimes also hypertrichotic lesion (Evanset al. 1992b, Parry et al. 1994, Short et al. 1994).

Skin findings such as café au lait spots are less common and less frequent than in NF1.Approximately half of the NF2 individuals have from one to four macules, but six macules areextremely exceptional, being reported only in a few cases. Freckles in the axillary or groinarea are not found in NF2 individuals (Eldridge 1981, Evans et al. 1992b, Parry et al. 1994).

The main ophthalmological features of NF2 are juvenile posterior subcapsular lenticularopacities or juvenile cortical cataract (Table 6). Posterior subcapsular lenticular opacitieswere reported in 80% of the 45 NF2 patients studied by Bouzas et al. (1992) and peripheralcortical cataracts in 38%. Additional ophthalmological findings in NF2 patients are anepiretinal membrane and combined pigment epithelial and retinal hamartomas. The presenceof pre-senile cataract or an epiretinal membrane in a young patient should give rise to asuspicion of NF2 even without any other signs or symptoms (Landau et al. 1990, Sivalingamet al.1991, Kaye et al. 1992, Bouzas et al. 1992, Landau & Yasargil 1993).

35

2.5.3. Epidemiology and genetics

Evans et al. (1992a), in their population-based study, calculated an incidence of about1/40 000, a prevalence of 1/210 000 and a mutation rate of 0.65 x 10-5 for NF2 north-westEngland. It is inherited in autosomal dominant fashion, with about half of the casesrepresenting a new mutation. The penetrance is estimated to be over 95%, and most of thepatients have symptoms before the age of 40 years. The gene is located in the 22q12.2chromosomal region and it is a tumour suppressor gene.

Two groups independently identified the NF2 gene by positional cloning in1993(Rouleau et al. 1993, Trofatter et al. 1993). It consists of 16 constitutive exons and onespliced exon, and the protein product, called schwannomin or merlin (named after theproteins moesin, ezrin and radixin), acts as a tumour suppressor. It shares significanthomology (45-47% identity) with the ezrin-radixin-moesin proteins (ERM proteins) and isinvolved in the association of actin filaments with plasma membrane proteins. It mayregulate growth-promoting or growth-inhibitory signals transmitted by cell-cell or cell-substrate contact (Zucman-Rossi et al. 1998, Gutmann et al. 1999).

The literature contains careful descriptions of more than 100 mutations of the NF2 gene.No major “hotspots“ for mutations have been found. Phenotype-genotype correlationstudies have shown that missense mutations are associated with a mild type of diseasewhile nonsense and frameshift mutations are mainly associated with a severe diseasephenotype (MacCollin et al. 1994, Bourn et al. 1995a, Parry et al. 1996, Ruttledge et al. 1996,Zucman-Rossi et al. 1998).

It has been suggested that post-zygotic mosaicism is a relatively common finding inNF2. This explains why mutations have more often been found in NF2 patients with a familyhistory than in sporadic cases and why mutations among sporadic cases are found morefrequently in patients with a severe phenotype. The conclusion has been that mosaicism ismost commonly found in sporadic cases with a mild phenotype. In addition, some patientswith insufficient clinical findings to fulfill the diagnostic criteria for NF2 may represent theoutcome of mosaicism with very small populations of mutation-bearing cells (Evans et al.1998, Kluwe & Mautner 1998, Zucman-Rossi et al. 1998).

36

3. Aims of the research

The purpose of this research was to use patient and family material collected from NorthernFinland to evaluate the occurrence, genetic aspects and clinical features of the differenttypes of NF. The specific aims were

1) to evaluate the frequencies and population genetic aspects of neurofibromatosesin Northern Finland,

2) to study the clinical features of neurofibromatoses and variability among thedifferent types,

3) to evaluate the use of the established diagnostic criteria and linkage analysis fordiagnosing NF1 patients, and

4) to assess the significance of NF1 as a progressive disease to the patient and family,and to discuss its treatment, genetic counselling and follow-up.

37

4. Patients and methods

4.1. Collection of patients

The geographical area concerned was the region served by Oulu University Hospital(OUH), consisting of the provinces of Oulu and Lapland and the northern part of that ofVaasa, with a total population of 733 037 as of December 31, 1996 (Central Statistical Office ofFinland 1997). The region is 155 211 square kilometres in area and covers 46% of the totalarea of Finland, with a population frequency of only about five inhabitants per squarekilometre.

Complete ascertainment of NF was attempted by tracing all patients treated for it inNorthern Finland during the years 1967-1996 through multiple independent sources. Thebasic material consisted of patients and their affected first degree or more remote relativesattending the Department of Clinical Genetics at OUH for genetic counselling from 1982onwards. Additional patients with a diagnosis of neurofibromatosis (von Recklinghausendisease) (ICD 8th revision diagnosis number 743.40), or neurofibromatosis (InternationalClassification of Diseases (ICD) 9th revision diagnosis numbers 2377A or 2251A) were tracedfrom the records of the University Hospital and the four Central Hospitals in the area; two inLapland, one in Kainuu and one in Central Ostrobothnia. In addition, patients were traced bycontacting paediatricians, neuropaediatricians, dermatologists, neurologists, ophthalmo-logists, oncologists, pathologists, audiologists, otologists, surgeons, neurosurgeons,paediatric surgeons and internists in the area by letter and asking them to report any knownpatient with NF. The OUH records concerning patients treated for neurofibroma, opticglioma, multiple meningiomas or vestibular schwannomas were reviewed. All patients withplexiform neurofibroma, or congenital pseudoarthrosis were traced in order to re-evaluatethem for NF. The histological surgical and autopsy specimens examined at the Departmentof Pathology, Oulu University Hospital, and the clinical and autopsy records were reviewed.The two private pathological laboratories in the area were also contacted and their recordswere searched for any surgical NF specimens.

The examination protocol was approved by the Ethical Committee of the MedicalFaculty, University of Oulu.

These combined sources provided information on 200 individuals from 181 families witheither confirmed or suspected NF in one or more family members. All the patients were then

38

contacted through their own physicians and asked to participate in the study. Seven familieswith a suspected NF1 patient refused to attend, but if the patient fulfilled the diagnosticcriteria for NF1 according to the hospital records they were nevertheless included in theseries. All the others were assessed clinically at the Department of Clinical Genetics, OUH,almost all by the author.

A total of 46 patients originally suspected as having NF were excluded due to adiagnosis distinctly different from NF or failure to fulfil the diagnostic criteria for NF1 or NF2.In addition, five patients originally diagnosed as having NF1, were found to represent thesegmental type of NF after clinical examination. The diagnoses of the patients excluded areshown in Table 7.

Table 7. Diagnosis of the patients excluded (n=46).

4.2. Diagnostic criteria and clinical examination

The clinical diagnostic criteria presented by the international collaboration consortium(Tables 3 and 6) were used throughout.

Clinical evaluation of the patients included a review of their medical history includinghospital records, autopsy reports and family history. A detailed clinical examination wasperformed, followed by genetic counselling. Clinical assessment of cutaneous,neurological, ocular, orthopaedic and cardiovascular manifestations was carried out.Endocrine, auditory and dysmorphic findings were systematically recorded according to themodified examination schedule used by the NNFF (National Neurofibromatosis Foundation)International Database (Friedman et al. 1993) (Appendix 1). Developmental, behaviouralfeatures and growth parameters were recorded. For those patients who were thought tohave NF2, audiometry and brainstem audiotory evoked potential (BAEP) were alsoperformed. Other specialists were consulted in selected cases.

A (neuro-)ophthalmological examination (paper II and V) was performed on 123 of the164 NF1 patients examined clinically and on all NF2 and segmental NF cases ascertained.

39

The neuro-ophthalmological examination included refraction, visual acuity far and near withbest correction, examination of strabismus with cover tests, Maddox wing, Maddox rod, andsynoptophore, ocular motility, pupils, biomicroscopy, corneal sensitivity, eyelids, orbitaland periorbital regions, facial nerve function, Hertel measurements of the exophthalmos,intra-ocular pressure, visual fields by the Goldmann and Friedmann methods and with theconfrontation and Harrington-Flocks methods in children, colour vision with the Hardy-Rand-Rittler test and the Farnsworth D 15 desaturated test, and fundi with direct and indirectophthalmoscopy after dilation of the pupils.

The neuroradiological examination included a CT or MRI scan of the cranium and spine(paper II, III and V). MRI was performed with a 1.0 T superconducting unit (Magnetom,Siemens) using T1 and T

2 weighted images. A cranial CT or MRI scan was performed on 121

of the 164 NF1 patients examined and a MRI scan of the spine on 77 of those patients. MRIscans of the cranium and spine were taken in four of the NF2 cases. The images through thebrain and spine were analysed by a neuroradiologist (to be reported separately) payingspecial attention to the optic nerves, chiasma area, vestibular schwannoma (in NF2 cases)and spinal tumours.

The clinical characteristics of the 14 patients with congenital pseudarthrosis of the tibia(CPT) treated in Northern Finland were examined by means of careful orthopedic andradiographic examinations. Prevalence and incidence figures were also evaluated (paper VI).

The level of severity of NF1 was recorded in grades from 1 to 4 according to theclassification suggested by Riccardi (1992), where grade 1 (minimal type) signifies mildcutaneous findings without cosmetic or functional problems, grade 2 (mild type) numerousneurofibromas with some cosmetic or functional problems, grade 3 (moderate type) variousneurofibromas, learning difficulties and pseudarthrosis or other skeletal problems requiringsurgery, and grade 4 (severe type) intracranial and spinal tumours, pheochromocytoma,malignancy or mental retardation.

All the traceable first-degree relatives and some more remote ones were examinedclinically for mild symptoms and signs, including ophthalmological ones. Linkage analysiswas used in familial cases to exclude or verify a risk haplotype signifying the presence orabsence of NF.

4.3. Laboratory examinations

The histological slides of the surgical and autopsy specimens from 69 NF1 patients, fiveNF2 patients and one segmental NF case were reviewed (paper IV), and the clinical andautopsy records were studied. The age at original presentation and the severity of afflictionwere recorded for each patient.

4.4. Epidemiology, genetics and statistics

Population data for the years 1960-1996 for computing prevalence and incidence wereobtained from the Central Statistical Office in Finland (Suomen Virallinen Tilasto 1960-1997).The prevalence figures were calculated from the number of affected individuals at aparticular time in relation to the total population. The point prevalence on December 31,

40

1996 was calculated based on the total number of affected living persons in the populationof 733 037 in the area (Central Statistical Office in Finland, 1997) and period prevalenceswere calculated for the 1960’s, 1970’s, 1980’s and 1990-1995 separately from thecorresponding figures. Birth incidence rate was calculated as the number of NF1 casesborn in Northern Finland between 1960 and 1995 divided by the number of live births in thesame period. The prevalence and birth incidence rates were calculated per 100 000inhabitants/live births in the region. Patients were classified as sporadic or familialaccording to their family history and clinical examination, and as index cases they were theaffected ones through whom the family was ascertained and secondary cases they wererelatives of an index case and found to be affected.

NF1 patients were considered to represent probable new mutations if the parents did notshow any signs of NF1 upon clinical examination by us or another specialist experienced inNF1.

The family studies included a family history, a pedigree for three to four generations,and, in selected cases, tracing of common ancestors with the help of church records. Thebirthplaces of the patients were used to depict the geographical distribution of the diseasein Northern Finland. The disease was considered autosomal dominant if one of the patient’sparents was affected too, and as a new mutation if both parents were healthy. The case wasconsidered sporadic if there was only one affected individual in the family.

The mutation rate was calculated only for those cases in which both parents were seen,and was obtained by calculating the ratio of new mutation cases in a given period to the totalnumber of live births. Genetic fitness was estimated by Tanaka’s method (Tanaka 1974), inwhich relative fitness was calculated as a fraction comparing the frequency of NF1 amongthe parents of index cases with the frequency of NF1 among the offspring of index cases.The parental age of those cases assumed to represent new mutations was compared withthe parental age of the fathers in the general population of Finland and the parental age ofthe mothers in the region concerned. The birth order effect in families representing newmutations was assessed by the method of Haldane and Smith (1947), in which the sum of thebirth orders of all the affected sibs in each family was compared with the theoretical valuecalculated on the assumption that there is no birth order effect. Segregation analysis wasperformed by comparing the number of affected offspring of an affected parent having ahealthy spouse with the expected number (Emery 1986).

For statistical analysis, the results are presented as means with standard deviation andrange unless otherwise indicated. Standard deviation (SD) is presented in ± values after themean. Comparisons of distributions between groups were made using the Chi-Square test.Differences between means were tested with Student’s unpaired or paired t-test when thevariables followed a normal distribution, and with the Mann-Whitney U-test in the case of askewed distribution or of ordinal variables. Two-sided p-values were calculated at asignificance level of 0.05.

4.5. DNA studies

For molecular genetic studies, DNA from peripheral venous blood was obtained from thepatients and their first degree relatives by standard procedures. Linkage analysis was

41

performed in 20 familial NF1 cases using the following tightly linked flanking DNA markersand intragenic microsatellite markers: p11.3C4.2 (Rodenheiser et al. 1993), pHHH202(Ainsworth & Rodenheiser 1991), EVI-20 (Charrow et al. 1993), AluNF1 (Xu et al. 1991),IVS27AC28.4 (Lázaro et al. 1994), IVS38GT53.0 (Lázaro et al. 1993), pEW206 (Rodenheiser etal. 1993). In addition, 7 other familial and 39 sporadic NF1 cases in informative families werescreened for deletions with linked markers. Intragenic markers (EVI-20, ALuNF1,IVS27AC28.4 and IVS38GT53.0) were also used to detect possible deletions and to study theparental origin of the mutation.

Four markers linked to 22q12 were assayed in the NF2 cases: NF2CA3-A (Bourn &Strachan 1995b), D22S273, D22S280 and D22S281 (Gyapay et al. 1994).

In the study of the family with hereditary spinal neurofibromatosis (paper III) the lodscores for the two-point linkage analyses were calculated using the MLINK program(version 5.1) of the LINKAGE program package (Lathrop & Lalouel 1984). The genefrequencies used were 0.003 in NF1 and 0.0002 in NF2, and a penetrance value of 95%.Female and male recombination rates and allele frequencies for all marker alleles wereconsidered to be equal (Rodenhiser et al. 1993).

The numbers of patients in the groups, as reported in the original papers I-VI aresummarized in Table 8.

Table 8. Numbers of patients(P) and families(F) considered in the original papers I-VI.

42

5. Results

A total of 197 NF1, five NF2 and eight segmental NF patients were identified. The NF1patients came from 119 families. For confidentiality reasons pedigrees are not shown but areavailable at request. A total of 77 NF1 cases were sporadic and 117 familial, while threeadditional NF1 cases involved mothers with segmental NF whose children suffered fromNF1. 119 patients were index cases. With 95 male patients and 102 females, the sex ratio was0.93. The five NF2 patients, all of whom were male, were found in three families, two of thesecases being sporadic and three familial. The segmental type of NF was found in eightpatients, two of whom were men and six women. The geographical distribution of the NF1patients, shown in Figure 5, roughly corresponds to that of the general population in thearea.


5.1.1. Population genetics of neurofibromatosis type 1 (I)

The point prevalence for NF1 in Northern Finland on 31 December, 1996, was 23/100 000(1/4436). Evaluation of the period prevalence calculations showed a peak prevalence of34/100 000 (1/2983) for the age group 10-19 years (see Figure 1 in paper I).The birth incidencecalculated for the period 1960-1995 was 27/100 000 (1/3647), with corresponding incidencesper decade of 22/100 000 (1/4545) for the 1960’s and 1970’s and 34/100 000 (1/2941) for the1980’s. The incidence for the last six-year period 1990-1995 was 37/100 000 (1/2703).

The age distribution of the NF1 patients ranged from three months to 73 years (mean 29years). The mean age at the time of diagnosis was 20 ± 16 years, with a range from 3 monthsto 60 years. This figure was significantly lower in males than in females, and lower in theyounger generations (see Table 1 in paper I).

Family data for the mutation rate were available for 116 families, and based on these amutation rate of 4.37 ± 0.72 x 10–5 was calculated for the NF1 gene. The mean maternal ageat the birth of a child with NF1 due to a new mutation was 30 ± 6 years (range 21 - 43 years),while the mean overall maternal age in the area during the same period was 27.5 years(p<0.01). The corresponding mean paternal ages were 33 ± 6 years (range 19 - 50 years) for

43

Figure 5. Geographical distribution of the 197 NF1 patients in Northern Finland according totheir birthplaces. The region served by Oulu University Hospital, is marked with thin linesand the birthplaces of the patients with NF1 with circles. indicates one person, 5 personsand 10 persons.

NF1 children and 30.0 years for fathers in Finland in general, respectively (p<0.01). Out of132 offspring, 56 (42%) were affected. The ratio does not differ significantly from theexpected 0.5 (p=0.08). The mean birth order for 92 of the 96 new mutation cases was 2.6 ±1.4. The relative genetic fitness of the individuals with NF1, analysed for 68 index cases inwhich both parents were seen by us and 46 adult index cases in which all the children wereseen, was 0.48 (0.22/0.46). Calculated for both sexes separately, it was 0.24 for males and 0.72for females.

44

Linkage studies were performed on 20 families. The inherited mutation was of paternalorigin in 6/7 of the familial cases. Two cases with a deletion of the NF1 gene were identified,one familial case shown to have a deletion encompassing the loci from EVI-20 to INT-38 anda sporadic case with a deletion of the INT-27 locus. Thus deletions were found in 2/66families analysed (3%). The sporadic deletion occurred in the maternally derivedchromosome 17. No recombinants were observed for NF1, nor could any linkagedisequilibrium be demonstrated with the linked polymorphic markers used here.

5.1.2. Clinical findings regarding neurofibromatosis type 1(II, III, IV, V, VI)

Out of the total of 197 NF1 patients identified in the area, full clinical data were available for164 (83%). Adequate clinical details were available for four of the five NF2 cases recognisedand for all eight segmental NF cases.

Diagnostic features according to the NIH diagnostic criteria (Table 3) could beevaluated in 164 NF1 patients, comprising 111 probands (68%) and 53 affected relatives(32%). All these cases were examined by the author. The figures does not include twofemales who had inherited either the family’s risk haplotype (a 15-year-old girl) or differenthaplotypes from their affected father ( a pair of eight and 11-year-old sisters) but showed noclinical signs of NF1 (paper I). 85 patients were males and 79 females. A total of 56% of the164 NF1 cases seen by the author had other affected family members (Table 9). Café au laitspots were the most common finding (96%), followed by freckles (87%), Lisch nodules(70%), neurofibromas (69%), plexiform neurofibromas (20%), optic glioma (20%) andpseudarthrosis (3%). The mean number of diagnostic features present in the NF1 patientswas four (range from two to six). The frequencies of the NIH criteria in the age-groups areshown in Table 9.

.

45

CLS were detected in all the patients in the age-groups from three months up to 30 years,but were less frequently visible in the older age-groups. A total of 129 of those who had CLS(82%) presented with between six and 50 macules, and 22 (14%) had more than 50. Frecklesin the axillary or inguinal region were seen in all the individuals aged 11 to 20 years, 64% ofthose under the age of 11 years and 95% of those aged 11 years or older. Even in the agegroup under 6 years, 52% of the individuals had freckles. The number of neurofibromasincreased from 4% in the age-group under six years to 100% in that over 41 years.Neurofibromas were recorded more often in females (75%) than in males (64%), but thedifference was not significant (p=0.7). Cutaneous and subcutaneous neurofibromas wereboth seen in about 70% of cases, and pendulous neurofibromas in about 40% of the affectedpatients. The number of cutaneous neurofibromas ranged from 10 to 100 in 45% of cases,and exceeded 100 in 18%. A total of 77 patients underwent a MRI scan of the spine, andspinal neurofibromas was observed in 26% of them. These included two patients from thefamily with familial spinal neurofibromatosis. A plexiform neurofibroma was diagnosed in33 patients (20%), the youngest being a boy aged one year and three months who had atumour at the back of the neck. In nine cases the skin was found to be hyperpigmented ontop of the plexiform neurofibroma, and in two the skin was also hairy.

A total of 123 NF1 patients were studied by a (neuro-)ophthalmologist, yielding findingsof Lisch nodules (≥ 2) in 92 cases (75%). In six children they were found only in one eye, atages of four, six, six, seven, eight and ten years. The youngest child with a diagnosticnumber of iris Lisch nodules (≥ 2) was four years old. These nodules were seen in 81% ofcases in the age-group older than five years, and in 89% of those over the age of 20 years.The number of Lisch nodules ranged from two to ten in 47% of cases, and exceeded ten in50%. Optic glioma was found in 24/121 (20%) of patients. The mean age at diagnosis was 11± 11 years (range two to 40 years), and in the children under 16 years (n=18) it was 6 ± 2.8years (range two to 12 years). A total of 19 (79%) of the cases were symptomatic in theophthalmological examination and 5 (21%) asymptomatic.

14 cases of pseudarthrosis were ascertained and all were evaluated for other findingssuggestive of NF1. Pseudarthrosis was found in 5 males and 9 females. NF1 was furtherdiagnosed in 6 of them, all with the tibial bone affected and pseudarthrosis diagnosed at theage of one year or earlier (paper VI).

Additional dermatological findings included haemangiomas in 9/164 patients (5%),hypopigmented macules in 16/164 (10%) and hyperpigmented regions with a dark “dirtyskin“ impression in 50/164 cases (31%). Troublesome itching was reported by 30 subjects(18%), 21 of whom suffered from atopia. A total of 23 out of these 30 patients (77%)experienced itching of the skin above the neurofibroma or in areas where neurofibromasappeared later. Abnormalities of the teeth were seen in 68 cases (42%), including 39 withincreased caries, nine with positional abnormalities, and 20 cases reported to have aprosthesis by the age of about 30 years. Two patients had cysts in the jaw-bone.

Short stature ( ≤ -2SD for age and sex in Finnish children) was recorded in 18% of thepatients, 22% of the females and 14% of the males (p=0.3). A markedly asymmetric face wasseen in 13 patients, and other dysmorphic features in seven. Five patients had featurescharacteristic of Noonan syndrome, although other members of these families with NF1 didnot have any signs of Noonan syndrome. Macrocephaly (head circumference ≥ +2SD forage and sex) was diagnosed in 29% (46), this being more common in the males (36%) than inthe females (22%), (p=0.08) and there were 5 additional cases with macrocephaly who all had

46

hydrocephalus as well, 3 of them because of an aqueduct stenosis and 2 because of atumour. These 5 patients were excluded from the macrocephaly calculations.

The MRI scan of the brain showed high-signal-intensity lesions in T2-weighted images

in 21 of the 31 cases of macrocephaly (68%). Twelve patients had macrocephaly and learningdifficulties, and nine of these showed high-intensity lesions. A total of 35/72 NF1 patientswith normal head circumference (47%) showed high-signal-intensity lesions in T

2-weighted

MR images. These findings were significantly more frequent in the younger patients, beingseen in 84% of cases under 16 years of age and in 24% age 16 years or older (p=<0.001).Other major MRI findings included one previously undiagnosed patient withhydrocephalus and another with an adenoma of the hypophysis.

Malignant tumours were found in 11 cases (7%), the most common being a malignantperipheral nerve sheath tumour, which occurred in five patients at the site of a previousplexiform neurofibroma (paper II and IV). Of the total of 164 NF1 patients in the series, sevendied during the seven-year monitoring, and in three of them the cause of death was a malig-nant peripheral nerve sheath tumour and its metastases, at ages of 23, 26 and 56 years. Thefrequencies of symptoms and signs other than diagnostic features are reported in Table 10.

Learning difficulties at school and adult age and delayed developmental milestones atpre-school age were observed in 46 patients (28%), 33 of whom were males (72%) and 13females (28%) (p=0.001). Fine motor co-ordination problems were seen in 64 cases (39%),and the IQ was under 85 in 22 (13%), eight of these being mentally retarded (IQ<70) and theremaining 14 having a borderline intelligence with an IQ in the range 70-85. Only five of the91 adults (older than 20 years) had passed the school matriculation examination (5%),including one who was studying at university and one who had a degree in sociology.Forty-six percentages of the adults were employed, 12% were unemployed, 4% werehousewives, 29% were receiving a disability pension and 9% were receiving a retirementpension. One third of the adults had never worked outside their home. A psychiatricdiagnosis had been made in 17 adult patients (19%), pointing to chronic psychosis in fiveand depression in 12.

A total of 106 NF1 patients (65%) suffered from one or more symptoms related to NF1and needed either medical intervention, rehabilitation or observation. In 63 (38%) thepatient’s medical problems had been treated before NF1 was actually diagnosed. The meannumber of symptoms per patient was 2 ± 1, with a range from one to five. The findings areshown in Table 10. Pseudarthrosis proved to be a perplexing orthopedic problem in which itwas difficult to achieve and maintain bony union and good alignment after manifestation ofthe pseudarthrotic fracture. Numerous bone grafting operations and additional surgicalprocedures were needed. Nails, plates, screws and fixation devices were removed, largenumbers of plaster casts and glass fibre bandages were prepared postoperatively or afterremoval of the fixation devices, and various braces were used for early ambulation and theprevention of recurrent angulation (paper VI).

About half of the patients needed hospitalisation because of NF1, and 111 (68%) hadundergone related surgery. A total of 107 (65%) had used the primary health care service 3 ±6 times (range from one to 60) in the two years preceding the first genetic visit, and 146 (89%)had had 4 ± 7 (range from one to 80) visits to a specialist hospital outpatient department. Atpresent 50 (82%) of the patients who are alive and under 16 years and 36 (38%) of the olderones attend for regular follow-up examinations, at least once a year for children and onceevery second year for older patients (p=<0.001).

47

Table 10. Neurofibromatosis type 1: frequencies and percentages of symptomsand signs other than diagnostic features in 164 patients.

48

The mean severity grade of 164 NF1 patients on a scale where 2 denotes mild and 3moderate severity was 2.6 ± 0.9. 19/164 (11%) cases were classified as minimal, 57/164 (35%)cases as mild, 65/164 (40%) as moderate and 23/164 (14%) as severe. The severity was notrelated to sex (p=0.9), but the older patients (≥16 years) (n=101) had a more severe degree ofNF1 than the younger ones (n=63), (p=0.04), the sporadic cases (n=70) were significantlymore severe than the familial ones (n=91), (p=0.008), and the probands (n=111) had a moresevere type than their affected relatives (n=53), (p<0.001) (Table 11).

Intrafamilial variability was assessed in 13 sibships including two to four affected sibs(n=31). Two of these cases were classified as having the most severe degree of the disease(Table 11), but they came from different families and there was no relationship between theaffected sibships and the grade of severity or number of diagnostic criteria. There werealtogether nine affected individuals belonging to two families, who had only cutaneousfeatures. Parental clinical status could be analysed in 107 cases. The father was affected in21 cases and the mother in 47 cases, while in 39 cases both parents were considered healthy.The grade of severity in the offspring was not related to whether the gene was inherited fromthe mother (2.4 ± 0.9) or the father (2.4 ± 0.8) (p=0.8).

In one three-generation family seven members were affected with a rare spinalneurofibromatosis (for pedigree, see Figure 1 in paper III) also fulfilling the NF1 clinicalcriteria. The affected adults, at ages of 32, 37, 38, and 61, showed multiple spinalneurofibromas symmetrically affecting all spinal roots (Figure 2 in paper III). In addition, allthe patients had several café au lait spots, one had numerous freckles in the axillary area, andtwo had possible dermal neurofibromas, but iris Lisch nodules were not present. Other signsof NF1 or NF2 were absent. Linkage study suggested a close linkage to the NF1 locus andexcluded it from the NF2 locus.

Altogether 64% of the adults (n=91) were or had been married, and 88% (n=51) of thesehad altogether 132 children (mean 2.6, range from 1 to 11). 73% of the 34 affected women witha history of pregnancy had reported new neurofibromas and growing of the existingcutaneous neurofibromas during their pregnancies, and four had had toxaemia ofpregnancy. Three families had asked for a prenatal diagnosis, but none had eventuallyconsidered the disease to be severe enough for any further prenatal analysis, although allthe cases were indicative in terms of linkage analysis.

49

5.2. Neurofibromatosis type 5 (Segmental type) (II, IV, V)

Segmental NF was found in eight patients, two of whom were men and six women. Based onthese eight cases, the prevalence estimate for the segmental type of NF was 1.1/100 000 (1/91500). The clinical findings for the eight adult individuals are shown in Table 12. In six of theeight cases the segmental area was on the left side of the body and in four of them it was theupper quadrant. In four patients the lesions were CLS and/or freckles, and in one these wereaccompanied by neurofibromas. Three additional individuals had only a few neurofibromasand one patient had multiple Lisch nodules as the only lesion. Three patients were mothersof daughters suffering from classical NF1.

Table 12. Clinical findings in 8 patients with segmental neurofibromatosis (NF5).

.

5.3. Neurofibromatosis type 2 (IV, V)

Five NF2 patients, all male, were found in three families. Two of these cases were sporadicand three were familial. Full clinical data was available for both sporadic cases and two of thefamilial ones. The prevalence estimate for NF2 in Northern Finland on 31.12.1996 was0.5/100 000 (1/183 000). In the two familial cases the first symptom was loss of balance, atages of 28 and 31 years. In the two sporadic cases the first sign was a cervical schwannomaat the age of 14 years in one and a recurrent orbital meningioma from the age of 10 years inthe other one (see Table 3 in paper IV). A cranial MRI showed bilateral vestibular schwan-

50

nomas and a spinal MRI numerous spinal tumours in all the patients. Three of the four caseshad several cutaneous tumours and all four between one and three café au lait spots ofdiameter ≥ 1.5 cm, but no freckles. The ophthalmological examination showed an epiretinalmembrane, presenile cataract, nystagmus and corneal opacity, but no Lisch nodules.

51

6. Discussion

The search for patients with NF in the Oulu University Hospital region covered ageographically well defined area of Northern Finland with a total population of 733 037 andan area of 155 211 square kilometres (14% of the country’s total population and 46% of itstotal area) (Figure 5). Several methods of ascertainment were used and many of the indexcases were ascertained through various sources indicating that the patient search coveredthe region relatively well. Mildly affected patients diagnosed and treated only in localhealth centres may have been missed, however, and some undiagnosed cases and possiblysome unreported cases must certainly exist. Thus the incidences and prevalences of NFcalculated here should be taken as minimum values for Northern Finland.

6.1. Population genetics of neurofibromatosis type 1 (I)

The estimated overall prevalence of NF1 in Northern Finland was 1/4436, with the highestperiod prevalence 1/2983, in the age group 10 to 19 years. Only four population-basedepidemiological studies of NF1 have been produced previously, and the prevalences in twoof them (Table 5) are of the same magnitude, being 1/4600 in Gothenburg, Sweden(Samuelsson 1981) and 1/4150-1/4950 in Wales (Huson et al. 1989a). The third study, fromNew Zealand, reported the highest overall prevalence (1/2190) (Fuller et al. 1989), whichcould partly be explained by a small population effect, with a base population of 113 700.The fourth report, from Italy allows the prevalence to be estimated, giving a lower rate thanelsewhere (1/6711) (Clementi et al. 1990). This finding is explained by the different methodsof ascertainment, as the primary material consisted only of probands referred for geneticcounselling because of CLS. In the Welsh series the observed figures were corrected byadding extra cases to compensate for the loss of mildly affected young individuals with nofamily history of NF, whereas in the present work use is made of period prevalencecalculations for different age-groups to find the most reliable prevalence estimates.Increased mortality will obviously reduce the number of NF1 cases in the older age groups.

The overall birth incidence of NF1 was estimated to be 1/3647, but it varied from one timeperiod to another, being highest during the six-year period 1990-1995 (1/2703). The overallbirth incidence estimate in Wales was in the same range 1/2558 (Huson et al. 1989a). Given

52

about 9000 live births per year in Northern Finland, one can estimate that there will be threenew NF1 children born annually.

The diagnosis of NF1 has clearly improved with time as shown by a drop in the mean ageat diagnosis in sporadic cases in the present material by an average of 10 years 16 ± 9 yearsin the 1960’s to 6 ± 4 years in the 1980’s. The age-dependent prevalence/incidence figureswere also significantly higher in the younger age groups. This may reflect a betterawareness of the importance of diagnosing NF in suspected cases, improved knowledge ofthe diagnostic features and a greater willingness to refer suspected cases for detailedexamination. The findings suggest that the incidence figures may be most reliable forpatients under 20 years of age, and the age-related period prevalences more accurate in thesame age-groups.

The mutation rate for the NF1 gene, 4.37 ± 0.72 x 10-5, is comparable to those observed inthe literature (Table 5), and confirms the very high figures reported previously. A birth ordereffect, that sibs born later are more likely to be affected, was also found here suggesting thatparental age has an effect on the mutation rate. The mean paternal age in the cases with anew mutation was significantly higher than in the general population, and the maternal agewas also much higher among the NF1 mothers as compared with the average NorthernFinland population. Similar findings have been reported previously by Riccardi (1992) andBunin et al. (1997), but the present linkage analysis and those in the literature have shownthat most of the new mutations are of paternal origin, suggesting that an elevated maternalage in new mutation cases might not be so important. In the informative families examined bylinkage study, six out of the seven new mutations were of paternal origin, as has also beenobserved in 28 out of the 30 previously published cases (93%) (Jadayel et al. 1990, Stephenset al, 1992, Elyakim et al. 1994). On the other hand, more recent reports on families withintragenic gross NF1 gene deletions have shown that the majority of them arise on thematernal chromosome 17 (Upadhyaya et al. 1998).

As expected, no indication was found here of a founder effect for NF1 in northernFinland. Most of the families with more than one affected individual were small, about 2/3 ofthem being two-generation families, and only four four-generation families were found, anobservation wich is in accordance with previous studies (Samuelsson 1981, Huson et al.1989a, Riccardi 1992). No uneven geographical clustering was observed, either. Studies withclose linked markers gave no indication of linkage disequilibrium.

An effect of NF1 on propagation was observed here, as previously. Genetic fitness in thepresent material had decreased to about half of the expected in the presence of NF1 (0.48),the effect being more marked in males (0.24) than in females (0.72). Interestingly, more markedreductions in males were also found by Crowe et al. (1956) and Huson et al. (1989a), wherethe observed fitnesses for all males were 0.41 and 0.31, respectively, those for all females 0.75and 0.60 and the overall fitnesses 0.53 and 0.47. It has been suggested that the overallreduction in genetic fitness in terms of fecundity is due partly to biological factors and partlyto non-biological factors such us selection against affected subjects marrying (Crowe et al.1956, Huson et al. 1989a).

The segregation analysis of the NF1 families in Northern Finland agrees with autosomaldominant inheritance, with a segregation ratio of 0.44, which is comparable to the figuresquoted by Huson et al. (1989a) and Clementi et al. (1990).

New mutations accounted for a maximum of 49% of the 197 NF1 patients whenanamnestic data obtained from the family and/or from the hospital records were also used, a

53

figure that is in agreement with those published earlier (Samuelsson 1981, Huson et al.1989a, Clementi et al. 1990). On the other hand, a group of cases identified during the period1960-1995 in whom both parents of the index patient could be examined included only 32%new mutations. This figure is much lower than the previously estimated overall mutationfigure of 49% and can partly be explained by the small number of parents examined (37 pairs).However, it might also suggest that familial cases are more common than had been thought,because mildly affected family members are easily overlooked without a systematic clinicalexamination. In order to accept or reject the proposition of new mutations being lesscommon than was previously thought, more families with both parents available should bestudied.

In three families a daughter had classical NF1 while her mother had segmental NF. Thisfinding is exceptional and has been described earlier in only seven families (Rubenstein etal. 1983, Boltshauser et al. 1989, Riccardi 1992, Moss & Green 1994, Huson & Ruggieri 1996),in four of which the affected parent was the father and in three the mother. The findings inthese families suggest that the disorder in the parent-child pairs are of same origin and that,in addition to segmental NF, the affected parent has mosaicism of the NF1 gene. If so theindividuals with segmental NF represent mosaics for the NF1 gene (Huson 1998).

In the linkage study of 20 NF1 families the presence/absence of the linked haplotypeswas concordant with the clinical phenotypes. Exceptional observations were seen, however,in two families where linkage analysis showed the NF1 risk haplotype in a healthy child of anaffected parent. One explanation is that the children/adolescents were still at theasymptomatic stage. This would be exceptional, as all the affected cases in this series hadalready developed a diagnostic number of café au lait macules or some other diagnosticfeatures by the age of five years, as also in earlier reports (Huson et al. 1989a, Riccardi 1992,Wolkenstein et al. 1996). Another explanation would be that the disease was not linked toNF1 gene in these families. This is unlikely, because the affected parent in both familiesfulfilled the NIH diagnostic criteria for NF1 (CLS and freckles in the first family, andneurofibromas and Lisch nodules in the second family). A third explanation would bemosaicism in a parent with NF1 in whom part of the germ-cells do not carry the NF1 mutation.A fourth explanation would be non-penetrance of the NF1 mutation in the individuals inquestion. Non-penetrance has previously been reported in three cases, a 50-year-oldunaffected woman who had an affected brother, son and grandson, an 45-year-oldunaffected man with an affected mother and daughter (Spence et al. 1983), and a healthyindividual with an affected father and two affected daughters (Spence et al. 1986). If thepresent individuals represent non-penetrance its frequency can be estimated to be high as3.7% (2/54). Although non-paternity is not probable one should exclude it with othermarkers. Further molecular genetic studies are needed to resolve the role of this finding andit is going to be finally answered only after the families NF1 mutation has been found.

The present intragenic linkage studies pointed to two cases with a deletion, 3% of thoseinvestigated (n=66). Elsewhere such a mutation has been reported in only 10-20% of casesin spite of the use of more modern techniques based on both DNA and RNA analysis (Shenet al. 1996). A total of 246 mutations involved in NF1 had been reported to the NNFFinternational NF1 mutation analysis consortium up to February 1999, 45% of them deletions.

54

6.2. Clinical findings in neurofibromatosis type 1 (II-VI)

NF1 in Northern Finland showed mostly the same phenotype and clinical heterogeneity asobserved in other clinical studies.

The mean age at diagnosis in the whole material was relatively high (20 ± 16 years), butwas significantly lower in the younger generations (10 ± 9 years), those born in 1960 to 1995(p<0.001). This is most likely due to improved diagnostic methods. According to the presentresults, NF1 should be well diagnosable before the age of six years, with only a fewexceptional cases, as has been reported previously by others (Huson et al. 1988, Obringer etal. 1989, Riccardi 1992, Cnossen et al. 1998a, Friedman & Birch 1997). The youngest patientin this series was only three months of age, but there were no cases in the discharge recordsin which NF1 was detected or reported in a newborn infant. All 27 children belonging to theyoungest age group (under six years) presented with a diagnostic number of CLS, about ahalf had freckles and one out of five had Lisch nodules. It is interesting that 70% of thechildren in this age group had an affected parent, as compared with the 56% in the wholematerial, indicating that children of affected parents may not at present be affected any moreoften earlier, but the diagnosis is suspected earlier.

It was possible to diagnose NF1 and differentiate it from NF2 and segmental NF purelyon clinical grounds and by taking into consideration the information on first degreerelatives. Additional clinical and imaging studies were useful to diagnose NF1 in somecases, however, and in many cases they were used to extend the phenotype. The minimumnumber of six CLS was observed in 96% of the cases in the present material, as comparedwith 84 to 100% in the other populations studied (Table 4). CLS are important diagnosticcriteria but not pathognomonic for NF1; e.g. the prevalence of one to five CLS amonginstitutionalised unselected adult individuals has been reported to be as high as 10%(Crowe & Schull 1953). On the other hand, freckles in the axillary or inguinal area were seenin 87% of the present patients, and these have not been reported in healthy individuals andonly rarely in other diseases, with cutaneous manifestations such as Addison’s disease,Leopard’s syndrome or Watson’s syndrome.

A total of 14 cases with pseudarthrosis were identified in Northern Finland, and six ofthem had NF1, the percentage among the NF1 patients being 3%, as compared with 1% to4% in other studies. The remaining eight cases found here confirmed that pseudarthrosis isan independent disease which is connected with NF1 in about half of all cases, a finding alsoreported in the literature (Fairbank 1994).

A common neuroimaging finding in the younger age-groups was high intensity T2

signals, noted in 94 % of the children under the age of six years, and 84% of those under 16years who had had a MRI head scan. This finding is so common that it could even be usedas a diagnostic criteria, as reported by Itoh et al. (1994) and North (1997). The presentfindings indicate that the agreed diagnostic criteria are valid and help to achieve an earlydiagnosis, but that additional features such as high intensity T

2 signals may accelerate an

early diagnosis if accepted as diagnostic criteria.In addition, molecular genetic analysis either by linkage analysis in familial cases or

though the finding of a mutation may be used to confirm the diagnosis in special situations.In the present linkage study of 20 NF1 families there was concordance with the clinicalphenotype and with the family’s risk haplotype in 52 individuals. Only in two cases didunaffected healthy young individuals who carried the family’s risk haplotype fail to show

55

any signs of NF at the clinical examinations. Such cases need monitoring andcharacterisation of the NF1 gene mutation in the family in question to find out whether theyin fact have NF1 or whether they represent cases of non-penetrance of the NF1 gene defect.At present mutation analysis is not diagnostic in the major of families, and more advancedmolecular genetic techniques await development (Huson 1999). On the other hand, thepossibility of performing linkage analysis in familial cases also raises ethical questions ofpredictive diagnosis in a clinically unaffected family member. Especially where a small childor fetus is concerned, we are able to tell the family the disease status but neither the severityof the disorder nor its prognosis.

The family in which seven members in three generations were affected with a rare spinalneurofibromatosis (paper III) provides additional evidence that the rare form of NF1 firstreported in two families by Pulst (1991), and recently in another three-generation Spanishfamily by Ars (1998) actually exists. One of the families observed by Pulst, like the familydescribed here, showed linkage to the NF1 locus, while a frameshift mutation in the NF1gene was found in the Spanish family. Spinal neurofibromas were seen in MRI in 20/77(26%) of the present cases, which was more often than had been suspected in the clinicalexamination. The presence of a wide, symmetrical distribution of spinal neurofibromas israre, however, and the multiple spinal neurofibromas affecting all spinal roots symmetricallyin all adult members of this family are exceptional. In families with symmetrical appearanceof neurofibromas it is possible that the gene defect has a special effect on the neural crestcells and their precursors in the nerve roots. In a case of suspected spinalneurofibromatosis it is advisable to perform spinal MRI on all first degree relativessuspected of having a precocious appearance of neurofibromas in the spinal areas evenwithout any other common features of NF1.

Another unusual clinical observation concerned eight cases with mainly cutaneousfeatures of NF1 in one segment or in a few dermatomes of the body. They all showed thesegmental type of NF, which has previously been described in only about 150 cases (Huson1998).

NF1 has been considered to be a slowly progressive disorder with hamartomatouslesions increasing in number and size with age, and the increasing appearance of diagnosticfeatures in older age groups found in the present material supports this opinion. In spite ofthe progressive nature of the disorder in Northern Finland, its mean grade of severity wasbetween mild and moderate (2.6), and in about half of the individuals (47%) it remainedeither a cosmetic disorder or a relatively well accepted familial disorder (Table 11). Similarresults have been reported in Wales by Huson et al. (1989a), where the mean severity wascloser to mild (2.2), and in Houston, Texas, by Riccardi (1992), with a mean severity similarto the current result (2.6).

Even though it has been suggested that NF1 could already be diagnosable in newborninfants, there are no case reports of this in the literature. Similarly, the youngest patient inthe present series was three months of age. On the whole, the main reason for the furtherexamination of young infants at the University Hospital was to confirm the diagnosis, andonly very few infants had any symptoms related to the disorder that needed medicaltreatment, with the exception of those with pseudarthrosis. All the cases withpseudarthrosis of the tibia were found before or at one year of age. The medical problemsentailed in the rare congenital plexiform neurofibroma are usually cosmetic, with possibleassociated hemihypertrophy (Gutmann et al. 1997). A plexiform neurofibroma at the back of

56

the neck in one infant in this series caused both a cosmetic and a diagnostic problem, butwithout any symptoms at the age of one year and four months.

Optic glioma and learning difficulties were the main complications of NF1 in childhoodand adolescence with frequencies of 11/36 (31%) and 19/36 (53%), respectively. Of thechildren with an optic glioma, 8/11 (73%) had vision problems. These were mild in five cases,causing reduced visual acuity and bilateral visual field defects, but in three children thesymptoms were so serious that operative and/or radiation therapy were needed. Learningdifficulties were a common problem in childhood and adolescence and were found in 53% ofcases. Learning difficulties were observed in 28% of cases in the whole material, a lowerfigure than that reported in North America (70%; Riccardi 1992), and in about the same rangeas in Wales (33%; Huson 1994), or in the NNFF international database (40%; Friedman &Birch 1997) (see Table 4 in paper II). The overall frequency of intellectual handicap (IQ<85)seemed to be higher than in the previous reports, while that of mental retardation (IQ<70)was 5%, i.e. comparable to that found elsewhere (Riccardi 1992, Huson 1994, Friedman &Birch 1997).

The main features needing a medical follow-up in adulthood were psychiatric problems,problems related to pregnancy, and the increased risk of malignant tumours. Many patientsappeared to have problems with coping with NF, including cosmetic problems. 17/91 adults(19%) were treated for a psychiatric diagnosis, the most common being chronic depression(12/17). Zöller (1997), in her recent 12-year follow-up study of 70 adult NF1 patients, reportsthat about 1/3 of them had a mental illness. She also found that the disease had a significantinfluence on psychiatric symptoms and personality variables, dysthymia being the mostcommon diagnosis. The present psychosocial variables describing how an affectedindividual is coping with NF showed that 26/91 (29%) were receiving a disability pension, asituation that was more common than in the general population (8%), 58/91 (64%) were orhad been married and 55/91 (60%) had children, both being in about the same range as in thegeneral population. The NF1 individuals as a group were less educated than the generalpopulation, only 5/91 (5%) having passed the school matriculation examination. Thepregnancies of 34 NF1 women were mostly normal, with the exception of four cases wheretoxaemia of pregnancy was reported, suggesting that there might be a small subgroup ofNF1 women who need special monitoring of pregnancy, a finding also observed earlier(Dugoff & Sujansky 1996).

The most difficult complications were malignant tumours, which were observed in 8/91of the adults. They were most common in the age group 21-30 years, where 3/22 (14%) diedof complications of NF during the seven-year monitoring period, indicating an increased riskof mortality. The most common cause of death was malignant peripheral nerve sheathtumour (MPNST), in five cases, i.e. one out of every seven plexiform neurofibromas in thismaterial showed malignant transformation (see paper II and IV). Similar findings are reportedin other two Scandinavian long-term follow-up studies, where a 42-year follow-up pointed toa relative risk of 4.0 (95 % confidence limits 2.8 to 5.6) of probands with NF1 developing amalignant neoplasm or a benign central nervous system tumour (Sorensen et al. 1986) and atwelve-year follow-up study in Gothenburg, Sweden, by Zöller et al. (1995) reported 22deaths among 70 adult NF1 patients and a malignancy in 55% of them, whereas 5.1 deathswere expected in the general Swedish population (p<0.001).

57

The latter authors also reported a reduction of about 15 years in the mean age at death,to 61.6 years as compared with 75 years in the population statistics. These results reveal thatNF1 is a progressive disease with a higher risk of premature death.

6.3. Neurofibromatosis type 2 (IV, V)

The prevalence figures obtained here and the figures from England (Evans et al. 1992b) lie inapproximately the same range. The calculated prevalence of NF2 was 0.5/100 000, but thisfigure must be considered only a rough estimate, due to the small number of cases.

The diagnostic criteria and features of NF2 are more flexible than those described forNF1, and include a subgroup of individuals who have possible NF2 and need furthermonitoring to resolve the situation. In addition, it has been shown that molecular geneticanalysis is at present more helpful in the question of NF2 than with NF1, and it has beenestimated that the mutation can be found in about 50% of NF2 cases (Gutmann et al. 1997,Huson 1999). Similar difficulties were encountered in diagnosing NF2 here as elsewhere(Evans et al. 1992b, Parry et al. 1994). In both sporadic cases the diagnosis was reachedaccidentally while treating one of the rare features of NF2 that needed medical attention. Adiagnosis was similarly reached in the familial cases only after a complication connectedwith NF2 had been recognised. A clinician who is experienced with NF2 is needed to suspectit, especially when the first sign is not a vestibular schwannoma.

NF2 as a disease in general shows remarkable interfamilial variation, but it is usuallysimilar in all affected family members of the same family.

6.4. Genetic counselling, care and follow-up of neurofibromatoses (II)

Due to the complexity and wide variability of NF, all affected individuals and their familiesneed a proper clinical examination and genetic counselling, preferably by an experiencedclinical genetic team. A special examination form as used in the present study (Appendix 1)and by others (Friedman et al. 1993) will guarantee that all the points involved in NF areexamined carefully. It also allows a systematic follow-up of the various featuresprospectively during monitoring. Careful clinical examination of first degree relatives wasshown to be the best way to acquire adequate information on their NF status. In addition itwas shown to be the only way to find those relatives who are mildly affected and those whomight present with segmental NF, but who in all probability still have the NF1 gene and thusare at risk of passing it to the next generation. It is suggested that only if both parents areclinically examined and found healthy is it possible to be fairly sure that the affectedindividual represents a new mutation. The risk for the next child may be low, but the risk ofnon-penetrance might be worth discussing. Even though non-penetrance has very seldombeen reported with NF1 (Spence et al. 1983, Spence et al. 1986), the present findings of twopossible non-penetrance cases raise the question again, and more research is needed toreject or prove the role of possible non-penetrance in NF1.

58

Early diagnosis is needed to alleviate unnecessary worries, and collaboration betweenspecialists such as a paediatrician, neuropaediatrician, dermatologist, neurologist,ophthalmologist, oncologist, pathologist, audiologist, neurosurgeon, paediatric surgeon,internist and clinical geneticist is needed to recognise those NF cases that need immediatemedical attention and to organise monitoring. In agreement with earlier reports (Riccardi1992, Huson et al. 1989b), it was found that patients and their families appreciated geneticcounselling and follow-up visits, during which the whole picture of NF with its variabilityand unpredictability was clearly explained. In an ideal system it should be every patient’sand family’s option to visit a NF clinic at least once, while follow-up examinations canusually be organised in the patient’s own local hospital following recommendations from theNF team.

The results gained here show that individually scheduled monitoring is needed from theviewpoint of both the patient and the family. In the diagnostic phase a follow-up examinationmay be needed half-yearly or yearly to confirm the diagnosis and to help the family adapt tothe situation. Later on, monitoring is needed in cases of NF1 in particular, for the earlydetection of possible developmental delay, macrocephaly or optic glioma (Huson et al. 1988,Wolkenstein et al. 1996, Gutmann et al. 1997, Cnossen et al. 1998a). The value of a cranialMRI and/or regular neuro-ophthalmological follow-up examinations should be carefullyevaluated (Listernick et al. 1997). In young adulthood or adolescence patients may needindividual counselling on the inheritance and genetic aspects of NF. Adult patients shouldbe instructed to note any changes in their condition, such as a sudden increase in the size ofthe existing tumour, and for this purpose individually scheduled monitoring is needed.

59

7. Summary and conclusions

NF1 appeared to be as common in Northern Finland as in other geographical areas, with anoverall prevalence of about 1/4000 and a birth incidence of 1/3500. These are probablyminimum estimates, because there will always remain some undiagnosed cases and somepatients may die prematurely, and if these factors are taken into account, more reliableprevalence and birth incidence figures would be 1/3000 and 1/2700. No clusters of patientsor founder effect were detected, and most of the families identified were small.

Genetic fitness was reduced. About half of the cases were new mutations, even thougha closer estimate might give lower figures, most of the new mutation cases were of paternalorigin and the mutation rate was similar to those reported elsewhere. The penetrance of theNF1 gene was apparently very high, but the observed possible non-penetrant casessuggest that it was probably not 100%. Nearly all of the cases examined had classical NF1,with a few exceptional phenotypes observed, among them one family with spinalneurofibromatosis, pointing to the usefulness of MRI for diagnosing NF, and eight raresegmental NF cases, including three families in which a mother had segmental NF and herdaughter classical NF1, suggesting that these three affected parents may show a uniquemosaic form of NF1.

The established clinical diagnostic criteria for NF1 were found to be excellent. Theproblem is that they are not commonly enough known, so that the diagnosis of NF is stillmade too late, even though the age at diagnosis is lower nowadays than it used to be. Of thefeatures other than diagnostic criteria, high intensity T

2 lesions were so common that they

could be used for early diagnostic purposes, especially in early childhood. The usefulnessof close polymorphic NF1 gene markers was also observed, confirming their profitability insupporting the diagnosis and making prenatal diagnosis possible in informative families.One problem is possible reduced penetrance of the NF1 gene.

NF1 was found to be of same severity in Northern Finland as elsewhere, most of thecases being mildly or moderately affected. A constant finding was a progression of thedisease over the years, reflected in an increase in the number of clinical diagnostic criteria.Interfamilial and intrafamilial variation was high and a number of patients needed regularmedical treatment. In addition, it was significant how much NF affected the individual’s lifethrough regular hospital visits, the achievement of a lower education than in the generalpopulation, difficulties in finding a permanent job and effects on family planning. In

60

addition, a higher risk of malignancy and mortality was found. In conclusion, NF proved tobe a complicated and remarkable disease even beyond its somatic symptoms.

Because of its pattern of inheritance and its great clinical and individual variability, NF1should be recognised as early as possible, already in early childhood if possible, and thepatient and family should be conducted to genetic counselling. The above observationsconsidered together confirm that a multidisciplinary NF team is needed to take care of NFpatients and their families.

61

8. References

Ahn MS, Jackler RK & Lustig LR (1996) The early history of the neurofibromatoses: evolution ofthe concept of neurofibromatosis type 2. Arch Otolaryngol - Head & Neck Surg 122: 1240-1249.

Ainsworth PJ & Rodenhiser DI (1991) Rapid nonradioactive detection by PCR of pHHH202/RsaIRFLP linked to neurofibromatosis type I. Am J Hum Genet 49: 1098-1099.

Akbarnia BA, Gabriel KR, Beckman E & Chalk D (1992) Prevalence of scoliosis in neurofibromatosis.Spine 17: 244-248.

Allanson JE, Hall JG & van Allen MI (1985) Noonan phenotype associated with neurofibromatosis.Am J Med Genet 21: 457-462.

Allanson JE, Upadhyaya M, Watson GH, Partington M, MacKenzie A, Lahey D, MacLeod H,Sarfarazi M, Broadhead W, Harper PS & Huson SM (1991) Watson syndrome: is it asubtypeof type 1 neurofibromatosis? J Med Genet 28: 752-756.

Aloi FG & Massobrio R (1989) Solitary plexiform neurofibroma. Dermatologica 179: 84-86.Andersen KS (1978) Congenital pseudarthrosis of the tibia. Thesis. Copenhagen.Ars E, Kruyer H, Gaona A, Casquero P, Rosell J, Volpini V, Serra E, Lázaro C & Estivill X (1998) A

clinical variant of neurofibromatosis type 1: familial spinal neurofibromatosis with a frameshiftmutation in the NF1 gene. Am J Hum Genet 62: 834-841.

Bahuau M, Hoydayer C, Assouline B, Blanchet-Bardon C, Le Merre M, Lyonnet S, Giraud S, RécanD, Lakhdar H, Vidaud M & Vidaud V (1998) Novel recurrent nonsense mutation causing neurofibromatosis type 1 (NF1) in a family segregating both NF1 and Noonan syndrome.Am J Med Genet 75: 265-272.

Baptiste M, Nasca P, Metzger B, Field N MacCubbin P, Greenwald P, Armbrustmacher V, WaldmanJ & Carlton K (1989) Neurofibromatosis and other disorders among children with CNStumours and their families. Neurology 39: 487-492.

Barker D, Wright E, Nguyen K, Cannon L, Fain P, Goldgar D, Bishop DT, Carey J, Baty B, Kivlin J,Willard H, Waye JS, Greig G, Leinwand L, Nakamura Y, O’Connell P, Leppert M, Lalouel J-M,White R & Skolnick M (1987) Gene for von Recklinghausen neurofibromatosis is in thepericentromeric region of chromosome 17. Science 236: 1100-1102.

Benjamin CA, Colley A, Donnai D, Kingston H, Harris R & Kerzin-Storrar L (1993) Neurofibro-matosis type 1 (NF1): knowledge, experience, and reproductive decisions of affected patients andfamilies. J Med Genet 30: 567-574.

Blanquet V, Turleau C, Gross-Morand MS, Sénamaud-Beaufort C, Doz F & Besmond C (1995)Spectrum of germline mutations in the RB1 gene: a study of 232 patients with hereditary and nonhereditary retinoblastoma. Hum Mol Genet 4: 383-388.

Boltshauser E, Stocker H & Machler M (1989) Neurofibromatosis type 1 in a child of a parent withsegmental neurofibromatosis (NF-5). Neurofibromatosis 2: 244-245.

Borberg A (1951) Clinical and genetic investigations into Tuberous Sclerosis and Recklinghausen’sNeurofibromatosis. Acta Psychiatr Neurol Scand 71: 1-239.

Bourn D, Evans G, Mason S, Tekes S, Trueman L & Strachan T (1995a) Eleven novel mutations in theNF2 tumour suppressor gene. Hum Genet 95: 572-574.

62

Bourn D & Strachan T (1995b) Highly polymorhic dinucleotide repeat at the NF2 gene. Hum Genet95: 712

Bouzas EA, Mastorakos G, Chrousos GP & Kaiser-Kupfer MI (1993) Lisch nodules in Cushing’sdisease. Arch Ophthalmol 111: 439-440.

Bouzas EA, Parry DM, Eldridge R & Kaiser-Kupfer MI (1992) Familial occurrence of combinedpigment epithelial and retinal hamartomas associated with neurofibromatosis 2. Retina 12:103-107.

Boyd HB (1982) Pathology and natural history of congenital pseudarthrosis of the tibia. Clin OrthopRel Res 166: 5-13.

Brannan CI, Perkins AS, Vogel KS, Ratner N, Nordlund ML, Reid SW, Buchberg AM, Jenkins NA,Parada LF & Copeland NG (1994) Targeted disruption of the neurofibromatosis type-1gene leads to developmental abnormalities in heart and various neural crest-derived tissues.Gen Develop 8: 1019-1029.

Bunin GR, Needle M & Riccardi VM (1997) Paternal Age and Sporadic Neurofibromatosis 1: A case-control study and consideration of the methodologic issues. Genet Epidemiol 14: 507-516.

Carey JC, Laub JM & Hall BD (1979) Penetrance and variability in neurofibromatosis: a genetic studyof 60 families. Birth defects XV: 271-281.

Carey JC & Vickochil DH (1999) Neurofibromatosis Type 1: A model condition for the study of themolecular basis of variable expressivity in human disorders. Am J Med Genet 89: 7-13.

Cawthon RM, Andersen LB, Buchberg AM, Xu G, O’Connell P, Viskochil D, Weiss RB, Wallace MR,Marchuk DA, Culver M, Stevens J, Jenkins NA, Copeland NG, Collins FS & White R(1991) cDNA sequence and genomic structure of EVI2B, a gene lying within an intron of theneurofibromatosis type 1 gene. Genomics 9: 446-460.

Cawthon RM, Weiss R, Xu G, Viskochil D, Culver M, Stevens J, Robertson M, Dunn D, GestelandR, O’Connell P & White R (1990) A major segment of the neurofibromatosis type 1 gene:cDNA sequence, genomic structure, and point mutations. Cell 62: 193-201.

Central Statistical Office of Finland. Official statistics of Finland. Structure of Population and VitalStatistics. Whole country and Provinces. Population 1960-1996. Helsinki: Valtion

Painatuskeskus, 1960-1997.Charrow J, Listernick R & Ward K (1993) Autosomal dominant multiple café-au-lait spots and

neurofibromatosis-1: evidence on non-linkage. Am J Med Genet 45: 606-608.Charles SJ, Moore AT, Yates JRW & Ferguson-Smith MA (1989) Lisch nodules in neurofibromatosis

type 2. Arch Ophthalmol 107: 1571-1572.Chauffard MA (1896) Dermato-fibromatose pigmentaire (ou neurofibromatose généralisée). Soc Med

Hop Paris 13:777-784.Clark RD (1994) Proteus Syndrome. In: Huson SM & Hughes RAC, eds. The neurofibromatoses. A

Pathogenetic and Clinical overview. 17: 402-413. University Press, Cambridge, UK.Clementi M, Barbujani G, Turolla L & Tenconi R (1990) Neurofibromatosis-1: a maximum likelihood

estimation of mutation rate. Hum Genet 84: 116-118.Cnossen MH, de Goede-Bolder A, van den Broek KM, Waasdorp CME, Oranje AP, Stroink H,

Simonsz HJ, van den Ouweland AMW, Halley DJJ & Niermeijer MF (1998a) Aprospective 10 year follow-up study of patients with neurofibromatosis type 1. Arch DisChild 78: 408-412.

Cnossen MH, Moons KGM, Garssen MPJ, Pasmans NMT, de Goede-Bolder A, Niermeijer MF,& Grobbee DE and the neurofibromatosis team of Sophia Children’s Hospital (1998b)Minor disease features in neurofibromatosis type 1 (NF1) and their possible value indiagnosis of NF1 in children £6 years and clinically suspected of having NF1. J Med Gen 35:624-627.

Cnossen MH, van der Est MN, Breuning MH, van Asperen CJ, Breslau-Siderius EJ, van der PloegAT, de Goede-Bolder A, van den Ouweland AM, Halley DJ & Niermeijer MF (1997)Deletions spanning the neurofibromatosis type 1 gene: implications for genotype-phenotype correlations in neurofibromatosis type 1? Hum Mut 9: 458-464.

Colley A, Donnai D & Evans DGR (1996) Neurofibromatosis/Noonan phenotype: a variable featureof type 1 neurofibromatosis. Clin Genet 49: 59-64.

Colman SD, Williams CA & Wallace MR (1995) Benign neurofibromas in type 1 neurofibromatosis(NF1) show somatic deletions of the NF1 gene. Nat Genet 11: 90-92.

Crawford AH & Bagamery N (1986) Osseous manifestations of neurofibromatosis in childhood. JPediatr Orthop 6: 72-88.

63

Crowe FW (1964) Axillary freckling as a diagnostic aid in neurofibromatosis. Ann Intern Med 61:1142-1143.

Crowe FW & Schull WJ (1953) Diagnostic importance of the café au lait spot in neurofibromatosis.Arch Intern Med 91: 758-766.

Crowe FW, Schull WJ & Neel JV (1956) A clinical Pathological and Genetic study of multipleneurofibromatosis. Charles C Thomas, Springfields.

Crump T (1981) Translation of case reports in Ueber die multiplen Fibrome der Haut und ihreBeziehung zu den multiplen Neuromen by F. v Recklinghausen. Adv Neurol 29: 259-275.

Ducatman BS, Scheithauer BW, Piepgras DG, Reiman HM & Ilstrup DM (1986) Malignantperipheral nerve sheath tumours. A clinicopathologic study of 120 cases. Cancer 57: 2006-2021.

Duffner PK, Cohen ME, Seidel FG & Shucard DW (1989) The significance of MRI abnormalities inchildren with neurofibromatosis. Neurology 39: 373-378.

Dugoff L & Sujansky E (1996) Neurofibromatosis type 1 and pregnancy. Am J Med Genet 66: 7-10.Dunn DW & Roos KL (1989) Magnetic resonance imaging evaluation of learning difficulties and

incoordination in neurofibromatosis. Neurofibromatosis 2: 1-5.Edwards JNT, Fooks M & Davey DA (1983) Neurofibromatosis and severe hypertension in

pregnancy. Br J Obstet Gynaecol 90: 528-531.Eldridge R (1981) Central neurofibromatosis with bilateral acoustic neuroma. Adv Neurol 29: 57-65.Elyakim S, Lerer I, Zlotogora J, Sagi M, Gelman-Kohan Z, Merin S & Abeliovich D (1994)

Neurofibromatosis type I (NFI) in Israeli families: linkage analysis as a diagnostic tool.Am J Med Genet 53: 325-334.

Emery AEH. Methodology in genetics. Second edition. Edinburg: Churchill Livingstone, 1986.Evans DGR, Huson SM, Donnai D, Neary W, Blair V, Newton V & Harris R (1992b) A clinical study

of type 2 neurofibromatosis. Quart J Med 84: 603-618.Evans DGR, Huson SM, Donnai D, Neary W, Blair V, Teare D, Newton V, Strachan T, Ramsden R &

Harris R (1992a) A genetic study of type 2 neurofibromatosis in the United Kingdom. I.Prevalence, mutation rate, fitness, and confirmation of maternal transmission effect onseverity. J Med Genet 29:841-846.

Evans DGR, Trueman L, Wallace A, Collins S & Strachan T (1998) Genotype/phenotype correlationsin type 2 neurofibromatosis (NF2): evidence for more severe disease associated withtruncating mutations. J Med Genet 35: 450-455.

Fairbank J (1994) Orthopaedic Manifestations of Neurofibromatosis. In: Huson SM & Hughes RAC,eds. The Neurofibromatoses: A Pathogenic and Clinical Overview 11: 275-303. UniversityPress, Cambridge, UK.

Ferner RE, Chaudhuri R, Bingham J, Cox T & Hughes RAC (1993) MRI in neurofibromatosis 1. Thenature and evolution of increased intensity T

2 weighted lesions and their relationship to

intellectual impairment. J Neurol Neurosurg Psychiatr 56: 492-495.Fisher DA, Chu P & McCalmont T (1997) Solitary plexiform neurofibroma is not pathognomonic of

von Recklinghausen’s neurofibromatosis: a report of a case. Int J Dermat 36: 435-442.Fountain JW, Wallace MR, Bruce MA, Seizinger BR, Menon AG, Gusella JF, Michels VV, Schmidt

MA, Dewald GW & Collins FS (1989) Physical mapping of a translocation breakpoint inneurofibromatosis. Science 244: 1085-1087.

Friedman JM & Birch PH (1997) Type 1 neurofibromatosis: a descriptive analysis of the disorder in1728 patients. Am J Med Genet 70: 138-143.

Friedman JM, Birch P & Greene C and the NNFF International Database Participants (1993) NationalNeurofibromatosis Foundation International Database. Am J Med Genet 45: 88-91.

Fuller LC, Cox B & Gardner RJM (1989) Prevalence of von Recklinghausen Neurofibromatosis inDunedin, New Zealand. Neurofibromatosis 2: 278-283.

Fuller CE & Williams GT (1991) Gastrointestinal manifestations of type 1 neurofibromatosis (vonRecklinghausen´s disease). Histopathology 19: 1-11.

Gammel JA (1931) Localized neurofibromatosis. Arch Dermatil Syphilol 24: 712.Gardner WJ & Frazier CH (1930) Bilateral acoustic neurofibromas: a clinical study and field survey of

five generations with bilateral deafness in thirty-eight members. Arch Neurol Psychiat 23:266-302.

Garretto NS, Ameriso S, Molina HA, Arberas C, Salvat J, Monteverde D & Sica REP (1989) Type 2neurofibromatosis with Lisch nodules. Neurofibromatosis 2: 315-321.

Garty BZ, Laor A & Danon YL (1994) Neurofibromatosis type 1 in Israel: survey of young adults. JMed Genet 1994;31:853-857.

64

Granroth T (1946) A case of von Recklinghausen’s neurofibromatosis. Acta Dermato-Venereologica26: 321-328.

Gutmann DH, Aylsworth A, Carey JC, Korf B, Marks J, Pyeritz RE, Rubenstein A & Viskochil D(1997) The diagnostic evaluation and multidisciplinary management of neurofibromatosis 1and neurofibromatosis 2. JAMA 278: 51-57.

Gutmann DH, Gurney JG & Shannon KM (1996) Juvenile xanthogranuloma, neurofibromatosis 1 andjuvenile chronic myeloid leukemia. Arch Dermatol 132: 1390-1391.

Gutmann DH, Sherman L, Seftor L, Haipek C, Lu KH & Hendrix M (1999) Increased expression ofthe NF2 tumour suppressor gene product, merlin, impairs cell motility, adhesion andspreading. Hum Mol Genet 8: 267-275.

Gutmann DH, Wright DE, Geist RT & Snider WD (1995) Expression of the neurofibromatosis 2(NF2) gene isoforms during rat embryonic development. Human Molec Genet 4: 471-478.

Gyapay G, Morissette J, Vignal A, Dib C, Fizames C, Millasseau P, Marc S, Bernardi G, Lathrop M& Weissenbach J (1994) The 1993-1994 Généthon human genetic linkage map. Nat Genet 7:246-339.

Hagymásy L, Tóth M, Szücs N & Rigó J (1998) Neurofibromatosis type 1 with pregnancy-associatedrenovascular hypertension and the syndrome of hemolysis, elevated liver enzymes, and lowplatelets. Am J Obstet Gynecol 179: 272-274.

Haldane JBS & Smith CAB (1947) A simple exact test for birth-order effect. Ann of Eugenics 14: 117-124.

Harkin JC & Reed RJ (1969) Tumors of the peripheral nervous system. In: Atlas of tumor pathology67-106. AFIP, Washington, D.C.

Hirvonen O, Lakkakorpi J, Aaltonen V, Hirvonen H, Rossi M, Karvonen S-L, Ylä-Outinen H, KalimoH & Peltonen J (1999) Developmental regulation of NF1 tumor suppressor gene in humanperipheral nerve. J Neurocytol (in press)

Holt JF & Wright EM (1948) The radiologic features of neurofibromatosis. Radiology 51: 647-663.Hughes RAC (1994) Neurological complications of neurofibromatosis 1. In: Huson SM & Hughes

RAC, eds. The Neurofibromatoses: A Pathogenic and Clinical Overview 8: 204-232.University Press, Cambridge, UK.

Huson SM (1994) Neurofibromatosis 1: a clinical and genetic overview. In: Huson SM & HughesRAC, eds. The Neurofibromatoses: A Pathogenic and Clinical Overview 7: 160-203.University Press, Cambridge, UK.

Huson SM (1998) Neurofibromatosis type 1: historical perspective and introductory overview. In:Upadhyaya M, Cooper DN eds. Neurofibromatosis type 1 from genotype to phenotype 1:1-20. Biddles Ltd, Guilford.

Huson SM (1999) What level of care for the neurofibromatoses? Lancet 353: 1114-1116.Huson SM, Compston DAS, Clark P & Harper PS (1989a) A genetic study of von Recklinghausen

neurofibromatosis in South East Wales. I Prevalence, fitness, mutation rate, and effect ofparental transmission on severity. J Med Genet 26: 704-711.

Huson SM, Compston DAS & Harper PS (1989b) A genetic study of von Recklinghausenneurofibromatosis in south east Wales. II Guidelines for genetic counselling. J Med Genet26: 712-721.

Huson SM, Harper PS & Compston DAS (1988) Von Recklinghausen neurofibromatosis: a clinicaland population study in south east Wales. Brain 111: 1355-1381.

Huson S, Jones D & Beck L (1987) Ophthalmologic manifestations of neurofibromatosis. Br JOphthalmol 71: 235-238.

Huson SM & Ruggieri M (1996) Segmental NF: more common and more variable than the literaturesuggests?. Brazilian J Genet 19 (Suppl.2): 179.

Ilgren EB, Kinnier-Wilson LM & Stiller CA (1985) Gliomas in neurofibromatosis: a series of 89 caseswith evidence for enhanced malignancy in associated cerebellar astrocytomas. Pathol Ann 20: 331-358.

Itoh T, Marmaldi S, White RM, Denckla MB, Hofman K, Naidu S & Bryan RN (1994)Neurofibromatosis type 1: the evolution of deep gray and white matter MR abnormalities.Am J Neuroradiol 15:1513-1519.

Jaakkola S, Muona P, James WD, Hannuksela M, Karvonen J, Peltonen J & Uitto J (1990) Segmentalneurofibromatosis: immunohistochemical analysis of the cutaneous lesions. J Am AcadDermatol 22: 617-621.

Jacks T, Shih TS, Schmitt EM, Bronson RT, Bernards A & Weinberg RA (1994) Tumourpredisposition in mice heterozygous for a targeted mutation in Nf1. Nat Genet 7: 353-361

65

Jadayel D, Fain P, Updadhyaya M, Ponder MA, Huson SM, Carey J, Fryer A, Mathew CGP, BarkerDF & Ponder BAJ (1990) Paternal origin of new mutations in von Recklinghausenneurofibromatosis. Nature 343: 558-559.

Jensen LN, Fenger K, Olsen JH, Mulvihill JJ & Sorensen SA (1998) Cancer and mortality inneurofibromatosis1 (NF1): a 54 year follow-up of a nationwide cohort in Denmark. Am JHum Genet 63: A114 (suppl).

Karjalainen AO (1964) Recklinghausen tauti ja raskaus. Duodecim 80: 565-567.Katila O & Granroth T (1955) Ett interessant fall av neurofibromatos. FLH 98: 175-180.Kaye LD, Rothner AD, Beauchamp GR, Meyers SM & Estes ML (1992) Ocular findings associated

with neurofibromatosis type II. Ophthalmology 99: 1424-1429.Kayes LM, Riccardi VM, Burke W, Bennet RL & Stephens K (1992) Large de novo DNA deletion in

a patient with sporadic neurofibromatosis 1, mental retardation and dysmorphism. J MedGenet 29: 686-687.

Kluwe L & Mautner V-F (1998) Mosaicism in sporadic neurofibromatosis 2 patients. Hum MolGenet 7: 2051-2055.

Knudson AJ (1971) Mutation and cancer: statistical study of retinoblastoma. Proc Nat Acad Sci USA68: 820-823.

Kopf AW, Levine LJ & Rigel DS (1985) Prevalence of congenital-nevus-like nevi, nevi spilli and caféau lait spots. Arch Dermatol 121: 766-769.

Korf BR (1998) The NF1 Genetic Analysis Consortium. In: Upadhyaya M& Cooper DN eds.Neurofibromatosis Type 1 from genotype to phenotype 4: 57-62. Biddles Ltd, Guilford,UK.

Korf BR (1999) Plexiform neurofibromas. Am J Med Genet 89: 31-37.Landau K, Dosseter FM, Hoyt WF & Muci-Mendoza R (1990) Retinal hamartoma in

neurofibromatosis 2. Arch Ophthalmol 108: 328-329.Landau K & Yasargil GM (1993) Ocular fundus in neurofibromatosis type 2. Br J Ophthalmol 77:

646-649.Lathrop GM & Lalouel JM (1984) Easy calculations of lod scores and genetic risks on small

computers. Am J Hum Genet 36: 460-465.Lázaro C, Gaona A & Estivill X (1994) Two CA/GT repeat polymorphisms in intron 27 of the human

neurofibromatosis type 1 (NF1) gene. Hum Genet 93: 351-352.Lázaro C, Gaona A, Ravella A, Volpini V & Estivill X (1995a) Prenatal diagnosis of neurofibromatosis

type 1: from flanking RFLPs to intragenic microsatellite markers. Prenat diag 15: 129-134.Lázaro C, Gaona A, Xu G, Weiss R & Estivill X (1993) A highly informative CA/GT repeat

polymorphisms in intron 38 of the human neurofibromatosis type 1 (NF1) gene. HumGenet 92: 429-430.

Lázaro C, Kruyer H, Gaona A & Estivill X (1995b) Two further cases of mutation R1947X in the NF1gene: screening for a relatively common recurrent mutation. Hum Genet 96: 361-363.

Ledbetter DH, Rich DC, O’Connell P, Leppert M & Carey JC (1989) Precise localization of NF1 to17q11.2 by balanced translocation. Am J Hum Genet 44: 20-24.

Legius E, Descheemaeker MJ, Spaepen A, Casaer P & Fryns JP (1994) Neurofibromatosis type 1 inchildhood: a study of the neuropsychological profile in 45 children. Genet Couns 5: 51-60.

Legius E, Marchuk DA, Collins FS & Glover TW (1993) Somatic deletion of the Neurofibromatosistype 1 gene in a neurofibrosarcoma supports a tumour suppressor gene hypothesis. NatGenet 3: 122-126.

Lewis RA, Gerson LP, Axelson KA, Riccardi VM & Whitford RP (1984) Von Recklinghausenneurofibromatosis: II. Incidence of optic gliomata. Ophthalmology 91: 929-935.

Lewis RA & Riccardi VM (1981) Von Recklinghausen neurofibromatosis: incidence of irishamartomata. Ophthalmology 88: 348-354.

Lisch K (1937) Ueber Beteiligung der Augen, insbesondere das Vorkommen von Irisknötchen bei derNeurofibromatose (Recklinghausen). Z Augenheilkd 93: 137-143.

Listernick R, Charrow J Greenwald M & Mets M (1994) Natural history of optic pathway tumors inchildren with neurofibromatosis type 1: a longitudinal study. J Pediatr 125: 63-66.

Listernick R, Charrow J & Gutmann DH (1999) Intracranial gliomas in neurofibromatosis type 1. AmJ Med Genet 89: 38-44.

Listernick R, Darling C, Greenwald M, Strauss L & Charrow J (1995) Optic pathway tumors inchildren: the effect of neurofibromatosis 1 on clinical manifestations and natural history. JPediatr 127: 718-722.

66

Listernick R, Louis DN, Packer RJ & Gutmann DH (1997) Optic pathway gliomas in children withneurofibromatosis 1: consensus statement from the NF1 optic pathway glioma task force.Ann Neurol 41: 143-149.

Louis DN, Ramesh V & Gusella JF (1995) Neuropathology and molecular genetics ofneurofibromatosis 2 and related tumors. Brain Pathology 5: 163-172.

Lubs M-LE, Bauer MS, Formas ME & Djokic B (1991) Lisch nodules in neurofibromatosis type 1. NEngl J Med 324: 1264-1266.

MacCollin M, Ramesh V, Jacoby LB, Louis DN, Rubio M-P, Pulaski K, Trofatter JA, Short MP,Bove C, Eldridge R, Parry DM & Gusella JF (1994) Mutational analysis of patients withneurofibromatosis 2. Am J Med Genet 55: 314-320.

MacEwen GD (1990) Orthopedic aspects of neurofibromatosis. In: Rubenstein AE & Korf BR, eds.Neurofibromatoses. A handbook for patients, families, and health-care professionals. 12:125-141. Thieme Medical Publishers, New York, USA.

Marie P & Bernard A (1896) Présentation d’un malade atteint de neurofibromatose generalisée. SocMed Hop Paris 13: 200-203.

Matsui I, Tanimura M, Kobayashi N, Sawada T, Nagahara N & Akatsuka J (1993) Neurofibromatosistype 1 and childhood cancer. Cancer 72: 2746-2754.

McGaughran JM, Harris DI, Donnai D, Teare D, MacLeod R, Westerbeek R, Kingston H, Super M,Harris R & Evans DGR (1999) A clinical study of type 1 neurofibromatosis in north westEngland. J Med Genet 36:197-203.

Michels VV, Whisnant JP, Garrity JA & Miller GM (1989) Neurofibromatosis type I with bilateralacoustic neuromas. Neurofibromatosis 2: 213-217.

Moore BD, Ater JL, Needle MN, Slopis J & Copeland DR (1994) Neuropsychological profile ofchildren with neurofibromatosis, brain tumor, or both. J Child Neurol 9: 368-377.

Moore BD, Slopis JM, Schomer D, Jackson EF & Levy BM (1996) Neuropsychological significanceof areas of high signal intensity on brain MRIs of children with neurofibromatosis.

Neurology 46: 1660-1668.Morier P, Merot Y, Paccaud D, Beck D & Frenk E (1990) Juvenile chronic granulocytic leukemia,

juvenile xanthogranulomas, and neurofibromatosis. Case report and review of the literature.J Am Acad Dermatol 22: 962-965.

Moss C & Green SH (1994) What is segmental neurofibromatosis? Br J Dermatol 130: 106-110.Mulvihill JJ (1988) Neurofibromatosis: History, Nomenclature, and Natural History.

Neurofibromatosis 1: 124-131.Mulvihill JJ (1990a) Introduction and History. In: Rubenstein AE & Korf BR, eds.

Neurofibromatoses. A handbook for patients, families, and health-care professionals. 1: 1-12. Thieme Medical Publishers, New York, USA.

Mulvihill JJ (1994) Malignancy: Epidemiologically associated cancers. In: Huson SM & HughesRAC, eds. The Neurofibromatoses: A Pathogenic and Clinical Overview 12: 305-315.University Press, Cambridge, UK.

Mulvihill JJ, Parry DM, Sherman JL, Pikus A, Kaiser-Kupfer MI & Eldridge R (1990b)Neurofibromatosis 1 (Recklinghausen Disease) and Neurofibromatosis 2 (BilateralAcoustic Neurofibromatosis). NIH conference. An Update. Ann Intern Med 113: 39-52.

North K (1993) Neurofibromatosis type 1: Review of the first 200 patients in an Australian clinic. JChild Neurol 8: 395-402.

North K (1997) Neurofibromatosis type 1 in childhood. Mac Keith Press, London, UK.Obringer AC, Meadows AT & Zackai EH (1989) The diagnosis of neurofibromatosis-1 in the child

under the age of 6 years. AJDC 143: 717-719.Oravisto T (1949) On neurofibromatosis in the region of the eye. Acta Ophthalmol 27: 89-98.Parry DM, Eldridge R, Kaiser-Kupfer MI, Bouzas EA, Pikus A & Patronas N (1994)

Neurofibromatosis 2 (NF2): Clinical characteristics of 63 affected individuals and clinicalevidence for heterogeneity. Am J Med Genet 52: 450-461.

Parry DM, MacCollin MM, Kaiser-Kupfer MI, Pulaski K, Nicholson HS, Bolesta M, Eldridge R &Gusella JF (1996) Germ-line mutations in the neurofibromatosis 2 gene: correlations withdisease severity and retinal abnormalities. Am J Hum Genet 59: 529-539.

Peltonen J, Foidart J-M & Aho HJ (1984) Type IV and V collagens in von Recklinghausen’sneurofibromas. An immunohistochemical and electrophoretical study. Virchows Arch (CellPathol) 47:291-301.

Peltonen J, Jaakkola S, Lebwohl M, Renvall S, Risteli L, Virtanen I & Uitto J (1988) Cellular differentiation and expression of matrix genes in type 1 neurofibromatosis. Lab Invest 59: 760-771.

67

Ponder M, Murton F, Hallowell N, Statham H, Green J & Richards M (1998) Genetic counselling,reproductive behavior and future reproductive intentions of perople with neurofibromatosis type1 (NF1). J Gen Couns 7: 331-344.

Pulst S-M, Riccardi VM, Fain P & Korenberg JR (1991) Familial spinal neurofibromatosis: clinicaland DNA linkage analysis. Neurology 41: 1923-1927.

Ragge NK (1993b) Clinical and genetic pattern of neurofibromatosis 1 and 2. Br J Ophthalmol 77: 662-672

Ragge NK, Falk RE, Cohen WE & Murphree AL (1993a) Images of Lisch nodules across thespectrum. Eye 7: 95-101.

Riccardi VM (1981) Von Recklinghausen neurofibromatosis. N Engl J Med 305: 1617-1627.Riccardi VM (1982) Neurofibromatosis: clinical heterogeneity. Curr probl cancer 7: 1-34.Riccardi VM (1992) Neurofibromatosis: Phenotype, Natural History, and Pathogenesis. 2nd edn. The

Johns Hopkins University Press, Baltimore.Riccardi VM, Dobson CE, Chakraborty R & Bontke C (1984) The pathophysiology of neurofibro

matosis. IX. Paternal age as a factor in the origin of new mutations. Am J Med Genet 18: 169-176.Riccardi VM & Powell PP (1989) Neurofibrosarcoma as a complication of von Recklinghausen

neurofibromatosis. Neurofibromatosis 2: 152-165.Rodenhiser DI, Ainsworth PJ, Coulter-Mackie MB, Singh SM & Jung JH (1993) A genetic study of

neurofibromatosis type 1 (NF1) in south-western Ontario. II A PCR based approach tomolecular and prenatal diagnosis using linkage. J Med Genet 30: 363-368.

Rodenhiser DI, Coulter-Mackie MB, Jung JH & Singh SM (1991) A genetic study of neurofibro-matosis 1 in south-western Ontario. I Population, familial segregation of phenotype, andmolecular linkage. J Med Genet 28: 746-751.

Roth RR, Martines R & James WD (1987). Segmental neurofibromatosis. Arch Dermatol 123: 917-920.

Rouleau GA, Merel P, Lutchman M, Sanson M, Zucman J, Marineau C, Hoang-Xuan K, Demczuk S,Desmaze C, Plougastel B, Pulst SM, Lenoir G, Bijlsma E, Fashold R, Dumanski J, de JongP, Parry D, Eldridge R, Aurias A, Delattre O & Thomas G (1993) Alteration in a new geneencoding a putative membrane-organizing protein causes neuro-fibromatosis type 2. Nature363: 515-521.

Rubenstein AE, Bader JL, Aron AA & Wallace S (1983) Familial transmission of segmental neuro-fibromatosis. Neurology 33 (Suppl.2): 76.

Ruttledge MH, Andermann AA, Phelan CM, Claudio JO, Han F, Chretien N, Rangaratnam S,MacCollin M, Short P, Parry D, Michels V, Riccardi VM, Weksberg R, Kitamura K, Bradburn JM,Hall BD, Propping P & Rouleau GA (1996) Type of mutation in the neurofibromatosis type 2gene (NF2) frequently determines severity of disease. Am J Hum Genet 59: 331-342.

Sainio M, Strachan T, Blomstedt G, Salonen O, Setälä K, Palotie A, Palo J, Pyykkö I, Peltonen L, &Jääskeläinen J (1994) Presymptomatic DNA and MRI diagnosis of neurofibromatosis 2 with mildclinical course in an extended pedigree. Neurol 45: 1314-1322.

Sakurai T (1935) Multiple neurofibroma patient showing multiple flecks on the anterior surface of theiris. Acta Soc Ophthalmol Jpn 39: 87-93.

Samuelsson B (1981) Neurofibromatosis (v. Recklinghausen’s disease) A Clinical – Psychiatric andGenetic Study. Thesis, University of Gothenburg, Dept Psychiatry.

Sawada S, Florell S, Purandare SM, Ota M, Stephens K & Viskochil D (1996) Identification of NF1mutations in both alleles of a dermal neurofibroma. Nat Genet 14: 110-112.

Schmidt MA, Michels VV & Dewald GW (1987) Cases of neurofibromatosis with rearrangements ofchromosome 17 involving band 17q11.2. Am J Med Genet 28: 771-775.

Seizinger BR (1993) NF1: a prevalent cause of tumorigenesis in human cancers? Nat Genet 3: 97-99.Sergeyev, AS (1975) On the mutation rate of neurofibromatosis. Hum Genet 28: 129-138.Serra E, Puig S, Otero D, Gaona A, Kruyer H, Ars E, Estivill X & Lázaro C (1997) Confirmation of a

double-hit model for the NF1 gene in benign neurofibromas. Am J Hum Genet 61: 512-519.Seward GR (1994) Did the Elephant Man have Neurofibromatosis 1? In: Huson SM & Hughes RAC,

eds. The neurofibromatoses. A Pathogenetic and Clinical overview. 16: 382-401. UniversityPress, Cambridge, UK.

Shannon KM, O’Connell P, Martin GA, Paderanga D, Olson K, Dinndorf P & McCormick F (1994)Loss of the normal NF1 allele from the bone marrow of children with type 1 neurofibromatosisand malignant myeloid disorders. N Engl J Med 330: 597-601.

Sharma JB, Gulati N & Malik S (1991) Maternal and perinatal complications in neurofibromatosisduring pregnancy. Int J Gynecol Obstet 34: 221-227.

68

Shen MH, Harper PS & Upadhyaya M (1996) Molecular genetics of neurofibromatosis type 1 (NF1)J Med Genet 33: 2-17.

Short PM, Martuza RL & Huson SM (1994) Neurofibromatosis 2: clinical features, geneticcounselling and management issues. In: Huson SM & Hughes RAC, eds. TheNeurofibromatoses: A Pathogenic and Clinical Overview 18: 415-444. University Press,Cambridge, UK.

Silva AJ, Frankland PW, Marowitz Z, Friedman E, Lazlo G, Cioffi D, Jacks T & Bourtchuladze R(1997) A mouse model for the learning and memory deficits associated with neurofibromatosistype I. Nat Genet 15: 281-284.

Sivalingam A, Augsburger J, Perilongo G, Zimmerman R & Barabas G (1991) Combined hamartoma ofthe retina and pigment epithelium in a patient with neurofibromatosis type 2. J PediatrOphthalmol Strab 28: 320-322.

Skuse GR & Ludlow JW (1995) Tumor suppressor genes in disease and therapy. Lancet 345: 902-906.Sorensen SA, Mulvihill JJ & Nielsen A (1986) Long term follow-up of von Recklinghausen

neurofibromatosis. N Engl J Med 314: 1010-1015.Spence MA, Bader JL, Parry DM, Field LL, Funderburk SJ, Rubenstein AE, Gilman PA & Sparkes

RS (1983) Linkage analysis of neurofibromatosis (von Recklinghausen disease). J MedGenet 20: 334-337.

Spence MA, Parry DM, Bader JL, Marazita ML, Bocian M, Funderburk SJ, Mulvihill JJ & SparkesRS (1986) Genetic linkage analysis of neurofibromatosis. Ann NY Acad Sci 486: 287-92.

Stay EJ & Vawter G (1977) The relationship between nephroblastoma and neurofibromatosis (vonRecklinghausen´s disease). Cancer 39: 2550-2555.

Stephens K, Kayes L, Riccardi VM, Rising M, Sybert VP & Pagon RA (1992) Preferential mutationof the neurofibromatosis type 1 gene in paternally derived chromosomes. Hum Genet 88: 279-282.

Stevenson DA, Birch PA, Friedman JM, Viskochil DH, Balestrazzi P, Boni S, Buske A, Korf B,Niimura M, Pivnick EK, Schorry EK, Short MP, Tenconi R, Tonsgard JH & Carey JC (1999)Descriptive analysis of tibial pseudarthrosis in patients with neurofibromatosis 1. Am J MedGenet 84: 413-419.

Stumpf DA, Alksne JF, Annegers JF, Brown SS, Conneally PM, Housman D, Leppert MF, Miller JP,Moss ML, Pileggi AJ, Rapin I, Strohman RC, Swanson LW & Zimmerman A (1988) Neurofibromatosis conference statement. National Institutea of Health consensus development conference.Arch Neurol 45: 575-578.

Suomen Virallinen Tilasto 1960-1997. Väestömuutokset/Väestö VI A. The Central Statistical Officeof Finland, Helsinki. Valtion Painatuskeskus.

Szudek J, Birch PH & Friedman JM (1998) Height and head circumference in patients withneurofibromatosis type 1 (NF1). Am J Hum Genet 63 (suppl): A122.

Takano T, Kawashima T, Yamanouchi Y, Kitayama K, Baba T, Ueno K & Hamaguchi H (1992)Genetics of neurofibromatosis 1 in Japan: mutation rate and paternal age effect. Hum Gen89: 281-286.

Tanaka K (1974) A new simplified method for estimating relative fitness in man. Japanese J HumGenet 19: 195-202.

Tassabehji M, Strachan T, Sharland M, Colley A, Donnai D, Harris R & Thakker N (1993) Tandemduplication within a neurofibromatosis type 1 (NF1) gene exon in a family with features ofWatson syndrome and Noonan syndrome. Am J Hum Genet 53: 90-95.

Trofatter JA, MacCollin MM, Rutter JL, Murrell JR, Duyao MP, Parry DM, Eldridge R, Kley N,Menon AG, Pulaski K, Haase VH, Ambrose CM, Munroe D, Bove C, Haines JL, Martuza RL,MacDonald ME, Seizinger BR, Short MP, Buckler AJ & Gusella JF (1993) A novel moesin-,ezrin-, radixin-like gene is a candidate for the neurofibromatosis 2 tumor suppressor. Cell 72:791-800.

Upadhyaya M, Fryer A, MacMillan J, Broadhead W, Huson SM & Harper PS (1992) Prenataldiagnosis and presymptomatic detection of neurofibromatosis type 1. J Med Genet 29:180-183.

Upadhyaya M, Ruggieri M, Maynard J, Osborn M, Hartog C, Mudd S, Penttinen M, Cordeiro I,Ponder M, Ponder BAJ, Krawczak M & Cooper DN (1998) Gross deletions of the neurofibromatosis type 1 (NF1) gene are predominantly of maternal origin and commonly associated withlearning disability, dysmorphic features and developmental delay. Hum Genet 102: 591-597

69

Walden PAM, Johnson AG & Bagshawe KD (1977) Wilms‘s tumour and neurofibromatosis. BMJ6064: 813.

Wallace MR, Marchuk DA, Andersen LB, Letcher R, Odeh HM, Saulino AM, Fountain JW, BreretonA, Nicholson J, Mitchell AL, Brownstein BH & Collins FS (1990) Type 1 neurofibromatosisgene: identification of a large transcript disrupted in three NF1 patients. Science 249: 181-186.

Wanebo JE, Malik JM, VandenBerg SR, Wanebo HJ, Driesen N & Persing JA (1993) Malignantperipheral nerve sheath tumors. A clinicopathologic study of 28 Cases. Cancer 71: 1247-1253.

Weissman A, Jakobi P, Zaidise I & Drugan A (1993) Neurofibromatosis and pregnancy: An update. JReprod Med 38: 890-896.

Weleber RG & Zonana J (1983) Iris hamartomas (Lisch nodules) in a case of segmental neurofibro-matosis. Am J Ophthalmol 96: 740-743.

Wellwood JM, Bulmer JH & Graff DJC (1971) Congenital defects of the tibia in siblings with neuro-fibromatosis. J Bone Joint Surg (Br) 53: 314-319.

Westerhof W & Kondrad K (1982) Blue-red macules and pseudoatrophic macules: additionalcutaneous signs in neurofibromatosis. Arch Dermatol 118: 577-581.

Wiestler OD & Radner H (1994) Pathology of neurofibromatosis 1 and 2. In: Huson SM & HughesRAC, eds. The neurofibromatoses. A Pathogenetic and Clinical overview. 6: 135-159.University Press, Cambridge, UK.

Viskochil DH (1998) Gene structure and expression. In: Upadhyaya M & Cooper DN, eds. Neuro-fibromatoses type 1 from genotype to phenotype 3: 39-56. Biddles Ltd, Guilford, UK.

Viskochil D, Buchberg A, Xu G, Cawthon RM, Stevens J, Wolff RK, Culver M, Carey JC, CopelandNG, Jenkins NA, White R & O’Connell P (1990) Deletions and a translocation interrupt acloned gene at the neurofibromatosis type 1 locus. Cell 62: 187-192.

Viskochil D & Carey JC (1994) Alternate and related forms of the neurofibromatoses. In: Huson SM& Hughes RAC, eds. The Neurofibromatoses: A Pathogenic and Clinical Overview 19: 445-474. University Press, Cambridge, UK.

Viskochil D, Cawthon R, O’Connell P, Xu G, Stevens J, Culver M, Carey J & White R (1991) Thegene encoding the oligodendrocyte-myelin glycoprotein is embedded within theneurofibromatosis type 1 gene. Mol Cell Biol 11: 906-912.

Wolkenstein P, Frèche B, Zeller J & Revuz J (1996) Usefulness of screening investigations inneurofibromatosis type 1. A study of 152 patients. Arch Dermatol 132: 1333-1336.

von Deimling A, Krone W & Menon AG (1995) Neurofibromatosis type 1: pathology, clinicalfeatures and molecular genetics. Brain Pathology 5: 153-162.

von Deimling A & Krone W (1997) Neurofibromatosis type 1. In: Kleihues P & Cavenee WK, eds.Pathology and genetics of tumours of the nervous system 172-174. IARC, Lyon, France.

Wu B-L, Austin MA, Schneider GH, Boles RG & Korf BR (1995) Deletion of the entire NF1 genedetected by FISH: four deletion patients associated with severe manifestations. Am J MedGenet 59: 528-535.

Wu B-L, Schneider GH & Korf BR (1997) Deletion of the entire NF1 gene causing distinctmanifestations in a family. Am J Hum Genet 69: 98-101.

Xu G, Nelson L, O’Connell P & White R (1991) An Alu polymorphism intragenic to theneurofibromatosis type 1 gene (NF1). Nucleic Acid Research 19: 3764.

Xu W, Mulligan LM, Ponder MA, Liu L, Smith BA, Mathew CGP & Ponder BAJ (1992) Loss of NF1alleles in phaeochromocytomas from patients with type 1 neurofibromatosis. GenesChromosomes Cancer 4: 337-342.

Zacharin M (1997) Precocious puberty in two children with Neurofibromatosis type I in the absenceof optic chiasmal glioma. J Pediatr 130: 155-157.

Zucman-Rossi J, Legoix P, Der Sarkissian H, Cheret G, Sor F, Bernardi A, Cazes L, Giraud S, OllagnonE, Lenoir G & Thomas G (1998) NF2 gene in neurofibromatosis type 2 patients. Hum MolGenet 7: 2095-2101.

Zvulunov A, Barak Y & Metzker A (1995) Juvenile xanthogranuloma, neurofibromatosis, and juvenilechronic myelogenous leukemia. World statistical analysis. Arch Dermatol 131: 904-908.

Zöller MET (1997) Neurofibromatosis type 1 Psychiatric and somatic aspects. A 12-year follow-upof adult patients in Sweden. Thesis, University of Gothenburg, Institute of Clinical Neuro-science, Dept Psychiatry.

Zöller M, Rembeck B, Åkesson HO & Angervall L (1995) Life Expectancy, Mortality and PrognosticFactors in Neurofibromatosis Type 1 Acta Derm Venereol (Stockh) 75: 136-140.

70

Appendix 1

NF examination form [modified from Friedman et al. (1993)]

Name Date of examine

GrowthHeight Weight Head circumference

Café au lait spotsSize Number

FrecklingAxillar Other

NeurofibromasSubcutaneous Cutaneous PendulousPlexiform Internal

Other skin problemsHaemangiomas Xanthogranulomas DepigmentationHyperpigmentation Troublesome itching Other

Ocular abnormalitiesLisch nodules Proptosis StrabismusCataract/opacity Optic glioma Other

Auditory abnormalitiesSensory neuronal hearing loss Age at onset Other

Cardiovascular abnormalitiesCongenital heart disease Hypertension Other

Gastrointestinal problems

Dentition problems

Endocrine abnormalitiesNormal age at puberty Menarche Other

NeurologicalSeizures Headache HydrocephalusOther

Intellectual/educationalHighest education completed Developmental examine resultsIQ Full scale IQ Verbal IQ Performance

Orthopaedic abnormalitiesPseudarthrosis Congenitally bowed tibia Scoliosis/KyphosisDysplastic vertebrae Spinal compression Dysplastic sphenoid wingLong bone thinning Long bone bowing Bone cystsLong bone cortical atrophy Bony overgrowth Bony asymmetryOther

CT/MRI ScanBrain Orbits Optic nerveSpine Other area

Dysmorphic featuresPhoto on file Noonan phenotype Facial asymmetrySoft tissue asymmetry Other

epidemiological, clinical and genetic aspects of ...jultika.oulu.fi/files/isbn951425418x.pdf ·...

Documents