ORIGINAL ARTICLE

Survival after incidental prostate cancer diagnosisat transurethral resection of prostate: 10-year outcomes

S. Ahmad F. OKelly R. P. Manecksha

I. M. Cullen R. J. Flynn T. E. D. McDermott

R. Grainger J. A. Thornhill

Received: 30 April 2011 / Accepted: 26 August 2011 / Published online: 11 September 2011

Royal Academy of Medicine in Ireland 2011

Abstract

Background The most appropriate management of inci-

dental prostate cancers diagnosed at transurethral resection

of prostate has been debated. It is important to determine

the long-term outcomes to establish an appropriate man-

agement in patients with incidental prostate cancer.

Aims We aim to determine 10-year survival and to

identify the factors of worse prognosis of incidental pros-

tate cancers diagnosed at transurethral resection of prostate.

Methods A retrospective analysis of patients with pT1a

pT1b prostate cancers diagnosed between 1998 and 2003.

Medical notes, PSA and pathology results were reviewed.

Overall and cancer specific survival was calculated at mean

10-year follow-up.

Results Sixty patients with incidental prostate cancer were

identified (pT1a = 18, pT1b = 42). Fifty-one percents of the

patients were managed on a watchful waiting strategy with

overall 84% survival and 9.7% cancer specific mortality.

Twenty patients (all with pT1b) received hormone therapy.

Overall survival in this cohort was 50% with 20% cancer

specific mortality. Nine patients received curative therapy

(Radical prostatectomy = 4, Radiotherapy = 5). In this

group, overall survival was 88% with no cancer specific

mortality.

Conclusions Stage pT1a disease and preoperative low

PSA were associated with favourable survival. However,

for pT1b and/or high Gleason score (C7), mortality was

comparatively higher. Hence, patients with high Gleason

score and/or pT1b disease should be considered for cura-

tive therapy. Additionally, active surveillance may have a

role in selected men with incidental prostate cancer.

Keywords Hormone therapy Prostate cancer PSA Watchful waiting

Introduction

Detection of incidental prostate cancers (PCa) in pathology

specimens from transurethral resection of prostate (TURP)

has decreased in the modern PSA era. These incidental PCa

are categorized as stage pT1a (PCa in \5% of the TURPspecimen) and pT1b (PCa in [5% of the TURP speci-men).These low volume tumours progress only in 1025%

of the patients in 10 years and seldom advance signifi-

cantly in 5 years [1]. The pT1b PCa is associated with

higher Gleason score and has significant risk of progression

[2]. Patients with pT1a disease generally do well but recent

reports identified subsets of patients with pT1a disease

having adverse pathological features on subsequent radical

prostatectomy specimens [3, 4]. Guidelines by European

Association of Urology recommend watchful waiting

(WW) for these incidental tumours (pT1a and pT1b) if the

Gleason score is 6 or less and life expectancy is less than

10 years. However, for pT1b disease and life expectancy of

more than 10 years, radical surgery is recommended. The

prediction of residual disease (post TURP) and its

S. Ahmad F. OKelly R. P. Manecksha I. M. Cullen R. J. Flynn T. E. D. McDermott R. Grainger J. A. Thornhill

Department of Urology, The Adelaide and Meath Hospital

incorporating the National Childrens Hospital, Tallaght, Dublin,

Ireland

Present Address:S. Ahmad (&)Department of Urology, Ninewells Hospital, Dundee

DD2 9SY, UK

e-mail: drsarfrazrana@hotmail.com

123

Ir J Med Sci (2012) 181:2731

DOI 10.1007/s11845-011-0753-x

significance are also debated as TURP may remove the

entire tumour [5]. Hence, it is vital to identify sub-group of

patients who need aggressive treatment.

We conducted a retrospective study of pT1apT1b PCa

in our institute, to review management, survival and factors

associated with disease progression.

Patients and methods

We performed a retrospective analysis of patients diag-

nosed with incidental PCa (pT1apT1b) between 1998 and

2003. All of these patients had TURP for obstructive lower

urinary tract symptoms including urinary retention. Only

patients with preoperative benign prostate on digital rectal

examination (DRE) were included in the study. Patients

with abnormal DRE, DRE not recorded and previously

diagnosed PCa were excluded. Medical notes were

reviewed and age, PSA, T stage, volume of disease and

Gleason score were analysed. These tumours were classi-

fied in accordance with TNM staging system (1992).

Overall and cancer specific mortality was calculated at

mean 10 years (range: 6.812) after diagnosis of PCa.

Survival at 6.8 years (minimum duration of follow-up) was

also analysed.

Results

Patients sub-groups and management

Sixty patients were identified according to the inclusion

criteria. The mean age at TURP was 69.7 years (range;

5586). Pathological staging revealed 18 patients (30%)

with pT1a disease and 42 patients (70%) with pT1b PCa.

Thirty-one patients (51.6%) were managed by WW. Mean

age in this sub-group was 72 years (range; 5886) and

median pre TURP PSA (available for 16 patients) was

5.4 ng/ml (range; 0.313.7). Histological analysis of TURP

chips categorized 15 patients (48%) with pT1a PCa and 16

(52%) with pT1b. The Gleason score was 6 in 24 patients,

7 in 3 and 5 in 2 patients. One patient in this sub-group had

Gleason score of 9 and another one had 3. Median pre

TURP PSA was 3.1 ng/ml (range; 0.311.7) in patients

with pT1a disease and 3.3 ng/ml (range; 2.413.7) in men

with pT1b PCa.

Twenty patients (33%) were treated with hormone

therapy (4 weeks of biclutamide 50 mg, and 3 monthly

LHRH analogue) after cancer diagnosis at TURP. Mean

age in this sub-group was 73.3 years (range; 6085).

Median pre TURP PSA (available for 12 patients) was

9.2 ng/ml (range; 149). All of these patients had pT1b

disease. The Gleason score was 6 in 6 patients, 7 in 6, 9 in

3 and 10 in 2 patients. While two patients had the Gleason

score of 8 and another one had 5.

Fifteen percent of the patients were treated with curative

intent. Four (7%) of these had open radical prostatectomy

while five (8%) received external beam radical radiother-

apy. In radical prostatectomy sub-group, mean age of

patients was 66 years (range; 5576) and 75% (n = 3) of

these had pT1b disease. The Gleason score was 6 in 3

patients and 4 in 1 (TURP specimen) with median pre

TURP PSA (available for all 4 patients) of 7 ng/ml (range;

0.66). Final pathology (radical prostatectomy specimen)

showed pT2b in all patients with negative margins. In

radical radiotherapy sub-group, mean age of the patients

was 67.4 years (range 6669) and the median pre TURP

PSA (available for all 5 patients) was 7 ng/ml (range;

216.4). Forty percent (n = 2) of these patients had pT1a

PCa and 60% (n = 3) pT1b disease, while the Gleason

score was 6 in four patients and 5 in one.

Overall and cancer specific mortality

At mean 10-year follow-up, overall mortality (death due to

PCa ? death due to other causes) in this cohort was 27%

(n = 16). In WW sub-group, the cancer specific mortality

was 9.7% (n = 3) while overall mortality was 16%

(n = 5). Hormone therapy sub-group had the highest

overall mortality (50%, n = 10) while the cancer specific

mortality was 20% (n = 4). No cancer specific death was

reported in radical prostatectomy and radical radiotherapy

sub-groups. However, one patient in radical prostatectomy

sub-group died after 5.8 years of TURP due to an acute

cardiac event.

All patients in this study had minimum follow-up of

6.8 years. Hence, survival was also analysed at 6.8 years

after initial diagnosis of PCa. At this follow-up, overall

mortality was 17% (n = 10). The cancer specific mortality

was 10% (n = 2) in hormone therapy sub-group but there

was no cancer specific deaths in other two sub-groups.

However, overall mortality was 7 (n = 2), 35 (n = 7) and

25% (n = 1) in WW, hormone therapy and radical pro-

statectomy sub-groups, respectively.

Prognostic factors and survival

Overall survival at mean 10-year follow-up was 73%

(n = 44). Prognostic factors of PCa were analysed in all

patients who died of PCa (Tables 1, 3). Patients with high

volume disease and high Gleason score had higher mor-

tality. Preoperative PSA values were not available for all

patients, hence in this cohort statistical significance of

preoperative PSA could not be assessed. However, it was

observed that patients with high preoperative PSA had

increased risk of disease progression.

28

123

Pathological features in patients died of PCa were ana-

lysed and compared with survivors in treated sub-groups

(Tables 2, 3). Mean survival of three patients died of PCa

in WW sub-group was 10.4 years (post diagnosis of PCa)

and the mean age at death was 84 years. The current mean

age of surviving patients (n = 26) in this sub-group is

80 years (range; 66.897.3), statistically different from

patients who died of PCa (p = 0.01). In patients treated

with hormone therapy, two patients died of PCa at 3 and

4.3 years, respectively after diagnosis of PCa. Both had

high grade pT1b but age of death was 78 and 86 years.

Another 2 patients died of PCa at mean 10 years after

initial diagnosis at age of 78 and 83 years. Both of these

had high grade pT1b disease. The current age of alive

patients in this sub-group is 84 years (range; 71.594.4),

not statistically different from mean age of patients who

died of PCa (p = 0.3). One patient in radical prostatectomy

who died of acute cardiac event had preoperative PSA of

6 ng/ml, pT1b disease and the Gleason score of 6 (TURP

specimen). All five patients treated with external beam

radiotherapy were alive at the time of study (mean age

77.5 years, range; 7680.6).

Discussion

We have shown variable survival amongst different man-

agement modalities for incidental PCa. WW appeared an

acceptable option for patients with low volume incidental

PCa. However, high volume and high grade PCa need

aggressive management. It is observed that PSA testing at

time of TURP (19982003) was not as common as it is in

last few years, hence pre TURP PSA was not available for

39% patients (n = 23). Additionally, the mean PSA was

higher than accepted normal in all sub-groups. This can

affect categorization in stages pT1a and pT1b. Current

TNM staging system (1992) did not include PSA to dif-

ferentiate between pT1a and pT1b. As in our study the

mean PSA was higher, hence it is possible these patients

may not be suitable for pT1a or pT1b categorization rather

may fall into T1c or higher staging. However, the indica-

tion for TURP in all of these patients was severe LUTS

urinary retentions and mean age was 69.7 years. These

factors can cause high PSA levels in benign disease.

Management of these incidental cancers is debatable.

Retrospective studies, with 510 years follow-up, question

the need for radical approaches in stage pT1 PCa [6, 7].

Furthermore, no large randomised trial exists and treatment

for these tumours is based on retrospective observational

studies[4]. Residual disease is an important consideration

for the management of pT1apT1b PCa as upstaging of

these tumours has been reported previously [8]. In this

situation, prediction of disease progression and selection of

high risk patients is vital. The risk of progression of

untreated pT1a PCa after 5 years is only 5%, but can

increase to 50% after 1013 years. In contrast, most

patients with pT1b tumours progress and show lower

cancer specific survival after 5 years [5]. Hence, aggressive

treatment is often warranted in selected patients. However,

WW and symptom control have also become viable options

for older men due to the morbidity and mortality associated

with treatment as well as the presence of underlying

co-morbidities [9]. But, the advancing age may not be a

limiting factor in deciding curative treatments as cancer

specific survival benefit following radical prostatectomy

was also reported in older men [10].

Table 1 Watchful waiting sub-group: Prognostic and pathological factors in patients died of prostate cancer

Age at diagnosis

(years)

Pre TURP

PSA (ng/ml)

T stage % volume of

PCa in TURP chips

Gleason

score

Age at death

due to PCa (years)

Survival after

TURP (years)

74 N/A T1a \5 6 84 10.175 15 T1b 56 6 85 9.8

71 0.7 T1b 10 7 83 11.5

TURP Transurethral resection of prostate, N/A not available, PCa Prostate cancer

Table 2 Hormone therapy sub-group: Prognostic and pathological factors in patients died of prostate cancer

Age at diagnosis

(years)

Pre TURP

PSA (ng/ml)

T Stage % volume of

PCa in TURP chips

Gleason

score

Age at death

due to PCa (years)

Survival after

TURP (years)

68 N/A T1b 75 8 78 9.8

72 8 T1b 50 8 83 10.2

83 N/A T1b 70 9 86 3

74 14 T1b 90 8 78 4.3

TURP Transurethral resection of prostate, N/A not available, PCa Prostate cancer

29

123

In the present study, men with incidental PCa showed an

overall survival of 73% and cancer specific mortality of

11% at mean 10-year follow-up. Over half of these patients

were placed on a WW with 83% overall survival and only

10% cancer specific mortality. Of the remainder, 15%

received treatment with curative intent resulted in an

overall survival rate of 88%. Patients treated with hormone

therapy have only 50% overall survival however, cancer

specific mortality in this sub-group was 20%. Interestingly,

more than half of the deaths (n = 10) occurred in the first

6.8 years of follow-up. However, cancer specific mortality

was low at that point (n = 2). Active surveillance for PCa

is relatively a new concept which involves the postpone-

ment of immediate therapy, with definitive treatment used

if there is evidence that the patient is at increased risk for

disease progression. In our study cohort, no active sur-

veillance was offered, as these patients were diagnosed

between 1998 and 2003 and the concept of active sur-

veillance was not widely accepted at that time. Given that

WW is a viable therapeutic option for those with low-risk

PCa, and that deferred curative surgical treatment is pos-

sible, it is important to determine at what point that the

change in treatment algorithms (from WW to intervention)

should take place.

Patients in our study were divided into pT1a and pT1b

PCa based on TNM classification (1992). There has been

no change to the TNM classification for nearly 20 years for

T1 disease and many reports have questioned the accuracy

of pT1a and pT1b classification [4, 11]. PSA values with

prostate volume, weight of resected tissue and the Gleason

score can predict progression of pT1a PCa [4]. Our results

indicate that the preoperative PSA was not available for all

patients however, cancer specific mortality was associated

with high stage/grade and high pre operative PSA (where

available). Capitanio et al. [4] showed that for incidental

PCa, the PSA levels before and after surgery with the

Gleason score can estimate biochemical failure with an

accuracy of 87.5%. While Rajab et al. [11] recently

reported that the percentage of positive chips was highly

predictive of PCa death and they suggested that the current

TNM staging model cut-off of 5% was sub-optimal.

These data suggest that men considering surgical or

medical management of benign prostatic hyperplasia

should be informed that they may harbour clinically sig-

nificant PCa which may need additional treatment. Clinical

T1 stage PCa can no longer be viewed as a single entity

with relatively equivalent rates of progression and modes

of behaviour across its sub-classification. The above results

further strengthen the multi-faceted nature of T1 PCa and

show that current clinical algorithms may be refined to

tailor both the disease and the patient. WW can be used as a

viable option for those with pT1a disease, however, active

surveillance can also be offered depending on individual

patients. Men with pT1b disease must be informed that

they have a higher rate of progression and hence curative

treatment can reasonably be offered. Furthermore, T1

staging requires further refinement to identify sub-groups

of patients with high risk of progression.

References

1. Epstein JI, Paull G, Eggleston JC, Walsh PC et al (1986) Prog-

nosis of untreated stage A1 prostatic carcinoma: a study of 94

cases with extended follow-up. J Urol 136(4):837839

2. Tombal B, De Visccher L, Cosyns JP et al (1999) Assessing the

risk of unsuspected prostate cancer in patients with benign

prostatic hypertrophy: a 13-year retrospective study of the inci-

dence and natural history of T1aT1b prostate cancers. BJU Int

84(9):10151020

3. Descazeaud A, Peyromaure M, Salin A et al (2008) Predictive

factors for progression in patients with clinical stage T1a prostate

cancer in the PSA era. Eur Urol 53(2):355361

4. Capitanio U, Scattoni V, Freschi M et al (2008) Radical prosta-

tectomy for incidental (stage T1aT1b) prostate cancer: analysis

of predictors for residual disease and biochemical recurrence. Eur

Urol 54(1):118125

Table 3 Comparison of surviving patients with those died of prostate cancer in watchful waiting and hormone therapy sub-groups

Sub-groups Survival

status (n)Mean age

at diagnosis

(years)

Median pre

TURP PSA

(ng/ml)

T stage (n) % meanvolume of

PCa in

TURP chips

Mean

Gleason

score

Mean age

at death

due to PCa

(years)

Mean survival

after TURP

(years)

Watchful waiting

sub-group

Patients died of PCa

(n = 3)73.3 7.85 T1a (n = 1),

T1b (n = 2)23.3 6.3 81.25 10.4

Alive patients

(n = 26)73 9.2 T1a (n = 13),

T1b (n = 13)22.5 6.03 All alive 9.2

Hormone therapy

sub-group

Patients died of PCa

(n = 4)74.25 11 pT1b (n = 4) 71.25 8.25 81.25 6.8

Alive patients

(n = 10)73 9.2 pT1b (n = 10) 70 7.5 All alive 9.4

TURP Transurethral resection of prostate, PCa Prostate cancer

30

123

5. Helfand BT, Mongiu AK, Kan D et al (2009) Outcomes of radical

prostatectomy for patients with clinical stage T1a and T1b dis-

ease. BJU Int 104(3):304309

6. Chodak GW (1994) The role of conservative management in

localized prostate cancer. Cancer 74(Suppl (7)):21782181

7. Aus G, Hugosson J, Norlen J et al (1995) Long-term survival and

mortality in prostate cancer treated with noncurative intent.

J Urol 154(2 Pt 1):460465

8. Talcott JA, Rossi C, Shipley WU et al (2010) Patient-reported

long-term outcomes after conventional and high-dose combined

proton and photon radiation for early prostate cancer. JAMA

303(11):10461053

9. Adolfsson J (2008) Watchful waiting and active surveillance: the

current position. BJU Int 102(1):1014

10. Pierorazio PM, Humphreys E, Walsh PC, Partin AW, Han M

(2010) Radical prostatectomy in older men: survival outcomes in

septuagenarians and octogenarians. BJU Int 106(6):791795

11. Rajab R, Fisher G, Kattan MW et al (2011) An improved prog-

nostic model for stage T1a and T1b prostate cancer by assess-

ments of cancer extent. Mod Pathol 24(1):5863

31

123

Copyright of Irish Journal of Medical Science is the property of Springer Science & Business Media B.V. andits content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder'sexpress written permission. However, users may print, download, or email articles for individual use.