Download The PCPT: New Findings, New Insights, and Clinical - EU-ACME

european urology supplements 5 (2006) 634–639
available at www.sciencedirect.com
journal homepage: www.europeanurology.com
The PCPT: New Findings, New Insights, and Clinical
Implications for the Prevention of Prostate Cancer
Bulent Akduman a,*, E. David Crawford b
a
b
Zonguldak Karaelmas University, School of Medicine Zonguldak, Turkey
University of Colorado Health Sciences, Denver, CO, USA
Article info
Abstract
Keywords:
Finasteride
Gleason score
Prostate cancer
Prostate Cancer Prevention
Trial (PCPT)
The Prostate Cancer Prevention Trial (PCPT) demonstrated that finasteride therapy significantly reduced the risk of prostate cancer by 24.8%
( p < 0.001) compared with placebo. Controversially, there was an
increased incidence of high-grade (Gleason score 7) tumours in the
finasteride arm compared with placebo. The increase in incidence of
high-grade disease observed in finasteride-treated subjects does not
appear to be a histopathologic effect. A number of potential biases have
been identified, including increased detection rate due to prostate
volume reduction and improved prostate-specific antigen specificity
and sensitivity for detecting prostate cancer. This would suggest that
there was an improved detection of overall prostate cancer as well as
high-grade prostate cancer in men treated with finasteride, rather than
an increase in risk compared with placebo. Further analyses of the data
from the PCPT together with other clinical findings strongly suggest that
the increase in high-grade tumours in the finasteride arm is an artefact.
Please visit
www.eu-acme.org to read
and answer the EU-ACME
questions on-line. The
EU-ACME credits will then
be attributed automatically.
# 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved.
* Corresponding author. Zonguldak Karaelmas University, School of Medicine Zonguldak,
67600, Turkey. Tel. +90 372 261 0169; Fax: +90 372 261 0155.
E-mail address: [email protected] (B. Akduman).
1.
Introduction
The Prostate Cancer Prevention Trial (PCPT) was a
randomised, placebo-controlled study of the effect
of finasteride on the 7-yr prevalence of prostate
cancer [1–3]. The study was based on the observation that androgens are not only involved in the
growth of the prostate, but also play a key role in
development of prostate cancer [4–6]. 5a-Reductase
is involved in the conversion of testosterone to the
more potent androgen dihydrotestosterone (DHT).
There are two forms of the enzyme, type 1 and type
2, and the latter is the predominant isoenzyme
found in the prostate. Individuals with congenital
deficiency of type 2 5a-reductase have small
prostates, undetectable prostate-specific antigen
(PSA) levels, and no evidence of prostatic epithelium; they do not develop benign prostatic hyperplasia (BPH) or prostatic cancer [7]. Finasteride
selectively inhibits type 2 5a-reductase, reducing
both serum [8] and intraprostatic DHT [9].
The PCPT revealed that finasteride significantly
reduced the 7-yr period-prevalence of prostate
cancer compared with men receiving placebo.
However, data from the study showed an unexpected increase in the prevalence of high-grade
1569-9056/$ – see front matter # 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.eursup.2006.05.003
635
european urology supplements 5 (2006) 634–639
tumours in the finasteride arm. This finding has
been the topic of much discussion and further
analyses of data from the PCPT have been conducted to determine whether it is a true effect or an
artefact.
2.
The PCPT
The primary end point of the PCPT was the
prevalence of prostate cancer in men receiving
finasteride or placebo over the 7-yr trial period.
Secondary end points included associated mortality (including disease-specific mortality), adverse
events (AEs), quality of life (also including sexual
function), grade and stage of diagnosed cancers,
accuracy of PSA and digital rectal examination
(DRE) for diagnosing prostate cancer, and the
incidence of BPH. During the study period, subjects
underwent an annual DRE and PSA measurement;
an abnormal DRE or raised PSA level (4.0 ng/ml)
during the study resulted in the recommendation
of a biopsy (termed ‘‘for-cause’’ biopsy). However,
because of the known effect of finasteride on
reducing PSA levels [10], an end-of-study biopsy
was also offered to all patients who did not have a
diagnosis of prostate cancer. Results indicated a 7yr prevalence of prostate cancer of 18.4% in the
finasteride arm and 24.4% in the placebo arm
among the 9060 subjects included in the final
analysis of the study [1]. This represents a risk
reduction of 24.8% between the two treatment
arms (95% confidence interval [CI], 18.6–30.6;
p < 0.001). There was a 6% absolute reduction in
prostate cancer incidence in patients treated with
finasteride. Irrespective of whether detected by
for-cause or by end-of-study biopsy, rates of
prostate cancer were lower in the finasteride
arm. Although there was some variance in the
magnitude of the risk reduction effect, all subgroups benefited from finasteride therapy, irrespective of age, race, baseline PSA level, or family
history of prostate cancer.
Although a clear overall benefit was seen, biopsy
results revealed an unexpected trend: the rate of
high-grade prostate cancer (Gleason score 7) was
slightly, though significantly, raised in the finasteride group compared with placebo (6.4% vs. 5.1%,
respectively, p < 0.005). This effect was only seen in
the for-cause biopsy group and not in the end-ofstudy biopsy. Given the implications of this increase
in high-grade tumours on finasteride therapy for
both chemoprevention of prostate cancer and
management of BPH, data from the PCPT have been
studied further.
Fig. 1 – The cumulative proportional incidence of high
Gleason grade tumours in the Prostate Cancer Prevention
Trial as assessed by for-cause biopsy. Reproduced with
permission from Marberger M et al. [12].
3.
Does finasteride induce high-grade cancer?
If finasteride had really induced high-grade prostate
cancer, there would have been a continued divergence in the risk for high-grade tumours between
the finasteride and placebo arms. This effect has
been observed in clinical trials on tamoxifen in
breast cancer, where the rate of endometrial cancers
increased over time [11]. This was not the case in the
PCPT. A closer examination of the results from the
study revealed that although the incidence of highgrade prostate cancer was greater in the finasteride
arm, this effect was seen within 1 yr of study
commencement and the cumulative proportional
incidence rose at a similar rate to that of placebo
throughout the study (Fig. 1) [12].
Evidence of increased biologic aggressiveness of
tumours detected in the finasteride group was
explored through the histologic examination of
high- grade tumours. Results showed that subjects
with high-grade prostate cancer who received
finasteride were less likely to have bilateral disease,
despite a significantly smaller prostate sampled
Table 1 – Secondary indicators of aggressive tumour
growth in the biopsy samples from the Prostate Cancer
Prevention Trial
Mean no. of biopsy cores
Median number cores positive
Median percent cores positive
Greatest linear extent (mm)
Bilateral cancer
Perineural invasion
Median prostate volume (ml)
Adapted from Lucia S et al. [13].
Finasteride
N = 98 Mean
Placebo
N = 61 Mean
6
2
33%
4.4
26.5%
9.2%
24.4
6
3
40%
4.4
44.3%
16.4%
38.8
636
european urology supplements 5 (2006) 634–639
with an equivalent number of biopsy cores (Table 1)
[13]. Other surrogate markers of tumour aggressiveness indicated no significant differences between
finasteride and placebo samples.
Given these findings, what other factors might
account for this increase in the observed incidence
of high-grade cancer in the finasteride arm? A
number of potential explanations have been
explored including (1) the influence of finasteride
therapy on histologic grading, (2) the effect of
prostate size on cancer detection, and (3) the effect
of finasteride on the performance of PSA in prostate
cancer screening. These aspects are discussed in
further detail below.
for 6 mo [17]. Histologic effects were compared with
control prostate samples and samples from patients
treated with a luteinising hormone-releasing hormone agonist (LHRHa) for 3 mo. Results indicated that
finasteride did not induce a greater degree of
morphologic change to that observed in the control
group. In addition, no consistent hormonal therapy
effect with finasteride treatment was observed
compared with that seen with an LHRHa. The authors
concluded that the morphologic changes due to longterm finasteride are not a likely cause of the greater
percentage of high-grade tumours in the study arm.
5.
4.
Histologic effect
Alterations in the microscopic appearance and
immunophenotype of cells in the prostate and in
adenocarcinomas, described as ‘‘substantial and
characteristic’’ have been reported following androgen-deprivation therapy [14]. The authors observed
that these changes were not often seen in untreated
cancers and were distinct enough to be consistently
identified, suggesting that the changes might be
specifically linked to the treatment. However, one
study on the effects of finasteride on Gleason score in
45 patients reported no effect of treatment on
histologic grading [15]. Likewise, a review of data
from the Proscar Long-Term Efficacy and Safety
Study (PLESS) found no significant histologic differences between the finasteride and placebo arms in
patients who developed prostate cancer [16]. More
recent data come from a study of 56 patients with
prostate cancer who had previously taken finasteride
Prostate volume effect
The effect of finasteride on reducing prostate
volume has been well documented [18,19]. A 6-yr
study with finasteride in 487 patients with BPH
resulted in a mean reduction in prostate volume of
24% [20]. This reduction matches closely that
observed in the PCPT: median prostate volumes
were 25.5 ml in the finasteride arm and 33.6 ml in
the placebo arm, a relative difference of 24.1% [1]. It
has been shown that cancer detection rates
decrease as prostate volume increases, as demonstrated in Fig. 2 [21–23] and that the accuracy of
tumour detection and grading of biopsy samples
improves with decreasing tumour size [24–26].
Further evidence comes from a study by Kulkarni
et al who found that prostate volume was a
significant predictor of high-grade disease on
prostate biopsy, that is, smaller prostates were
correlated with high-grade cancer [27]. This effect
was lost when true histologic assessment was
performed in radical prostatectomy specimens.
Fig. 2 – Theoretical model showing prostate cancer detection bias associated with prostate volume.
european urology supplements 5 (2006) 634–639
A mathematical model has examined the effect
of prostate volume on cancer detection using data
from the PCPT. Assuming a 1-ml total tumour
volume presenting as four tumours, a prostate
volume of 30 ml, six biopsy cores, and a 25%
reduction in prostate volume with finasteride, the
predicted increase in cancer detection due to
change in prostate volume was 20% [28]. The
authors of the study suggest that this compares
well with the 25% increase in high-grade tumours
seen in the PCPT. Based on published findings,
it is likely that the increased detection of highgrade tumours in the finasteride arm (vs. placebo)
is an artefact, resulting from a detection bias
caused by the significant reduction in prostate
volume.
6.
637
Fig. 3 – Prevalence of all tumours and high-grade tumours
in the placebo group of the Prostate Cancer Prevention
Trial stratified by prostate-specific antigen (PSA) level [31].
PSA effect
Although the prevalence of high-grade prostate
cancer was significantly greater in diagnosed
tumours in the finasteride arm compared with
placebo (37.0% [280 of 757] vs. 22.2% [237 of 1068]
for placebo; p < 0.001), this effect was primarily
driven by the PSA-initiated for-cause biopsy subset
of subjects. Differences between the two arms were
not significant in for-cause biopsies recommended
due to either abnormal DRE or abnormal DRE plus
elevated PSA level. Likewise, differences in the endof-study biopsies were also not significant. These
findings raise the issue of a possible bias in the study
design with respect to the role of PSA in detecting
prostate cancer.
High PSA levels >4.0 ng/ml were established as
predictors of prostate cancer in the early 1990s [29];
since then, detection of prostate cancer has
increased markedly [30]. With the advent of the
PCPT, the opportunity arose to assess prospectively
the predictability of low PSA levels (0–4.0 ng/ml)
with respect to all prostate tumours including
those of high grade. This sub-analysis concentrated on the placebo arm of the PCPT and selected
participants whose PSA levels had remained
4.0 ng/ml, had a normal DRE, had not undergone
a for-cause biopsy or transurethral resection, but
had undergone an end-of-study biopsy [31]. A total
of 2950 of the 9459 men who were randomised to
the placebo arm in the PCPT were eligible and
included in the analysis. The results revealed that
there was a natural progression of prostate cancer
with increasing PSA levels, which also applied to
high-grade tumours. As the PSA level increased, so
did the prevalence of all prostate cancers including
high-grade cancer (Fig. 3). Thus, a PSA level
0.5 ng/ml equated to a 6.6% risk of developing
prostate cancer and a 0.8% risk of developing highgrade disease, whereas values of 3.1–4.0 ng/ml
were associated with a 26.9% risk of developing
cancer and a 6.7% risk of developing high-grade
disease.
More recently, analyses of the receiver operating
characteristics (ROC) of PSA have been conducted,
suggesting that PSA performance is improved in
patients receiving finasteride, in terms of overall
detection of cancer, as well as for high-grade
cancers. In an analysis of the 5112 men in the
placebo arm who underwent PSA screening and a
prostate biopsy in the PCPT, 1111 cases of prostate
cancer were detected, of which 240 were Gleason
score 7 and 55 were Gleason score 8 [32]. Of the
4579 men in the finasteride arm, 695 were diagnosed
with cancer; 264 were Gleason score 7 and 81 were
Gleason score 8. Analysis showed that area under
the ROC curve for finasteride was greater than for
placebo, both for cancer detection and for highgrade disease. These data suggest that there was a
bias towards greater detection of prostate cancer,
including high-grade prostate cancer, in the finasteride group compared with placebo. Therefore, the
improved performance of PSA in men receiving
finasteride may have played a role in the increased
detection of observations regarding high-grade
cancer in the PCPT.
From a clinical perspective, increased specificity
of PSA testing may translate into fewer falsepositive results and consequently a reduction in
unnecessary biopsies. Similarly, increased sensitivity of PSA testing may result in fewer false-negative
biopsies and fewer men with prostate cancers left
undetected.
638
european urology supplements 5 (2006) 634–639
Table 2 – Incidence of medical events and adverse events
in the Prostate Cancer Prevention Trial
Finasteride
(n = 9423)
n (%)
Effects on sexual functioning or endocrine effects
Reduced volume of ejaculate
5690 (60.4)
Erectile dysfunction
6349 (67.4)
Loss of libido
6163 (65.4)
Gynaecomastia
426 (4.5)
Genitourinary effects
BPH
Increased urinary
urgency or frequency
Urinary retention
Transurethral resection
of the prostate performed
Prostatitis
Urinary tract infection
Placebo
(n = 9457)
n (%)
4473
5816
5635
261
(47.3)
(61.5)
(59.6)
(2.8)
488 (5.2)
1214 (12.9)
823 (8.7)
1474 (15.6)
398 (4.2)
96 (1.0)
597 (6.3)
180 (1.9)
418 (4.4)
90 (1.0)
576 (6.1)
126 (1.3)
BPH = benign prostatic hyperplasia.
Reproduced with permission from Thompson et al. [1].
7.
Urologic effects of finasteride
In the PCPT, finasteride was associated with
significantly fewer genitourinary effects, with fewer
BPH diagnoses (5.2% vs. 8.7% for placebo). Similarly,
there were significantly fewer cases of acute urinary
retention, urinary urgency/frequency, transurethral
resections of the prostate, prostatitis, and urinary
tract infections compared with placebo (all
p < 0.001). Finasteride was well tolerated over the
7-yr study period of the PCPT. It was associated with
a slightly greater incidence of sexual or endocrine
AEs ( p < 0.001 vs. placebo; Table 2) [1].
8.
Conclusions
The PCPT demonstrated that finasteride treatment
decreased the risk of prostate cancer by 25% over a 7yr period. The accumulated body of evidence
strongly suggests that the increase in high-grade
tumours in the finasteride arm is an artefact
resulting from a detection bias driven by two
principle factors. First, finasteride treatment leads
to a significant reduction in prostate volume, and
second, treatment is associated with improved
performance of PSA to detect prostate cancer (and
high-grade prostate cancer). Importantly, the prostate volume reduction seen in finasteride-treated
subjects probably led to increased detection of all
prostate cancers, indicating that the risk reduction
with this drug was probably greater than that
actually reported.
References
[1] Thompson IM, Goodman PJ, Tangen CM, et al. The influence of finasteride on the development of prostate cancer.
N Engl J Med 2003;349:215–24.
[2] Coltman Jr CA, Thompson Jr IM, Feigl P. Prostate Cancer
Prevention Trial (PCPT) update. Eur Urol 1999;35:544–7.
[3] Thompson IM, Klein EA, Lippman SM, Coltman CA, Djavan B. Prevention of prostate cancer with finasteride: US/
European perspective. Eur Urol 2003;44:650–5.
[4] Hsing AW, Reichardt JK, Stanczyk FZ. Hormones and
prostate cancer: Current perspectives and future directions. Prostate 2002;52:213–35.
[5] Ross RK, Bernstein L, Lobo RA, et al. 5-Alpha-reductase
activity and risk of prostate cancer among Japanese and
US white and black males. Lancet 1992;339:887–9.
[6] Giovannucci E, Stampfer MJ, Krithivas K, et al. The CAG
repeat within the androgen receptor gene and its relationship to prostate cancer. Proc Natl Acad Sci U S A 1997;
94:3320–3.
[7] Imperato-McGinley J, Gautier T, Zirinsky K, et al. Prostate
visualization studies in males homozygous and heterozygous for 5 alpha-reductase deficiency. J Clin Endocrinol
Metab 1992;75:1022–6.
[8] One-year experience in the treatment of benign prostatic
hyperplasia with finasteride. The MK-906 (Finasteride)
Study Group. J Androl 1991; 12:372–5.
[9] Marks LS, Partin AW, Gormley GJ, et al. Prostate tissue
composition and response to finasteride in men with
symptomatic benign prostatic hyperplasia. J Urol
1997;157:2171–8.
[10] Andriole GL, Guess HA, Epstein JI, et al. Treatment with
finasteride preserves usefulness of prostate-specific antigen in the detection of prostate cancer: results of a randomized, double-blind, placebo-controlled clinical trial.
PLESS Study Group. Proscar Long-term Efficacy and Safety
Study. Urology 1998;52:195–201.
[11] Burke C. Endometrial cancer and tamoxifen. J Clin Oncol
Nurse 2005;9:247–9.
[12] Marberger M, Adolfsson J, Borkowski A, et al. The clinical
implications of the prostate cancer prevention trial. BJU
Int 2003;92:667–71.
[13] Lucia S, Darke A, Tangen C, Coltman CA, Thompson IM.
Pathological assessment of high grade tumors in the
Prostate Cancer Prevention Trial (PCPT). J Urol 2005;
173:451 (abstract no. 1664).
[14] Bostwick DG, Ramnani D, Cheng L. Treatment changes in
prostatic hyperplasia and cancer, including androgen
deprivation therapy and radiotherapy. Urol Clin North
Am 1999;26:465–79.
[15] Carver BS, Kattan MW, Scardino PT, Eastham JA. Gleason
score remains an important prognostic predictor in men
diagnosed with prostate cancer while on finasteride therapy. BJU Int 2005;95:509–12.
[16] Yang XJ, Lecksell K, Short K, et al. Does long-term finasteride therapy affect the histologic features of benign
prostatic tissue and prostate cancer on needle biopsy?
PLESS Study Group. Proscar Long-Term Efficacy and
Safety Study. Urology 1999;53:696–700.
european urology supplements 5 (2006) 634–639
[17] Rubin MA, Allory Y, Molinié V, et al. Effects of finasteride
treatment on prostate cancer morphology and clinical
outcome. Urology 2005;66:930–4.
[18] Schulman CC. Long-term aspects of medical treatment of
BPH. Eur Urol 2001;40:8–12.
[19] Marks LS, Partin AW, Dorey FJ, et al. Long-term effects of
finasteride on prostate tissue composition. Urology 1999;
53:574–80.
[20] Lowe FC, McConnell JD, Hudson PB, et al. Long-term 6year experience with finasteride in patients with benign
prostatic hyperplasia. Urology 2003;61:791–6.
[21] Basillote JB, Armenakas NA, Hochberg DA, Fracchia JA.
Influence of prostate volume in the detection of prostate
cancer. Urology 2003;61:167–71.
[22] Ung JO, San Francisco IF, Regan MM, DeWolf WC, Olumi
AF. The relationship of prostate gland volume to extended
needle biopsy on prostate cancer detection. J Urol 2003;
169:130–5.
[23] Ochiai A, Babaian RJ. Update on prostate biopsy technique. Curr Opin Urol 2004;14:157–62.
[24] Crawford ED, Hirano D, Werahera PN, et al. Computer
modeling of prostate biopsy: tumor size and location–not
clinical significance–determine cancer detection. J Urol
1998;159:1260–4.
[25] Mariappan P, Chong WL, Sundram M, Mohamed SR.
Increasing prostate biopsy cores based on volume vs
the sextant biopsy: a prospective randomized controlled
[26]
[27]
[28]
[29]
[30]
[31]
[32]
639
clinical study on cancer detection rates and morbidity.
BJU Int 2004;94:307–10.
Makhlouf AA, Krupski TL, Kunkle D, Theodorescu D. The
effect of sampling more cores on the predictive accuracy
of pathological grade and tumour distribution in the
prostate biopsy. BJU Int 2004;93:271–4.
Kulkarni GS, Al-Azab R, Lockwood G, et al. Evidence of
biopsy derived grade artefact among larger prostate
glands. J Urol 2006;175:505–9.
Serfling R, Schulman M, Thompson GL, et al. Cancer
detection rate and grade shift in the PCPT are reflections
of finasteride-induced changes in prostate volume and
tumor shrinkage: results of a mathematical modelling
study. J Urol 2006:174 (abstract no. 1493).
Catalona WJ, Smith DS, Ratliff TL, et al. Measurement of
prostate-specific antigen in serum as a screening test for
prostate cancer. N Engl J Med 1991;324:1156–61.
Brawer MK, Chetner MP, Beatie J, Buchner DM, Vessella
RL, Lange PH. Screening for prostatic carcinoma with
prostate specific antigen. J Urol 1992;147:841–5.
Thompson IM, Pauler DK, Goodman PJ, et al. Prevalence of
prostate cancer among men with a prostate-specificantigen
level 40 ng per milliliter. N Engl J Med 2004;350:2239–46.
Thompson IM, Chi C, Goodman P, et al. Comparison of the
operating characteristics of PSA for prostate cancer detection for finasteride and placebo in the Prostate Cancer
Prevention Trial. J Urol 2006:174 (abstract no. 470).