Reports
Relearning and Refining the Lessons from the Prostate Cancer Prevention Trial
This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.
PRIORITY PRESS - 104th Annual Meeting of the American Urological Society
Chicago, Illinois / April 25-30, 2009
To a large extent, the increased incidence of high-grade cancers in the finasteride arm of the PCPT (Prostate Cancer Prevention Trial) overshadowed the overall benefit on cancer prevention. The finding has contributed directly to many clinicians’ reluctance to recommend the 5-alpha reductase inhibitor (5-ARI) for chemoprevention of prostate cancer.
Evidence accumulated since publication of the PCPT results has shown that finasteride improves cancer detection by enhancing the performance characteristics of prostate-specific antigen (PSA) testing and digital rectal exam (DRE). In particular, the data have shown that chemoprevention with the 5-ARI leads to more accurate identification of high-grade cancer on biopsy (Cohen et al. J Natl Cancer Inst 2007;99(18):1366-74, Kulkarni et al. J Urol 2006;175(2):505-9).
PCPT principal investigator Dr. Ian M. Thompson, University of Texas at San Antonio, stated that the improved PSA metrics in the active treatment arm “would be expected to contribute to greater detection of all grades of prostate cancer with finasteride” (Thompson et al. J Natl Cancer Inst 2006;98(16)1128-33). Subsequent studies corroborated his prediction.
A comparison of biopsy and radical prostatectomy specimens from the PCPT provided additional evidence that finasteride improved detection of high-grade cancer (Lucia et al. J Natl Cancer Inst 2007;99(18):1375-83). The relative risk of finding high-grade cancer in treated patients declined substantially in prostatectomy specimens (relative risk [RR] 1.20) vs. biopsy specimens (RR 1.68).
Moreover, among treated patients, high-grade cancer in a radical prostatectomy specimen predicted a significantly greater likelihood of high-grade cancer in the biopsy (70% vs. 50% with placebo, P=0.01).
Adjusting for Bias
A bias-adjusted analysis that corrected for the increased biopsy accuracy in the active arm of the PCPT resulted in an even greater reduction in prostate cancer risk compared with placebo (RR 0.70, P=0.0001) and a non-significant increase in the relative risk of high-grade cancer (RR 1.14, P=0.12). An analysis that incorporated grading data from 500 radical prostatectomy specimens revealed a 28% decrease in the incidence of high-grade cancer in the active arm (Redman et al. Cancer Prev Res 2008;1(3):174-81).
Another recent analysis corrected for the known variability in Gleason score between biopsy and prostatectomy specimens (Pinsky et al. Cancer Prev Res 2008;1(3):182-6). That analysis showed that chemoprevention with the 5-ARI reduced the incidence of all grades of prostate cancer (Gleason score 2-6, 7 and 7-10) by 16% to 39% compared with placebo.
When used to treat benign prostatic hyperplasia, finasteride improves symptoms, in part, by reducing prostate volume. In the PCPT, men in the active arm had a smaller prostate volume at biopsy (25 cc vs. 33 cc). The difference offered yet another plausible explanation for the higher rate of high-grade cancer in the finasteride arm of the PCPT: A smaller prostate would increase the chances of detecting cancer, including high-grade cancer, particularly “for-cause” biopsies performed before the end of the trial.
Prostate Volume and Cancer Diagnosis
Two studies reported here at the AUA meeting examined the relationships among prostate volume, PSA metrics and prostate cancer diagnosis. Investigators at Stanford University in California studied 1304 men referred for a first prostate biopsy due to an elevated PSA level (>4 ng/mL to <u><</u>10 ng/mL) or abnormal DRE, a population that mimicked characteristics of PCPT participants who had for-cause biopsies. None of the men in the Stanford study had used finasteride or other hormone-altering therapy.
Receiver-operator characteristic (ROC) curves were determined for PSA after stratification by prostate volume and the presence of cancer or high-grade cancer (Gleason <u>></u>7). For detection of any cancer, the AUC for a PSA <u><</u>10 ng/mL increased from 0.520 for a prostate volume >50 cc to 0.758 for a prostate volume <30 cc. For detection of high-grade cancer, the AUC increased from 0.497 to 0.712 with decreasing prostate volume, reported Dr. Christopher Elliott. In both cases, the difference in PSA metrics was statistically significant (P=0.01).
“Decreases in prostate volume over time and the resultant change in PSA performance characteristics may have contributed to a bias towards the detection of high-grade disease in the finasteride arm of the PCPT,” Dr. Elliott and colleagues concluded.
The second AUA presentation focused on the relationship between prostate volume and pathologic grade of prostate cancer. The study involved 2880 patients who underwent radical retropubic prostatectomy or robotic-assisted laparoscopic prostatectomy. High-grade cancer was defined as a Gleason score <u>></u>7, according to Dr. Mark Newton, Vanderbilt University, Nashville, Tennessee.
High-grade cancer was associated with a median prostate volume of 43.6 cc compared with a median of 46.5 cc for low-grade disease (P<0.0001). Moreover, the likelihood of upgrading from biopsy to radical prostatectomy increased with decreasing quartile of prostate volume. In addition, low prostate volume was associated with positive surgical margins and extracapsular extension. In a multivariate analysis, older age, increasing PSA level and higher pathologic stage were associated with an increased risk of high-grade cancer. In contrast, prostate volume had an inverse association with diagnosis of high-grade cancer (OR 0.78-0.39 for quartiles 2 to 4 vs. 1, P=0.043 to P<0.001). “Low prostate volume is an independent predictor of high-grade disease,” Dr. Newton and colleagues concluded.
More Data on Chemoprevention
The AUA meeting also provided the forum for the initial presentation of the data from the REDUCE (Reduction by Dutasteride of Prostate Cancer Events) trial, which evaluated chemoprevention with the 5-ARI dutasteride. The trial differed from the PCPT in that study participants had an increased risk of prostate cancer by virtue of an elevated PSA value at enrolment. Nonetheless, both trials tested the hypothesis that a 5-ARI can reduce the risk of developing prostate cancer.
REDUCE included 8200 men ages 50 to 75 with baseline PSA values of 2.5 to 10 ng/mL for those younger than 60 years of age or 3 to 10 ng/mL for older men. All participants had a negative prostate biopsy in the six months prior to enrolment and no evidence of prostate cancer. The men were randomized to dutasteride 0.5 mg/day or to placebo and followed for four years. All participants had prostate biopsies after two years and at the end of the study. The primary end point was biopsy-detectable prostate cancer.
During the trial, prostate cancer was diagnosed in 1516 (22.5%), 1419 cases by protocol-mandated biopsy and 97 cases by for-cause biopsy. Placebo-treated patients had 857 cases of prostate cancer compared with 659 in the dutasteride group, representing a 23% risk reduction (P<0.0001). The two patient groups had similar rates of high-grade cancer, whether high grade was defined as Gleason 7-10 or Gleason 8-10.
Subgroup analysis favoured treatment with the 5-ARI, regardless of patient age, family history of prostate cancer or baseline prostate symptom score, prostate volume or PSA level, ROC analysis showed that chemoprevention with dutasteride improved the diagnostic utility of PSA. The AUC was significantly greater in the 5-ARI group for all tumours that were diagnosed (P<0.0001) and for diagnosis of high-grade cancer (P<0.02). The most common adverse events were loss of libido, erectile dysfunction, decreased semen volume and gynecomastia, all of which occurred more often in the dutasteride group (P<0.05).
“There was a consistent effect across time and key baseline subgroups,” reported Dr. Gerald L. Andriole, Chief of Urologic Surgery, Washington University School of Medicine, St. Louis, Missouri, and chair of the REDUCE steering committee. “Dutasteride enhanced the utility of PSA as a diagnostic test for prostate cancer—overall and for high-grade disease—had beneficial on benign prostatic hyperplasia outcomes and was well tolerated.”