This article is more than 5 years old. Information may no longer be current.
Prostate cancer trials: Waiting for Godot
Click Here to Manage Email Alerts
In an era in which we are ever more hurrying to find truth, and the FDA is exploring a range of surrogate endpoints to support early introduction of new anticancer agents for prostate cancer, two important communications recently found their way into The New England Journal of Medicine.
Each one reflects decades of follow-up of seminal randomized trials that now seal the data as representing “truth.”
No routine role for finasteride
Many years ago, SWOG published the results of a randomized trial of 18,882 men testing prophylactic finasteride — a 5-alpha-reductase inhibitor that blocks the uptake of androgens by prostate cells — vs. placebo to prevent prostate cancer in high-risk patients.
The initial results were promising, with nearly a 25% reduction in cases in the finasteride arm, but follow-up was modest.
An important study from University of Southern California raised the question of whether those cancers that did occur were of higher grade and histological aggressiveness than those that arose in the control population, and similar results occurred in the randomized study. As a result, although the data were accepted and viewed as important, the study was not paradigm-shifting, and most patients at risk for prostate cancer were not offered finasteride prevention.
Goodman and colleagues updated the data and showed that with median follow-up of more than 18 years, prostate cancer deaths on placebo (n = 56 of 9,457) slightly exceeded those on finasteride (n = 42 of 9,423), but that total deaths on placebo (n = 2,979 of 9,457) were slightly less than those on finasteride (n = 3,048 of 9,423). High Gleason scores appeared evenly distributed.
I don’t believe this study was set up to report long-term complications of finasteride, but that would have been important as there are reports of sexual dysfunction and perhaps an increase in diabetes in patients taking that agent for benign prostate hyperplasia. Whatever the P values (not published), it is clear from the HRs that finasteride has no routine role in prostate cancer prevention in the dose schedule employed, and we can put that concept to bed.
What is more interesting is that the incidence of prostate cancer was 0.6% vs. 0.43% in these two subgroups of high-risk patients, and that should set into context some of the rhetoric advanced by those who advocate broad-based PSA screening.
Surgery vs. watchful waiting
In parallel, Bill-Axelson and colleagues reported 29-year follow-up of an important randomized trial on the management of early prostate cancer, also published in NEJM.
This Scandinavian group first reported their comparison of radical prostatectomy vs. watchful waiting in men with localized prostate cancer in 2002, 13 years after commencing recruitment.
As we all know, prostate cancer can be a slowly evolving disease, and we have come to recognize that far more men are diagnosed with the illness than actually succumb to it. This has led to recycling confusion in recommendations on screening, early detection and watchful waiting over the years.
These researchers recruited 695 men between 1989 and 1999, with entry criteria including age younger than 75 years, an anticipated life expectancy of more than 10 years, PSA less than 50 ng/ml and histology that was moderately to poorly differentiated. These patients predominantly had clinically detected rather than PSA-detected disease, constituting a somewhat different population from some of the other contemporary studies.
Of importance, a committee that was ignorant of the randomization class reviewed cause of death.
In an early report, results suggested that radical prostatectomy reduced only disease-specific mortality, without a significant impact on overall mortality; 3 years later, they published data to show that radical prostatectomy reduced disease-specific and overall mortality as well as the risks for metastasis, although the absolute reduction in risk for death after 10 years was small.
In their current study, researchers reported that 261 of 347 men in the radical prostatectomy group have died, in comparison with 292 of 348 men in the watchful-waiting group. This included 71 tumor-specific deaths in the radical prostatectomy group and 110 in the watchful-waiting group.
In the surgery group, extracapsular extension and Gleason score higher than 7 were strong adverse survival predictors.
With complete data at 23 years, the surgery group achieved a 2.9-year survival benefit. Patients should be able to decide whether that benefit is of sufficient magnitude to justify the side effects of surgery.
Implications for disease management
So, what do we learn from these studies?
Firstly, most prostate cancer is a relatively slowly evolving disease, and we must still be very careful in identifying whether surrogate markers are truly predictive.
It does seem that, for clinically detected prostate cancer, active surgery in healthy patients is a safer option than watchful waiting, although there will be some variation depending on the clinical circumstances. Whether the current vogue for early reporting of surrogate markers in metastatic prostate cancer is truly predictive of outcome in the long-term remains to be seen.
Of importance, the study by Bill-Axelson and colleagues is not a surrogate for PSA-detected — nonclinically identified — disease, and mature data from other studies will govern future management algorithms.
With respect to finasteride chemoprevention, very mature data suggest that the concern about increasing the Gleason score of cancers that did evolve was a false lead, but also confirm that finasteride does not seem to prevent prostate cancer.
Perhaps of more importance, there may have been a missed opportunity with the absence of reporting of the long-term implications of finasteride treatment — hundreds of thousands of men take this agent every day for management of benign prostatic hyperplasia, and it would be helpful for them to know the extent of iatrogenic consequences compared with a placebo-treated cohort, especially as there are alternative surgical treatment options.
References:
Bill-Axelson A, et al. N Engl J Med. 2018;doi:10.1056/NEJMoa1807801.
Cote RJ, et al. Br J Cancer. 1998;78:413-418.
Goodman P, et al. N Engl J Med. 2019;doi:10.1056/NEJMc1809961.
Thompson IM, et al. N Engl J Med. 2003;doi:10.1056/NEJMoa030660.
For more information:
Derek Raghavan, MD, PhD, FACP, FRACP, FASCO, is HemOnc Today’s Chief Medical Editor for Oncology. He also is president of Levine Cancer Institute at Atrium Health. He can be reached at derek.raghavan@atriumhealth.org.
Disclosure: Raghavan reports no relevant financial disclosures.