Survival in the Prostate Cancer Prevention Trial

To the Editor:

In their article on long-term survival in the Prostate Cancer Prevention Trial (PCPT), Thompson et al. (Aug. 15 issue)1 comment on the aggressiveness of high-grade disease (which they defined as a tumor having a Gleason score of 7 to 10). These authors have always argued that the increase in high-grade disease in men receiving finasteride was an artifact.2 If this were true, men with high-grade disease who are treated with finasteride should have a better survival rate than men in the placebo group. However, the survival rates in these groups were virtually the same. This finding indicates that the increase in high-grade disease is not an artifact but is real. The Food and Drug Administration rejected the use of finasteride for the prevention of prostate cancer on the basis of a related increase in the most aggressive, potentially lethal form of high-grade disease (Gleason score, 8 to 10), excluding the less aggressive pattern of disease (Gleason score of 7).3,4 In another recent report, the PCPT investigators stated that tumors with a Gleason score of 8 to 10 may cause “a small increase in prostate cancer mortality.”5 In the study published in the Journal, which defined a high-grade tumor as one having a Gleason score of 7 to 10, there is no report on prevalence or mortality for cancers with Gleason scores of 8 to 10. What are they?

Patrick C. Walsh, M.D.
Johns Hopkins Medical Institutions, Baltimore, MD
pwalsh@jhmi.edu

No potential conflict of interest relevant to this letter was reported.
5 References

To the Editor:

The 5-α reductase inhibitor finasteride has found considerable clinical use, not only in the treatment of prostate disease but also as a treatment for male-pattern alopecia. Concerns have been raised about the potential for an increased risk of male breast cancer among patients receiving finasteride.1,2 Given that the PCPT involved a very large sample of more than 18,000 men, the investigators have the opportunity to study the incidence patterns of male breast cancer among the patients enrolled in the trial.

Swaroop Revannasiddaiah, M.D.
Swami Rama Cancer Hospital and Research Institute, Haldwani, India
swarooptheone@gmail.com
Sridhar P. Susheela, M.D.
Bangalore Institute of Oncology, Bengaluru, India

No potential conflict of interest relevant to this letter was reported.
2 References

The authors reply: In response to Walsh: the outcomes according to Gleason score in our recent report, with scores of 2 to 6 indicating low-grade tumors and scores of 7 to 10 indicating high-grade tumors, were consistent with those in our initial 2003 report.1 Table 1Table 1Estimates of Postdiagnostic Survival for Men with Tumors with Gleason Scores of 8 to 10 in the Prostate Cancer Prevention Trial. shows postdiagnostic Kaplan–Meier survival estimates for patients with Gleason scores between 8 and 10. In this small subset of men from the PCPT, confidence intervals are wide and overlap. Modeling the time from randomization to death with the use of time-dependent covariates to account for the time at which the diagnosis of prostate cancer occurred (an approach that is less biased than one incorporating postdiagnostic survival), we calculated the between-group difference in the hazard ratios for death when Gleason scores of 8 to 10 rather than 7 to 10 were used to indicate high-grade cancer. After adjusting for age and race, we found that the P value for the comparison was 0.59, indicating that there was no statistically significant difference in survival. However, this test is also underpowered and represents another reason why we chose not to highlight this subset in our article.
Revannasiddaiah and Susheela ask about the incidence of male breast cancer in the PCPT trial. With 141,009 person-years of follow-up, there have been two cases, one in each of the two study groups.

Catherine M. Tangen, Dr.P.H.
Phyllis J. Goodman, M.S.
Fred Hutchinson Cancer Research Center, Seattle, WA
Ian M. Thompson, Jr., M.D.
Cancer Therapy and Research Center, San Antonio, TX
thompsoni@uthscsa.edu

Since publication of their article, the authors report no further potential conflict of interest.
1 Reference

Related Article
 

Long-Term Survival of Participants in the Prostate Cancer Prevention Trial

August 15, 2013 | I.M. Thompson and Others