PHILADELPHIA—In a study of high-risk prostate cancer treated with radical prostatectomy or radiation therapy, more than 6 months of androgen deprivation therapy was associated with higher mortality than shorter durations, Cliff Robinson, MD, reported at the 48th Annual Meeting of the American Society for Therapeutic Radiology and Oncology (ASTRO abstract 2302).
The retrospective analysis showed that treatment failure rates were similar with longer and shorter durations. Androgen deprivation for longer than 6 months was associated with a twofold increase in mortality, said Dr. Robinson, a radiation oncology resident at the Cleveland Clinic.
The current recommendation for at least 2 years of androgen deprivation as part of the definitive treatment of high-risk prostate cancer is based on older studies in patient groups no longer representative of the prostate cancer population, Dr. Robinson said. The studies were completed in the pre-PSA era, mostly in patients with locally advanced prostate cancer. Today, few patients have locally advanced cancer at diagnosis, and risk classification is based instead on initial PSA level, Gleason score, and American Joint Committee on Cancer (AJCC) stage.
He added that the current study included surgery patients, which other previous studies did not, and the radiation therapy patients in the current study received a higher dose than in prior trials.
The analysis was conducted in 579 high-risk (by the modern definition) patients treated at the Cleveland Clinic between 1998 and 2003. Patients received external beam radiation (67%), radical prostatectomy (29%), or a permanent prostate radioactive seed implant (4%). Patients were followed a mean of 5 years after definitive treatment. Androgen deprivation therapy was used in 74% of patients and lasted longer than 6 months (median, 15 months) in 17%.
The incidence of biochemical and clinical failure was unrelated to the duration of androgen deprivation. In a multivariate analysis, more than 6 months of androgen deprivation was an independent predictor of death (relative risk 2.1 vs no treatment, P = .03), as were pack-years of cigarette smoking and Gleason score. "The reasons patients receiving longer term hormone therapy may do worse are unclear," he said. "A number of factors could complicate the issue, and this area needs further investigation before any conclusions can be drawn." The next step, he said, will be to investigate medical comorbidity as a possible confounder.
