Long-term comparative toxicity of LDR versus HDR prostate brachytherapy ± EBRT and evaluation of risk hazard over time: A SEER-Medicare analysis

108 Background: Severe urinary adverse events (UAEs) include surgical treatment of urethral stricture, urinary incontinence and radiation cystitis. Our objective is to compare the incidence of late UAEs after low dose rate BT (LDR) and high dose rate BT (HDR) as well as LDR+EBRT and HDR+EBRT. Methods: We identified men treated with LDR (n=12,801), HDR (n=685), LDR+EBRT (8,518) and HDR+EBRT (n=2,392) from the SEER-Medicare Database. The populations were balanced by propensity weighting and the Kaplan-Meier incidence of severe UAEs was compared. Propensity-weighted Cox proportional hazards models were used to compare the adjusted hazard of UAEs. These UAEs were compared to a cohort of men not treated for prostate cancer. Results: Median follow-up was 4.3 years. At 8 years, the propensity weighted cumulative UAE incidence was highest after HDR+EBRT (28%) and lowest after LDR (17%; see Figure). The absolute excess risk over non-treated controls of a UAE at 8 years was 1.9%, 3.8%, 8.4% and 12.9% for the LDR, HDR, LDR + EBRT, and HDR + EBRT respectively. This translates into a number needed to harm of 53, 26, 12, and 8 persons. There is no statistical difference in severe UAE risk between HDR vs. LDR or between HDR+EBRT vs. LDR+EBRT. The additional risk for developing a UAE related to treatment for LDR, LDR+EBRT, and HDR+EBRT, was greatest within the 2 years following treatment, and continued to decline over time. For HDR monotherapy, the risk was greatest within the first 4 years, and then declined. The risk of developing a severe UAE matched the baseline risk of the control population for all treatments at 4 years following therapy. Conclusions: LDR and HDR brachytherapy are statistically indistinguishable for late severe urinary adverse events. However, combination radiotherapy (either HDR+EBRT or LDR+EBRT) increases the risk of severe UAEs compared to HDR alone or LDR alone. In the 8 years following brachytherapy treatment, the increased risk of urinary toxicity occurs almost exclusively within the 2 years following therapy, and then declines to a baseline hazard. The hypothesis that late urinary radiation toxicity accelerates over time is not supported by this study.