Utility of Long-Term Follow-Up to Determine Safety in Radiotherapy-Specific Trials for Localized Prostate Cancer: Meta-Analysis of 29 Randomized Trials

PURPOSE/OBJECTIVE(S): Guideline and clinical adoption of new radiotherapeutic approaches commonly requires 5-10 years of follow-up in randomized trials to ensure long-term safety. This is resource intensive and slows the pace of advancement. We hypothesized that the relative risk (RR) of radiation-induced gastrointestinal (GI) or genitourinary (GU) toxicity would remain similar or decrease overtime with long-term follow-up. MATERIALS/METHODS: PubMed and Embase were systematically searched for randomized trials evaluating changes in radiotherapy (RT; hypofractionation, dose escalation, change in field size) for localized prostate cancer published between 1970 and 2020. Yearly toxicity estimates were digitally extracted for GI and GU toxicity. Short- and long-term toxicity were defined as toxicity arising in years 1-3 and 5-10 post-RT, respectively. RR differences of toxicity between trial arms were calculated, and differences between short- and long-term relative risk measures of toxicity were compared using paired sample Wilcoxon signed-rank tests weighted by trial size. RESULTS: Median follow-up was 6 years (range 3-16) of the included 29 trials (n = 18,069 men). Of trials that reported risk group (67%), 90% of patients had intermediate and high-risk disease, and 57% of patients received ADT. There was no evidence of an increase in the RR of any grade of GI or GU toxicity with follow-up beyond 3 years. The mean RR point estimate of short- vs long-term GI and GU toxicity consistently decreased over time, which was represented by negative values for all grades. Mean differences (late RR - early RR) for grade 1, 2+ and 3+ GI toxicity were (-0.12, P < 0.01), (-0.12, P = 0.02), and (-0.08, P = 1.0), respectively. There were no significant differences in the mean relative risk difference (late RR - early RR) over time for late grade 1 (-0.05, P = 0.3), 2+ (-0.04, P = 0.5), or 3+ (-0.03, P = 0.8) GU toxicity. CONCLUSION: While toxicity events may continue to occur over time, the relative difference at later time points is similar or lower compared to early time points. Thus, our data does not support concerns for very late relative increases in radiation-induced toxicity. Physician reported toxicity assessed up to 3 years post-treatment is likely sufficient to establish safety following alternative radiotherapy dose/fractionation schedules and target volumes in prostate cancer.