Trial Participation is Not Associated with Better Biochemical Recurrence-free Survival in a Large Cohort of External Beam Radiotherapy-Treated Intermediate- and High-Risk Prostate Cancer Patients.

ABSTRACT

AIMS: There is a widespread belief that outcomes of cancer patients treated within clinical trials might not be representative of the outcomes obtained within standard clinical settings. We sought to investigate the effect of trial participation on biochemical recurrence (BCR) in localised, D'Amico intermediate- and high-risk prostate cancer patients treated with external beam radiotherapy (EBRT). MATERIALS AND METHODS: We relied on a study population treated with EBRT between January 2001 and January 2021 at a single tertiary care centre, stratified according to trial enrolment. Separate Kaplan-Meier and multivariable Cox regression models tested BCR-free survival at 60 months within intermediate- and high-risk EBRT patients, after adjustment for covariables. Additionally, the analyses were refitted after inverse probability treatment weighting was performed separately for both risk subgroups. RESULTS: Of 932 eligible patients, 635 (68%) and 297 (32%) had intermediate- and high-risk prostate cancer, respectively. Overall, 53% of patients were trial participants. BCR rates were 11 versus 5% (P = 0.27) and 12 versus 14% (P = 0.08) in trial participants versus non-participants for intermediate- and high-risk subgroups, respectively. Differences in patient and clinical characteristics were recorded. Trial participation status failed to reach predictor status in multivariable Cox regression models for BCR in both intermediate-risk (hazard ratio 1.34; 95% confidence interval 0.71-2.49; P = 0.4) and high-risk patients (hazard ratio 1.03; 95% confidence interval 0.45-2.34; P = 0.9). Virtually the same results were recorded in inverse probability treatment weighting cohorts. CONCLUSIONS: Relying on a large cohort of EBRT-treated intermediate- and high-risk patients, no BCR differences were recorded between trial participants and non-participants after accounting for confounders.