Evaluation of disease-free survival as an intermediate metric of overall survival in patients with localized renal cell carcinoma: A trial-level meta-analysis

Background Overall survival (OS) is a critical endpoint in adjuvant trials but requires long durations to events and significant patient resources. In the current study, the authors assessed whether disease-free survival (DFS) can be an early clinical surrogate for OS in the adjuvant setting for localized renal cell carcinoma (RCC). Methods Using Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, the authors performed a systematic literature review of PubMed and the American Society of Clinical Oncology, European Society for Medical Oncology, and ClinicalTrial.gov Web sites (1996-2016). Inclusion in the current study required randomized controlled trials (RCTs) of adjuvant systemic therapy for localized RCC after nephrectomy with ≥3 years of outcomes data. Data regarding hazard ratios (HRs) and 5-year event-free rates from Kaplan-Meier estimates were extracted. A trial-level meta-analysis correlated estimates of 5-year DFS and 5-year OS as well as treatment effects (HRs) on these endpoints, weighted by the number of DFS events. R-squared ≥ 0.7 was prespecified as being indicative of a strong correlation and the potential for surrogacy. Results Thirteen RCTs encompassing 6473 patients who were treated with a variety of systemic therapies met eligibility. Only a modest correlation was observed between 5-year DFS and 5-year OS rates (R-squared, 0.48; 95% confidence interval, 0.14-0.67) and between treatment effects as measured by DFS and OS HRs (R-squared, 0.44; 95% confidence interval, 0.00-0.69). Conclusions Across RCTs of adjuvant systemic therapy for localized RCC, there was no strong correlation noted between 5-year DFS and 5-year OS rates or between treatment effects on these endpoints. These results highlight the need to identify alternative and more rapid clinical or biologic endpoints to hasten drug development and improve clinical outcomes. Cancer 2018;124:925-33. © 2017 American Cancer Society.