Abstract Background Randomized controlled trials are the gold standard for demonstrating treatment efficacy. When infeasible, external control arm (ECA) analysis is an effective way to interpret treatment arm results. We developed an ECA for a single-arm trial (LS-004) for a HIF-2α inhibitor (belzutifan) in 61 patients with VHL renal cell carcinoma (RCC) to help interpret results. With a median follow-up of 37.8 months in LS-004, the ORR was 64% (95% CI = 50.6 to 75.8); median time to surgery was not reached. Methods The ECA was developed using natural history study data for VHL RCC patients undergoing active surveillance with ≤5 years of follow-up. Key LS-004 eligibility criteria were applied. Propensity score (PS) weighting was used to balance prognostic factors, with balance evaluated using standardized mean difference (SMD). PS adjusted point estimates and 95% CI for ORR and time to surgery (TTS) are presented. For the ECA, ORR was evaluated among patients with ≥3 scans to optimize for confirmed responses. Results The ECA included 244 patients (167 for ORR analysis). Prognostic factors were balanced with SMD 0.1 for all covariates. PS adjusted ORR for LS-004 and ECA was 63.9% (95% CI = 51.9 to 76.0) and 1.5% (95% CI = 0.0 to 3.3), respectively. Median TTS was 51.3 (95% CI = 43.8 to not yet reached) months in ECA; not yet reached in LS-004. Conclusions The belzutifan treatment effect is large compared with the ECA, supporting belzutifan efficacy in VHL RCC. Although residual confounding is possible, the large effect is unlikely due to chance.
Linehan et al. (Mon,) studied this question.