OBJECTIVE: To estimate comparative effectiveness, in terms of objective response rate (ORR), progression free and overall survival (PFS, OS) using unanchored MAIC of single-arm LIBRETTO-001 (selpercatinib) and EXAM trials (cabozantinib vs placebo) in advanced/metastatic RET-mutation positive MTC.
METHODS: Patient-level data from LIBRETTO-001 were weighted to match summary data from the cabozantinib arm of the EXAM trial. ORR and PFS were available for the RET mutation-positive cohort but OS only for the RET M918T subgroup in EXAM. Cohorts were balanced on all available baseline covariates (age, weight, performance status, sex, smoking status, prior tyrosine kinase inhibitor therapy, and RET M918T mutation status). Hazard ratios (HR) for PFS and OS, odds ratios (OR) for ORR, and related 95% CIs were estimated. Multiple imputation marginalization was performed as a sensitivity analysis for PFS and OS outcomes.
RESULTS: A total of 295 who received selpercatinib (LIBRETTO-001) and 107 patients who received cabozantinib (EXAM) were included. Before weighting, ORR, PFS and OS comparisons suggested favoring selpercatinib. Weighting resulted in balanced distributions of baseline covariates. After weighting, ORR was 82.9% and 31.7% (OR=10.46[95%CI 6.24,17.55, p<.0001]) for selpercatinib vs cabozantinib, respectively. 12-month PFS rates are 89% and 54%, and HR=0.08 (95%CI:0.05,0.13, p<0.001) for selpercatinib vs cabozantinib. Respective 24-month OS rates are 92% and 66%, and HR=0.20 (95%CI: 0.13,0.32, p=0.001). Sensitivity analyses were consistent with the primary analysis.
CONCLUSION: In weighted comparison, selpercatinib demonstrated a significant improvement in ORR, PFS and OS versus cabozantinib. A limited number of covariates were available for adjusting selpercatinib data. Hence, there is a risk of unmeasured confounding that could influence these findings. Additional uncertainty is due to EXAM OS data in RET M918T subgroup while MAIC used covariates, including proportion of RET M918T, from RET mutation-positive patients. These findings should be interpreted with caution considering the limitations of an unanchored MAIC.
