The combination of cabozantinib plus nivolumab (CaboNivo) is approved for the first-line (1L) treatment of adults with advanced renal cell carcinoma (aRCC). We conducted a systematic literature review to assess the methodological quality of network meta-analyses (NMAs) including 1L CaboNivo in patients with aRCC. Outcomes for CaboNivo versus alternative combinations were also assessed.

METHODS:

Systematic searches were run (April 27 [updated August 23], 2021) in MEDLINE® and Embase to identify NMAs featuring 1L CaboNivo in aRCC populations. Publication titles/abstracts were screened for relevance. The analysis methods used in eligible NMAs were characterized.

RESULTS:

Twenty-eight NMAs involving aRCC populations were identified, of which four NMAs (6 publications) compared 1L CaboNivo versus various alternatives. All NMAs included comparisons of 1L CaboNivo versus axitinib/pembrolizumab and sunitinib, three versus lenvatinib/pembrolizumab, atezolizumab/bevacizumab, ipilimumab/nivolumab, and avelumab/axitinib, and one versus single-agent tyrosine kinase inhibitors, including cabozantinib. All four NMAs analyzed efficacy outcomes (overall survival [OS], progression-free survival [PFS], objective response rate [ORR]). Safety endpoints varied among NMAs. Three NMAs performed analyses for prognostic risk and programmed death ligand-1 expression subgroups. All NMAs were based on hazard ratio comparisons (3 used Frequentist methods; 1 used Bayesian analyses). All four NMAs assumed proportional hazards (PH), despite frequent violation of the assumption for at least one CaboNivo comparator. CaboNivo was consistently ranked first for OS, and second or third for PFS versus alternative combination therapies, but its rank order varied for the ORR, safety and subgroup analyses.

CONCLUSIONS: The validity of published NMA results comparing 1L CaboNivo versus alternative combination therapies for aRCC are limited by their reliance on the PH assumption. A more robust NMA that uses survival models with parameters for shape and scale, using the full Kaplan–Meier curves, is warranted to offer more reliable comparative outcome estimates, including long-term estimates, for 1L combination aRCC therapies.