Nivolumab Plus Ipilimumab vs Nivolumab Alone in Advanced Cancers Other Than Melanoma: A Meta-Analysis

IMPORTANCE: Although the combination of nivolumab plus ipilimumab has unquestionable benefit over nivolumab monotherapy in advanced melanoma, currently no summative analyses have compared the combination with nivolumab monotherapy for advanced cancers other than melanoma. OBJECTIVE: To examine whether the addition of ipilimumab to standard-dose nivolumab safely improves clinical outcomes in patients with advanced cancers other than melanoma. DATA SOURCES: Electronic databases (PubMed, EBSCO Information Services, Embase, and Cochrane Library) were systematically searched for studies of standard-dose nivolumab plus ipilimumab vs nivolumab alone in the treatment of advanced cancers other than melanoma published from database inception to October 31, 2022. STUDY SELECTION: Eight studies (total patients, 1727; nivolumab plus ipilimumab group, 854; nivolumab monotherapy group, 873) met the selection criteria. Patients had squamous cell lung cancer, non–small cell lung cancer with programmed death ligand 1 level of 1% or higher, small cell lung cancer, pleural mesothelioma, urothelial carcinoma, esophagogastric carcinoma, sarcoma, or glioblastoma multiforme. DATA EXTRACTION AND SYNTHESIS: For comparison of overall survival (OS) and progression-free survival (PFS) outcomes, estimation of log(hazard ratios [HRs]) and SEs was initially performed for OS and PFS of each included study based on summary statistics extracted from individual Kaplan-Meier curves. Inverse-variance weighting was then used to compute pooled HRs (95% CIs). For comparison of dichotomous data (treatment-related grade 3 to 4 adverse events and discontinuations), odds ratios (ORs) were used, and the Mantel-Haenszel method was used to estimate pooled ORs (95% CIs). RESULTS: Treatment with nivolumab plus ipilimumab was not associated with improvement in OS over treatment with nivolumab alone (pooled HR, 0.95; 95% CI, 0.85-1.06; P = .36), with 4 of the 8 studies having numerically lower median OS with the combination. Nivolumab plus ipilimumab combination therapy was associated with marginal, but not clinically meaningful, improvement in PFS over nivolumab alone (pooled HR, 0.88; 95% CI, 0.79-0.98; P = .02). The combination was associated with substantially higher treatment-related grade 3 to 4 adverse events (pooled OR, 1.84; 95% CI, 1.47-2.31; P < .001) and treatment-related discontinuations (pooled OR, 1.96; 95% CI, 1.44-2.65; P < .001). This finding was recapitulated in meta-analyses of individual grade 3 to 4 adverse events of hepatotoxicity, gastrointestinal toxicity, pneumonitis, endocrine dysfunction, dermatitis, fatigue. CONCLUSIONS AND RELEVANCE: In this meta-analysis of 8 advanced cancers other than melanoma, the differences detected in OS and PFS between nivolumab plus ipilimumab and nivolumab were not clinically meaningful (even though statistical significance was detected in PFS). Treatment-related higher-grade toxicity and discontinuations were substantially higher with the combination therapy. The data indicate that investigations of anti–programmed death 1 (PD1) plus anti–cytotoxic T-lymphocyte–associated protein 4 (CTLA-4) therapies in any nonmelanoma advanced cancer should be conducted along with anti-PD1 monotherapy to ensure that the net effect of the addition of anti-CTLA-4 to anti-PD1 can be clearly established for that cancer and setting and that unnecessary CTLA-4 inhibition with related toxic effects (both clinical and financial) can be avoided.