Relative efficacy of varicella vaccines: network meta-analysis of randomized controlled trials.

Although varicella vaccination is highly effective, no head-to-head randomized controlled trials (RCTs) have compared the efficacy of different vaccine formulations. This study assessed the relative efficacy of different varicella vaccines using network meta-analysis (NMA). We estimated the relative efficacies of varicella vaccines and dosing regimens from RCTs using Bayesian NMA. Modeling-based time-series NMA (MBNMA) was performed, accounting for differences in time since vaccination, to extrapolate long-term vaccine efficacy (VE). Eight RCTs were included based on systematic review of biomedical databases. Efficacy data were reported for four varicella-containing vaccines: Varivax (V-MSD, one and two dose), Varilrix (V-GSK, one dose), Priorix-Tetra (MMRV-GSK, one dose), and Sinovac (V-Sinovac, one dose). All varicella vaccines were effective versus no vaccination. Two-dose V-MSD (98.29%, 95% credible interval [CrI] 96.08-99.23) showed significantly higher VE versus all one-dose varicella-containing vaccines, but no significant difference versus two-dose MMRV-GSK (95.19%, 95% CrI 90.3-97.63). Two-dose MMRV-GSK showed higher VE than one-dose V-GSK (66.47%; 95% CrI 43.02-79.43), but no significant differences in VE versus one-dose V-MSD or one-dose V-Sinovac. In one-dose comparisons, V-MSD showed significantly higher VE (93.09%, 95% CrI 89.13-95.96) than V-GSK, but no significant difference versus V-Sinovac (89.22%; 95% CrI 67.1-96.5). MBNMA indicated that protection against varicella was sustained without waning over the 10 year follow-up. Our study reported higher VE for two-dose V-MSD and MMRV-GSK. Among one-dose formulations, one-dose V-MSD was more efficacious than one-dose V-GSK. Policymakers should take into consideration differences in VE when implementing one- versus two-dose strategies in universal vaccination programs.