Immunogenicity, efficacy, and safety of different dose schedules of human papillomavirus vaccines: a Bayesian network meta–analysis of randomized controlled trials
Background To enhance global vaccine coverage and reduce costs, Human papillomavirus (HPV) immunization schedules are shifting from the standard three–dose (3D) regimen toward reduced–dose (1D or 2D) schedules. We aimed to evaluate the comparative immunogenicity, efficacy, and safety of diverse dose schedules and vaccine types using a Bayesian network meta–analysis. Methods Two reviewers independently searched PubMed, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), CNKI, and Wanfang Data from inception to February 15, 2026, for randomized controlled trials (RCTs) evaluating different dose schedules of HPV vaccines in females aged 9–26 years. Primary outcomes were HPV–16/18 Seroconversion rates (SCR) and 6–month Persistent infection (PI). A Bayesian random–effects model was constructed using the multinma package in R, and treatments were ranked using the Surface Under the Cumulative Ranking (SUCRA) values. Results For immunogenicity, the 9v–3D regimen demonstrated the highest SCR compared with placebo (OR 62.45, 95% CrI 15.20–158.30; SUCRA 0.96). However, these large effect estimates are largely driven by near–zero events in placebo arms. Regarding protective efficacy against PI, while 3–dose regimens ranked highest (9v–3D SUCRA 0.95; 2v–3D SUCRA 0.89), single–dose schedules (1D) maintained clinically significant protection (e.g., 2v–1D vs. placebo: OR 0.32, 95% CrI 0.07–0.96; SUCRA 0.61). Safety rankings identified 1D schedules as having the lowest risk of injection–site pain (SUCRA: 2v–1D 88.0%, 9v–1D 82.0%). Subgroup analysis suggested comparable efficacy of 2–dose regimens in adolescents aged 9–14 years compared to the standard 3D regimen (P–interaction = 0.04). Conclusions Although reduced–dose schedules induce lower antibody titers than the standard 3–dose regimen, they provide robust protective efficacy against persistent HPV infection with a superior safety profile. Our findings support the widespread adoption of 2–dose schedules in adolescents and suggest that a single–dose strategy is a viable alternative in resource–constrained settings to accelerate the elimination of cervical cancer. However, given the surrogate nature of the study endpoints and the limitations of network meta-analysis for establishing non-inferiority, cautious interpretation of these results is warranted, and further long-term clinical endpoint data are needed.