Resumen
-
Background
The extent of cross-protection provided by currently licensed bivalent and quadrivalent HPV vaccines versus direct protection against HPV 31-, 33-, 45-, 52-, and 58-related disease is debated. A systematic literature review was conducted to establish the duration and magnitude of cross-protection in interventional and observational studies.
Methods
PubMed and Embase databases were searched to identify randomized controlled trials (RCT) and observational studies published between 2008 and 2019 reporting on efficacy and effectiveness of HPV vaccines in women against non-vaccine types 31, 33, 45, 52, 58, and 6 and 11 (non-bivalent types). Key outcomes of interest were vaccine efficacy against 6- and 12-month persistent infection or genital lesions, and type-specific genital HPV prevalence or incidence. RCT data were analyzed for the according-to-protocol (bivalent vaccine) or negative-for-14-HPV-types (quadrivalent vaccine) efficacy cohorts.
Results
Data from 23 RCTs and 33 observational studies evaluating cross-protection were extracted. RCTs assessed cross-protection in post-hoc analyses of small size subgroups. Among fully vaccinated, baseline HPV-naïve women, the bivalent vaccine showed statistically significant cross-protective efficacy, although with wide confidence intervals, against 6-month and 12-month persistent cervical infections and CIN2+ only consistently for HPV 31 and 45, with the highest effect observed for HPV 31 (range 64.6% [95% CI: 27.6 to 83.9] to 79.1% [97.7% CI: 27.6 to 95.9] for 6-month persistent infection; maximal follow-up 4.7 years). No cross-protection was shown in extended follow-up. The quadrivalent vaccine efficacy reached statistical significance for HPV 31 (46.2% [15.3–66.4]; follow-up: 3.6 years). Similarly, observational studies found consistently significant effectiveness only against HPV 31 and 45 with both vaccines.
Conclusions
RCTs and observational studies show that cross-protection is inconsistent across non-vaccine HPV types and is largely driven by HPV 31 and 45. Furthermore, existing data suggest that it wanes over time; its long-term durability has not been established.