Comparative analysis of the impact of chickenpox and herpes zoster vaccination in Belgium under two different exogenous boosting mechanisms

Background: Chickenpox (CP) and herpes zoster (HZ), both caused by the varicella-zoster virus (VZV), present a significant public health burden in unvaccinated populations. Universal CP vaccination has been long debated due to concerns about a potential increase in HZ incidence as a consequence of the exogenous boosting hypothesis. Methods: We performed a cost-utility analysis on two deterministic compartmental dynamic transmission models, each of which employed a different underlying mechanism of exogenous boosting: temporal or progressive immunity. We considered four vaccination strategies: the current practice of no widespread vaccination and three strategies involving CP and HZ vaccination, either alone or in combination. The CP vaccines considered were Varivax and ProQuad, while the HZ vaccine considered was the recombinant zoster vaccine (RZV), Shingrix. The vaccine prices per dose were as follows: Varivax - €52.52, ProQuad - €73.69, recombinant zoster vaccine - €170.26. The clinical and economic impact of vaccination on both CP and HZ outcomes were evaluated. The main health outcome of interest was the quality-adjusted life year (QALY) which was used to compare the strategy yielding the highest average net monetary benefits (i.e., the optimal strategy) across a range of willingness-to-pay (WTP) values. Costs and health outcomes were discounted at 3.0% and 1.5% annually, respectively. We used 3 time horizons (i.e., 50, 75 & 100 years) and implemented the healthcare payer perspective throughout the analysis. Results: CP vaccination led to a substantial reduction in CP incidence in both models. In the temporary immunity boosting (Temp) model, strategies that included HZ vaccination showed a decrease in HZ incidence. For the CP vaccination strategy in the Temp model, and for all CP and HZ vaccination strategies in the progressive immunity boosting (Prog) model, we observed both short- and medium-term increases in HZ, followed by a decrease to levels below the no-vaccination scenario. From the healthcare payer’s perspective, using a WTP of €40,000 per QALY gained, the Temp model indicated that the three vaccination strategies were cost-effective when considering time horizons of 50, 75, and 100 years. For the Prog model, only strategies combining both CP and HZ vaccination were cost-effective given a 100-year time horizon. Vaccination strategies under the Temp model became cost-effective at lower values of WTP compared to those under the Prog model. Conclusion: Both models predicted that universal CP vaccination would result in significant reductions in the burden of CP disease, however, the HZ disease burden impact varied significantly depending on the assumed boosting mechanism. Hence, the choice of modeled exogenous boosting mechanism leads to different optimal vaccination strategies. Ascertaining the relative accuracy of these structural model choices will require continued research on the mechanism of VZV boosting.