Cost-effectiveness Analysis of Maternal Immunization with RSVpreF Vaccine for the Prevention of Respiratory Syncytial Virus Among Infants in Spain

Respiratory syncytial virus (RSV) is the leading cause of acute lower respiratory infections (ALRI) in children under 1 year of age, mainly those related to bronchiolitis and pneumonia. All infants are at risk of acquiring an RSV infection with severe complications, which are unpredictable and cause hospitalization [1]. In particular, about 1–2% of bronchiolitis are severe enough to require hospital admission, and of these, about 10–20% require care in intensive care units (ICU) [2, 3]. Worldwide, RSV is the second leading cause of death in children under 1 year of age. While RSV mortality in developed countries is lower relative to developing countries, high-risk infants with comorbidities experience higher rates of mortality. In Spain, mortality rates of 1.47 deaths per 100,000 in children aged < 5 years and 3.40 deaths per 100,000 in children aged < 2 years have been reported, with an average of 30 deaths per year [4]. An observational study in Spain recently found an in-hospital case fatality rate of 0.14% [1].

In high-income countries like Spain, RSV infections cause a significant overload of care every winter, both in primary care services and in hospitals, imposing a significant burden to the healthcare system. In 2019, in Spain, the annual cost of hospitalizations due to RSV in children was €44.8 million, and infants aged < 6 months accounted for almost 60% of these costs [5]. Consequently, RSV disease is associated with an important public health impact and substantial economic burden, which has made the development of RSV prevention strategies a major global health priority [6, 7].

In this context, a new bivalent RSV prefusion F protein-based vaccine (RSVpreF) has recently been approved by the European Medicines Agency (EMA) [8] and reimbursed in Spain. This new strategy aims to prevent RSV-associated illness in infants in the first 6 months of life by active immunization of pregnant women between 24 and 36 weeks gestational age, providing passive protection of the newborn through transplacental antibody transfer [8]. The efficacy, safety, and immunogenicity of RSVpreF in infants born to people vaccinated during pregnancy has been evaluated in a placebo-controlled phase III clinical trial (MATernal Immunization Study for Safety and Efficacy [MATISSE]) [9]. In this study, RSVpreF efficacy against severe medically attended ALRI due to RSV was 81.8% (99.5% confidence interval [CI] 40.6–96.3) through the first 90 days of life and 69.4% (97.6% CI 44.3–84.1) over the 6-month follow-up period. Against RSV-related medically attended ALRI, vaccine efficacy was 57.1% (99.5% CI 128 14.7–79.8) through the first 90 days of life and 51.3% (97.6% CI 29.4–66.8) over the 6-month follow-up period [9].

While the efficacy and safety of RSVpreF vaccine have been demonstrated based on MATISSE, public health discussions should also consider if maternal vaccination against RSV is an efficient allocation of resources. This study has been used to inform healthcare national authorities in order to evaluate the efficiency of vaccinating pregnant women with the RSVpreF vaccine for an RSV immunization program against the standard of care on that moment, which was no intervention, by performing a cost-effectiveness analysis from the Spanish National Healthcare System (NHS) perspective.

Our model suggests that maternal vaccination with RSVpreF would be a cost-effective option that could bring savings to the Spanish NHS when compared to no intervention.

Several studies have been published on the cost-effectiveness of maternal vaccination with RSVpreF (none of them in Spain). Compared to no intervention, the cost-effectiveness of immunization during pregnancy ranged from dominant (more effective and less costly) to €200,000 per QALY gained, depending on the modeled region, efficacy, pricing, and severity of the RSV season. Researchers from the European Respiratory Syncytial Virus Consortium in Europe (RESCEU) programme conducted a multi-country analysis in six European countries (Denmark, England, Finland, Italy, The Netherlands, and Scotland). This analysis found that RSVpreF maternal vaccination was dominant from the Finnish NHS perspective and cost-effective from the Scottish NHS perspective (€28,600/QALY) considering a WTP of £20–30,000 (€23–34,500) per QALY gained, compared to no-intervention [30]. Another evaluation was taken from the Norwegian NHS perspective [31]. In this analysis, maternal immunization was considered not to be cost-effective when compared to no intervention (~ €118,000/QALY). However, these analyses did not incorporate deaths avoided due to RSV prevention strategies because of very low rates of RSV-related deaths in high-income countries [30, 31]. This may underestimate the cost-effectiveness of RSVpreF should a portion of these deaths be avoided in clinical practice. Furthermore, data from the MATISSE trial were not published at the time of these analyses, so the authors assumed the World Health Organization (WHO) preferred product characteristics, 70% efficacy and 4-month protection. Higher efficacy rates have been reported from the MATISSE trial, especially in younger infants, which in turn are those with higher rates of RSV infection. As such, the efficacy inputs used in the previously mentioned studies may not be reflective of the efficacy of RSVpreF in preventing severe RSV outcomes in very young infants. Another study from the Canadian Immunisation Research Network (CIRN) found that maternal immunization was cost-effective at the WTP threshold of $50,000 per QALY gained at a maximum vaccine price of $160 (€146) from a healthcare perspective or $200 (€183) from a societal perspective [32]. Finally, another study published by RESCEU applied the same parameters and data for a hypothetical cohort of 100,000 births in five different models (three static and two dynamic), and interventions were compared against non-immunization [33]. The three static models estimated comparable medically attended cases averted by maternal vaccination versus no intervention (1.019–1.073 cases), saving ∼€1 million medical and €0.3 million nonmedical costs while gaining 4–5 discounted quality-adjusted life years (QALYs) annually in < 1-year-olds, being a dominant intervention [33].

It should be noted that, in all studies, published ICERs have been calculated versus no intervention and none of them have carried out a cost-effectiveness analysis of maternal vaccination versus the approved monoclonal antibodies nirsevimab and palivizumab. This is mainly due to the absence of indirect comparisons and the limitations encountered when comparing both strategies due to the different clinical endpoints and populations included, as in the case of palivizumab, only used in high risk and/or extreme preterm infants.

A study evaluating maternal vaccination's impact on the prevention of RSV cases in the UK and Wales has been recently published. The results of this study are very much aligned with the results we obtained from the pharmacoeconomic study carried out for our NHS. In the UK study, a year-round maternal vaccination program with 60% coverage would reduce hospitalizations by 32%; in our study, with a 70% coverage, maternal vaccination would reduce hospitalizations by almost 40%. However, unlike our analysis, the UK study uses a dynamic transmission model that allows investigators also to estimate the benefit of vaccination in pregnant women, which they estimate at 20% of cases averted [34].

When assessing the impact of the reduction in vaccination coverage, we see a decrease in RSV cases avoided. Specifically, reducing vaccination coverage from 70 to 50% could avoid 2535 fewer hospitalizations, 2181 fewer emergency department visits, and 2563 fewer primary care visits. Moreover, savings in medical care would decrease from 44.1 to 31.5 million euros, annually. These findings highlight the importance of achieving higher levels of vaccination coverage in order to have a greater impact on reducing the burden of disease, both clinically and economically.

The present study has some limitations to be considered when interpreting the results. Firstly, while initial VE and waning through age 6 months were based on data from MATISSE, assumptions for VE waning during the months thereafter (i.e., ages 6 to < 10 months) were informed by evidence on the kinetics and decay of maternal transferred antibodies following natural infection and vaccination [35‐37]. Secondly, while we assumed vaccine provided some protection against RSV among late preterm infants (based on an ongoing observational seroepidemiology study of naturally acquired RSV antibody transplacental transfer [20]), we conservatively assumed that VE among early and extreme preterm infants would be 0%. Thirdly, in the model we assumed homogeneity in both disease incidence and vaccine uptake across all the Spanish territory, therefore, not reflecting the possible inequalities that may occur as a result of social and geographical particularities in certain populations. To address this aspect, the RSV incidence was obtained from a study including two Spanish regions, which would account for variability between regions, and variations of ± 25% in disease incidence rates were analyzed in sensitivity analysis, which showed that vaccine would still be cost-effective in a worst-case scenario. Moreover, a reduced vaccine uptake of 50% was also evaluated in an alternative scenario, demonstrating that vaccine would still be dominant. In addition, a literature review was performed to obtain the model inputs to be adapted to the Spanish environment and when not available, assumptions or data adjustments had to be made. Sensitivity analyses mitigated the uncertainty around the magnitude of the model key parameters, and expert opinion confirmed the totality of the assumptions.

There are also several assumptions and limitations which may confer a conservative bias to our model. For example, our model does not capture the potential direct effects of immunization on RSV among vaccinated pregnant people since vaccine efficacy among pregnant women was not an outcome in MATISSE. However, the recent study previously mentioned that did include health benefits for pregnant women as an outcome showed that the benefit accounted for 20% of the population-level health burden averted [34]. In this regard, a recent systematic review also summarizes the evidence about the RSV burden among pregnant women, and up to 10.7% of women with respiratory infections tested positive for RSV. Those RSV-positive pregnant women had higher odds of preterm delivery (OR 3.6 [1.3; 10.3]) [38]. Moreover, we use a static model that does not account for the potential indirect impact of maternal vaccination on other populations (e.g., other susceptible individuals who may reside in the same household as the infant). Nevertheless, as Pitman et al. 2012 [38] indicated, static models are acceptable when their projections suggest that an intervention is cost-effective, and dynamic effects would further enhance this (e.g., via prevention of secondary cases). In addition, due to the lack of data, there are also a number of potential short- and long-term sequelae of RSV that were not captured in this model as secondary bacterial infections, subsequent respiratory infections, recurrent wheezing, worsening of chronic lung disease among preterm infants, exacerbation of congenital heart disease, and asthma.

In conclusion, the results of this economic evaluation suggest that, in Spain, maternal vaccination with bivalent RSVpreF could avoid a high number of hospitalizations, emergency room visits and primary care visits, while generating savings for the NHS, being a more efficient strategy (dominant) when compared to a non-prevention strategy.