Joukje Willemsen

RSV is one of the leading causes of death among young children worldwide. Although most children experience only mild symptoms, the virus can lead to severe lower respiratory tract infections such as bronchiolitis or pneumonia. Annually, an estimated 100,000 children die from RSV, the vast majority (97%) of whom live in low- and middle-income countries (LMICs), and nearly half of these deaths occur in the first six months of life.

This thesis investigates the potential impact of maternal vaccination—where pregnant women are vaccinated to transfer antibodies to their unborn child via the placenta—on reducing RSV-related mortality in LMICs.

Part I focuses on the age distribution of children who die from RSV. In Chapter 2, we compare children who die in the hospital with those who die outside the hospital (in the community). We find that children in the community die at a younger age (median 1.5 months) than those in the hospital (median 2.4 months). In Chapter 3, we present the first detailed age distribution of RSV mortality specifically for Gavi-eligible countries. This shows that the burden of mortality is concentrated at a very young age, peaking around one month.

Part II describes the development of a mathematical model to predict the impact of vaccination. In Chapter 4, we use this model to calculate how many deaths in 51 Gavi countries can be prevented. It is estimated that maternal vaccination can prevent 55% to 63% of hospital deaths among children under six months of age. In Chapter 5, we show that small policy differences between the US FDA and the European EMA—such as limiting the vaccination window to 32-36 weeks of pregnancy to minimize the risk of preterm birth—can have major consequences. Such a restriction could lead to an estimated 3,300 additional deaths per year in LMICs.

Part III tests and refines the immunological assumptions in the maternal vaccination model, such as how quickly antibody levels decline in young children and how placental health affects the transfer of maternal antibodies. In Chapter 6, we investigate the half-life of antibodies in preterm and term-born children. Further analysis shows that the difference in protection between preterm and term-born children is likely smaller than previously thought. In Chapter 7, we examine whether placental abnormalities affect the transfer of antibodies from mother to child. We find no clear link.

Part IV reflects on how clinical research can be better designed so that results are more useful for impact models and policy decisions. Chapter 8 describes a meeting on the design of clinical trials for RSV prevention, which revealed that there is often a gap between evidence required for approval and evidence required for effective implementation in LMICs. In Chapter 9, we introduce a new statistical method to analyze the effectiveness of interventions with multiple endpoints.

In the General Discussion, we combine the key insights from this thesis and conclude that models combining improved disease burden estimates with immunologically informed protection and realistic implementation assumptions are essential to optimally shape the use of RSV vaccination and other prevention strategies.