Policy-relevant estimates of oral rotavirus vaccine performance in low- and middle-income countries

Oral rotavirus vaccine efficacy is lower in low- and middle-income countries (LMICs) than in high-income countries. The degree to which antibiotic use impacts rotavirus vaccine immunogenicity in LMICs is unknown. Also, few studies have assessed whether efficacy estimates from rotavirus vaccine randomized controlled trials (RCTs) are applicable in real-world settings. To address these gaps, we examined the association between early life antibiotic use on immune response to oral rotavirus vaccines and generalized results from a monovalent rotavirus vaccine trial in Malawi to all trial-eligible children in Malawi. First, we estimated adjusted effect estimates for the association between early life antibiotic use and rotavirus vaccine immunogenicity in the MAL-ED birth cohort study using a modified Poisson regression. Using the parametric g-formula, we then estimated the impact of hypothetical interventions that treated all children and alternatively prevented inappropriate antibiotic treatments on vaccine seropositivity. Finally, we used a weight-based method to generalize results from a monovalent rotavirus vaccine trial in Malawi to a target population of trial-eligible children in the 2015-2016 Malawi Demographic and Health Survey. We found that the prevalence of seropositivity was higher among children who received at least one course of antibiotics than those with no antibiotic exposure. There was no significant difference by the number of antibiotic courses received or total duration of antibiotics. We also found that vaccine efficacy against severe and any-severity rotavirus gastroenteritis was higher after weighting to the target population, in part due to differences in nutritional status. These findings illustrate that further investigation into the underlying mechanisms of antibiotic use on the infant gut microbiome and immune response to rotavirus vaccines is needed to identify potential points of intervention. They also demonstrate that rotavirus vaccine efficacy may differ between RCT samples and target populations when these populations differ and there is treatment effect heterogeneity across population subgroups. Overall, findings from this dissertation can help public health officials in LMICs to plan more effective vaccine implementation programs and ensure that vaccines for rotavirus and other disease indications are maximally effective.Doctor of Philosoph