RTS,S/AS01 malaria vaccine efficacy and its interaction with seasonal precipitation: Results from a phase 3 randomized controlled trial in Lilongwe, Malawi

Background: Phase 3 trials of the efficacy and safety of the candidate malaria vaccine RTS,S/AS01 are complete, but site-specific data are limited. This study assesses the interaction of precipitation and vaccine efficacy and corresponding estimates of clinical malaria episodes averted in a seasonal-transmission region of sub-Saharan Africa. Methods: We followed children (5-17 months of age) and infants (6-12 weeks of age) who had been randomly assigned to one of three groups: 1) a group that received 3 doses of RTS,S and a booster dose 18 months after the third dose (R3R group); 2) a group that received the same schedule without the booster (R3C group); 3) a control group –the rabies vaccine (VeroRab, Sanofi-Pasteur) was administered to children 5-17 months of age and the meningococcal serogroup C conjugate vaccine (Menjugate, Novartis) was administered to infants 6-12 weeks of age (C3C group). Primary efficacy was defined as development of clinical malaria (fever ≥37.5°C and Plasmodium falciparum parasitemia density >5,000 parasites per microliter) for both age groups. Patients were followed passively but were told to seek medical care at a designated Health Center for any illness. Precipitation data were obtained from the Chitedze Agricultural Center in Lilongwe, Malawi (-13.97S, 33.63E, 3770 ft). Findings: Over the duration of follow-up, 744 of 1513 (49.1%) children and infants had at least 1 episode of clinical malaria. Among children, vaccine efficacies were 42.7% (95% CI 25.7%, 55.8%, P<0.0001) for the R3R group and 33.1% (95% CI 14.5%, 47.7%, P=0.001) for the R3C group for first malaria case. Among infants, vaccine efficacies were 41.8% (95% CI 24.4%,53.7%, P<0.0001) for the R3R group and 29.9% (95% CI 11.3%, 44.6%, P=0.003) for the R3C group. Precipitation was significantly associated with increased malaria incidence, with each 1-inch increase in rainfall per month elevating the hazard of malaria by 12.6% (95% CI 9.6%, 15.6%) among children and by 15.9% (95% CI 12.8%, 18.9%) among infants. Vaccine efficacy was not affected by seasonal variations in precipitation. Interpretations: Policy discussions for the expanded use of RTS,S/AS01 are underway. It will be critical to determine the performance of the vaccine according to the seasonal pattern of malaria transmission. We show that vaccine efficacy does not change during the rainy or dry season in Lilongwe, Malawi, an area of mesoendemic, seasonal transmission.