Fig. 5

Complement-fixing antibodies against merozoite antigens are associated with protection from malaria. a Adjusted Hazard Ratios (aHR) for each antigen are plotted, ranked from highest to lowest based on analysis of samples and data from the Mugil cohort; the aHR for whole merozoites in also shown. aHRs were calculated using the Cox proportional hazards model, comparing high responders versus low responders, and were adjusted for confounders. Error bars indicate 95% confidence intervals. The red line indicates 0% protection (HR = 1.0). b Antigens were classified into tertiles according to weak (n = 7), intermediate (n = 7) and strong associations with protection (n = 6) from clinical malaria episodes; pie charts show the proportion of surface proteins or apical proteins associated with weak, intermediate or strong protection levels. Total n = 20 antigens; MSP1–42 was excluded since it is part of the same antigen as MSP1–19, and only one allele of MSP2 (3D7) was included. c The median breadth score of antibodies is shown stratified by age (< or >9 years of age, n = 91 or n = 115) or d PCR status at enrolment, PCR+ or PCR− (n = 67, n = 139). e shows the breadth score for all antigens tested for C1q fixation, respectively, stratified by the level of protection in the follow up period, where ‘protected’ is malaria-episode free during the follow up period (n = 116), ‘susceptible 1’ one malaria episode (n = 27) or ‘susceptible 2′ 2 or more malaria episodes (n = 54). P-values were determined by Mann-Whitney test (C + D) or Kruskal–Wallis test (E). Error bars represent the interquartile range