Allele-specific efficacy of two malaria subunit vaccines following immunization with the polymorphic antigen apical membrane antigen-1 (AMA1)

Abstract

Background: Antigenic diversity of malaria parasites poses a major obstacle to the development of an effective malaria vaccine. Population genetics analyses suggest that polymorphic regions of the vaccine antigen apical membrane antigen 1 (AMA1) are targets for protective immune responses. Objectives: The study exploited a unique opportunity to assess strain-specific efficacy in clinical trials of two AMA1-based vaccines, with the ultimate goal of developing an AMA1 malaria vaccine that protects against genetically diverse parasites. Methods: We used data and samples collected during two randomized, double-blind, controlled phase 2 trials (Bancoumana AMA1-C1 and Bandiagara FMP2.1/AS02A vaccine trials) of malaria vaccine safety and efficacy to assess allele-specific in AMA1 vaccinees compared to control groups. In each study, we sequenced the Plasmodium falciparum ama1 gene from samples collected at baseline, during clinical episodes, at scheduled intervals during post-vaccination follow-up period, and whenever blood smears were collected for malaria diagnosis. Using a homology definition based on clusters c1, c1L, c2, c3 and domains 1, 2 and 3, we assessed the time to first malaria episode with an AMA1 sequence identical to that of the vaccine strains 3D7 or FVO type. In addition, we measured the hazard of having a clinical episode with the vaccine strain in the two treatment groups. Finally, genetic diversity parameters were assessed and compared between treatment arms. Results: The bivalent malaria vaccine showed no overall efficacy. AMA1 vaccine group and controls groups were found to have genetic parameters comparable at baseline and during the follow-up period. No difference was found between the 2 treatment arms regarding the time to first clinical episode with a 3D7 or FVO AMA1 c1L haplotype. In Bandiagara, the monovalent vaccine has a 20% overall efficacy against malaria first clinical episodes. The time to first clinical episode with a 3D7 c1L allele was longer in the AMA1 vaccine group compared to the rabies vaccine, Likelihood Ratio p-value respectively equal to 0.025 (ITT) and 0.018 (PP). The vaccine efficacy against c1L allele was 64% (ITT) and 72% (PP). Conclusion: It may be possible to design a polyvalent or chimeric AMA1 vaccine that provides broader cross-protection.