ChAd63-MVA-vectored blood-stage malaria vaccines targeting MSP1 and AMA1: assessment of efficacy against mosquito bite challenge in humans

Abstract

The induction of cellular immunity, in conjunction with antibodies, may be essential for vaccines to protect against blood-stage infection with the human malaria parasite Plasmodium falciparum. We have shown that prime-boost delivery of P. falciparum blood-stage antigens by chimpanzee adenovirus 63 (ChAd63) followed by the attenuated orthopoxvirus MVA is safe and immunogenic in healthy adults. Here, we report on vaccine efficacy against controlled human malaria infection delivered by mosquito bites. The blood-stage malaria vaccines were administered alone, or together (MSP1+AMA1), or with a pre-erythrocytic malaria vaccine candidate (MSP1+ME-TRAP). In this first human use of coadministered ChAd63-MVA regimes, we demonstrate immune interference whereby responses against merozoite surface protein 1 (MSP1) are dominant over apical membrane antigen 1 (AMA1) and ME-TRAP. We also show that induction of strong cellular immunity against MSP1 and AMA1 is safe, but does not impact on parasite growth rates in the blood. In a subset of vaccinated volunteers, a delay in time to diagnosis was observed and sterilizing protection was observed in one volunteer coimmunized with MSP1+AMA1-results consistent with vaccine-induced pre-erythrocytic, rather than blood-stage, immunity. These data call into question the utility of T cell-inducing blood-stage malaria vaccines and suggest that the focus should remain on high-titer antibody induction against susceptible antigen targets.

Figure 1

Pilot safety Phase IIa study VAC037: efficacy of ChAd63-MVA MSP1 immunization following Plasmodium falciparum 3D7 strain sporozoite challenge. (a) Kaplan–Meier survival analysis of time to patent parasitemia in days (calculated from hours between mosquito bite and diagnosis) for the VAC037 pilot challenge study: MSP1 vaccinated volunteers (n = 3) versus unvaccinated controls (n = 6). Median time to patent parasitemia = 13.9 days for vaccinees versus 11.4 days for unvaccinated controls (P = 0.035, log-rank test). (b) Geomean quantitative PCR data for each group. Data for each individual volunteer are shown in Supplementary Figure S2a and Supplementary Table S2. The lower limit of detection is indicated by the dotted line at 20 parasites/ml. ChAd63, chimpanzee adenovirus 63; MSP1, merozoite surface protein 1.

Figure 2

VAC039 flow chart of study design and volunteer recruitment. Eleven volunteers were excluded following screening for the following reasons: proteinuria, hypertension (two individuals), unexplained “jerking” episode, past medical history of Hodgkin's lymphoma, history of recreational drug use (three individuals), family history of sudden cardiac death, difficult venous access, and lack of response to medical screening letter by general practitioner. Recruitment complete = eligible volunteers not enrolled in the trial once enrollment was complete. All immunizations were administered intramuscularly. In Groups 1 and 2, ChAd63- and MVA-vectored vaccines were administered into separate arms. In Groups 3 and 4, ChAd63-vectored vaccines were coadministered, one vaccine in each deltoid. MVA-vectored vaccines were also coadministered in separate arms (but the same arm as the corresponding ChAd63 vaccine). One volunteer in Group 1 was withdrawn from the study 10 days after MVA MSP1. One volunteer in Group 1 was withdrawn from the study 6 days after CHMI. ChAd63, chimpanzee adenovirus 63; MSP1, merozoite surface protein 1; pfu, plaque-forming unit; vp, vector particle.

Figure 3

Systemic AEs deemed definitely, probably or possibly related to coadministration of ChAd63- or MVA-vectored vaccines (Groups 3 and 4). Only the highest intensity of each AE per subject is listed. Data are combined for all AEs for all volunteers receiving the same vaccine at the stated dose. There were no immunization-related serious AEs. (a) Systemic AEs after coadministration of ChAd63 MSP1 and ChAd63 AMA1 (Group 3) and after coadministration of ChAd63 MSP1 and ChAd63 ME-TRAP (Group 4). “Other” systemic AEs in Group 3 were mild loss of appetite and mild tender cervical lymphadenopathy. “Other” systemic AEs in Group 4 were moderate dizziness and mild exacerbation of pre-existing psoriasis, mild abdominal pain, mild dizziness, and mild loss of appetite. (b) Systemic AEs after coadministration of MVA MSP1 and MVA MSP1 (Group 3) and MVA MSP1 and MVA ME-TRAP (Group 4). “Other” systemic AEs in Group 4 were mild rhinitis and moderate light-headedness. AE, adverse event; AMA1, apical membrane antigen 1; ChAd63, chimpanzee adenovirus 63; MSP1, merozoite surface protein 1.

Figure 4

Cellular immunogenicity of ChAd63-MVA immunization regimens. (a) Median ex-vivo IFN-γ ELISPOT responses (summed response across all the individual peptide pools) in PBMCs are shown for each relevant Group to the MSP1, AMA1, ME-TRAP antigens. The total response (summed response to transgene inserts for Groups 3 and 4) is also shown. (b) Individual and median IFN-γ ELISPOT responses are shown for each antigen and each relevant Group at the day 14, day 63, and dC-1 timepoints. *P < 0.05, **P < 0.01, ***P < 0.001 by Mann–Whitney test (AMA1) and Kruskal–Wallis test with Dunn's multiple comparison test (MSP1 and total). Group 5 controls were excluded from the dC-1 analysis. (c) PBMCs from dC-1 were restimulated with separate pools of antigen-specific peptides and assayed by ICS. Box and whisker plots show the range, median, and interquartile range for the % CD4⁺ and CD8⁺ T cells positive for CD107a, IFN-γ, IL-2 or TNFα. Significance testing as in b. AMA1, apical membrane antigen 1; ChAd63, chimpanzee adenovirus 63; ICS, intracellular cytokine staining; IFN, interferon; IL, interleukin; MSP1, merozoite surface protein 1; PBMC, peripheral blood mononuclear cell; SFU, spot-forming unit; TNF, tumor necrosis factor.

Figure 5

IgG antibody responses and functional GIA induced by ChAd63-MVA immunization regimens. Geomean serum IgG ELISA responses are shown for all Groups to the 3D7 allele of the (a) MSP1₁₉ and (b) AMA1 antigens. (c) Individual and geomean responses are shown for each antigen and each relevant Group at day 28, day 56, and day C-1 timepoints. **P < 0.01 by Mann–Whitney test (AMA1) and Kruskal–Wallis test with Dunn's multiple comparison test (MSP1). Group 5 and Groups not vaccinated with the antigen were excluded from the dC-1 (day before challenge) analysis. (d) In vitro GIA of purified IgG was assessed at 10 mg/ml. Individual data and medians are shown for each group at the dC-1 timepoint. Pre-immunization (d0) sera were also pooled and the GIA tested for each of the four vaccinated groups. Responses >20% are typically regarded as positive. AMA1, apical membrane antigen 1; ChAd63, chimpanzee adenovirus 63; ELISA, enzyme-linked immunosorbent assay; GIA, growth inhibitory activity; MSP1, merozoite surface protein 1.

Figure 6

Efficacy of ChAd63-MVA immunization following Plasmodium falciparum 3D7 strain sporozoite challenge. Kaplan–Meier survival analysis of time to patent parasitemia in days (calculated from hours between mosquito bite and diagnosis) for the VAC039 challenge study: MSP1 (n = 8), AMA1 (n = 9), MSP1+AMA1 (n = 9), MSP1+ME-TRAP (n = 10) vaccinated volunteers versus the unvaccinated controls (n = 6). Log-rank test P values are shown. Median time to patent parasitemia = 10.8 days for MSP1, 10.8 days for AMA1, 11.0 days for MSP1+AMA1, 10.8 days for MSP1+ME-TRAP, and 9.7 days for controls. AMA1, apical membrane antigen 1; ChAd63, chimpanzee adenovirus 63; MSP1, merozoite surface protein 1.

Figure 7

Sine-wave modelling of qPCR data. (a) Liver-to-blood inocula (LBI) and (b) parasite multiplication rates (PMR) as calculated using the model based upon a sine-wave. Values are indicated for both Phase IIa challenge studies (VAC039 to the left of the dotted line, and VAC037 to the right). Individual data points with median are shown for each immunization group and infectivity controls. The one volunteer who was sterilely protected (LBI and PMR = 0) in the MSP1+AMA1 immunization group is indicated by the half-filled triangle symbol. An arbitrary threshold line is indicated in a with the sterilely protected volunteer plotted below this. (c) Geomean qPCR data for each group in the VAC039 challenge study (until day 11). Data for each individual volunteer are shown in Supplementary Figure S6 and Supplementary Table S3. The lower limit of detection is indicated by the dotted line at 20 parasites/ml. The relationships between time to diagnosis and (d) LBI and (e) PMR estimated by the sine-wave model are shown along with Spearman's rank correlation coefficient and P value. AMA1, apical membrane antigen 1; MSP1, merozoite surface protein 1; qPCR, quantitative PCR.

Figure 8

Analysis of CHMI clinical data. (a) Comparison of the number of symptomatic days before malaria diagnosis between vaccinees who underwent CHMI and were diagnosed with malaria (n = 35) and unvaccinated controls (n = 6) (P = 0.92, Mann–Whitney test). Number of symptomatic days pre-diagnosis in vaccinees: mean 0.94 days, median 1.00 days. Number of symptomatic days pre-diagnosis in controls: mean 1.42 days, median 0.50 days. Median values for each group are indicated on the figure. (b) Comparison of the number of symptoms present at diagnosis between vaccinees who underwent CHMI and were diagnosed with malaria (n = 35) and unvaccinated controls (n = 6) (P = 0.22, Mann–Whitney test). Number of symptoms at diagnosis in vaccinees: mean 4.28, median 4.00. Number of symptoms at diagnosis in controls: mean 6.33, median 8.50. Median values for each group are indicated on the figure. (c) Comparison of duration of symptoms of malaria post-diagnosis between vaccinees who underwent CHMI and were diagnosed with malaria (n = 35) and unvaccinated controls (n = 6) (P = 0.58, Mann–Whitney test). Duration of symptoms post-diagnosis in vaccinees: mean 5.51, median 5.00. Duration of symptoms post-diagnosis in controls: mean 4.50, median 3.50. Median values for each group are indicated on the figure. (d) Comparison of maximum severity of any symptom of malaria infection between vaccinees diagnosed with malaria (n = 35) and unvaccinated controls (n = 6). (e) Laboratory AEs after CHMI deemed possibly, probably or definitely related to Plasmodium falciparum infection. For “any laboratory abnormality” only the highest intensity AE per subject is counted. AE, adverse event; ALT, alanine transaminase; CHMI, controlled human malaria infection.