Two chemoattenuated PfSPZ malaria vaccines induce sterile hepatic immunity

Abstract

The global decline in malaria has stalled¹, emphasizing the need for vaccines that induce durable sterilizing immunity. Here we optimized regimens for chemoprophylaxis vaccination (CVac), for which aseptic, purified, cryopreserved, infectious Plasmodium falciparum sporozoites (PfSPZ) were inoculated under prophylactic cover with pyrimethamine (PYR) (Sanaria PfSPZ-CVac(PYR)) or chloroquine (CQ) (PfSPZ-CVac(CQ))-which kill liver-stage and blood-stage parasites, respectively-and we assessed vaccine efficacy against homologous (that is, the same strain as the vaccine) and heterologous (a different strain) controlled human malaria infection (CHMI) three months after immunization ( https://clinicaltrials.gov/ , NCT02511054 and NCT03083847). We report that a fourfold increase in the dose of PfSPZ-CVac(PYR) from 5.12 × 10⁴ to 2 × 10⁵ PfSPZs transformed a minimal vaccine efficacy (low dose, two out of nine (22.2%) participants protected against homologous CHMI), to a high-level vaccine efficacy with seven out of eight (87.5%) individuals protected against homologous and seven out of nine (77.8%) protected against heterologous CHMI. Increased protection was associated with Vδ2 γδ T cell and antibody responses. At the higher dose, PfSPZ-CVac(CQ) protected six out of six (100%) participants against heterologous CHMI three months after immunization. All homologous (four out of four) and heterologous (eight out of eight) infectivity control participants showed parasitaemia. PfSPZ-CVac(CQ) and PfSPZ-CVac(PYR) induced a durable, sterile vaccine efficacy against a heterologous South American strain of P. falciparum, which has a genome and predicted CD8 T cell immunome that differs more strongly from the African vaccine strain than other analysed African P. falciparum strains.

Extended Data Fig. 1 |. Trial profile for PfSPZ-CVac low-dose study.

^αOne participant was treated at day 16 after CHMI although that participant did complete the end-of-study visit. ^βThe serious adverse event (SAE) and adverse event (AE) in the PYR/CQ group were encephalopathy and eye irritation, respectively. The non-compliance incidents include a participant’s unwillingness to remain on pregnancy prevention (PYR/CQ group) and a work conflict preventing the participant from attending scheduled visits (CQ group).

Extended Data Fig. 2 |. Detection of parasitaemia in volunteers by qPCR after low-dose PfSPZ-CVac.

Parasitaemia was measured in participants on days 6–10 after each PfSPZ-CVac dose. The median parasitaemia is displayed for positive participants and error bars indicate the IQR. Vax, PfSPZ Challenge + CQ or PYR/CQ administration and follow-up until day 14 after PfSPZ Challenge 1 and day 10 after PfSPZ Challenge doses 2 and 3. The table shows (from left to right in each cell): the number of participants who were positive by qPCR/the injected number of participants; the median peak parasite density of positive participants (parasites per ml); and the average day of peak parasite density after each dose of PfSPZ-CVac. ND, not detected; n/a, not applicable.

Extended Data Fig. 3 |. Protective efficacy of PfSPZ-CVac(CQ) and PfSPZ-CVac(PYR + CQ) against homologous PfSPZ CHMI in the PfSPZ-CVac low-dose study.

Participants were followed for 27 days after CHMI, and a survival curve is presented displaying the number of participants that remained protected throughout follow-up after homologous CHMI. To assess vaccine efficacy, the unvaccinated group was compared with the vaccine group for time to first detectable parasitaemia after the final challenge using a log-rank test, and the presence or absence of parasitaemia using a two-tailed Barnard’s test. In the pilot PfSPZ-CVac(PYR + CQ) group (to test safety), 0/2 participants were protected; in the main PfSPZ-CVac(PYR+CQ) group, 2/9 (22.2%, P = 0.8) participants were protected; in the PfSPZ-CVac(CQ) group, 4/5 (80%, P = 0.048; 95% confidence interval, 1–99%) participants were protected.

Extended Data Fig. 4 |. Trial profile for the PfSPZ-CVac high-dose study.

^αOther, 15 withdrew consent, 6 deemed eligible after study was fully enrolled, 3 lost to follow-up, 2 schedule conflict, 1 withdrawn before randomization. ^βOne individual enrolled in the PYR group was withdrawn before receiving the first vaccination owing to acute illness, and enrolled in the control group. This person is counted twice. ^γBoth participants were withdrawn before receiving vaccination 1. ^δThe two serious adverse events in the CQ group were mental status change and pneumothorax.

Extended Data Fig. 5 |. Vδ2 γδ T cells at baseline in protected and infected individuals who received the vaccine.

Filled circles indicate vaccinated participants in the high-dose study, and open circles indicate vaccinated participants in the low-dose study. Median values are displayed and error bars indicate the IQR. For protected individuals in the high-dose study, n = 20; for protected individuals in the low-dose study, n = 6. For infected individuals in the high-dose study, n = 3; for infected individuals in the low-dose study, n = 8.

Extended Data Fig. 6 |. Anti-PfCSP IgM antibodies in vaccinated participants in the PfSPZ-CVac high-dose study analysed using ELISA.

IgM antibodies against PfCSP (net OD 1.0) 2 weeks after the third dose of PfSPZ-CVac(PYR) and PfSPZ-CVac(CQ); and immediately before CHMI. Median values are displayed and error bars indicate the IQR. At both time points, sample sizes for the PYR high-dose study, n = 14 for protected and n = 3 for infected vaccinated participants; for the CQ high-dose study, n = 6 for protected vaccinated individuals at 2 weeks after the third CVac and n = 5 for protected vaccinated individuals at before CHMI. No vaccinated participants were infected in the CQ high-dose study.

Extended Data Fig. 7 |. Antibody responses in vaccinated participants of the PfSPZ-CVac low-dose and PfSPZ-CVac high-dose studies analysed using automated immunofluorescence and automated inhibition of sporozoite invasion assays.

For each panel, filled circles are protected participants, open circles are infected participants who received homologous CHMI, filled triangles are uninfected (protected) participants and open triangles are infected participants who received heterologous CHMI. Median values are displayed and error bars indicate the IQR. P values were calculated using two-sided Wilcoxon–Mann–Whitney tests and group differences with P > 0.07 are not indicated. a, b, Antibodies against PfCSP by automated immunofluorescence assay were analysed 2 weeks after the third dose of PfSPZ-CVac(PYR) and PfSPZ-CVac(CQ) (a) and immediately before CHMI (b). c, d, Antibodies against PfCSP by automated inhibition of sporozoite invasion were analysed 2 weeks after the third dose of PfSPZ-CVac(PYR) and PfSPZ-CVac(CQ) (c) and immediately before CHMI (d). At both time points, sample sizes for the PYR low-dose study, n = 2 for protected and n = 7 for infected vaccinated participants; for the PYR high-dose study, n = 14 for protected and n = 3 for infected vaccinated participants; for the CQ low-dose study, n = 4 for protected and n = 1 for infected vaccinated participants; for the CQ high-dose study, n = 6 for protected participants in a and c (2 weeks after the third CVac), and n = 5 for protected participants in b and d (before CHMI). No vaccinated participants were infected in the CQ high-dose study. Anti-PfCSP antibodies at both 2 weeks after the last dose (high-dose 1,941 versus low-dose 382, P = 0.043) and immediately before CHMI (high-dose 1,149 versus low-dose 95, P = 0.009) were significantly higher in the high-dose PfSPZ-CVac(PYR) participants compared with the low-dose PfSPZ-CVac(PYR) participants. In PfPSZ-CVac(CQ) participants, the anti-PfSPZ antibodies were higher but not significant (2 weeks after the third dose, 300 versus 4,272, P = 0.176; before CHMI, 245 versus 910, P = 0.310). The inhibition of sporozoite invasion assay showed a statistically significant increase after vaccination for the high-dose PfSPZ-CVac(PYR) versus low-dose PfSPZ-CVac(PYR) when measured 2 weeks after the third dose (45.00 versus 3.81, P < 0.001) and immediately before CHMI (39.87 versus 1.00, P < 0.001). In PfPSZ-CVac(CQ) participants, the anti-PfCSP antibodies were significantly higher in the high-dose group when measured 2 weeks after the third dose (29.79 versus 8.79, P = 0.054) but not significantly different before CHMI (5.67 versus 7.32, P = 0.671).

Extended Data Fig. 8 |. Detection of parasitaemia by qPCR in each individual after the first, second and third dose of PfSPZ-CVac in the high-dose study.

Parasitaemia was measured in participants on days 6–10 after each PfSPZ-CVac dose. Vax, PfSPZ Challenge + CQ or PYR administration and follow-up until day 14 after PfSPZ Challenge 1 and day 10 after PfSPZ Challenge doses 2 and 3.

Extended Data Fig. 9 |. In vitro PYR activity against liver-stage parasites.

PYR was added to the cultures (blue) at three concentrations (1 μM, 10 μM and 100 μM), starting 2 days after infection and replaced with daily medium changes until day 4, at which point the cultures were fixed. The diameter of representative parasites of each group on day 4 are shown. Ctl, control. Statistical significance was determined by one-sided ANOVA. ***P < 0.001 (exact P = 0.0006); ****P < 0.0001; ns, not significant (exact P = 0.2335).

Extended Data Fig. 10 |. Flow cytometry gating strategy used to enumerate T cell subsets using flow cytometry ex vivo staining of whole blood.

The FSC/SSC parameters were used to define the lymphocyte gate as the first gate in the whole blood ex vivo assay. Within the lymphocyte gate, CD3-Alexa-700-positive events (T cell gate) were gated against CD56-PE-Cy7-positive events (NK cells). In the T cell gate, plotting Vδ2-FITC versus gdTCR-PE identified discrete Vδ2⁺, Vδ2⁻ and CD3⁺γδTCR⁻ populations. The T cell gate was used to further delineate CD4- and CD8-positive events.

Fig. 1 |. Parasitaemia detected by qPCR after the first, second and third dose of 2 × 10⁵ PfSPZ for the PfSPZ-CVac high-dose study and PYR activity against liver-stage parasites.

a, Median parasitaemia values and IQRs are shown for positive PfSPZ-CVac(CQ) participants. Dose 1, PfSPZ Challenge inoculation under CQ or PYR treatment cover with follow-up for 14 days; doses 2 and 3, PfSPZ Challenge inoculation under CQ or PYR treatment cover with follow-up for 10 days. The table shows (from left to right in each cell): the number of participants who were positive by qPCR/the number of injected participants; the median peak parasite density of positive participants (parasites per ml); and the mean day of peak parasite density (positive participants). NA, not applicable; ND, not detected. Individual data are shown in Extended Data Fig. 8b. b, c, PYR was added to cultures (1 μM, 10 μM or 100 μM) on days 2 and 3 after infection with daily medium changes until day 4, at which point cultures were fixed. b, On day 4, the remaining liver-stage parasites were counted. Data are mean ± s.e.m., n = 3 independent wells. Statistical significance was determined using a one-way ANOVA with post hoc pairwise comparisons. ****P < 0.0001. Parasite diameters are shown in Extended Data Fig. 9c. c, Representative images of P. falciparum from control (top) and PYR-treated (bottom left, 1 μM; bottom right, 10 μM) groups. Parasites (red) were identified by anti-HSP70. Scale bars, 5 μm.

Fig. 2 |. Vaccine efficacy of PfSPZ-CVac(PYR) against homologous and heterologous CHMI and PfSPZ-CVac(CQ) against heterologous CHMI, 3 months after the last immunization.

Survival curves display the percentage of study participants who remained protected throughout 27 days of follow-up after inoculation with a high dose of PfSPZ (2 × 10⁵). Vaccine efficacy compared the vaccinated to unvaccinated groups for the time to first detectable parasitaemia after inoculation using a log-rank test. In the PfSPZ-CVac(PYR) group, 7/8 participants (vaccine efficacy = 87.5%, P = 0.003; 95% confidence interval, 42.5–100%) were protected from homologous CHMI and 7/9 participants (vaccine efficacy = 78%, P = 0.001; 95% confidence interval, 39.8–100%) from heterologous CHMI. In the PfSPZ-CVac(CQ) group, 6/6 participants (vaccine efficacy = 100%, P = 0.001; 95% confidence interval, 54.1–100%) were protected from heterologous CHMI.

Fig. 3 |. Vδ2 γδ T cells related to protective immunity.

Vδ2 γδ T cells in vaccinated individuals stratified by protection status. Median fold changes and IQRs using unadjusted %Vδ2 values are shown. a, For the low-dose study, n = 4 PfSPZ-CVac(CQ) protected vaccinated participants; n = 1 PfSPZ-CVac(CQ) infected vaccinated participant; n = 2 PfSPZ-CVac(PYR + CQ) protected vaccinated participants; n = 7 PfSPZ-CVac(PYR + CQ) infected vaccinated participants. b, For the high-dose study, n = 6 PfSPZ-CVac(CQ) protected vaccinated participants; n = 14 PfSPZ-CVac(PYR) protected vaccinated participants; n = 3 PfSPZ-CVac(PYR) infected vaccinated participants. Black dashed lines indicate vaccination days; red dashed line indicates day of PfSPZ Challenge inoculation.

Fig. 4 |. Anti-PfCSP IgG levels by ELISA in vaccinated participants of the low-dose and high-dose studies.

Filled circles are protected participants and open circles are infected participants in homologous CHMI; filled triangles are protected participants and open triangles are infected participants in heterologous CHMI. Median values and IQRs are displayed. P values were calculated using two-sided Wilcoxon–Mann–Whitney tests and group differences with P > 0.07 are not indicated. a, b, IgG antibodies against PfCSP (net OD 1.0) 2 weeks (day 70) (a) or approximately 3 months (before CHMI) after the third PfSPZ-CVac dose (b). PYR low-dose study, n = 2 protected and n = 7 infected vaccinated participants; PYR high-dose study, n = 14 protected and n = 3 infected vaccinated participants; CQ low-dose study, n = 4 protected and n = 1 infected vaccinated participant; CQ high-dose study, n = 6 protected vaccinated participants in a (day 70) and n = 5 protected vaccinated participants in b (before CHMI). No vaccinated participants were infected in the CQ high-dose study.