The effect of intermittent preventive treatment for malaria with dihydroartemisinin-piperaquine on vaccine-specific responses among schoolchildren in rural Uganda (POPVAC B): a double-blind, randomised controlled trial

Abstract

Background: Several important vaccines differ in immunogenicity and efficacy between populations. We hypothesised that malaria suppresses responses to unrelated vaccines and that this effect can be reversed-at least partially-by monthly malaria intermittent preventive treatment (IPT) in high-transmission settings.

Methods: We conducted an individually randomised, double-blind, placebo-controlled trial of the effect of malaria IPT with dihydroartemisinin-piperaquine on vaccine responses among schoolchildren aged 9-17 years in Jinja district, Uganda. Participants were recruited from two schools and did not have exposure to vaccines of interest after the age of 5 years, with the exception of human papillomavirus (HPV). Computer-generated 1:1 randomisation was implemented in REDCap. 3-day courses of dihydroartemisinin-piperaquine (dosage by weight) or placebo were administered monthly, including twice before the first vaccination. Trial participants were vaccinated with the live parenteral BCG vaccine (Serum Institute of India, Pune, India) at week 0; yellow fever vaccine (YF-17D; Sanofi Pasteur, Lyon, France); live oral typhoid vaccine (Ty21a; PaxVax, London, UK), and quadrivalent virus-like particle HPV vaccine (Merck, Rahway, NJ, USA) at week 4; and toxoid vaccines (tetanus-diphtheria; Serum Institute of India) and an HPV booster at week 28. An additional HPV vaccination at week 8 was provided to female participants older than 14 years who had not previously been vaccinated, and a tetanus-diphtheria booster was given after completion of the trial at week 52. Primary outcomes were vaccine responses at week 8 and, for tetanus-diphtheria, at week 52, and analysis was done in the intention-to-treat population. Malaria parasite prevalence at enrolment and during follow-up was determined retrospectively by PCR. The safety population comprised all randomly allocated participants. The trial was registered at the ISRCTN Registry (ISRCTN62041885) and is complete.

Findings: Between May 25 and July 14, 2021, we assessed 388 potential participants for eligibility. We enrolled and randomly allocated 341 participants to the two groups (170 [50%] to dihydroartemisinin-piperaquine and 171 [50%] to placebo); 192 (56%) were female and 149 (44%) participants were male. 145 (85%) participants in the dihydroartemisinin-piperaquine group and 140 participants (82%) in the placebo group were followed up until the week 52 endpoint. At enrolment, 109 (64%) of all participants in the dihydroartemisinin-piperaquine group and 99 (58%) of 170 participants in the placebo group had malaria; this reduced to 6% or lower at all follow-up visits in the dihydroartemisinin-piperaquine group. There was no effect of dihydroartemisinin-piperaquine versus placebo on primary outcomes: BCG-specific IFNγ ELISpot response had a geometric mean ratio (GMR) of 1·09 (95% CI 0·93-1·29), p=0·28; yellow fever neutralising antibody was 1·19 (0·91-1·54), p=0·20 for plaque reduction neutralising reference tests (PRNT₅₀) titres (the reciprocal of the last plasma dilution that reduced by 50%) and 1·24 (0·97-1·58), p=0·09 for PRNT₉₀ titres (reciprocal of the last plasma dilution that reduced by 90%); and IgG to Salmonella enterica serovar Typhi O-lipopolysaccharide was 1·09 (0·81-1·46), p=0·58, HPV-16 was 0·72 (0·44-1·77), p=0·19, HPV-18 was 0·71 (0·47-1·09), p=0·11; tetanus toxoid was 1·22 (0·91-1·62), p=0·18, and diphtheria toxoid was 0·97 (0·83-1·13), p=0·72. There was some evidence that dihydroartemisinin-piperaquine reduced waning of the yellow fever response.

Interpretation: IPT for malaria with dihydroartemisinin-piperaquine did not improve peak vaccine responses, despite reducing malaria prevalence. Possible longer-term effects on response waning should be further explored.

Funding: UK Medical Research Council.

Translation: For the Luganda translation of the abstract see Supplementary Materials section.

Figure 1

Trial schedule Samples were collected before vaccinations or anthelminthic treatment at relevant timepoints. D–P=dihydroartemisinin–piperaquine. HPV=virus-like particle human papillomavirus vaccine. T–D=tetanus–diphtheria vaccine. YF-17D=yellow fever vaccine. Ty21a=live oral typhoid vaccine. *Primary endpoint following BCG, YF-17D, Ty21a, and HPV vaccination; additionally, an HPV dose was given to previously unvaccinated girls aged 14 years or older. †Secondary endpoint following BCG, YF-17D, Ty21a, and HPV vaccination. ‡A T–D boost was given to comply with the Uganda National Expanded Program on Immunization guidelines.

Figure 2

Trial profile D–P=dihydroartemisinin–piperaquine. YF-17D=yellow fever vaccine. Ty21a=live oral typhoid vaccine. HPV=virus-like particle human papillomavirus vaccine. PBMCs=peripheral blood mononuclear cells. SFU=spot-forming units. *Some participants had multiple reasons for exclusion †Reasons for not being included in the final analysis included no sample at the primary endpoint, or the sample was available, but the participant did not receive the correctly allocated vaccine. ‡BCG INFγ gamma ELISpot assay—samples that had more than 83·3 SFUs per a million PBMCs in the unstimulated well were considered invalid and not included in the final analysis. §In the D–P group, 25 participants did not complete follow-up, while 31 did not complete follow-up in the placebo group, totalling 56 participants. The reasons for withdrawal were being lost to follow-up (n=31), became pregnant during follow-up (n=1) and withdrawal of consent (n=24). ¶Reasons for not being included in the final analysis were no sample at week 52, or the sample was available but participant did not receive a tetanus–diphtheria vaccine at week 28.

Figure 3

Malaria infection status by visit Malaria prevalence at each timepoint was compared between trial groups using χ² tests. Linear mixed models were fitted to compare malaria prevalence (p<0·0001) between trial groups, from week –6 to week 52. The numbers of participants with PCR results available at each timepoint were 340 [99%] of 342 at week –6; 303 [100%] of 303 at week 0; 305 [100%] of 305 at week 4, 300 [100%] of 300 at week 8, 294 [100%] of 295 at week 28, and 284 [100%] of 284 at week 52. D–P=dihydroartemisinin–piperaquine.