Peripheral and Placental Prevalence of Sulfadoxine–Pyrimethamine Resistance Markers in Plasmodium falciparum among Pregnant Women in Southern Province, Rwanda

ABSTRACT. Intermittent preventive therapy during pregnancy (IPTp) with sulfadoxine–pyrimethamine (SP) is recommended in areas of moderate to high malaria transmission intensity. As a result of the increasing prevalence of SP resistance markers, IPTp-SP was withdrawn from Rwanda in 2008. Nonetheless, more recent findings suggest that SP may improve birthweight even in the face of parasite resistance, through alternative mechanisms that are independent of antimalarial effects. The prevalence of single nucleotide polymorphisms in Plasmodium falciparum dihydropteroate synthase (pfdhps) and dihydrofolate reductase (pfdhfr) genes associated with SP resistance among 148 pregnant women from 2016 to 2018 within Rwanda’s Southern Province (Huye and Kamonyi districts) was measured using a ligase detection reaction–fluorescent microsphere assay. The frequency of pfdhps K540E, A581G, and the quintuple (pfdhfr N51I + C59R + S108N/pfdhps A437G + K540E) and sextuple (pfdhfr N51I + C59R + S108N/pfdhps A437G + K540E + A581G) mutant genotypes was 90%, 38%, 75%, and 28%, respectively. No significant genotype difference was seen between the two districts, which are approximately 50 km apart. Observed agreements for matched peripheral to placental blood were reported and found to be 207 of 208 (99%) for pfdhfr and 239 of 260 (92%) for pfdhps. The peripheral blood sample did not miss any pfdhfr drug-resistant mutants or pfdhps except at the S436 loci. At this level of the sextuple mutant, the antimalarial efficacy of SP for preventing low birthweight is reduced, although overall SP still exerts a nonmalarial benefit during pregnancy. This study further reveals the need to intensify preventive measures to sustain malaria control in Rwanda to keep the overall incidence of malaria during pregnancy low.


INTRODUCTION
An estimated 247 million clinical cases and 619,000 malaria-related deaths were reported worldwide in 2021. 1 Plasmodium falciparum is the most prevalent and virulent of the human malaria species. 1 Antimalarial drug resistance in P. falciparum parasites poses one of the greatest threats to malaria control.As a result of the widespread resistance to chloroquine (CQ) and sulfadoxine-pyrimethamine (SP), artemisinin-based combination therapies have been recommended for P. falciparum malaria treatment since 2006. 2 Sulfadoxine-pyrimethamine plus amodiaquine (AQ) was used as a first-line antimalarial treatment of uncomplicated malaria in Rwanda from 2001 to 2006, at which time artemether-lumefantrine (AL) became the recommended first-line treatment. 2Resistance to SP is associated with polymorphisms in the P. falciparum dihydrofolate reductase (pfdhfr) and dihydropteroate synthase (pfdhps) genes. 3,4ntimalarial drug policy change may influence drug sensitivity as a result of the withdrawal of drug pressure on parasites.This was demonstrated in a 2003 study 5 in Malawi where, 12 years after CQ withdrawal, no evidence of the P. falciparum CQ resistance transporter (pfcrt K76T) mutation was detected among isolates, compared with 85% reported in 1992, indicating that CQ efficacy may have been restored.In Rwanda, a slower recovery rate of CQ sensitivity has been reported after 14 years of cessation. 2However, despite 7 years of the presumed absence of SP pressure, a high level of SP resistance was reported in samples collected from Rwanda in 2015, 2 which persisted in 2020. 6ulfadoxine-pyrimethamine remains the only drug recommended by the WHO for intermittent preventive treatment during pregnancy (IPTp) to prevent the adverse consequences of malaria.Intermittent preventive treatment during pregnancy-SP is a recommended intervention in 35 sub-Saharan African countries. 1,7lthough the quintuple mutant genotype, consisting of triple mutations of pfdhfr (N51I/S59R/S108N) and double mutations of pfdhps (A437G/K540E), is associated with clinical and parasitological SP treatment failure, IPTp-SP remains efficacious and confers some protection, particularly against low birthweight, even in areas with a high level of quintuple mutant genotypes, 8 presumably because of the nonantimalarial effects of SP. 9,10 However, IPTp-SP efficacy for preventing low birthweight is reduced when a high proportion of P. falciparum parasites acquire an additional pfdhps A581G mutation, leading to the sextuple mutant ( pfdhfr N51I 1 C59R 1 S108N/pfdhps A437G 1 K540E 1 A581G). 8,11s a result of the documented high level of resistance, SP was withdrawn completely from all uses in Rwanda 2008, with cessation of the IPTp policy. 2 Rwanda experienced a dramatic increase in malaria cases between 2012 and 2017, from 564,407 in 2012 to 4,746,985 in 2017-a more than an 8-fold increase. 12More than 79% of the malaria burden was reported from the eastern and southern provinces.Because it is not ethically appropriate to conduct an SP efficacy trial in regions with known high rates of treatment failure, resistance marker detection may be the only evidence to support the notion of reintroducing drugs after their withdrawal.We report SP resistance markers in P. falciparum isolates from pregnant women in a relatively high malaria transmission zone in Rwanda, collected during a cluster randomized controlled trial investigating intermittent malaria screening and treatment of positive patients during pregnancy, compared with routine antenatal care for reducing malaria prevalence at delivery.We also compared samples obtained from placental or peripheral blood to determine whether peripheral blood captured drug-resistant genotypes accurately.

MATERIALS AND METHODS
Study area and study participants.The study was conducted in two districts (Huye and Kamonyi), both of which are in the Southern Province of Rwanda (Figure 1).The Southern Province encompasses an area of 5,963 km 2 with a population of about 2.6 million.Although malaria cases have declined in recent years, 1.8 million cases were reported in Rwanda in 2020 compared with 4.7 million cases in 2017.The Southern Province has the highest incidence in the country, with more than 450 cases per 1,000 people, compared with the national average of 198 per 1,000 people. 12In our study, we used clinical samples collected from pregnant women attending an antenatal care clinic in Rwanda between 2016 and 2018, and participating in a cluster-level randomized study to estimate the effectiveness of intermittent screening and treatment of malaria in pregnancy on maternal and birth outcomes (ClinicalTrials.govidentifier, NCT03508349). 13A total of 1,441 pregnant women participated in the study, which included rapid diagnostic testing locally.At the time of delivery, placental and peripheral blood samples for polymerase chain reaction (PCR) testing at later time were collected using a filter paper card. 13All available 1,267 placental or 1,347 peripheral samples were processed for PCR nucleic acid testing, with the 192 placental and 140 peripheral positive P. falciparum samples processed for further drug-resistant testing.Genotyping drug-resistant genes.We genotyped the P. falciparum dihydrofolate reductase (pfdhfr) and dihydropteroate synthase (pfdhps) genes that confer resistance to pyrimethamine and sulfadoxine, respectively, using placental samples.Peripheral samples were compared with placental samples in matched pairs of PCR amplification from both sample locations.These genes were amplified by nested PCR and then a ligase detection reaction-fluorescent microsphere assay was performed to identify single nucleotide polymorphisms (SNPs) in the pfdhfr and pfdhps genes as described previously. 14We targeted four SNPs within the pfdhfr gene (N51I, C59R, S108N, and I164L) and five SNPs within the pfdhps gene (S436A, A437G, K540E, A581G, and A613S).
Statistical analysis.The prevalence of each SNP was calculated by dividing the number of samples carrying the SNP by the total number of samples genotyped successfully at that SNP.Each isolate was coded based on the presence or absence of a mutant allele, for which three categories were considered: wildtype, mutant, and mixed infection.In the final analysis, both mutant and mixed infection were coded as mutant alleles to generate the number of mutant alleles per codon.All statistical analyses were conducted using IBM SPSS Statistics 28.0 (SPSS Inc., Chicago, IL).
Kappa statistics were used to determine the degree of agreement between placental and peripheral matched samples for each locus using the formula k 5 P 0 2 P e 1 2 P e , where P 0 represents the overall proportion of observed agreement and P e represents the overall proportion of expected agreement.

RESULTS
From the original study, placental samples were collected and analyzed successfully by reverse transcription-PCR from 1,267 women.The prevalence of placental infection at delivery by PCR from placental blood was 15% (192 of 1,267).whereas fingerstick peripheral prevalence was 10% (140 of 1,347).A total of 148 of 192 placental blood samples were genotyped successfully for pfdhfr and pfdhps genes, and 58 were matched to fully genotyped peripheral fingerstick blood.
pfdhfr gene.The prevalence of mutant alleles (pure mutant and mixed) in placental samples for N51I, C59R, and S108N was 97%, 93%, and 99%, respectively (Table 1).All samples were found to carry the wildtype allele at pfdhfr I164L.The prevalence of the pfdhfr triple (51 1 59 1 108) haplotype was 89% in the Huye District and 90% in the Kamonyi District; however, the distribution of haplotype combinations was comparable across the two sites (Table 2).

DISCUSSION
Despite transitioning from SP 1 AQ to AL as the first-line treatment in 2006, and discontinuing IPTp-SP in 2008, there continued to be a relatively high proportion of both quintuple (75%) and sextuple (28%) mutants in samples collected in southern Rwanda from 2016 to 2018, despite the limitation of collection 5 to 7 years ago. 2 Overall, the prevalence of pfdhps A581G was 38%.Data from neighboring Uganda highlight the focal nature of the A581G mutant, as prevalence varied from 5% to 46% in 2018 and 2019, depending on the province. 15Although the prevalence of the quintuple mutant in our study, conducted in 2016 to 2018, is similar to that reported from Rwanda's Ruhuha (Eastern Province) and Mubuga (Western Province) in 2015 (74%), the prevalence of the sextuple mutant genotype in our sample was greater (28% compared with 18%), with a corresponding greater prevalence of pfdhps A581G (from 24% in the 2015 study to 38% in our sample). 2Because resistance markers may be highly focal, the geographic difference in study sites may account for the difference in the prevalence between our findings and the previous study from Rwanda. 2 Thus, it is not possible to state that this is definitive evidence of increasing resistance over time.However, older data from the  Democratic Republic of the Congo, which shares a border with Rwanda, demonstrated an increase in pfdhps K540E/ A581G (not defined as a sextuple mutant) from 2% in 2007 to 18% in 2013, suggesting that the increase in Rwanda may be real. 16The sextuple mutant has been associated with increased treatment failure, as well as delivery of lowbirthweight infants among women taking IPTp with SP. 17,18 A recent, large clinical trial with nearly 1,500 participants in each arm compared SP to dihydroartemisinin-piperaquine alone or combined with a single dose of azithromycin in settings with 20% to 30% prevalence of the sextuple mutant.Despite the relatively high prevalence of the sextuple mutant, adverse pregnancy outcomes were reported more frequently in the groups receiving dihydroartemisinin-piperaquine. 10 Recovery of wildtype (susceptible) alleles after withdrawal of antimalarial drugs has been reported in several countries, although timelines have varied. 2,5For instance, high rates of recovery of CQ susceptibility have been reported in Malawi, where parasites reverted to wildtype at the pfcrt K76T codon responsible for CQ resistance 12 years after cessation of CQ use. 5 In Rwanda, a slower recovery rate has been reported, with the prevalence of mutant pfcrt K76T declining from 74% in 2010 to 49% in 2015, suggesting an approximate 25% recovery of wildtype pfcrt K76T after a decade of CQ withdrawal. 2,19][22] The persistence of high levels of SP resistance in Rwanda even in the absence of SP pressure for almost a decade may be a result of gene flow from neighboring countries or the continued use of SP or other antifolate inhibitors such trimethoprim and sulfamethoxazole, which may have crossresistance with SP. 23 Alternatively, the SP mutations may not have the same fitness cost as those for CQ, and there may thus be less pressure to revert.
We found a high level of agreement between placental and peripheral blood SP drug-resistant mutants.Importantly, testing of peripheral blood did not miss mutant SP P. falciparum parasites for pfdhfr and all but one of the pfdhps genes, suggesting that maternal peripheral blood would be sufficient and provide a good measurement for the prevalence of SP resistance markers in pregnant women.A small study in Gabon 24 saw greater than 90% agreement in SP drugresistant mutations from 22 matched samples collected in 2005 and 2006.A similar study in Ghana 25 with 294 matched pairs collected in 2000 and 2001 indicated 83% agreement between peripheral and placental blood samples that were influenced by parasite density, number of mutations, and fever.The agreement increases inversely with the number of alleles, such that the multiallelic MSP2 indicated less agreement between placental and peripheral samples. 26

CONCLUSION
Our study demonstrated a high level of focal SP resistance, with a 23% and 30% prevalence of the sextuple mutant in the Huye and Kamonyi districts, respectively, despite nearly a decade of a presumed absence of SP pressure.At this level of resistance, the efficacy of SP for prevention of low birthweight is likely compromised.As drug resistance can be quite focal, resistance testing would be indicated prior to implementing any SP-containing preventive strategy in Rwanda.

FIGURE 1 .
FIGURE 1. Map of Rwanda showing the study sites of the Huye and Kamonyi districts in Southern Province, Rwanda.

TABLE 1
The prevalence of single nucleotide polymorphisms in the pfdhps and pfdhfr genes in Southern Province, Rwanda, (N 5 148)

TABLE 2
Percentage of mutant (mixed and mutant) alleles

TABLE 3
Placental and peripheral blood agreement on drug resistance genotyping