Home » Projects


Malaria chemoprevention in children with sickle cell anemia in Western Kenya

Sickle cell anemia (SCA) afflicts over 240,000 newborns in sub-Saharan Africa annually. These children are at high risk of death owing, in part, to their high risk for falciparum malaria. Currently, daily malaria prophylaxis with Proguanil is recommended for children with SCA, but little evidence exists to support this preventive strategy. Intermittent preventive therapy against malaria, which consists of scheduled administration of antimalarials, is recommended for other high risk groups in Africa, including pregnant women and infants, based upon large, comparative efficacy trials. These trials have not been conducted in the expanding population of African children with SCA.

Hospital and community partners in Homa Bay, Kenya, September 2017.

With partners from Moi University and DGHI, we are conducting the EPiTOMISE study: Enhancing Preventive Therapy Of Malaria in children with Sickle cell anemia in East Africa. This is a randomized, open-label trial of three malaria chemoprevention regimens in young children with sickle cell anemia in Western Kenya. We are testing the comparative efficacies of monthly sulfadoxine-pyrimethamine/amodiaquine and monthly dihydroartemisinin-piperaquine with that of daily proguanil to prevent malaria in these children. In addition, we are testing the relative efficacies of these antimalarial regimens to prevent SCA-related morbidity.  For more details see our DGHI coverage.

Shared research clinic with partners in Western Kenya Clinical Research Group on the grounds of Homa Bay County Hospital.

This study is registered with clinicaltrials.gov as NCT03178643 and with the Pan African Clinical Trials Registry PACTR201707002371165.

EPiTOMISE began enrolling participants in January 2018 in Homa Bay and concluded followup in December 2020. We expect results in early 2022.

Partners in this project include:

moiuniversitylogo   dghilogourllogo-large


This project is funded by an R01 award from NHLBI.



Mechanisms of protection from severe falciparum malaria in sickle-trait

Figure. Possible mechanisms by which sickle-trait can neutralize P. falciparum parasites organize broadly into four categories: A. reduction of parasite intraerythrocytic maturation and growth, B. attenuation of pathogenic phenotypes which result from extracellular binding, C. modulation of the balance of innate immune responses to blood stage parasites, and D. disruption of the acquisition of adaptive immunity.

Sickle-trait hemoglobin confers dramatic and consistent protection from falciparum malaria: in clinical studies, heterozygous hemoglobin S (HbAS, or sickle-trait) reduces the risk of severe, life-threatening falciparum by 91% in African children. The mechanisms of this protection remain obscure. In molecular investigations, sickle-trait is associated with attenuated cytoadherence of infected red blood cells (iRBCs) to microvascular endothelium and aberrant trafficking of parasite-derived proteins to the iRBC surface. The mechanisms by which sickle-trait disrupts these parasite phenotypes and thereby protects children from life-threatening falciparum malaria remain incompletely understood, and may include elements from four broad mechanisms: attenuation of parasite growth, disruption of extracellular binding, modulation of innate immune responses, and attenuation of adaptive immunity (see Figure).

We have observed that parasites growing in sickle-trait cells fail almost completely to bind to the host receptor EPCR, an interaction which is closely linked to severe malaria. Interestingly, sickle-trait does not prevent the transcription of mRNA transcripts encoding PfEMP1 in general or specifically those variants that participate in this binding interaction. In contrat, sickle-trait very much does reduce the presentation on the RBC surface of translated PfEMP1 protein. Collectively these findings suggest a post-transcriptional mechanism for the defect in EPCR binding in sickle-trait cells. These findings were reported here in PLOS Pathogens.

In parallel , we performed an unbiased parasite comparative transcriptomics screen in normal and sickle-trait cells, and did so both in vitro using reference lines and in vivo using fresh parasites from Malian children with malaria. In vitro, using lines 3D7 and FUP, transcriptional programs were overall very conserved in sickle-trait cells, though they did diverge late in the parasite life-cycle suggesting defects in egress or merozoite formation. In fresh parasites, we observed significant differential expression of in trophozoite-stage parasites, including significant overexpression in sickle-trait children of transcripts encoding the putative protein phosphatase PF3D7_1127000 of over 30-fold. Full results are reported here in mSphere and the data are deposited here in NCBI as PRJNA685106. These results were fortified subsequently in a GWAS of parasite variation in people with and without HbS, in which variants in PF3D7_1127000 were 1 of 3 parasite loci enriched in people with sickle-trait. The mechanism of this is unknown, as is the link between this parasite adaptation to growth in sickle-trait cells and the attenuated pathogenesis of parasites in sickle-trait cells.

Partners in this project have included:

  • Steve Haase PhD, Dept of Biology and University Program in Genetics and Genomics, Duke University
  • Nirmish Shah MD, Division of Hematology, Duke University Medical Center
  • Rick Fairhurst MD PhD, Laboratory of Malaria and Vector Research, NIAID
  • Mahamadou Diakite PharmD DPhil, University of Sciences, Techniques and Technologies of Bamako, Mali
  • Thomas Lavstsen PhD, Center for Medical Parasitology, University of Copenhagen, Denmark
  • Gavin Band PhD, Wellcome Centre for Human Genetics, University of Oxford, UK

niaid upgg duke-biology-logo-139x139


This project was funded by an R21 award from NIAID.



Prevention of pregnancy-associated malaria in sub-Saharan Africa

In sub-Saharan Africa, antenatal maternal malaria infections are a major cause of low birthweight and neonatal mortality. Therefore, in most malaria-endemic African countries, pregnant women receive a suite of malaria-prevention measures, including insecticide-treated bednets and intermittent preventive therapy (IPTp) with antimalarials. The primary antimalarial used is the combination sulfadoxine-pyrimethamine, to which partial resistance is highly prevalent across Africa. Therefore, new approaches to preventing antenatal malaria are continually needed.

With partners from multiple Western and African institutions, the Malaria in Pregnancy Consortium tested alternate strategies to mitigate the effects of antenatal malaria and measured the efficacy of SP to prevent low birthweight. Follow-on studies from Mipc-sponsored field studies are focused on testing the impact of subpatent parasites on birth outcomes, describing the epidemiology of subpatent parasitemias during pregnancy, and testing for associations between emerging resistance markers and birth outcomes.

Partners in this project include:

  • Mwayiwawo Madanitsa MD PhD, Liverpool School of Tropical Medicine
  • Victor Mwapasa MD PhD, University of Malawi College of Medicine
  • Feiko O ter Kuile MD PhD, Liverpool School of Tropical Medicine and KEMRI

lstmlogo universityofmalawishieldunc_gillings_542_transp_blu


This project has been funded by an award to the Liverpool School of Tropical Medicine by the Bill and Melinda Gates Foundation.



Epidemiology of syphilis in pregnancy in Western Kenya

WASP Research Nurse Violet explains the study procedures to Duke Chancellor Dr Eugene Washington in the Webuye County Hospital ANC clinic.

In 2012, there were an estimated 930,000 maternal syphilis infections globally that caused 350,000 adverse pregnancy outcomes. Congenital syphilis remains prevalent in many LMICs: compared to the global average, the African region has a three-fold higher antenatal syphilis seroprevalence. In Kenya, congenital syphilis prevention is dependent solely on one-time screening of pregnant women at antenatal booking, but testing rates are low and congenital syphilis prevention is likely compromised by incident infections during pregnancy owing to the absence of contact tracing or repeat syphilis testing.

The Webuye Antenatal Syphilis Project (WASP) is a pilot project in which we will estimate the epidemiology and impact of incident syphilis in pregnant women at a single site in Western Kenya, in order to generate scientific premises and research infrastructure that are necessary to test new programmatic interventions to mitigate the risk of congenital syphilis.

The observational WASP study began enrolling participants in April 2018 and concluded followup in early 2020. We expect results to be final in early 2022.

Partners in this project include:

This project is funded by a Priority Partnership Location Pilot Research grant from the Duke Global Health Institute.

Molecular epidemiology of P. falciparum transmission in Western Kenya

Team Mozzie, Webuye, Kenya, June 2018

The science of identifying `hotspots’ of malaria or foci of transmission is a growing area that promises to help target interventions more effectively. However, it has not been shown whether infected individuals in close physical proximity (i.e. in the same household) are jointly infected due to simply living in a risky place, or because an infected household member is a risk factor for nearby susceptible individuals.

If the former, then targeting hotspots should focus on reducing environmental risk factors in the area around a hotpsot. If the latter, then interventions to identify and treat `transmitters’ will reduce transmission and reduce the incidence of new cases. Therefore, we need to understand the spatial scale of malaria transmission to predict the impact of community case detection and hotspot targeting.

Sharing of parasite csp haplotypes in a single household in Webuye between parasites collected in humans black, mosquito abdomens gray, and mosquito head/thoraces light gray. Each chord represents one shared haplotype. Image rendered by Kelsey Sumner using migest package implemented in R.

For this project, we will measure the genetic relatedness of infections within the same household compared to the relatedness of infections at further distances. This will provide evidence for local, focal transmission if nearby infections are more closely related or will point to mixed transmission whereby infections only begin to differ as you reach the distance of mosquito flying ranges.

Our second objective is to trap malaria mosquito vectors and identify infected mosquitoes. By doing so, we can find out whether infections are being transmitted at a household scale or transmission is `well mixed’ geographically and only limited by the range of the mosquito.

The ability to track infections from human to mosquito and back again would allow us to understand the dynamics and scale of transmission in a way that has not previously been possible. These results will provide insight into the expected impact of interventions designed to target hotspots.

Dr Andrew Obala and the Webuye team explain the process of mosquito morphologic assessment to Duke Chancellor for Health Affairs Dr Eugene Washington and former DGHI Director Dr Chris Plowe.

Partners in this project include:

  • Wendy O’Meara PhD, Duke Global Health Institute
  • Andrew Obala PhD, Moi University School of Medicine
  • Amy Wesolowski PhD, Johns Hopkins University

This project is funded by an R01 award from NIAID.



Importation of malaria parasites into an epidemic-prone zone of Northwestern Kenya

Turkana County in the arid Northwest of Kenya is traditionally considered malaria-free owing to the low densities of humans and of mosquito vectors. However, there are little data to support this, and anecdotal reports indicate there are convincing, parasitologically-confirmed cases of malaria. In addition, the epidemiology of hosts is changing in Turkana County, owing to changes in land use patterns and increased migration from malaria-endemic areas. Therefore, Turkana county appears both suitable for malaria transmission and at risk of significant importation for the establishment of autochthonous transmission.

Team members Sophie, Godfrey, and George audit CRFs during a home visit in Nadoto, Kenya.

In this project, we are conducting reactive case detection of confirmed malaria cases at six health centers in and around Lodwar, Kenya, and screening departing travelers to Lodwar for malaria at the main transit centers in Kitale and Eldoret. The goals of these investigations are to understand the spatial clustering of infections in this low-transmission setting, to test for the association between recent travel and malaria risk, and to estimate the rate of importation of malaria overall and of specific strains into Lodwar.

Using these tools, we can better develop and apply tools to estimate parasite importation into suitable environments, and understand the factors that lead to successful establishment of local malaria transmission.

Partners in this project include:

    • Wendy O’Meara PhD, Duke Global Health Institute
    • Andrew Obala PhD, Moi University School of Medicine
    • Amy Wesolowski PhD, Johns Hopkins University

    This project was funded by an R21 award from NIAID.


Surveillance of malaria parasite drug resistance

Neighbor-joining network tree of P. falciparum dhps gene microsatellite haplotypes, coded by drug-resistant marker haplotypes (color) and country of origin (shape). The triple-mutant SGEG drug resistant haplotypes collected from pregnant women in Malawi and Tanzania cluster separately, suggesting that these mutations arose independently on different genetic background, rather than sharing an ancestral parasite line.

Parasite resistance to antimalarial drugs is prevalent and continually evolving (see this Prezi). The molecular markers of parasite drug resistance have been described for many antimalarials, including chloroquine, sulfadoxine, pyrimethamine, mefloquine, and others. Most recently, single nucleotide polymorphisms that confer the phenotype of delayed parasite clearance after artemesinin therapy have been described in southeast Asia, and copy-number variants have been described, also in Asia, that confer resistance to piperaquine. These mutations are not typically actionable in clinical practice owing to the need for molecular detection capacity and the turnaround time for the tests. These markers are useful, however, to enhance surveillance for drug resistance without the need for routine clinical or cellular phenotyping. In addition, these markers, when coupled with additional parasite gene markers and analyzed using population genetic approaches, can enhance the tracking of drug resistance alleles through parasite populations and quantify the degree of genetic relatedness of parasite gene (see NJ network, right)


Comparison of approaches to sequencing-based protocols to genotype parasites. A. Traditional approach where parasites are individually extracted, amplified by PCR, and Sanger sequenced. B. Pooled sequencing approach, where parasites are individually extracted but then pooled prior to PCR amplification of the gene of interest, followed by second-generation sequencing to generate many sequencing reads for analysis that reflect the complex PCR template in the original pool. C. Similar to B, excepting that parasites are pooled as specimens, and extracted as an aggregated pool.

We have been adapting deep-sequencing platforms to more rapidly screen parasite populations for known markers of antimalarial resistance to SP and to artemisinins (see schematic, right).

With partners in W and E Africa, we are currently adapting similar approaches to monitor for the emergence of drug-resistant parasites in response to seasonal malaria chemoprophylaxis with SP in Mali and to mass drug administration with dihydroartemisinin-piperaquine in Mozambique.

Partners in these projects include:

cism-logoisglobal_logo_enimgres   niaid


This project is funded by an R01 award from NIAID.