Evaluation of immuno-epidemiological markers for assessing malaria transmission intensity in the hypo-endemic highlands of Kenya and the accuracy of malaria diagnosis in the Holo-endemic forest zone of Ghana

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
Current tools for measuring malaria transmission intensity have limited sensitivity when transmission is low. Robust surveillance systems are needed to monitor reduced transmission and prevent the rapid reintroduction in areas undergoing elimination. Serologic tools based on the antibody response to parasite and vector antigens are potential tools for transmission measurements. The present study examined the serologic evidence of vector exposure and malaria endemicity in the western Kenyan highland sites along a transmission intensity cline. Total IgG levels to Plasmodium falciparum MSP-119 and Anopheles salivary gland peptide gSG6-P1 were measured in an age-stratified cohort (< 5, 5-14 and ≥ 15 years) during low and high malaria transmission seasons. Antibody prevalence and level were compared among different localities. Regression analysis was performed to examine the association between antibody prevalence and parasite prevalence. Age-specific MSP-119 seroprevalence data was fitted to a simple reversible catalytic model to investigate the relationship between parasite exposure and age. Higher MSP-119 seroprevalence and density were observed in the valley residents than in the uphill residents. Adults (> 15 years) recorded high and stable immune response in spite of changing seasons. Lower responses were observed in children (≤ 15 years) which, fluctuated with the seasons particularly in the valley residents. In the uphill population, annual seroconversion rate (SCR) was 8.3% and reversion rate was 3.0%, with seroprevalence reaching a plateau of 73.3% by age of 20. Contrary, in the valley bottom population, the annual SCR was 35.8% and the annual seroreversion rate was 3.5%, and seroprevalence in the population had reached 91.2% by age 10. Seroprevalence of gSG6-P1 in the uphill population was 36% while it was 50% in the valley bottom population (χ2 13.2 P= 0.0002). Median gSG6-P1 antibody levels in the Valley bottom were twice as high as that observed in the uphill population [4.50 (1.02) vs. 2.05 (0.92)] and showed seasonal variation. The odds of gSG6-P1 seropositives having MSP-119 antibodies were almost 3 times higher than the odds of seronegatives [OR =2.87, P < 0.001]. The observed parasite prevalence for Kisii, Kakamega and Kombewa were 4%, 19.7% and 44.6% whilst the equivalent gSG6-P1 seroprevalence were 28%, 34% and 54% respectively. The study reveals the micro-geographic variation in malaria endemicity in the highland eco-system; thus validating the usefulness of sero-epidemiological tools in assessing malaria endemicity in the era of decreasing sensitivity to conventional tools. The seroprevalence of IgG to gSG6-P1 was sensitive and robust in distinguishing between hypo, meso and hyper transmission settings and seasonal fluctuations. The study determined the accuracy (sensitivity and specificity) of PfHRP2- based malaria rapid diagnostic test (RDT) in Western Kenya and in the forest zone of Ghana. The performance of RDTs when microscopy is used as the gold standard is reported based on an operational study that took place at the Ghanaian study sites and a field trial conducted based on asymptomatic primary school children in Kenya. Implications for the choice of the RDT in an era of changing malaria epidemiology are discussed.
A thesis submitted to the Department of Theoretical and Applied Biology, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana in partial fulfillment of the requirements for the award of Doctor of Philosophy In Biological Sciences,