Challenges to the Treatment of Malaria Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 20, 2019
  • Porter, Kimberly
    • Affiliation: Gillings School of Global Public Health, Department of Epidemiology
  • Malaria remains a significant cause of morbidity and mortality. Successful treatment of malaria is threatened by widespread drug resistance and co-infections with HIV. This dissertation explored two challenges to malaria treatment. The first aim addressed outcome misclassification in antimalarial treatment trials. Without accurate classification of patients' outcomes, estimates of drug efficacy are flawed. We identified factors related to outcome misclassification: transmission intensity, the distribution of genetic variants in parasite populations, multiplicity of infection, and PCR-insensitivity to minority variants; then used our findings to develop a Monte Carlo uncertainty analysis. Using the uncertainty analysis, we found that misclassification of new infections as treatment failures was common and underestimated treatment efficacy in the high transmission area. The initial estimate of the cure rate in the high transmission area was 63.8%; after adjustment for uncertainty related to outcome misclassification, the 95% simulation interval of the cure rate was 74.6 to 83.3%. The initial estimate of the cure rate in the low transmission area was 94.0%; after the uncertainty adjustment the 95% simulation interval of the cure rate was 93.5 to 96.5%. The second aim was to assess the effect of a co-formulation of HIV protease inhibitors (PI) on incidence of clinical malaria among HIV-infected adults. Laboratory evidence has demonstrated that HIV PIs inhibit growth of <italic>Plasmodium falciparum</italic>, a malaria-causing parasite. We conducted an ancillary analysis of data collected by the Adult AIDS Clinical Trials Group in two trials comparing PI-based against non-nucleoside reverse transcriptase inhibitor (NNRTI)-based antiretroviral therapy on the incidence of clinical malaria in study participants residing in areas with endemic malaria. We used pooled logistic regression to calculate hazard ratios (HR) and 95% confidence intervals (CI). There was no effect of PI-based therapy on incidence of clinical malaria (HR = 1.03, 95% CI (0.74 - 1.44)), nor was there modification of the HR by seasonality and use of concomitant medications. Successful treatment of malaria is a global health priority. This dissertation provides a novel way to estimate treatment efficacy and proposes that HIV PIs may not have antimalarial action in HIV-infected patients at risk of co-infection.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Eron, Joseph
  • Juliano, Jonathan
  • Cole, Stephen
  • Meshnick, Steven R.
  • Poole, Charles
  • Burch, Christina L.
  • Doctor of Philosophy
Degree granting institution
  • University of North Carolina at Chapel Hill Graduate School
Graduation year
  • 2010
  • This item is restricted from public view for 1 year after publication.

This work has no parents.