PK/PD Modeling of Anti-Circumsporozoite Protein (CSP) Antibodies to Predict Dosing Requirements for Malaria Prevention


This poster features a collaboration with Bill & Melinda Gates Medical Research Institute and was presented at ACoP14.


In the context of infectious disease, such as malaria, monoclonal antibodies (mAbs) have the potential to provide protection to both adults and the pediatric population. Under development are half-life extended antibodies that bind to circumsporozoite protein (CSP) on the surface of the infectious sporozoites carried by mosquitoes. There are several antibodies currently in development that demonstrate protection against malaria infection as well as long circulating half-lives, which may provide season-long protection with only a single dose. Here we develop a PK/PD model of CSP entry and distribution, anti-CSP mAb PK, and mAb-CSP binding. The model was used to study mouse efficacy models, calibrated to cyno PK and translated to human to determine the extent to which the mAb prevents escape of CSP to the liver. Modeling and analysis is performed here for MAM01, a mAb currently in development at GatesMRI, and the performance is compared to CIS43LS.


  • A PK/PD model was built using published data related to P. falciparum sporozoite dynamics following a bite from an infected mosquito 1-3 , as well as preclinical data for MAM01 and published clinical data for CIS43LS 4.
    • The human model used PK parameters allometrically scaled from the cynomolgus monkey for MAM01
    • The model utilizes an escape-to-liver metric as the surrogate of efficacy.
  • A mouse PK/PD model was developed using data from efficacy models measuring sporozoite liver burden and blood parasitemia in mice after MAM01 dose followed by the administration of sporozoites via bites or through IV dose
    • The model was used to simulate the escape-to-liver efficacy criteria that corresponds to 95% protection
    • Predicted reduction in escape-to-liver was used as the metric for efficacy in human simulations
  • The human model, trained on CIS43LS data was used to predict reduction in escape-to-liver that results from efficacious doses
    • Determine the MAM01 dose needed to achieve the efficacy criteria established with mouse modeling
    • Compare response to the model-predicted performance of CIS43LS


Mouse modeling and analysis predicts > 99.99% reduction in escape-to-liver which corresponds to 95% protection from infection. This is in agreement with the predicted reduction achieved with 3 mg/kg CIS43LS in CHMI and 40 mg/kg dose in Ph II studies. Efficacy was predicted to be achieved with lower doses of MAM01 than CIS43LS due to its higher binding affinity. The model will be refined with additional data and predictions made for adult and pediatric populations. Head-to-head preclinical studies are underway to demonstrate superiority of MAM01 vs. competitors.

Download the Poster