Emma Hughes1, Erika Wallender1, Richard Kajubi2, Ali Mohamed1, Liusheng Huang1, John Ategeka2, Teddy Ochieng2, Abel Kakuru2, Prasanna Jagannathan3, Miriam Nakalembe4, Bishop Opira2, Patience Nayebare2, Tamara D Clark1, Moses Kamya2,4, Philip Rosenthal1, Grant Dorsey1, Francesca Aweeka1, Rada M. Savic1
1 University of California, San Francisco, San Francisco, CA, United States 2 Infectious Disease Research Collaboration, Kampala, Uganda 3 Stanford University, Palo Alto, CA, United States 4 Department of Obstetric and Gynecology Makerere University College of Health Sciences
Objectives:Malaria during pregnancy affects 11 million African women each year and, when parasites sequester in the placenta causing placental malaria, the infection can lead to adverse outcomes for the fetus including low birth weight, preterm birth and fetal death [1-4]. Although the World Health Organization recommends malaria chemoprevention during pregnancy, in East Africa there is widespread resistance to the approved drug, sulfadoxine pyrimethamine (SP) [5]. Dihydroartemisinin-piperaquine (DHA-PQ) is being studied to replace SP in these areas [6-8].
In Uganda, a recent clinical trial comparing monthly SP and DHA-PQ during pregnancy found DHA-PQ was 96% protective again malaria. However, subclinical parasitemia and placental malaria were still detected in women who received DHA-PQ and DHA-PQ was not associated with improved birth outcomes [6]. PQ is considered safe yet concerns regarding QTc prolongation remain. This suggests further evaluation of dosing regimens is needed.
Our goal is to use pharmacokinetic/pharmacodynamic (PK/PD) modeling to define optimal chemoprevention strategies for DHA-PQ in pregnant women. Specifically, we aimed 1) to identify protective PQ concentrations against parasitemia for pregnant women; 2) to quantify covariates which alter the PK and validate the PQ model; 3) to define optimal prevention regimens for pregnant women using a population PK/PD-Safety model and 4) to quantify and integrate the relationships between PQ levels, malaria parasitemia, placental malaria, and adverse birth outcomes.
Methods:Data from three Phase 3 malaria prevention trials (BC1-3) including 1022 pregnant Ugandan women were analyzed. Women were enrolled during the second trimester and received SP (1500mg S; 75mg P) every 4 weeks (375 HIV-uninfected women), DHA-PQ (120mg DHA; 960mg PQ) every 8 weeks (92 HIV-uninfected) or 4 weeks (472 HIV-uninfected, 83 HIV-infected receiving efavirenz based antiretroviral therapy). Longitudinal PQ troughs were taken monthly and a subset of women (30 HIV-uninfected, 28 HIV-infected and 30 HIV-uninfected postpartum women) underwent intensive sampling at times pre-dose, 0.5, 1, 2, 3, 4, 6, 8, 24 hours, and days 4, 7, 14, and 21 post last-dose.
Longitudinal parasitemia was evaluated monthly for all BC3 women by microscopy or quantitative PCR (qPCR). QTc measures were recorded in all BC3 women at weeks 20, 28, and 36 prior to their first dose and 3-4 hours after their last dose. Placental malaria was diagnosed by histopathology. A composite outcome was used for adverse birth outcomes and included low birth weight, preterm birth and small for gestational age.
1). PK model: The structural PK model was built using data from BC1 & BC2. A stepwise covariate modeling approach with forward inclusion (p-value <0.05) and backwards elimination (p-value <0.01) was used to identify demographic and clinical characteristics leading to PK variability. PK data from the BC3 cohort was used to externally validate this model.
2). PK/PD model: The probability of being parasite positive by qPCR was modeled with the BC3 data using longitudinal mixed effects logistic regression. PQ levels, gravidity and trimester were tested as covariates. Simulations were conducted to predict the minimum PQ concentration associated with parasitemia prevention. A sensitivity analysis was conducted to investigate how this concentration changed when different parasite density thresholds were used to classify a woman as parasite negative.
3). PK-safety model: The validated PK model was used to simulate the expected PK concentrations at the time QTc measures were recorded in the BC3 women. These concentrations were linked to the QTc measures.
4). Integrated PK-parasitemia-birth outcome model: Logistic regression models evaluated associations between PQ PK, clinical and demographic factors, and placental malaria. Logistic regression models were also used to evaluate risk factors for composite adverse birth outcomes.
Results:A total of 2218 PQ plasma concentrations contributed to the PK model. A three-compartment model with an absorption lag best fit the data. Pregnancy, BMI, and efavirenz use in HIV-infected women were independently associated with increased PQ clearance and therefore decreased plasma concentrations. Pregnancy was associated with a 72% increased PQ clearance compared to postpartum women. Low-BMI pregnant women had higher clearance (2% increase for every unit drop in BMI). Similarly, HIV-infected women taking efavirenz had clearance increased by 61% in comparison to HIV-uninfected pregnant women. While this model was successfully validated markedly higher trough levels were observed in the BC3 cohort. This difference was captured by using a separate bioavailability term and attributable to better adherence.
Higher PQ concentrations were associated with a lower probability of having parasitemia during pregnancy (Emax: 0.94). Protective PQ concentrations ranged by parasite density threshold, with a maximum of 74% protection from maternal parasitemia at concentrations >16.0 ng/mL for the most sensitive threshold of 0.001 parasites/uL.
Simulations of the final PK model identified that HIV-infected and low-BMI women consistently had the lowest PQ concentrations putting them at risk of sub-optimal exposure. More frequent dosing was necessary to achieve protective concentrations. One tab daily DHA-PQ resulted in the most HIV-uninfected women (>91%) above 16.0 ng/mL. Our PK-QTc model did not predict any prolongation greater than 30 msec.
For placental malaria, higher PQ concentrations and starting prevention earlier were both found to be protective for women receiving DHA-PQ, odds ratio (OR) 0.92 (95% CI 0.88-0.97) and OR 0.42 (95% CI 0.21-0.83), respectively. In addition, having >38% of qPCR samples positive for parasites was associated with greater odds of placental malaria for DHA-PQ and SP (DHA-PQ: OR 2.7 (95% CI, 1.5-5.0); SP OR 6.6 (95% CI; 2.8-15)) and being primagravida was also a risk factor (DHA-PQ OR 13.4 (95% CI; 6.4-28); SP 13.6 (95% CI: 4.8-38.5)). Women in the SP arm who had clinical malaria, especially during the third trimester, had an increased risk for placental malaria OR 4.2 (95% CI; 1.4-13). For adverse birth outcomes, placental malaria was a significant risk factor, OR 2.5 (95% CI;1.5-4) and greater weight gain was protective (OR 0.88 (95% CI;0.82-0.95)).
Conclusions: With placental malaria as a clear risk factor for adverse birth outcomes, we found that higher PQ concentrations and longer durations of chemoprevention with DHA-PQ were associated with decreased odds of placental malaria. By using qPCR, we found a maximum of 74% protection against parasitemia occurred at concentrations >16 ng/mL. Overall, >91% of HIV-uninfected women were predicted to maintain PQ troughs >16 ng/mL with daily dosing, suggesting that daily dosing and early initiation of chemoprevention could reduce placental malaria and improved birth outcomes.
References:
[1] World Health Organization. (2019). World Malaria Report 2019 (Geneva, Switzerland, 2019).
[2] Akinyemi, R.O. et al. Contribution of noncommunicable diseases to medical admissions of elderly adults in Africa: a prospective, cross-sectional study in Nigeria, Sudan, and Tanzania. J Am Geriatr Soc 62, 1460-6 (2014).
[3] Moore, K.A. et al. Mediation of the effect of malaria in pregnancy on stillbirth and neonatal death in an area of low transmission: observational data analysis. BMC Med 15, 98 (2017).
[4] Moore, K.A., Simpson, J.A., Scoullar, M.J.L., McGready, R. & Fowkes, F.J.I. Quantification of the association between malaria in pregnancy and stillbirth: a systematic review and meta-analysis. Lancet Glob Health 5, e1101-e12 (2017).
[5] Naidoo, I. & Roper, C. Drug resistance maps to guide intermittent preventive treatment of malaria in African infants. Parasitology 138, 1469-79 (2011).
[6] Kajubi, R. et al. Monthly sulfadoxine-pyrimethamine versus dihydroartemisinin-piperaquine for intermittent preventive treatment of malaria in pregnancy: a double-blind, randomised, controlled, superiority trial. Lancet 393, 1428-39 (2019).
[7] Kakuru, A. et al. Dihydroartemisinin-Piperaquine for the Prevention of Malaria in Pregnancy. N Engl J Med 374, 928-39 (2016).
[8] Natureeba, P. et al. Intermittent Preventive Treatment With Dihydroartemisinin-Piperaquine for the Prevention of Malaria Among HIV-Infected Pregnant Women. J Infect Dis 216, 29-35 (2017).
Reference: PAGE () Abstr 9567 [www.page-meeting.org/?abstract=9567]
Poster: Oral: Drug/Disease Modelling