Application of Physiologically Based Pharmacokinetic Model to Predict Cobicistat Concentration during Pregnancy
Ibtihel Hammami(1), Lisa Almond(1), Nikunjkumar Patel(1), Jia Ning(1), Xian Pan(1), Hannah Jones(1), Iain Gardner (1)and Khaled Abduljalil(1)
(1) Simcyp LTD (a Certara Company), Sheffield, UK
Objectives:
Cobicistat, a potent CYP3A4 inactivator that was developed specifically as a pharmacokinetic enhancer, is given in combinations with antiretrovirals to boost their concentrations. However, it has been reported that cobicistat is not an effective booster during pregnancy due to its decreasing exposure [1]. In 2018, the US Food and Drug Administration (FDA) amended labelling for cobicistat containing products to advise that they must not be prescribed in pregnancy [2]. The aim of this work is to use physiologically based pharmacokinetics (PBPK) modelling to simulate cobicistat concentrations in pregnant women during the second and third trimester of pregnancy and investigate the mechanisms at play.
Methods:
A PBPK model including prior in vitro and in vivo data describing the disposition of cobicistat was developed and simulations of 150 mg repeat oral doses were run using the Simcyp Simulator V21 and compared to observed data. The cobicistat PBPK model included autoinhibition due to its mechanism-based inhibition of CYP3A4, a major route for cobicistat metabolism. The verified model was then applied to the pregnancy population that considers physiological changes during pregnancy, including gestational-dependent increase in CYP3A4 and CYP2D6 activities [3], to predict maternal exposure during the second and third trimesters. The predicted geometric mean ratio (GMR) for pharmacokinetic parameters in pregnant versus age-matched nonpregnant women was then compared to the observed data from Momper et al 2018[4] and 2021[5].
Results:
Simulated PK parameters for cobicistat following multiple oral doses (150 mg) were within 1.5-fold of clinically observed data [6,7 and 8] except data from Custodio et al 2014[9] which was predicted within 2-fold. The model predicted the lower cobicistat exposure in both the second and the third trimester of pregnancy compared to non-pregnant women reasonably well with a marginally superior performance for the third trimester. The simulated over observed [5] ratios of AUCt and Cmax in pregnant versus non pregnant populations were 1.3, 1.0 and 1.3, 1.1 for the second and third trimester, respectively. The simulated over observed [4] ratios of AUCt and Cmax were 1.1, 1.1 and 1.0, 1.0 for the second and third trimester respectively. The PBPK model predicted 40% reduction of AUC during pregnancy toward term compared to the nonpregnant exposure.
Conclusions:
Simulations indicate that the longitudinal reduction of cobicistat exposure during pregnancy can be explained by the induction of CYP3A4 and CYP2D6 and the decreasing level of serum albumin during pregnancy. PBPK can be useful to better understand the effect of pregnancy on cobicistat disposition. Further work to model the impact of reduced cobicistat concentrations on the PK of the different partnered antiretrovirals is now warranted. As pregnant women are excluded from the majority of clinical studies, PBPK may help inform and assess treatment options in pregnancy to avoid the use of sub-optimal therapy.
References:
[1] Eke AC et al., J Clin Pharmacol. 2019;59(6):779-783.
[2] https://www.ascpt.org/Resources/ASCPT-News/View/ArticleId/22744/FDA-News-Issue-33-November-2018
[3] Abduljalil K et al., J Pharmacokinet Pharmacodyn. 2020 Aug;47(4):361-383.
[4] Momper JD et al., AIDS. 2018;32(17):2507-2516.
[5] Momper JD et al., AIDS. 2021;35(8):1191-1199.
[6] Custodio et al ., J Clin Pharmacol. 2014;54(6):649-656.
[7] https://www.ema.europa.eu/en/documents/assessment-report/rezolsta-epar-public-assessment-report_en.pdf (study GS-US-216-0115)
[8] Ramanathan et al., J Acquir Immune Defic Syndr. 2013;64(1):45-50.
[9] Custodio et al ., Antimicrob Agents Chemother. 2014;58(5):2564-2569.