Sulav Duwal, Vikram Sunkara, Max von kleist
Freie Universität Berlin
Objectives: While HIV-1 cannot be cured, current efforts focus on developing pharmaco-intervention strategies to end the epidemic. Two strategies have been proposed: (i) Treatment-as-prevention (TasP) decreases the patients’ virus load and thereby markedly reduces the infectivity of the potential transmitter. However, onwards transmission may preferentially occur early after infection [1], when the potential transmitter is unaware of his/her infection, arguing that TasP may only prevent a small fraction of transmission events in reality. (ii) The second intervention is called pre-exposure prophylaxis (PrEP). Here, an uninfected, exposed individual takes anti-viral drugs to prevent infection upon viral exposure.
Our objective was to use a predictive PK-coupled MOA model for nucleoside reverse transcriptase inhibitors (NRTIs) [2,3] to explore the potential of the NRTIs lamivudine (3TC), emtricitabine (FTC) and tenofovir (TDF) for repurposing in the context of PrEP. We assessed the mechanisms of prevention and pharmacological limitations and opportunities for various schedules.
Methods: We developed mechanistic models that relate viral load in the potential transmitter to viral exposure in the recipient and estimated the infection probability for unprotected sex. All steps were validated using published data. We estimated reduction in infection probability under prophylaxis, taking the individual PK of the analyzed NRTIs within the exposed person into account. This pipeline allowed us to analyze the efficacy of all NRTIs, integrating virus load in the transmitter, mode of exposure, timing of viral challenge, and dosing schedule.
Results: The viral load distribution in the potential transmitter population [1] was log-normal distributed. Predictions indicated that FTC and 3TC are superior to TDF, in contrast to current beliefs [4], which was particularly evident for ‘PrEP on demand’. Our pipeline revealed a dependency between reported efficacy endpoints in PrEP trials and the individual follow-up duration. This may explain contradicting reports with regard to PrEP efficacy.
Conclusions: We report the first approach to mechanistically model PrEP efficacy by integrating viral loads within the donor population, mode- and timing of viral challenge with respect to PK-PD in the exposed individual. The framework can be extended to other drug classes and also be used explore TasP efficacy with- and without PrEP. Furthermore it can inform epidemiologic models.
References:
[1] Yousef, KP, Meixenberger K, Smith MR, Somogyi S, Gromöller S, Schmidt D, Gunsenheimer-Bartmeyer B, Hamouda O, Kücherer C and von Kleist M (2016) JAIDS (under revision)
[2] von Kleist M, Metzner P, Marquet R and Schütte C (2012) PLoS Comp. Biol. 8: e1002359
[3] Duwal S, von Kleist M (2016) European Journal of Pharmaceutical Sciences, ahead of print
doi:10.1016/j.ejps.2016.01.016
[4] Hendrix CW (2013) Cell 155(3):515-8
Reference: PAGE 25 (2016) Abstr 5774 [www.page-meeting.org/?abstract=5774]
Poster: Drug/Disease modeling - Infection