Sara Iannuzzi
Robert Koch Institute
Objectives:
According to the The Joint United Nations Programme on HIV and AIDS (UNAIDS) estimates, approximately 37.9 million individuals worldwide are infected with the human immunodeficiency virus (HIV). HIV continues to spread, and the latest incidence estimates amount to about 1.7 million new infections, almost 5000 per day, in 2018. There is still no cure nor vaccine for HIV, and once infection takes place, the virus can persist for years in the system. This HIV reservoir feature gives rise to the need for a life-long therapy with antiretroviral drug in order to avoid viral rebound and the acquired immunodeficiency syndrome (AIDS). Multi-drug therapy is currently used to tackle early HIV infection. The two main therapeutic strategies can be outlined as treatment-as-prevention and HIV pre-exposure prophylaxis (PrEP). The former follows the rationale that successful treatment of HIV-infected individuals reduces their viral loads, rendering them virtually non-contagious and ideally leading to a reduction in the per-contact probability of infection [2]. The latter consists in promoting viral elimination before irreversible infection in uninfected individuals at risk, through antiviral medication. It is becoming well-established, and the two-drug combination Truvada has been approved for PrEP in the US and EU in 2012 and 2016 respectively.
As highlighted in previous work, HIV transmission probabilities are extremely low, making the computation of PrEP efficacy prone to chance events. As an example, the recent DISCOVER trial evaluating the efficacy of emtricitabine plus tenofovir alafenamide (Descovy) against Truvada was conducted over 8756 person-years. Despite this enormous number, there were as little as 17 infections and only 2 (one in each respective PrEP arm) in individuals with detectable drugs. Hence, the proportion of gathered data that actually informs about the clinical efficacy is marginally low. Furthermore, statistical empowering of clinical studies would result in costly efforts of both monetary and organizational nature. Moreover, estimating parameters of clinical efficacy is not feasible in vivo, due to ethical reasons. It follows that a translational tool would aid scientific advances in this framework, possibly overcoming some aspects of the above-mentioned limitations.
A first PrEP modelling framework has been established and adapted to a clinical context by following the workflow established by the systems biology/systems pharmacology community. In this framework, mechanistic modelling is applied to integrate available multidisciplinary data (pharmacokinetics, mechanisms of antiviral action, viral dynamics, viral exposure). In this approach, initial viral dynamics after exposure are modelled stochastically, subject to drug pharmacokinetics, to compute viral elimination probabilities for different PrEP regimen. Core assumptions of the model include:
A. Concentrations at the viral exposure site match the drug concentrations in peripheral mononuclear blood cells (PBMC).
B. There is no micro-compartmentalization with respect to initial viral replication, with little emphasis on putative differences in PrEP efficacy in different risk groups.
This highlights the need for a translational tool to predict PrEP efficacy.
Methods: In order to carry out the proposed study, integrative mathematical modelling is applied. Current models have to be implemented in order to test different adherence profiles and resulting plasma concentration profiles. Once the model has been implemented in Python programming language, simulations can be performed. Parameterization of the model will follow, and in turn used as an input for further simulations. At a later stage, adjustment of predictions and parameters with human explant data can be performed.
Results:
Implementation of hybrid stochastic pharmacological models for HIV PrEP was accomplished. A first prototype based on the available models is now ready, serving as an excellent starting point for further observations. This allowed to study HIV PrEP under both the pharmacological and mathematical aspects. A first prototype based on the available models is now ready, serving as an excellent starting point for the envisioned project.
Conclusions:
The implemented model is in agreement with the pharmacokinetic model developed by [1]. Further tests can be carried out upon receival of clinical data (Spring 2020).
References:
[1] Burns, R.N., C.W. Hendrix, and A. Chaturvedula, Population pharmacokinetics of tenofovir and tenofovir-diphosphate in healthy women. J Clin Pharmacol, 2015. 55(6): p. 629-38.
[2] Cohen, M.S., et al., Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med, 2011. 365(6): p. 493-505.
Reference: PAGE () Abstr 9561 [www.page-meeting.org/?abstract=9561]
Poster: Oral: Drug/Disease Modelling