Sulav Duwal(1), Max von Kleist(1)
(1) Dep. of Mathematics and Computer Science, Freie Universität Berlin, Germany
Objectives: Nucleoside reverse transcriptase inhibitors (NRTIs) are backbone compounds for modern HIV-1 treatment. They are pro-drugs, which, after intracellular phosphorylation, compete with endogenous nucleotides for incorporation into nascent viral DNA and thus prevent polymerization. Plasma- and effect-site PK are usually asynchronous and cell-specific for NRTIs, which is also true for the PD, due to their mechanism of action [1]. While the NRTIs tenofovir disoproxil fumarate (TDF) and emtricitabine (FTC) have recently been approved as PrEP compounds, we aim to assess the prophylactic utility of approved NRTIs and those that were never tested for this purpose in silico, when taken continuously, sporadically and on-demand [2].
Methods: We developed PK-PD models for FTC, TDF and lamivudine (3TC) and explored their efficacy in PrEP, based on stochastic simulation & analysis. Data was collected from various PK studies to link dosing, plasma- and intracellular PK for each drug. Depending on the data sources, we used different methods for parameters estimation, including non-linear mixed effect modelling. We coupled the intracellular PK to a HIV-1 model (PD) [3,4] using the Emax equation. The latter was mechanistically confirmed for NRTIs [1]. Using PD data, we estimated the IC50 value for each drug. These values were then compared to a mechanistic approach [1] that relies on in vitro (enzymatic) data. An analytical formula [4] was used to compute the prophylactic efficacy based on intracellular NRTIs levels, while drug administration schemes were assessed using hybrid stochastic-deterministic simulation.
Results: For all drugs, a two compartment model with a first order absorption best described the plasma PK. The intracellular PK was captured in terms of saturable uptake & anabolism and first order elimination. IC50 values estimated from PD profiles showed remarkable agreement with mechanistically derived IC50s [1] and suggest strong potency against wild type HIV at clinically achieved concentrations. The protection level provided by 3TC, FTC and TDF relative to the absence of drug was computed.
Conclusions: The PK-PD models for all drugs adequately characterized their clinical profiles. Our approach suggests strong potency of FTC and 3TC against HIV-1 in PrEP. The mechanistic IC50 [1] may be used to assess drug potency in various target cells. In the future, co-administration and synergistic effects of NRTIs, e.g. with respect to PrEP, will be explored.
References:
[1] von Kleist M, Metzner P, Marquet R and Schütte C (2012) PLoS Comp. Biol. 8: e1002359
[2] Cohen J (2015) Science 347:1055
[3] von Kleist M, Menz S and Huisinga W (2010) PLoS Comput. Biol. 6: e1000720
[4] Duwal S, Schütte C and von Kleist M (2012) PLoS One 8: e40382
Reference: PAGE 24 (2015) Abstr 3459 [www.page-meeting.org/?abstract=3459]
Poster: Drug/Disease modeling - Infection