In vitro Characterization of Efficacy and Toxicity of Nevirapine Analogs as Substitutes in HIV Treatment

Abstract

Nevirapine (NVP), a non-nucleoside HIV reverse transcriptase (RT) inhibitor, is generally a well-tolerated drug. However, certain toxicological issues have been reported, including idiosyncratic drug reactions (IDRs) such as skin rash and hepatotoxicity. Additionally, NVP is a known mechanism-based inactivator (MBI) and inducer of CYP3A cytochrome P450 enzymes (CYPs), with the potential to cause drug-drug interactions. It has been demonstrated previously that the 12-hydroxynevirapine metabolite and a reactive quinone methide species, produced by CYP3A4 and CYP2B6, are responsible for the observed IDRs. The reactive quinone methide species also leads to mechanism-based inactivation of CYP3A4, while NVP itself is an inducer of both CYP3A4 and CYP2B6. The goal of our study was to characterize a series of NVP analogs with respect to their safety and efficacy to treat HIV by examining target engagement efficiency and metabolic liability through in vitro characterization. Our hypothesis was that a NVP analog will be able to inhibit RT with similar efficacy, but with limited production of the toxic metabolites. Initial studies were conducted to determine the IC50 values of NVP and each analog in the RT inhibition assay, followed by determining the CYP induction/inhibition potential of NVP and its associated analogs in cryopreserved primary human hepatocytes. mRNA fold induction, CYP3A4 activity fold change, and % lactate dehydrogenase (LDH) release were all studied via qPCR, LC-MS/MS, and LDH assays respectively. In total, NVP and eight analogs were tested. The RT assay IC50 of NVP was 0.5 uM and the three analogs with the lowest IC50 values were: nevirapine-d3 (NVP-d3, 0.8 uM), 2-nitro nevirapine (2-N-NVP, 1.1 uM), and 3-bromo nevirapine (3-B-NVP, 3.0 uM). The induction experiments demonstrated that CYP3A4 was induced ~2 fold at 25 uM NVP, and ~6 fold at 100 uM NVP. CYP2B6 showed higher induction, with ~5 fold at 25 uM NVP and ~13 fold at 100 uM NVP. For CYP3A4 activity, 25 uM NVP showed ~1.4 fold increase in activity, while 100 uM NVP resulted in no fold change. This is likely due to the combined induction/inhibition effect of NVP; i.e., at higher concentrations more NVP is present to inactivate CYP3A4 and thus inhibition wins out over induction, while at lower concentrations of NVP induction dominates. %LDH release along with observation of the monolayer revealed low levels of hepatocyte mortality. Overall, three analogs were identified as potential candidates to replace NVP, due to their low IC50 values in the RT inhibition assay and reduced potential for CYP induction. Further characterization of these analogs is ongoing with the goal to identify an efficacious and safer NVP analog that can be used to treat HIV in pregnant women and children in developing countries, where NVP is still often used.