Preclinical profile of BI 224436, a novel HIV-1 non-catalytic-site integrase inhibitor

Abstract

BI 224436 is an HIV-1 integrase inhibitor with effective antiviral activity that acts through a mechanism that is distinct from that of integrase strand transfer inhibitors (INSTIs). This 3-quinolineacetic acid derivative series was identified using an enzymatic integrase long terminal repeat (LTR) DNA 3'-processing assay. A combination of medicinal chemistry, parallel synthesis, and structure-guided drug design led to the identification of BI 224436 as a candidate for preclinical profiling. It has antiviral 50% effective concentrations (EC50s) of <15 nM against different HIV-1 laboratory strains and cellular cytotoxicity of >90 μM. BI 224436 also has a low, ∼2.1-fold decrease in antiviral potency in the presence of 50% human serum and, by virtue of a steep dose-response curve slope, exhibits serum-shifted EC95 values ranging between 22 and 75 nM. Passage of virus in the presence of inhibitor selected for either A128T, A128N, or L102F primary resistance substitutions, all mapping to a conserved allosteric pocket on the catalytic core of integrase. BI 224436 also retains full antiviral activity against recombinant viruses encoding INSTI resistance substitutions N155S, Q148H, and E92Q. In drug combination studies performed in cellular antiviral assays, BI 224436 displays an additive effect in combination with most approved antiretrovirals, including INSTIs. BI 224436 has drug-like in vitro absorption, distribution, metabolism, and excretion (ADME) properties, including Caco-2 cell permeability, solubility, and low cytochrome P450 inhibition. It exhibited excellent pharmacokinetic profiles in rat (clearance as a percentage of hepatic flow [CL], 0.7%; bioavailability [F], 54%), monkey (CL, 23%; F, 82%), and dog (CL, 8%; F, 81%). Based on the excellent biological and pharmacokinetic profile, BI 224436 was advanced into phase 1 clinical trials.

FIG 1

Comparison of the antiviral inhibition curves of integrase strand transfer and non-catalytic-site integrase inhibitors. The HIV-1 NL4.3 virus was used to infect PBMCs in the presence of increasing concentrations (conc.) of either BI 224436 or raltegravir, representing the NCINI and INSTI classes of integrase inhibitors, respectively. The antiviral concentration-response curves were generated with 1.5-fold compound dilutions in order to more precisely evaluate the Hill slope values of approximately 4 for BI 224436 and 1 for raltegravir.

FIG 2

Fold decrease in mean EC₅₀s in the presence of different percentages of human serum with the in vitro antiviral activity determined using the C8166 LTR luciferase reporter cell line. The HIV-1 NL4.3 virus was used to infect the reporter cell line in RPMI medium containing 10% FBS that was supplemented with 0, 10, 20, 30, 40, or 50% human serum. Data represent the averages from three separate experiments with the standard deviations shown. Extrapolation to 100% human serum gives a 2.14-fold increase in the EC₅₀ for BI 224436.

FIG 3

Three-dimensional plot of synergy and antagonism at 95% confidence for the pairwise combination of BI 224436 with raltegravir (RAL). Analysis was performed with MacSynergy II software and is shown as a representative plot of the data shown in Table 3. The synergy volume is 10 nM²%, and the antagonism volume is 0 nM²%.