GSK1265744 Demonstrates Robust In Vitro Activity Against Various Clades of HIV-1

To the Editors: INTRODUCTION Antiretroviral agents have been approved by the US Food and Drug Administration for use as pre-exposure prophylaxis (PrEP) against acquisition of HIV-1 in high-risk populations.1 Studies of daily oral emtricitabine and tenofovir disoproxil (FTC/TDF) have demonstrated variable efficacy in preventing incident infections.2 This variability is thought due in large part to issues surrounding adherence to the prescribed daily oral medication regimen.3 Disappointingly, 2 trials, FEM-PrEP and VOICE, were stopped prematurely because of futility, and measurements of plasma drug concentrations revealed that fewer than 30% of participants were taking the drug appropriately.4,5 It has been hypothesized that the use of long-acting (LA) antiretroviral agents given by intramuscular injection in the clinic may overcome some obstacles to adherence. GSK1265744 (GSK744), an analog of the integrase inhibitor dolutegravir, is formulated as a LA injectable nanosuspension.6 When given intramuscularly, it has demonstrated a prolonged half-life in plasma of 21–50 days. When dosed at 800 mg, levels of drug have remained greater than 4 times the 90% protein-adjusted inhibitory concentration (PAIC90) for up to 3 months, supporting dosing every 12 weeks.7 GSK744 LA has shown 90%–100% efficacy in preventing SHIV162P3 transmission intrarectally and intravaginally in animal models.8,9 Given this pharmacokinetic profile and its activity in the macaque/SHIV model, this compound has entered the clinic for testing as a next-generation PrEP agent. It is critical that a next-generation global PrEP agent be active against a variety of HIV-1 clades. To this end, we determined the relative effective concentrations of GSK744 against a panel of recombinant viruses containing integrase coding regions derived from various clades of HIV-1. METHODS We assembled a panel of recombinant viruses derived from 5 prevalent clades: clade B viruses from the Primary Infection Cohort at the Aaron Diamond AIDS Research Center in New York,10 clade A1 and D viruses collected from patients in Uganda and Rwanda,11 clade AE viruses collected from Thai patients,12 and clade C viruses collected from patients in Zambia.13 All viral specimens were obtained from newly infected untreated patients and were de-identified. The integrase coding region (867 bp) was amplified from viral complementary DNA, proviral DNA, or plasmid DNA using nested polymerase chain reaction as previously described.14 Each integrase coding region was recombined into a plasmid, R7/3(X/S)Bsd, which was derived from HXB-2, and carries inactivated envelope.15 Pseudotyped viruses were prepared by cotransfecting proviral DNA and pCI-VSV-G into HEK293T using jetPRIME transfection reagent (PolyPlus, Illkirch, France). The cells were fed the next day and viral supernatant was harvested after 48 hours and saved at −80°C until use. Viral concentrations were measured by p24 enzyme-linked immunosorbent assay. The 50% and 90% effective concentrations (EC50 and EC90) of GSK1265744 for each integrase were determined by a single-cycle in vitro infectivity assay. As described previously,14 TZM-bl cells were plated at 8000 cells per well with 150 μL of Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum. After incubation for 24 hours at 37°C, culture medium was aspirated and TZM-bl cells were incubated for 30 minutes in DMEM/10% fetal bovine serum with 20 μg polybrene (Abbott Laboratories, Abbott Park, IL). A GSK744 stock solution was serially diluted to multiple concentrations. Polybrene-containing medium was replaced by 150 µL of DMEM/10%FCS and 50 μL of each of these preparations was added. After another 30 minutes of incubation, pseudotyped virus (5 ng of p24 in 50 μL media) was added to the appropriate wells. After a 48-hour incubation, the cells were processed to measure β-galactosidase using the Galacto-Star (Applied Biosystems, Framingham, MA) protocol and the light signal was measured on an MLX microplate luminometer (Dynex Technologies, Inc., Chantilly, VA). All assays were performed in triplicate and repeated to verify the results. The luminometer results were analyzed using GraFit (Erithacus Software, Horley, United Kingdom) and Prism (GraphPad Software, San Diego, CA) software to create dose–response curves and to calculate the EC50 and EC90 for each drug/virus combination (including the reference virus R7/3). We then expressed each combination as the ratio of n-fold change EC50 and EC90 over R7/3. Effective concentrations were compared using analysis of variance and pairwise with the Mann–Whitney U test. RESULTS In vitro infectivity assays were performed on 23 pseudoviruses containing the integrase genes from various clades of HIV-1 from newly infected untreated patients (Table 1). In our assay, GSK744 had an EC50 of 0.28 nM and EC90 of 0.85 nM against R7/3, and it demonstrated potent activity in vitro against pseudotyped viruses with integrase genes from all 5 clades of HIV-1. The mean EC50 fold change was 0.91 (range, 0.17–1.38); the mean EC90 fold change was 0.97 (range, 0.28–3.16). Analysis of variance revealed statistically significant differences between viruses in drug susceptibility relative to R7/3 (P = 0.0007). Notably, viral sample R462F, from a clade A1 virus, was found to be hypersusceptible to GSK744, with an EC50 fold change of 0.17, and this value is lower (more susceptible) than those of several other samples. All other patient-derived samples demonstrated consistent susceptibility to GSK744. As controls, we used 2 viruses with integrase mutations known to confer resistance to first-generation integrase inhibitors: p8070 (G140S, Q148H) and p4736 (E92Q, N155H).16 The fold change for EC50 and EC90 for raltegravir against p8070 and p4736 was 5307 and 194 and 444 and 64, respectively, when compared with the reference strain R7/3. In contrast, the EC50 for GSK744 was about 11-fold higher against p8070 and about 4-fold greater against p4736 when similarly compared with R7/3. These results suggest that similar to dolutegravir, GSK744 demonstrates activity against viral variants highly resistant to first-generation strand transfer inhibitors of HIV-1 integrase.TABLE 1: EC50 and EC90 of HIV-1 Pseudovirus Stocks Containing Integrase Genes From Various Clades, Relative to the Reference Virus, R7/3DISCUSSION To conclude, GSK744, an analog of dolutegravir, is an integrase inhibitor that is amenable to quarterly or every 12-week dosing.6 Our results demonstrate that this drug displays activity against a broad range of transmitted viruses derived from various clades across geographic locations. Given its pharmacokinetic and resistance profile as well as its broad activity, it seems to be a promising candidate as a next-generation PrEP agent. This being the case, intramuscular injections of GSK744 are currently undergoing phase II safety and tolerability testing in anticipation of large phase III efficacy trials. ACKNOWLEDGMENTS The authors thank all of the investigators who provided us with reagents for this study: Clade A1 and D viruses were from Dr. Joshua Baalwa (University of Alabama at Birmingham); Clade C viral clones were from Dr. George M. Shaw and Dr. Beatrice H. Hahn (University of Pennsylvania); Clade AE cDNA were from Dr. Sodsai Touanabutra (The Henry M. Jackso Foundation for the Advancement of Military Medicine, Inc.); and R7/3(X/S)Bsd was from Dr. Mark A. Muesing (ADARC). In addition, the following reagents were obtained through the NIH AIDS Reagent Program, Division of AIDS, NIAID, NIH: TZM-bl from Dr. John C. Kappes, Dr. Xiaoyun Wu, and Tranzyme, Inc., and p8070/M1380133.1 and p4736/H1059422-4 from Dr. Robert Shafer and Elizabeth Reuman, MS. The authors received raltegravir (ISENTRESS/MK-0518) from Merck & Co., Inc. (Whitehouse Station, NJ).