HIV attachment inhibitor BMS-663068 safe and effective in phase 2b study

Bristol-Myers Squibb's BMS-663068 or fostemsavir, a first-in-class HIV attachment inhibitor that stops the virus from binding to and entering cells, was well-tolerated and demonstrated good antiviral activity in a study presented at the Conference on Retroviruses and Opportunistic Infections (CROI 2015) last week in Seattle, USA. Related research showed that BMS-663068 can safely be taken with antiretrovirals commonly used by treatment-experienced patients. A phase 3 trial is now underway.

Combination antiretroviral therapy (ART) consists of drugs that target different steps of the HIV lifecycle. None of the currently approved agents block the very first step, initial attachment of the virus to a host cell. Drugs that work in novel ways could be particularly beneficial for highly treatment-experienced people with extensively resistant HIV.

BMS-663068 is a pro-drug or precursor of BMS-626529, which binds directly to the gp120 protein that makes up part of the 'spikes' on HIV's outer surface, thereby preventing viral attachment and entry into CD4 T-cells. CCR5 blockers like maraviroc (Celsentri/Selzentry) and fusion inhibitors like enfuvirtide (Fuzeon) work at slightly later steps; BMS-663068 is active regardless of whether an HIV strain uses CCR5 or CXCR4 co-receptors.

Melanie Thompson from the AIDS Research Consortium of Atlanta and colleagues conducted a phase 2b trial to investigate the safety, efficacy, and dose-response characteristics of BMS-663068 in treatment-experienced people with HIV.

This study included 254 randomised participants. With sites in South Africa and other middle-income countries, it had a higher proportion of women and non-white people than many antiretroviral drug trials. A majority (60%) were men, just over 30% were white, 30% were black and the median age was 39 years. Two-thirds had HIV subtype B.

At study entry, participants had HIV viral load of at least 1000 copies/ml, with about 40% having high viral loads above 100,000 copies/ml. Overall they had relatively advanced disease, with a mean CD4 count of approximately 230 cells/mm³ and nearly 40% having less than 200 cells/mm³.

Many participants had failed first- or second-line ART and about half had at least one major mutation conferring resistance to at least one widely used antiretroviral drug class. They were, however, required to still be sensitive to raltegravir (Isentress), tenofovir (Viread) and the comparator drug atazanavir (Reyataz). Pre-treatment phenotypic screening was done to ensure that their HIV was likely to be susceptible to BMS-626529.

Study participants were randomly allocated to five treatment arms. Four groups received BMS-663068 at doses of 400mg or 800mg twice daily, or 600mg or 1200mg once daily, while a control group received ritonavir-boosted atazanavir. Everyone also took tenofovir and raltegravir.

Results from the first 24 weeks of therapy were presented at last year's CROI, showing that BMS-663068 was generally safe and well-tolerated and demonstrated viral suppression rates across doses similar to those seen with atazanavir. This year's poster presentation described 48-week results.

At 48 weeks, all treatment arms had statistically similar virological response rates, with 61-82% of people taking BMS-663068 and 71% of those taking atazanavir/ritonavir achieving HIV RNA <50 copies/ml in a modified intent-to-treat 'snapshot' analysis. In an observed or as-treated analysis, BMS-663068 response rates rose to 69-91%. BMS-663068 response rates were somewhat higher for people starting treatment with low compared to high viral load. Mean CD4 cell gains were similar, in the 141-199 cells/mm³ range across all arms.

BMS-663068 was again generally safe and well-tolerated at all doses tested. A total of seven people discontinued treatment early due to adverse events, but none of these were deemed related to BMS-663068. The most common side-effects in the BMS-663068 arms were headache (one person in the 800mg arm) and abdominal pain (one person in the 1200mg arm). No notable trends in laboratory abnormalities were seen across the BMS-663068 arms. Atazanavir/ritonavir was associated with more adverse events and abnormalities, including elevated bilirubin.

"Virologic response rates...and immunologic responses appear to be generally similar across the BMS-663068 and atazanavir/ritonavir arms through week 48," the researchers concluded. "All BMS-663068 doses were generally well tolerated with no dose-response safety signals reported."

As a potential limitation, they noted that participants receiving BMS-663068 had a higher daily pill burden than those taking atazanavir/ritonavir, which could have an effect on adherence. In addition, the phenotypic assay used to determine BMS-663068 susceptibility was not able to provide results for about one-quarter of participants.

Given these promising results, a phase 3 clinical trial of BMS-663068 was started on February 23, according to a recent BMS press release. It will enrol highly treatment-experienced patients, defined as people who can no longer construct a viable standard antiretroviral regimen due to accumulation of drug resistance, past intolerability or contraindications. This study will enrol patients regardless of BMS-663068 susceptibility; a retrospective analysis will be done to determine whether a companion phenotypic assay is necessary.

Related studies

As described in a related poster at CROI, Ishani Savant Landry from BMS and colleagues analysed pharmacokinetic/pharmacodynamic data from more than 200 participants taking BMS-663068 in two phase 2 studies, including the one described above, in which a subgroup of patients took BMS-663068 as monotherapy for a 7-day 'lead-in' before starting combination therapy.

The researchers saw a relationship between BMS-663068 drug concentrations and antiviral activity, with no evidence of safety signals through 48 weeks. Baseline viral drug susceptibility was the strongest factor determining the magnitude of HIV RNA decline during monotherapy.

Simulation studies suggested that twice-daily BMS-663068 dosing conferred a slightly higher likelihood of achieving a greater than 1-log₁₀ decline in HIV RNA compared with once-daily dosing, although all tested doses taken either once or twice daily lowered viral load by at least 0.5 log₁₀.

A 600mg twice-daily dose was predicted to have the best risk-benefit profile. This dose maintained a higher steady level over time, but with a lower maximum or peak concentration than 1200mg taken once daily, which the researchers suggested would minimise the risk of QTc interval prolongation (a type of heart rhythm abnormality) if co-administered with ritonavir.

Landry's team also evaluated potential drug-drug interactions between BMS-663068 and ritonavir-boosted darunavir (Prezista) and the NNRTI etravirine (Intelence). The CYP3A4 pathway plays a role in metabolising the active drug BMS-626529, so there is potential for interactions that raise or lower drug levels.

This analysis of 42 healthy HIV-negative volunteers showed that co-administration of 600mg twice-daily BMS-663068 with darunavir/ritonavir increased BMS-626529 concentrations by 50-90%, consistent with ritonavir's inhibitory effect on CYP3A. Etravirine co-administration, in contrast, decreased BMS-626529 levels by 50% as it has a CYP3A-inducing effect. Taking all three drugs together led to smaller BMS-626529 increases of 30-50%, as the inhibitory effect of ritonavir predominated over the inducing effect of etravirine. BMS-626529, in turn, had little effect on levels of etravirine or darunavir/ritonavir.

Based on these findings, the researchers concluded that BMS-663068 can be safely co-administered with darunavir/ritonavir, etravirine, or both without the need for dose adjustment. Darunavir and etravirine (both produced by Janssen) are widely used by treatment-experienced HIV patients with drug resistance and may be potential partners in a combination ART regimen.