Reports

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

NEW FRONTIERS - 48th Annual Interscience Conference on Antimicrobial Agents and Chemotherapy/46th Annual Meeting of the Infectious Disease Society of America

Washington, DC / October 25-28, 2008

The ACTG5202 trial looking at the safety and efficacy of four different highly-active antiretroviral therapy (HAART) regimens in treatment-naïve HIV-infected adults raised questions about the comparative efficacy of regimens containing the combination of abacavir (ABC) and lamivudine (3TC). Sponsored by the U.S. National Institute of Allergy and Infectious Diseases, ACTG5202 is a randomized, partially blinded trial comparing four combinations of efavirenz, co-formulations of emtricitabine and tenofovir disoproxil fumarate (FTC/TDF) or ABC/3TC, and ritonavir-boosted atazanavir (atazanavir/r).

Investigators enrolled 1858 eligible men and women into two cohorts based on HIV viral loads at baseline: <100,000 or <u>></u>100,000 copies/mL. Patients were randomized to one of four trial arms: efavirenz, FTC/TDF, and placebo for ABC/3TC; efavirenz, placebo for FTC/TDF, and ABC/3TC; atazanavir/r, FTC/TDF, and placebo for ABC/3TC; or atazanavir/r, placebo for FTC/TDF, and ABC/3TC.

The regimens were comparably effective in most patients, but in February 2008, the data and safety monitoring board recommended an emendation to the ongoing trial based on a suggestion of more virologic failures among the subgroup of patients with higher baseline viral loads (<u>></u>100,000 copies/mL) who were started on regimens containing ABC/3TC compared with other combinations.

ARIES Trial Data

Data from other clinical trials, however, indicated a robust virologic response with ABC/3TC in patients with both high and low viral loads at baseline. For example, in 36-week results from the ARIES trial, the combination of ABC/3TC plus atazanavir/r showed potent antiviral activity through 36 weeks of therapy in treatment-naïve patients with viral loads <u>></u>1000 copies/mL at baseline.

The ARIES trial is evaluating the effectiveness and safety of a simplified treatment scheme involving induction with ABC/3TC plus atazanavir/r, followed by ABC/3TC maintenance and boosted or unboosted atazanavir. The trial was enrolled concurrently with the ACTG5202 study, and uses one of the regimens evaluated in that trial.

The investigators enrolled 515 antiretroviral therapy-naïve patients with HIV-1 infections who were negative for the HLA-B*5701 ABC hypersensitivity allele, and who had <u>></u>1000 copies/mL at screening and any CD4+ cell count. The patients were enrolled in an open-label fashion to the ABC/3TC plus atazanavr/r regimen, followed by randomization to either remain on boosted atazanavir or to drop the ritonavir.

The proportion of patients achieving <50 copies/mL at week 84 by a time to loss of virological response (TLOVR) analysis is the study’s primary end point. Virologic failure is defined as the failure to achieve <400 copies/mL by week 30, or a confirmed rebound of <u>></u>400 copies/mL after achieving <400 copies/mL. Investigators also conducted a second analysis using the ACTG5202 definition of failure as <u>></u>1000 copies/mL at or after week 16 and before week 24, or confirmed rebound <u>></u>200 copies/mL at week 24.

Good Biologic Responses

“In terms of efficacy, we saw very nice biologic responses through the first 36 weeks, with close to 90% of the patients responding by the intent-to-treat (ITT) and the TLOVR analysis,” reported lead author Dr. Kathleen E. Squires, Director, Division of Infectious Diseases and Environmental Medicine, Jefferson Medical College, Philadelphia, Pennsylvania.

They found that in an ITT population in the TLOVR analysis, 80% of patients overall had <50 copies/mL at 36 weeks, including 84% of those with a baseline viral load <100,000 copies/mL and 76% of those with baseline <u>></u>100,000 copies/mL. Nearly identical numbers were seen for the proportions of patients <200 and <400 copies/mL.

Using the ACTG5202 virological failure criteria, 97% of all patients had successful viral suppression, including 98% of those with lower baseline RNA counts and 95% of those with high baseline counts.

A total of 15 patients (3%) had virologic failure according to the trial protocol, with five patients failing to achieve <400 copies/mL by week 36, and 10 having confirmed rebound after dropping below 400 copies/mL. Eleven of the 15 were either non-compliant or had treatment interruption.

“After the result of the 5202 study came out—because this is one of the arms in that study and because almost 60% of patients had baseline viral loads greater than 100,000 copies/mL—we used ACTG5202 virologic failure definitions, and looked at 36 weeks at the proportion of patients achieving less than 50 copies in the under-100,000 and greater-than-100,000 groups. The outcomes were very good and not substantially different by baseline stratification,” Dr. Squires noted.

Grade 2-4 treatment-related adverse events occurred in 28% of patients, with hyperbilirubinemia being the most common, occurring in 13% of patients. Elevations in bilirubin are commonly seen in patients on atazanavir, Dr. Squires reminded delegates. There were four suspected cases of ABC hypersensitivity reactions, but skin-patch testing in all four cases were negative for hypersensitivity, she added.

Other Study Findings

Dr. Keith Pappa, Research Triangle Park, North Carolina, and colleagues presented data from six ABC/3TC clinical trials showing that in patients with baseline viral loads above and below 100,000 copies/mL, using ACTG5202 end points, virologic responses were strong and consistent between low and high viral load strata. In the studies, 87% to 95% of patients on ABC/3TC did not experience virologic failure after 48 weeks,

In addition, the safety end point outcome—time to first grade 3 or 4 sign, symptom, or laboratory anomaly—was similar regardless of the patient’s baseline viral load strata.

One of the studies included in the combined analysis was the HEAT trial, the first head-to-head study comparing the dual nucleoside reverse transcriptase inhibitor backbones ABC/3TC and TDF/FTC in combination with lopinavir/r. Dr. Benjamin Young, Denver Infectious Disease Consultants, Colorado, and colleagues presented a study characterizing virologic failure over 96 weeks in the HEAT study. They found that overall virologic failure rates were similar across both treatment arms, and that more patients with baseline <100,000 copies/mL who were on TDF/FTC experienced more virologic failure than patients on ABC/3TC, at 63% vs. 41%, respectively. The investigators did find, however, that among patients with <u>></u>100,000 copies/mL, more of those receiving ABC/3TC experienced virologic failure, at 59%, compared with 37% for TDF/FTC. The patients most likely to experience virologic failure were African Americans, the author noted, a finding revealed in both univariate and multivariate analyses predicting virologic failure. They also found that among patients who had virologic failure with treatment-emergent resistance mutations, mutations at the 184V codon were more common among patients on TDF/FTC than among patients who received ABC/3TC.

In their combined six-trial analysis, Dr. Pappa and colleagues suggested that the differences in high virologic load strata in ACTG5202 and their analysis could be due to differences in the third drugs used, differences in how the study and follow-up were conducted, use of end points different from those used in HEAT and most other HIV studies, and variables such as potential treatment interruptions, possible differences in patient adherence, and lack of resistance testing in some patients in ACTG5202 (40% of patients had no baseline genotyping). They also noted that when ACTG5202 end points were applied to HEAT data, the same magnitude of difference was not seen.

Hepatitis C Co-Infection

Also presented at the 2008 ICAAC/IDSA joint meeting was a study suggesting that contrary to two recent reports, ABC-containing regimens do not appear to have a negative interaction with pegylated interferons in patients with HIV and hepatitis C (HCV) co-infections.

Investigators from the University of Maryland School of Medicine in Baltimore observed HCV treatment responses among 27 HIV/HCV co-infections treated either with or without ABC, ritonavir, or any protease inhibitor (PI). They found that patients on ritonavir-containing regimens had lower early viral response and sustained viral response rates than patients on non-ritonavir-containing regimens. There were no differences in early or sustained viral responses among patients on ABC vs. non-ABC-containing regimens, however.

“This is limited by the small sample size, but our most interesting finding was that none of our patients who were on boosted PIs had a sustained virological response,” stated lead author Dr. Rohit Talwani, Assistant Professor of Medicine, University of Maryland. He suggested that it is possible that PIs interfere with proteasomes involved in regulatory cytokine signalling pathways, which in turn might have a negative affect on interferon-based therapy.

Summary

The disparate findings of the studies presented reinforce the idea that the key to successful initiation of HAART in treatment-naïve patients infected with HIV involves consideration of myriad factors such as viral load, mutations, co-infections, overall health status, and sociodemographic factors that may affect compliance. Ultimately, no single regimen or drug combination will be right for all patients.