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.

15th Conference on Retroviruses and Opportunistic Infections

Boston, Massachusetts / February 3-6, 2008

Based on objective studies, several treatment guidelines, including the latest from the Department of Health and Human Services (DHHS), have identified abacavir/lamivudine (ABC/3TC) and tenofovir/emtricitabine (TDF/FTC) to be preferred nucleoside/nucleotide reverse transcriptase inhibitor (NRTI) backbone combinations. Either is appropriate in combination with the currently accepted first-line protease inhibitors (PIs), which are ritonavir-boosted lopinavir (LPV/r), fosamprenavir (FPV/r) or atazanavir (ATV/r), or efavirenz, the only first-line non-NRTI. However, there are limited data with which to compare these NRTI combinations for their relative effect on important outcomes, such as virologic failure.

HEAT Study Design

In newly released 48-week data from a double-blind, randomized trial called HEAT (Head-to-Head Epzicom and Truvada), the ABC/3TC arm was associated with slightly greater immunologic recovery and showed non-inferiority for virologic control.

“ABC/3TC is comparable to TDF/FTC in virologic efficacy when combined with LPV/r through 48 weeks,” reported Dr. Kimberly Y. Smith, Associate Professor of Medicine, Rush University Medical Center, Chicago, Illinois. “Both treatment regimens were well tolerated with few discontinuations due to adverse events in either arm.”

In this phase IV segment, 688 HIV patients naïve to treatment were randomized at 78 sites in the U.S. and Puerto Rico to a LPV/r once-daily regimen that contained ABC/3TC or TDF/FTC. Each arm received a placebo of the opposite NRTI co-formulation. Although there were no restrictions for baseline viral load (other than HIV RNA >1000 copies/mL) or CD4+ cell count, patients were stratified for baseline viral load less than 100,000 copies/mL or greater. Importantly, no screening for HLA-B*5701, which detects ABC sensitivity, was allowed, and all comparisons were conducted on an intention-to-treat (ITT) analysis. In the BICOMBO study, a comparison of these two NRTI co-formulations in combination with efavirenz was criticized for several design features that complicated analysis.

The primary objective of the present study was to evaluate virologic non-inferiority for ABC/3TC vs. TDF/FTC at 48 weeks and to compare safety and tolerability over 96 weeks. In the newly released 48-week data, the proportion of patients with a viral load <50 HIV RNA copies/mL was 68% for ABC/3TC and 67% for TDF/FTC. When compared for viral load <400 HIV RNA copies/mL, the percentages for ABC/3TC and TDF/FTC were 75% and 71%, respectively. The median CD4+ cell count at week 48 was 429 cells/mL for ABC/3TC and 370 cells/mL for TDF/FTC, which represented an increase of 201 cells and 173 cells, respectively.

By definition of non-inferiority in this study, ABC/3TC not only had to be statistically comparable to TDF/FTC, but the lower boundary of the 95% confidence interval (CI) was not permitted to be more than 12% different from that of TDF/FTC, according to Dr. Smith. In fact, ABC/3TC provided a non-significant 1% absolute advantage over TDF/FTC, and the lower boundary of the 95% CI was less than 8% lower. Noting that the comparison was made with a strict ITT analysis in which missing data equaled failure and switches were included Dr. Smith reported, “The margin for non-inferiority was not equivocal.”

HEAT Findings

At the end of 48 weeks, 80% of the ABC/3TC patients and 76% of the TDF/FTC patients remained on treatment. Neither this result nor the differences in the reasons for premature withdrawal reached statistical significance. The low rate of discontinuations for adverse events, 4% with ABC/3TC and 6% with TDF/FTC, was reassuring. All of the ABC/3TC switches were due to suspected ABC hypersensitivity reactions vs. 1% of TDF/FTC patients. Proximal renal tube dysfunction was observed in 1% of TDF/FTC patients vs. 0% for ABC/3TC. In both groups, 1% discontinued therapy due to virologic failure. The rate of non-compliance for ABC/3TC and TDF/FTC was 2% and 3%, respectively. The rates of loss to follow-up (8% vs. 9%), protocol violation (1% vs. 0%), subject decision (3% vs. 4%) and other (2%, both groups) were similar. Not all virologic failures, which occurred in 12% of ABC/3TC patients and 11% of TDF/FTC patients, led to departure from the study.

One of the notable findings from the study with potential clinical implications was the difference in treatment-emergent mutations. Overall, these were more common in the TDF/FTC arm (53% vs. 34%) with a particularly increased rate of NRTI-associated mutations (44% vs. 20%). Most of these involved a M184V or a mixture of mutations containing M184V in the TDF/FTC arm.

“Twice as many subjects receiving TDF/FTC failed therapy with an M184V or mixture, which is a finding not previously reported,” Dr. Smith observed. She suggested that further analyses of this difference are needed for considering their relevance to drug selection in respect to preserving NRTI options in future regimens.

In this study, slightly more than one-third of the subjects were African-American and almost 20% were women. The mean age was 38 and the median baseline viral load was approximately 4.9 log10 copies/mL. Only 14% of the ABC/3TC patients and 10% of TDF/FTC patients had either hepatitis B or C. About half of patients had a CD4+ cell count of 200 cells/mL or greater. Less than 20% had baseline CD4+ cell counts less than 50 cells/mL. The two study arms were well matched for all of these variables.

The specific treatment-related side effects in the two study arms, like the proportion of patients who discontinued therapy for adverse events, were very similar. In both arms, diarrhea was the most common grade 2-4 adverse event, affecting 18% of those randomized to ABC/3TC and 19% of those randomized to TDF/FTC. No other grade 2-4 side effect was observed in more than 10% of patients in either arm, and all incidences were similar between the two study cohorts. When patients were compared for lipid changes, elevations in total cholesterol, triglycerides and HDL-C were observed in both arms, leading to a very modest reduction in the total cholesterol:HDL-C ratio (-0.26 vs. -0.56 for ABC/3TC vs. TDF/FTC, respectively).

BICOMBO: Nucleoside Backbones and a Non-NRTI

Although no major multicentre, randomized comparison of the two preferred NRTI regimens has been conducted previously in the context of a PI backbone, the same co-formulations have been compared in combination with efavirenz, as mentioned by Dr. Smith. The results of the BICOMBO study were presented July 2007 at the International AIDS Society conference in Sydney, Australia, by Martinez et al. (Abstract WESS102). In this multicentre study, ABC/3TC was non-inferior to TDF/FTC for virologic efficacy, but the ABC/3TC arm was reported inferior for treatment efficacy. This latter result was primarily due to an unusually high number of ABC/3TC patients suspected of ABC hypersensitivity. However, a retrospective analysis with HLA B*5701 testing confirmed ABC hypersensitivity in only one-third of these subjects. The study was also criticized for several design features, including randomization after a pre-treatment period with a regimen containing 3TC and using an end point of <200 HIV RNA copies/mL instead of the more sensitive <50 copies/mL.

Due to these design choices, there has been some debate about whether this served as a true direct comparison of the preferred NRTI co-formulations in treatment-naïve patients receiving a non-NRTI. Many of these problems in BICOMBO were specifically avoided in the HEAT study, which employed a blinded analysis of the 48-week data to protect the integrity of the 96-week comparisons.

“There are very limited direct comparative data between the recommended dual NRTI fixed-dose combinations, particularly using strict ITT criteria,” Dr. Smith noted. In HEAT, “consistent results were observed regardless of analysis method used, including when switches were counted as failures.”

Summary

The first comparison of the preferred NRTI combinations, ABC/3TC and TDF/FTC, has demonstrated non-inferiority for the critical end point of <50 copies/mL. Although the current DHHS recommendations suggest that ABC/3TC should be employed in those screened for ABC hypersensitivity with HLA B*5701, the study compared these two backbone NRTI combinations in patients without such screening. Although there was a notable difference in the types of resistance mutations observed at the end of the study, outcomes such as viral suppression, virologic failure, immunologic recovery and safety indicated that the regimens were comparable.