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.

PRIORITY PRESS - Fifth International AIDS Society (IAS)

Cape Town, South Africa / July 19-22, 2009

Due to the massive and rapid production of new virions from infected T lymphocytes, agents that block HIV from entering the host cells have been an attractive concept. The first entry inhibitor, enfuvirtide (T-20), which blocks HIV from attaching to host cells, has proven highly effective in treatment-experienced patients but requires daily injections because of the size of the molecule. Maraviroc (MVC) blocks CCR5, which along with CXCR4, is one of two co-receptors employed by HIV to gain entry into the cell. It is the first agent to validate the concept of entry inhibition through oral administration. The majority of HIV in North America employs the CCR5 co-receptor exclusively for target cell entry. HIV entry mediated by CXCR4 alone is uncommon, but a substantial minority of patients has dual tropic infection capable of using either receptor, making screening necessary.

The Importance of CCR5 Screening

Based on the mechanism, CCR5 tropism is considered a prerequisite for CCR5 inhibitors in standard regimens. Although these drugs may contribute an antiretroviral (ARV) effect even in the presence of CXCR4 when other options have been exhausted, the exclusion of patients with dual tropism has been an entry criterion in the major clinical trials with MVC. In fact, when more sensitive assays provided better exclusion of patients with dual tropic virus, relative efficacy rates improved. CCR5 screening was considered a potential burden for routine use of MVC, but a new study comparing V3 genotyping to Trofile, the previous standard, has found genotyping, which is far faster, easier and cheaper to perform, to be just as sensitive and specific.

“Both assays were comparable in predicting viral response to MVC in treatment-experienced patients. [This] observation was consistent in the subanalysis and responses were maintained over 24 weeks suggesting that V3 genotype is an attractive option for tropism screening to select candidates for maraviroc,” stated Pr. P. Richard Harrigan, British Columbia Centre for Excellence in HIV/AIDS, University of British Columbia, Vancouver.

These are important results because Trofile, a commercial recombinant phenotype assay that has been the standard for testing viral tropism, is expensive and it requires shipment of blood and plasma samples for remote evaluation, delaying the results needed to guide therapy. In contrast, screening of V3-loop sequencing of the viral protein envelope (where determinants of tropism are located) is performed by many laboratories and is already routine at many centres when switching agents of treatment-experienced patients. For those already being genotyped, there is no added cost from this method of verifying CCR5 tropism.

V3-loop Sequencing in the MOTIVATE Studies: A Re-Analysis

The study of the efficacy of V3-loop sequencing was based on a re-examination of blood samples collected in the phase III MOTIVATE (Maraviroc vs. Optimized Therapy In Viremic Antiretroviral Treatment-Experienced patients) trials. The two studies, MOTIVATE 1 and 2, randomized a total of 2,215 treatment-experienced patients to MVC or optimized background therapy (OBT). In the new analysis, V3-loop sequencing and Trofile were compared for their ability to predict response to MVC at 8 and 12 weeks. Although tropism is inferred by V3-loop sequencing, earlier studies have suggested that this approach has good sensitivity for tropism identification and may be at least as accurate as Trofile.

“There was no real large-scale data available to determine which one of these is superior in predicting a real gold standard which is biological outcome after going onto an R5 antagonist,” reported Pr. Harrigan, explaining the motivation for the study. He emphasized that combined data from the MOTIVATE studies provided an adequate sample for a viable comparison.

Independent of any patient characteristic, the accuracy of the screening methods proved to be remarkably similar both for detecting viral phenotype and in predicting virologic response. Both identified CCR5 tropism in 72% of patients. The specificity for predicting virologic suppression at eight weeks was 89% for V3-loop sequencing and 91% for Trofile. The specificity for predicting viral load (VL)<50 copies/mL at 24 weeks was 46.1% for V3-loop sequencing and 46.4% for Trofile (Figure 1). For median VL reduction over 24 weeks, the slopes of the curves for predicted efficacy from the two screening methods overlapped.

Figure 1.

“Early VL reductions in this treatment-experienced population were similar regardless of the assay used, suggesting the potential of genotyping as an accessible assay to select candidates for MVC,” Pr. Harrigan told delegates, indicating that this may substantially facilitate the use of MVC not only in treatment experienced patients but in treatment-naive patients.

Establishing CCR5 Tropism in MERIT

Even though genotyping is not routinely performed prior to initiating ARV therapy, it might be appropriate in some patients, and these results suggest it can now be more inexpensively performed when MVC is being considered. In fact, the value of establishing CCR5 tropism prior to MVC therapy was most strongly demonstrated in a trial for treatment-naive patients called MERIT (Maraviroc vs. Efavirenz Regimens as Initial Therapy). In this trial, patients were randomized to MVC 300 mg b.i.d. or efavirenz (EFV) 600 mg q.d. in combination with a co-formulation of zidovudine (ZDV) and lamivudine (3TC). Although MVC was initially reported to underperform EFV, a subsequent analysis that employed a much more sensitive assay (MERIT-ES) found MVC to be as effective as EFV for those with more accurately defined CCR5 tropism. Specifically, the proportion of patients achieving a VL <50 copies/mL at 48 weeks was 68% in both groups. For those who initiated therapy with a VL >100,000 copies/mL, the proportion with a VL <50 copies/mL was 64% and 63% for MVC and EFV, respectively. Immune reconstitution as measured by the increase in CD4+ cell count favoured MVC (170 vs. 144 cells/mm³).

As heard here at the IAS, there are a number of potential advantages for entry inhibitors. For example, preventing HIV from entering host cells to replicate eliminates the opportunity for new resistance mutations. In addition, the high degree of safety and tolerability of these agents may be due to their extracellular activity. With six years of clinical experience with T-20 and two years of experience with MVC, neither has been found to significantly influence lipid or glucose metabolism, affect liver or renal function, or impose other systemic activities that raise concern about the potential for a cumulative risk. The extracellular activity may also explain the absence of interactions with other ARV agents, an important attribute for an agent included in sustained combination regimens.

Summary

Entry inhibitors, which already serve as important options in the control of HIV, are expected to assume a more central role with the development of a more efficient method of selecting candidates for the oral agent. Although screening patients for CCR5 is an important prerequisite for the routine use of MVC, the V3-loop sequencing is a sensitive tool for identifying CCR5-only virus and predicting response to MVC. It is a relatively simple and inexpensive test to perform. For patients who are already undergoing genotyping to direct second-line therapy, this approach will eliminate extra steps and cost for evaluating susceptibility to MVC. It will also reduce the turn-around time for evaluating this strategy. Facilitating use of oral entry inhibitors is an important goal in maximizing the number of options available to maintain HIV VL below a level at which it can generate resistance mutations.