icon-folder.gif   Conference Reports for NATAP  
48th Annual ICAAC / IDSA 46th Annual Meeting
October 25-28, 2008
Washington, DC
Back grey_arrow_rt.gif
An enhanced assay to refine the use of CCR5 inhibitors
  David Margolis, MD
University of North Carolina
Chemokine receptors (CCRs) serve to regulate cellular responses by signaling in response to chemokines, the hormones of the immune system. HIV has co-opted the CCR system to gain access to CD4+ cells that are its preferred targets. In addition to binding a CD4 receptor to infect a cell, HIV also requires interaction with a CCR. Of the large family of CCRs, CCR5 is the CCR most often used by HIV.
New antivirals such as Maraviroc, licensed in last year, and Vicriviroc, under development, block HIV replication by blocking the use of CCR5 by HIV. However, although CCR5 is the preferred binding partner used by HIV to enter cells, some viruses are adapted to use alternate receptors, most often CXCR4. Why HIV prefers to use CCR5 is not completely clear, but virus that is CCR5-tropic is almost always the virus that establishes initial HIV infection, and dominates in most patients, sometimes for the entire course of disease. CXCR4-tropic viruses predominate In only about half the patients with late-stage disease, but may exist at low levels in some patients. Unlike drug resistance mutations, the envelopes of viruses that use CCR5 or CXCR4, or can use both (dual-tropic) are substantially different, and may only evolve after multiple, gradual changes, or perhaps more rapidly through viral recombination. However, such viruses appear, if a portion of the viral population growing in a patient is adapted to use CXCR4, it will be intrinsically resistant HIV CCR5 inhibitors. Treatment guidelines direct that a test to detect the presence or absence of viruses that can use CXCR4 vs. CCR5 be used to guide therapy when a CCR5 inhibitor is considered.
The Trofile assay from Monogram Biosciences is approved for clinical use to detect the presence of CCR5-tropic, CXCR4-tropic, or dual/mixed virus (which is either a mix of the two virus populations or "dual tropic" virus that can use either receptor). The originally licensed Trofile assay has been replaced by Trofile-ES, an improved version of the assay. The Trofile ES assay can detect X4 virus, resistant to CCR5 inhibitors, when they constitute a minor subpopulation of virus within a swarm of CCR5-using virus. Several studies presented at ICAAC/IDSA illustrated the potential benefit of the use of the newer, more sensitive test.
Laboratory validation data for this new assay was presented in a poster by Trihn and coworkers from Monogram (poster H-1219). Trofile ES increased detection sensitivity for CXCR4-using viruses by an average of 30-fold. 100% of the Trofile ES assays run detected the presence of X4 viruse when it was present in a viral population at a frequency of 0.3%. X4 detection was sometimes successful at frequencies down to 0.003%. Trofile (ES) accurately determined the tropism of 46 patient samples and isolates representing multiple subtypes including 14/18 with X4 variants below the detection limit of standard Trofile based on clonal analyses. Intra-assay precision (100%) and inter-assay reproducibility (99%) were demonstrated by concordant results for 135/135 and 228/230 pair-wise comparisons of R5, X4 and DM Envs and repeat testing of 46 patient Env populations, respectively.
Zhaohui Su presented the effect of the use of Trofile ES on a reanalysis of ACTG 5211, a a double blind, randomized phase 2 study of vicriviroc in treatment-experienced, patients (abstr. H-895). Using the original Trofile assay, patients were screened for the presence of dual/mixed or X4-tropic virus (DM), and only those with CCR5-tropic virus (R5) allowed to enter the study. Patients were treated with various doses of vicriviroc for 14 days in addition to their ongoing, failing antiviral therapy, after which the background antiretroviral therapy (ART) was optimized using resistance testing [Gulick JID 2007]. 118 patients with a median HIV-1 RNA level of 36,380 copies/mL and a median CD4 cell count of 146 cells/Ál were treated. At 14 days and 24 weeks, patients treated with vicriviroc had viral load declines of up to -1.86 logs greater than those that received placebo.
Su and colleagues analyzed the potential benefit of the use of the more sensitive Trofile ES; hypothesizing that better results would be obtained if patients with low-level X4 or dual/mixed HIV had been excluded from the study. They found that of the 114 patients whose samples could be retested, 25 of these had previously undetected X4 or dual/mixed HIV by the "old" Trofile. That is, 22% were found who were allowed into the vicriviroc study using the old Trofile, but would have been excluded by Trofile ES.
Of the 114 patients rescreened, 84 received vicriviroc in the study, and of these:
· 64 patients had R5 virus by Trofile ES at both screening and when retested at study entry. These patients had HIV RNA declines of -1.15 and -1.95 logs at day 4 of vicriviroc and after 24 weeks (background ART optimized at week 4).
· 5 patients had R5 virus at screening but DM virus at study entry. This finding does not necessarily mean that the size of the DM population substantially changed between screening and entry, but (like viral load measurements) the size of the population was more likely near the limit of detection of the assay, and found at one time but missed at another. These patients had HIV RNA declines of only -0.66 logs at day 4, but -1.20 logs at week 24.
· 15 patients had DM virus by Trofile ES at screening, and had no change in HIV RNA (change of -0.09 logs) at day 4, and only -0.57 logs at week 24.
These viral load declines were only significantly different when the 64 R5 patients were compared to the 15 DM patients, but the intermediate changes in the 5 patients with DM only at entry were not statistically different from the other groups.
Although the point is somewhat moot, as only Trofile ES is now available, this result shows that Trofile ES increases the detection of DM virus, and identifies a population who are likely to respond poorly to R5 inhibitor drugs. A few caveats bear mentioning. Even with the more sensitive assay, there appears to still be a population of patients with DM virus near the (now lower) level of detection of the Trofile ES assay. Such patients might respond to new optimized therapy given simultaneously with the additon of an R5 inhibitor. It is difficult to know if vicriviroc contributed anything to the -1.20 log decline of viremia in this handful of patients at week 24, or if this decline was due simply to potent activity of optimized background therapy. It should also be mentioned that this study design exposed patients to vicriviroc as a single new agent for 14 days, likely decreasing the durability of effect of the R5 inhibitor. However, is likely impractical to repeat Trofile assays in clinical practice in order to determine if patients belong to this small group.
Michael Saag presented a similar analysis (abstract H-1232a) of the MERIT study. This was a study of maraviroc bid (MVC) vs efavirenz qd (EFV), both with Combivir as initial therapy for ART-naive patients. In patients screened with the "old" Trofile, MVC failed to prove noninferiority to EFV for the endpoint of suppression to <50 copies/mL at week 48 (by intention to treat analysis), that is it could not be proven that MVC was "at least as good as" EFV in this study. Nearly half of virologic failures in the MVC arm were found to have CXCR4-using at failure.
Again, screening samples were reanalyzed to see if DM virus was present at screening and missed by the older Trofile assay. Trofile ES found that 107 of the 721 patients that could be rescreened had DM virus that was not detected by the original Trofile. This is nearly 15% of the study population, similar to the 22% of DM detected in failing, treatment-experienced population in ACTG 5211.
The table shows some of the results presented. All patients with a CCR5-tropic screening result by Trofile ES who received at least one dose of MVC BID or EFV were included; missing values are classified as failures/non-responders. The differences shown between MVC and EFV when Trofile ES was used to exclude patients are noninferior as defined by a lower inferiority bound of 1-sided 97.5% confidence interval of more than 10%.


So in this study of the use of MVC in ART-na´ve patients, the use of the more sensitive Trofile ES identified more patients with low levels of DM virus at entry. Excluding these patients improved the apparent efficacy of MVC bid such that it then appeared non-inferior to qd EFV when paired with Combivir. As seen in prior studies, the increase in CD4 cells seen following MVC therapy appeared to be slightly greater; it is not yet clear if there is a clinical benefit or significance to this.
So where does this leave us in regard to the use of CCR5 inhibitors and testing for the presence of clinically significant populations of X4 or DM virus that are resistant to CCR5 inhibitors? Assays of increasing sensitivity such as the Trofile-ES detect smaller populations of dual/mixed-tropic HIV in patients. The use of these more sensitive assays identify patients who are more likely to successfully achieve durable suppression of viremia, at least to 24 weeks in the setting for drug resistant HIV and 2nd/3rd-line therapy, and at least to 48 weeks in the case of initial therapy. As the proportion of patients discovered to have dual/mixed-tropic HIV was modest, the improvement in proportion of patients with clinical response was also modest. However, given the total costs of therapy and failure of therapy, this investment seems to be reasonable to guide the use of CCR5 inhibitors.
At the meeting, a member of the audience questioned the use of more sensitive assays, as they might exclude patients with X4/DM populations who would derive "residual benefit" from the use of MVC or another such drug. This might be the case for some patients with drug-resistant virus and few therapeutic options, but is clearly not the case in earlier use of MVC or other CCR5 inhibitors. Given the many options for early therapy, it seems currently unwise to use MVC without screening with the most sensitive test available. There is as yet no proven clinical benefit for this class of drugs over and above other classes, if equivalent viral suppression is obtained.
In the setting of drug resistance, a clinician should wish to utilize at least two drugs that were as active as possible. One can envision a desperate scenario in which MVC might be used in as part of "mega-ART" despite the presence X4/DM populations, but this would hopefully be an uncommon occurrence. As ever, careful clinical judgment should be used.
Su Z. et al. Response to vicriviroc (VCV) in HIV-infected treatment-experienced subjects using an enhanced Trofile HIV co-receptor tropism assay: Reanalysis of ACTG 5211 results. 48th Interscience Conference on Antimicrobial Agents and Chemotherapy, abstract H-895, Washington, 2008.
Saag M. et al. Reanalysis of the MERIT study with the enhanced Trofile assay (MERIT-ES). 48th Interscience Conference on Antimicrobial Agents and Chemotherapy, poster abstract H-1232a, Washington, 2008.
Trinh L. et al. Validation of an Enhanced Sensitivity Trofile˘ HIV-1 Co-receptor Tropism Assay, 48th Interscience Conference on Antimicrobial Agents and Chemotherapy, abstract H-1219, Washington, 2008.
Gulick RM et al. Phase 2 study of the safety and efficacy of vicriviroc, a CCR5 inhibitor, in HIV-1-Infected, treatment-experienced patients: AIDS clinical trials group 5211. J Infect Dis. 2007; 196:304-12