CCR5-tropic Resistance to Maraviroc is Uncommon Even Among Patients on Functional MVC Monotherapy or with Ongoing Low-level Replication
Reported by Jules Levin
CROI Feb 8-12 2009 Montreal
Jubb, M Lewis, P Simpson, C Craig, M Haddrick, M Perros, and M WestbyUK
Pfizer Global Research and Development, Sandwich, UK
115/160 (72%) of patients receiving MVC with a wOBTSS >/= 2 were TLOVR<50 responders with a viral load <50 HIV-1 RNA copies/mL at Week 48.
Excluding non-R5 failures and those without matched MVC susceptibility data, R5 virologic failure occurred in 62/331 (19%) patients with MVC-resistant HIV-1 seen in 22/62 (35%) of these.
Functional monotherapy or a single active NRTI (wOBTSS<1) accounted for 16/22 (73%) MVC resistance-associated failures, and no MVC-resistant virus was found in patients with a wOBTSS of >/= 2 who failed MVC treatment with an R5 tropism result.
16/80 (20%) patients who received functional monotherapy or a single active NRTI experienced MVC resistance-associated R5 virologic failure, and 50/80 (63%) of these patients responded through Week 48.
The low incidence of MVC resistance in patients failing with virologic rebound or who never achieved viral suppression is consistent with a high barrier to resistance in these patients. Virologic failure in these patients is likely to be associated with other factors.
Maraviroc (MVC) is a first-in-class CCR5 antagonist that is approved for use in the management of treatment-experienced patients with CCR5-tropic (R5) HIV-1 infection.
In the Phase 3 MOTIVATE 1 and 2 studies, MVC (QD or BID) plus optimized background therapy (OBT) demonstrated significantly greater virologic and immunologic efficacy and a similar safety profile compared to placebo + OBT at 48 weeks in treatment-experienced patients with R5 virus.1 These findings were durable through Week 96.2
As with other studies of new agents in treatment-experienced patients3,4 preplanned subgroup analyses of the MOTIVATE studies5 showed that more patients who received a higher number of potentially active drugs in their OBT according to baseline susceptibility scores achieved virologic suppression.
Subsequent post-hoc analyses of the MOTIVATE virologic outcomes population have shown that an alternative methodology, weighted OBT susceptibility score (wOBTSS), is a better measure of OBT activity than simply counting active drugs (X4 or D/M).6
Data from the MOTIVATE studies demonstrated that virologic failure in treatment-experienced patients receiving MVC-containing regimens can occur with either MVC-resistant R5 virus or as a result of 'unmasking' of pre-existing CXCR4-using virus.7
MVC resistance is different to resistance to ARVs that act on viral proteins; because MVC acts on a host protein rather than a viral protein the virus cannot mutate to prevent the drug from binding to its target. Rather, MVC resistance arises when the virus mutates to make use of drug-bound CCR5.
Plateaus in maximum drug inhibition, rather than shifts in IC50, are phenotypic markers of MVC resistance.8,9
A signature mutation for MVC has not been identified. V3 loop mutations were identified in viruses that showed a plateau in maximal percentage inhibition below 95%, but the pattern of amino acid changes was different between patients.9,10
Functional monotherapy, where a new agent is combined with an inactive background regimen, is typically associated with rapid emergence of resistance to new agents in treatment-experienced patients.
To investigate the influence of OBT regimen activity, as assessed by the wOBTSS algorithm, on the development of phenotypic resistance to MVC in antiretroviral-experienced patients who had virologic failure with R5 virus.
♦ The virologic outcomes (VO) population previously defined for wOBTSS-based analyses of MOTIVATE6 is a 636 patient subgroup of the full analysis set (Figures 1 and 2).
♦ The VO population excluded patients who failed study treatment through 48 weeks for non-virologic reasons (n=145) and patients for whom a baseline OBT activity score may not have accurately represented the drugs received throughout their period of randomized therapy (n=268).
♦ For the purpose of this substudy of MVC patients experiencing virologic failure with R5 virus, the VO population was further refined to exclude patients randomized to receive placebo (n=118), patients with non-R5 virologic failure (n=163)-including those with non-reportable tropism results and with viral load levels too low (<500 copies/mL) for assay at time of failure-and patients without paired baseline and at-failure phenotypic MVC susceptibility data (n=24).
Virologic response was determined by the TLOVR (time to loss of virologic response) <50 algorithm, which defined responders as those patients who achieved and maintained <50 copies/mL through Week 48.
Three patterns of virologic failure are defined by the algorithm:
♦ Non-responder (NR): Never achieved <50 copies/mL and discontinued before Week 48 for lack of efficacy.
♦ Never suppressed (NS): Never achieved <50 copies/mL but still on study at Week 48.
♦ Rebound (R): Two consecutive viral load measurements _50 copies/mL after a confirmed <50 copies/mL result.
The last active therapy date on study was the Week 48 visit for responders and those who never suppressed, and the date of study discontinuation for rebounders and non-responders.
The wOBTSS algorithm used was:
♦ Drugs in continuous use pre-screening to last active therapy date were scored as 0, since drugs already present in a failing baseline (pre-MOTIVATE) regimen do not contribute to post-baseline responses.
♦ Viral susceptibility or resistance was defined by phenotype only (PhenoSense HIV assay, Monogram Biosciences) or by genotype for enfuvirtide (BC Centre for Excellence in HIV/AIDS, Vancouver, Canada).
♦ Active NRTIs score 0.5 each, other active agents score 1.0. This is based on data showing active nucleosides contributing only about 50% of the virologic activity of drugs from other classes in treatment-experienced patients.11
Phenotypic MVC resistance was assessed in baseline and on-treatment samples at Week 48 or time of virologic failure, as appropriate, by the PhenoSense Entry assay (Monogram Biosciences).
The presence of reduced susceptibility to maraviroc among R5 failure samples was assessed by wOBTSS and by pattern of virologic failure. The analysis was descriptive and no formal statistical testing was performed.
Note: Numbers quoted in these analyses may differ slightly from those in the printed abstract due to a subsequent update to the wOBTSS censoring algorithm.
Virologic outcomes are shown in Figure 3a for the full MVC-treated set of 518 patients, classified by wOBTSS.
Patients who were either TLOVR<50 responders or who experienced virologic failure with R5 HIV-1 and had paired baseline and on-therapy MVC susceptibility data were then studied further (n=331).
Of these, 62/331 (18.7%) patients experienced virologic failure with R5 HIV-1, of whom 22 had virus with evidence of phenotypic resistance to MVC (Figure 3b).
In the analysis of the 331 patients, the percentage of responders increased with higher wOBTSS values; 88% of MVC patients with a wOBTSS >/= 2 were responders, compared to 63% of MVC patients with a wOBTSS <1 (i.e. patients receiving functional monotherapy or one active NRTI) (Figure 4).
The percentage of patients failing with MVC-resistant R5 virus was highest in the subgroup of patients with a wOBTSS <1, while no patient with a wOBTSS >/= 2 failed with MVC-resistant R5 virus (Figure 4).
16 patients failed with MVC-resistant vs 14 patients with MVC-susceptible R5 virus in the functional monotherapy or one active NRTI subgroup (wOBTSS<1).
Of 22 MVC-resistant R5 virologic failures by the TLOVR<50 algorithm, 16/22 (73%) were on functional maraviroc monotherapy or receiving a single active nucleoside in the OBT (wOBTSS <1) with the remaining 6/22 (27%) occurring in patients with an intermediate wOBTSS of 1 or 1.5 (Figure 5a).
By individual TLOVR<50 failure pattern, 15/34 (44%) non-responders (never <50 copies/mL and discontinued) and 7/20 (35%) patients who were never suppressed on treatment but were still on therapy at Week 48 had evidence of MVC resistance (Figure 5b). All eight patients who experienced virologic rebound following suppression to <50 RNA copies/mL had virus that was susceptible to MVC.
Only in the subgroup of patients who had a wOBTSS <1 and were non-responders was R5 virologic failure predominantly associated with MVC resistance (11 patients with MVC-resistant R5 virus vs 6 patients with MVC-susceptible R5 virus) (Figure 6).
1. Gulick RM, et al. N Engl J Med 2008; 359:1429-1441.
2. Hardy WD, et al. HIV9 2008. Presentation O425.
3. Clotet B, et al. Lancet 2007; 369:1169-1178.
4. Cooper DA, et al. N Engl J Med 2008; 359:355-365.
5. Fatkenheur G, et al. N Engl J Med 2008; 359:1442-1455.
6. Valdez H, et al. 48th ICAAC/46th IDSA 2008. Poster H-1221.
7. Lewis M, et al. XVI Int HIV Drug Res Wkshp 2007. Abstract 56.
8. Westby M, et al. J Virol 2007; 81:2359-2371.
9. Mori J, et al. XVI Int HIV Drug Res Wkshp 2007; Abstract 10.
10. Lewis M, et al. 15th CROI 2008; Poster 871.
11. Swanstrom R, et al. J Infect Dis 2004; 190:886-893.