icon-folder.gif   Conference Reports for NATAP  
  18th HIV Drug Resistance Workshop
June 9-12 2009
Ft Myers Florida
Back grey_arrow_rt.gif
Weighted Genotype and Phenotypic Scores Predict Response to Maraviroc
  Mark Mascolini
XVIII International Drug Resistance Workshop, June 9-13, 2009, Fort Myers, Florida
Weighted genotypic and phenotypic scores estimating the activity of background regimen drugs added to maraviroc in the MOTIVATE trials both proved strong predictors of response to the CCR5 antagonist after 48 weeks of treatment [1]. "Weighting" assigns different values to different antiretroviral classes (in this analysis protease inhibitors [PIs], nonnucleosides [NNRTIs], nucleosides [NRTIs], and enfuvirtide), depending on whether a patient's virus is sensitive to those agents according to genotypic or phenotypic rules.
This analysis by Jonathan Schapiro (National Hemophilia Center, Tel Hashomer, Israel) and colleagues determined that more than 70% of people with a weighted genotypic or phenotypic score of 2 or more had a viral load under 50 copies at week 48 of the MOTIVATE studies. People with either score at 2 or higher and a starting CD4 count above 50 had about an 80% sub-50 response rate at week 48. MOTIVATE 1 and 2 randomized people with heavy treatment experience to maraviroc or placebo plus optimized background therapy (OBT) that could not include darunavir, etravirine, or raltegravir [2]. At week 48 of these trials, just under half of people taking maraviroc had a viral load under 50 copies, compared with fewer than 20% in the placebo arms.
The weighted-score analysis included 597 of the original 1049 MOTIVATE participants, after the investigators eliminated patients who stopped treatment for nonvirologic reasons, people without matching genotypic and phenotypic data, and people whose OBT score did not accurately reflect the regimen they took throughout the trial Any antiretroviral, regardless of class, got scored 0 if it continued from the failing regimen a person was taking at enrollment into MOTIVATE. In the phenotypic score, any antiretroviral with a fold-change in susceptibility above the PhenoSense clinical cutoff for that drug scored 0. PIs, NNRTIs, and enfuvirtide scored 1 if they were below that cutoff, and NRTIs score 0.5 if they were below that cutoff. In other words, NRTIs are given less weight than drugs in the other three classes.
In the genotypic score, PIs scored 1 if rated sensitive by the French genotypic interpretation system [3], 0.5 if rated intermediate by the French system, and 0 if rated resistant. Respective scores for NNRTIs and enfuvirtide were 1, 0, and 0, and for NRTIs 0.5, 0, and 0.
Logistic regression analysis factored in the genotypic and phenotypic scores, starting CD4 count, starting viral load, and starting viral coreceptor use. (Everyone started the trial with R5 virus, according to the original Trofile assay.) The weighted scores and pretreatment CD4 count emerged as the strongest predictors of having a viral load below 50 copies at study week 48:
· Weighted genotypic score of 0 versus 2 or more: odds ratio [OR] 5.7 · Weighted phenotypic score of 0 versus 2 or more: OR 8.9 · Weighted genotypic score of 0 versus 1: OR 2.8 · Weighted phenotypic score of 0 versus 1: OR 2.8 · CD4 count below 50 versus over 200: OR 6.2 with genotypic score, 7.0 with phenotypic score · CD4 count below 50 versus 101 to 200: OR 3.7 with genotypic score, 3.8 with phenotypic score · CD4 count below 50 versus 50 to 100: OR 2.1 with genotypic score, 2.4 with phenotypic score
Pretreatment coreceptor use did not affect virologic response rates in this analysis, and pretreatment viral load had a borderline impact. Among people with a weighted genotypic score of 2 or more, 71% taking maraviroc twice daily had a sub-50 viral load at week 48. Among people with a weighted phenotypic score of 2 or more taking twice-daily maraviroc, 73% were under the 50-copy mark at week 48. Results were similar for people randomized to once-daily maraviroc. In contrast, 57% of placebo recipients with a genotypic score of 2 or higher were responders, as were 52% of those with a phenotypic score of 2 or more.
When the investigators focused only on patients starting twice-daily maraviroc with a CD4 count above 50, 77% with a genotypic score of 2 or more and 81% with a phenotypic score of 2 or more reached sub-50-copy territory. Response rates with placebo in this subanalysis were 62% with a genotypic score of 2 or more and 55% with a phenotypic score of 2 or more.
Maraviroc takers with a genotypic or phenotypic score or 2 or higher had the same small rates of emergent CXCR4-using virus at failure as placebo recipients did. But higher proportions of maraviroc-treated people than placebo takers with genotypic or phenotypic scores under 2 had CXCR4-using virus at failure.
1. Boucher C, Schapiro J, Kuritzkes D, et al. Genotypic- and phenotypic-weighted OBT susceptibility scores are similarly strong predictors of virologic response <50 copies/ml at week 48 in MOTIVATE 1 and 2. XVIII International Drug Resistance Workshop. June 9-13, 2009. Fort Myers, Florida. Abstract 48.
2. Gulick R, Lalezari J, Goodrich J, et al. Maraviroc for previously treated patients with R5 HIV-1 infection. N Engl J Med. 2008;359:1429-1441.
3. HIV French Resistance. HIV-1 genotypic drug resistance interpretation's algorithms. http://www.hivfrenchresistance.org/2002/tab2.html.