 |
|
|
| |
Comparison of Quest Diagnostics to Trofile and Trofile-ES Coreceptor Tropism Assays for Predicting Virologic Response to Maraviroc (MVC), a CCR5 Antagonist
|
| |
| |
Reported by Jules Levin
ICAAC Sept 14 2010 Boston
T J Wilkin1, Z Su2, R Kagan3, J Heera4, J M Schapiro5
1Weill Cornell Medical College, New York, NY; 2Outcome Sciences Inc, Cambridge, MA; 3Quest Diagnostics, Nichols Institute, San Juan Capistrano, CA;
4Pfizer Global R&D, New London, CT; 5National Hemophilia Center, Tel Hashomer, Israel.
ABSTRACT
Background: An accurate assessment of coreceptor tropism is
essential before initiating CCR5 antagonists.
Methods: We constructed a cohort of subjects receiving MVC without regard to coreceptor tropism by combining data from 326 subjects randomized to MVC in two clinical trials enrolling subjects with CCR5-using HIV (R5) determined by Trofile (TF) and one enrolling subjects with dual or mixed infections (DM). Screening TF samples were retested by Trofile-Enhanced Sensitivity (TF-ES) and the Quest Diagnostics assay that combines sequencing and heteroduplex analysis (CE-HDA) of the V3 loop codons 1-35. The primary objective
was to show non-inferiority (< -0.5 log10 cp/mL between groups with discordant tropism results) of the CE-HDA assay compared to TF for change in HIV-1 RNA 8 weeks after starting a MVC-containing regimen. Comparisons are adjusted for weighted sensitivity score of the background regimen and baseline HIV-1 RNA.
Results: Median CD4 was 98 cells/µL; median HIV-1 RNA was 5.0 log10 cp/mL. The concordance between CE-HDA and TF was 70% with modest agreement (kappa=0.40); 19 (6%) were R5 by CE-HDA/DM by TF and 79 (24%) were DM by CE-HDA/R5 by TF. The unadjusted mean change in HIV-1 RNA at week 8 was -1.4 and -1.7 log10 cp/mL in the 2 discordant groups respectively. The adjusted difference was -.23 [95% CI -0.89, 0.44]. When comparing discordant CE-HDA and TFES results, the unadjusted mean change in HIV-1 RNA at week 8 was - 1.4 and -2.1 log10 cp/mL in the R5 by CE-HDA/DM by TF-ES and DM by CE-HDA/R5 by TF-ES groups respectively. The adjusted difference was -.70 [95% CI -1.2, -.2].
Conclusions: The CE-HDA tropism assay did not achieve noninferiority to TF, and appeared inferior to TF-ES. These data have been used to change the HDA cutoffs for optimization of the Quest Diagnostics assay.
BACKGROUND
METHODS
Study Population: Patients were screened for the Phase 3 studies of maraviroc in R5 ART-experienced subjects (MOTIVATE trials) and submitted plasma samples for coreceptor tropism by Trofile. If found to be exclusively R5, they were eligible for one of the MOTIVATE trials; if found to be D/M, X4 or tropism
not determined, they were offered enrollment in A4001029. The A4001029 completed enrollment when the MOTIVATE studies were about 25% accrued.
The inclusion criteria for this analysis were as follows:
- Patients were enrolled in A4001029 or MOTIVATE while the A4001029 study was open to accrual.
- Patients received at least one dose of maraviroc during A4001029 or MOTIVATE study.
- Patients whose tropism at screening was indeterminate or missing at screening are not included in this analysis.
- Taken together, this population represents a cohort who received maraviroc without regard to coreceptor tropism testing.
Stored screening plasma specimens were tested for coreceptor tropism using the CE-HDA assay as described in the background. Prior to conducting the analysis, the protocol team decided to retest the screening specimens using the newer version of the Trofile assay with enhanced sensitivity for X4 virus (TF-ES). After the primary analysis, a subset of the study samples (N=77) using ultra-deep
pyrosequencing data obtained on the 454 GS FLX platform (Dr. Richard Harrigan, University of British Columbia) was used to recalibrate the CE-HDA cutoffs used in the Quest Diagnostics assay (revCE-HDA) which were applied to the whole dataset.
Primary Objective: To examine whether the Quest Diagnostics CE-HDA assay is comparable to the Monogram original TrofileTM assay for predicting
virologic response in this population of maraviroc-treated patients.
Statistical Analysis:
· The primary endpoint is change in HIV-1 RNA from baseline to week 8. The non-inferiority margin is -0.5 log10 HIV-1 RNA comparing the two discordant groups.
· In the calculation of mean change in HIV-1 RNA from baseline, HIV-1 RNA measurements below 50 cp/mL were replaced with 25 cp/mL.
· The primary analysis models change in HIV-1 RNA as a function of treatment group, baseline HIV-1 RNA and weighted sensitivity score based on genotypic and phenotypic testing. HIV-1 RNA measurements below 50 cp/mL were censored at 50 cp/mL. The Lifereg procedure in SAS (version 9.2, Cary, NC) was used for the analysis.
Study Cohort Characteristics
· 326 participants met the eligibility criteria
· Median CD4 was 98 cells/µL
· Median HIV-1 RNA was 5.0 log10 cp/mL.
· Median Age 44 years
· 92% male; 80% white
· Median weighted sensitivity score 1.0
RESULTS
|
| |
|
|
|
|
|
