icon-    folder.gif   Conference Reports for NATAP  
 
  18th CROI
Conference on Retroviruses
and Opportunistic Infections
Boston, MA
February 27 - March 2, 2011
Back grey_arrow_rt.gif
 
 
 
Relative performance of ESTA, Trofile, 454 deep sequencing and "reflex" testing for HIV tropism in the MOTIVATE screening population of therapy experienced patients
 
 
  Reported by Jules Levin
CROI 2011 Boston March 2
 
Chanson Brumme1,Timothy Wilkin2, Zhaohui Su3, Jonathan Schapiro4, Ron Kagan5, Douglass Chapman6, Jayvant Heera7, Hernan Valdez6, P Richard Harrigan1
 
1BC Centre for Excellence in HIV/AIDS, Vancouver, BC, Canada 2Weill Cornell Medical College, New York, NY 3Outcome Sciences Inc, Cambridge, MA 4National Hemophilia Center, Tel Hashomer, Israel 5Quest Diagnostics Inc, San Juan Capistrano, CA 6Pfizer, Inc., New York, NY 7Pfizer Global Research and Development, New London, CT, USA
 
Author Conclusions
 
In an unselected, treatment-experienced patient population, all methodologies predicted a greater week 8 viral response in individuals identified as having R5 virus versus those identified as having non-R5 virus
 
In settings where population sequencing and either ESTA or 454 technologies are available in-house, a reflex testing approach could have several benefits:
 
- Substantial improvement in turnaround times in >25% of cases
 
- Decreased testing costs
 
- Little impact on the performance of short-term outcome predictions
 
All methods performed well at assessing HIV coreceptor tropism prior to initiation of maraviroc in treatment-experienced patients, including those with <2 active background drugs. However, the ESTA, 454 and reflex strategies had slightly greater discriminating ability in separating responders from non-responders in this dataset
 
Background

 
The original Monogram Trofile assay was used to assess HIV-1 co-receptor tropism prior to randomization into the maraviroc (MVC) clinical trials, but has been discontinued and replaced by the Enhanced Sensitivity Trofile Assay (ESTA)
 
OTA remains the only HIV tropism assay used prospectively to screen patients in randomized clinical trials of MVC
 
Previous comparisons of HIV tropism screening methods have often used patient populations pre-selected for CCR5-using HIV by Trofile
 
In a retrospective analysis of the MOTIVATE/1029 clinical trials, we compare the relative performance of ESTA versus other genotype-based tropism assays in predicting short-term virological response to maraviroc in an unselected treatment-experienced population
 
Objective
 
Compare the concordance of co-receptor tropism predictions of several tropism determination assays with ESTA and 454 technology
 
Evaluate the relative performance of ESTA versus other genotype-based tropism assays for predicting short term viral response after maraviroc initiation in a retrospective analysis of an unselected, treatment-experienced population (MOTIVATE)

apasl1.gif

apasl2.gif

1st chart below - Individual assay concordance with results from Enhanced Sensitivity Trofile (ESTA) and 454 sequencing assays in an unselected treatment-experienced patient population. Categories are expressed as ESTA result/Alternate Assay result and 454 result/Alternate Assay result, respectively. 2nd chart below - Assay performance metrics: Concordance, Sensitivity, Specificity, Positive Predictive Value (correct identification of non-R5 samples) and Negative Predictive Value (correct identification of R5 samples) of various tropism assays vs ESTA or 454.
 
"Reflex" Testing
 
In the "reflex" testing approach, N=83 samples were identified as non-R5 by the population sequencing (5.75 FPR) assay. These samples would have their non-R5 results reported immediately. The remaining N=229 samples would be tested by 454 or ESTA. Following the "Reflex" testing algorithm as outlined would result in a >25% reduction in the number of 454 or ESTA tests performed.

apasl3.gif

Virological response at week 8 post-MVC: Assay prediction accuracy
 
Proportion of patients achieving virological response (<50 copies/mL, or ≥2 log pVL decline from baseline) at week 8 stratified by tropism assay readout (R5, non-R5). Numbers (%) above the bars indicate assay prediction accuracy of week 8 virological response (Total percentage of R5 and non-R5 patients correctly classified as having a response or not). Panel A) includes results from all patients Panel B) includes only patients with <2 active background drugs as measured by weighted optimized background therapy susceptibility score (wOBTSS). ESTA, 454 and the reflex strategies showed the largest differences in the proportion of patients achieving virological response between the R5 and non-R5 populations.

apasl4.gif

apasl5.gif