



Viral Load and Genotypic Resistance Score Predict Response to 2Drug ART Switch



HIV Drug Therapy, Glasgow 2018, October 2831, 2018, Glasgow
Mark Mascolini
Viral load upon switching to a 2drug antiretroviral regimen and a genotypic susceptibility score (GSS) estimating number of fully active antiretrovirals in the new regimen proved the best predictors of virologic failure with the new 2drug combo [1]. Researchers working with the Italian ARCA database found the most recent GSS a better response predictor than cumulative GSS.
As 2drug regimens gain popularity, clinicians need tools to help them predict which combinations will work best for which people. Such tools would prove especially useful when picking a 2drug medley for someone who has already taken 1 or more failing regimens and could have circulating or archived HIV resistant to individual antiretrovirals or entire antiretroviral classes.
From the ARCA database (https://www.dbarca.net/) the researchers selected antiretroviralexperienced people starting a 2drug regimen from 2007 to 2017. Everyone had a preswitch genotype and at least 1 postswitch viral load measure. Among participants with virologic failure on their new regimen (defined as viral above 400 copies), the researchers used survival analysis to identify predictors of virologic failure. They calculated GSS for each person according to the Stanford University HIV drug resistance database (https://hivdb.stanford.edu/). GSS equals the sum of genotypic resistance scores for each drug in a regimen. A GSS of 1.0 indicates viral susceptibility to the regimen, 0.5 indicates possible resistance, and 0 indicates resistance [2].
The analysis included 2149 switchers with a median age of 50, a median CD4 count of 577, and a median viral load of 1.57 log (under 40 copies). Almost three quarters of participants (71%) had a baseline viral load below 50 copies, so most of these people were not switching from a failing regimen to 2 drugs. Median time since last viral load above 50 copies stood at 3 years (interquartile range [IQR] 1 to 6). Two thirds of the study group (68%) were men, 80% came from Western Europe, and 6% were nonCaucasian.
Almost all participants (99.3%) had taken a nucleoside, 82.4% had used a protease inhibitor (PI), 67.9% a nonnucleoside, and 32.2% an integrase inhibitor. The most frequently used 2drug combinations were lamivudine (3TC) plus a boosted PI (28%), a PI plus an integrase inhibitor (24.3%), an integrase inhibitor plus a nucleoside (17%), and an integrase inhibitor plus a nonnucleoside (11.4%).
Median GSS stood at 2 (IQR 1 to 2), 41% of participants has a GSS below 2, and 5.2% had a GSS below 1. Median cumulative GSS was also 2 (IQR 1 to 2), 50% of participants had a cumulative GSS below 2, and 11.6% had a cumulative GSS below 1. Estimated probability of virologic failure at 48 weeks stood at 5.1% with a GSS of 2, at 6.8% with a GSS of 1 to 1.99, and at 26.3% with a GSS below 1 (betweengroup difference P < 0.001). Probability of virologic failure at 48 weeks was 4.5% with a cumulative GSS of 2, 7.2% with a cumulative GSS of 1 to 1.99, and 13% with a cumulative GSS below 1 (betweengroup difference not significant).
Multivariate analysis to pinpoint predictors of virologic failure adjusted for antiretroviral regimen, viremia copyyears (a cumulative measure of time with a detectable viral load), CD4 count nadir, and GSS found that only viral load independently predicted failure. Every 10fold higher viral load at the switch almost doubled the odds of failure (adjusted odds ratio [aHR] 1.81, 95% confidence interval [CI] 1.38 to 2.38, P < 0.001). There was a trend toward an association between higher (better) GSS and lower virologic failure risk: A GSS of 1 to 1.99 (versus below 1) lowered failure risk 75% (aHR 0.25, 95% CI 0.05 to 1.22, P = 0.09). Factors not linked to virologic failure in this analysis included type of 2drug regimen, viremia copy years, nadir (lowestever) CD4 count, and cumulative GSS.
The researchers concluded that in antiretroviralexperienced people switching to a 2drug regimen, higher viral load at the switch and having less than 1 fully active drug in the new regimen (GSS below 1) "strongly influence virologic efficacy" of the 2drug combination. Most recent GSS proved a better failure predictor than cumulative GSS.
References
1. Rossetti B, D Redi D, Ciccullo A, et al. Predicting twodrug antiretroviral regimen efficacy by genotypic susceptibility score: results from a cohort study. HIV Drug Therapy, Glasgow 2018, October 2831, 2018, Glasgow. Abstract P299.
2. Villacian J. Basics of HIV resistance testing. TREAT Asia Annual Network meeting. Bangkok. October 11, 2009.






