Compared with completely undetectable HIV RNA (target not detected), a detectable viral load below 50 copies/mL independently boosted odds of rebound above 50 copies/mL in two years, according to a study in Italy. Hepatitis C (HCV) coinfection, an antiretroviral regimen containing more than three pills and briefer viral suppression also independently raised rebound odds.
Many consider a plasma viral load below 50 copies/mL "undetectable." But University of Torino researchers who conducted this retrospective 1055-person analysis note that commercial assays now detect plasma loads as low as 40, 37 or 20 copies/mL. Previous studies usually found that higher levels of sub-50-copy viremia raised chances of rebound above 50 copies/mL. Because factors associated with rebound risk in people with a load below 50 copies/mL remain poorly understood, the researchers undertook this analysis to address that question.
The retrospective study involved patients with two consecutive viral loads below 50 copies/mL and at least six months of follow-up at the University of Torino outpatient clinic in the first half of 2010. The investigators used the second sub-50 load as the baseline load, and they defined failure as two consecutive loads above 50 copies/mL at least seven days apart or one load above 50 copies/mL at the last follow-up visit. Using a commercial assay with a 20-copy cutoff Cobas AmpliPrep-Cobas TaqMan, the Torino team divided participants into three groups: (1) undetectable viral load (target not detected), (2) target detected <20 copies/mL and (3) 20 to 50 copies/mL.
Of the 1055 study participants, 753 (71%) were male, 922 (87%) were white and 295 (28%) had HCV infection. Median age stood at 46.7 years. While 41% were taking a protease inhibitor regimen, 40% were taking a nonnucleoside and 4% were taking raltegravir. Median duration of viral suppression measured 39.7 months, and participants had taken their current regimen for a median of 23.7 months.
Three-quarters of participants (74%) had no HIV RNA detectable, 18% had a load below 20 copies/mL and 8% had a load between 20 and 50 copies/mL. Multivariate logistic regression identified three factors independently associated with higher odds of a target-not-detectable viral load result: female gender (adjusted odds ratio [aOR 1.68, 95% confidence interval [CI] 1.20 to 2.35, P = 0.002), nadir CD4+ count above 200 cells/mm3 (aOR 1.56, 95% CI 1.18 to 2.08, P = 0.002) and viral suppression for more than two years (aOR 1.40, 95% CI 1.04 to 1.89, P = 0.026).
After 24 months of follow-up, 81 of 1055 people (7.7%) had a detectable viral load with a median level of 1057 copies/mL. Logistic regression analysis pinpointed four variables that independently raised the odds of rebound above 50 copies/mL. A detectable sub-50-copy load at baseline raised the odds about 70% (aOR 1.71, 95% CI 1.10 to 2.68, P = .017), as did HCV coinfection (aOR 1.69, 95% CI 1.08 to 2.62, P = .020). An antiretroviral regimen including more than three pills doubled the odds of rebound (aOR 1.97, 95% CI 1.26 to 3.06, P = .003). And fewer than two years of viral suppression more than doubled the odds of rebound (aOR 2.47, 95% CI 1.58 to 3.86, P < 0.001). Results were similar among people who never switched their regimen during 24 months of follow-up.
Among 50 adherent rebounders in care, 37 had successful resistance genotyping. Fourteen of those 37 (38%) had new resistance-associated mutations. Ten of those 14 patients had new reverse transcriptase mutations; four of six people taking raltegravir had integrase inhibitor mutations.
Although some previous research linked detectable sub-50-copy viremia to later rebound, the authors believe their study adds to the understanding on this outcome because (1) they used a commercially available 20-copy assay, (2) the study group included 42 patients taking raltegravir, (3) they analyzed factors associated with rebound and (4) they characterized emergence of resistance mutations after rebound. They propose that the target-not-detected output of their assay "may be relevant in the long-term management of HIV-positive patients."
The investigators highlight their finding that HCV coinfection boosts the odds of rebounding HIV RNA. They surmise that multiple comorbidities, immune dysfunction, polypharmacy and incomplete adherence in people with HIV/HCV may contribute to this result.
Finally the researchers suggest that a rebound-prediction score assigning one point to each of the four independently predictive variables may be useful in forecasting a rebound from below 50 copies/mL. Viral rebound incidence was 4.5% in people with zero or one predictor versus 13.7% in those with two or more predictors.