CROI 2014: Cumulative Viral Load and HIV Disease Progression


Periodic low-level viral load appears to predict virological failure, but not progression to AIDS or death, and cumulative viral load over time may be a risk factor for HIV disease progression and mortality, according to studies presented at the 21st Conference on Retroviruses and Opportunistic Infections (CROI 2014) this month in Boston.

Low-Level Viral Load

Marie-Anne Vandenhendefrom the University of Bordeaux and colleagues assessed the impact of low-level viral load on virological and clinical outcomes among people with HIV on antiretroviral therapy (ART).

The researchers analyzed data from nearly 18,000 people with HIV in 18 North American and European cohorts participating in the Antiretroviral Therapy Cohort Collaboration. About three-quarters were men, the median age was 40 years, and the mean baseline CD4 T-cell count was approximately 230 cells/mm3.

Participants started ART regimens containing 2 nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs) plus either a non-nucleoside reverse transcriptase inhibitors (NNRTI) or a ritonavir-boosted protease inhibitor; about 60% were on NNRTI-based regimens. They continued on ART for at least
6 months and reached a viral load <50 copies/mL 3-9 months after initiating therapy.

Investigators looked at the frequency of 2 consecutive low-level viral load measurements between 50 and 199 copies/mL or 200 and 499 copies/mL occurring after viral suppression, and whether this was associated with virological failure (2 measurements >500 copies/mL or 1 such measurement leading to ART modification), AIDS-defining events, or death.

During follow-up, 624 participants (3.5%) experienced at least 1 episode of low-level viral load between 50-199 copies/mL, with a mean total duration of 5.8 months. In addition, 482 people (2.7%) had at least 1 episode of low-level viral load between 200-499 copies/mL, with a mean total duration of 6.4 months.

There were a total of 1903 cases of virological failure. Rates were 10.4% among people without low-level viral load, 7.9% among people with 50-199 copies/mL, and 22.6% among people with 200-499 copies/mL.

Low-level viral load between 200-499 copies/mL was strongly associated with higher risk of virological failure after adjusting for other factors (adjusted hazard ratio [HR] 3.97, or about 4-fold higher risk). Low-level viral load between 50-199 copies/mL, however, was only weakly associated with virological failure (adjusted HR 1.38). Neither type of ART regimen, modification of ART, nor cumulative duration of low-level viral load were associated with virological failure.

There were a total of 532 post-ART AIDS events and 480 deaths during follow-up. Death rates were 2.6% among people without low-level viral load, 3.8% among people with 50-199 copies/mL, and 5.0% among people with 200-499 copies/mL.

There was little evidence that low-level viral load either between 50-199 or 200-499 copies/mL was associated with AIDS events (adjusted HR 1.11 and 0.81, respectively) or death (1.19 and 1.11, respectively), as compared with continued viral suppression. 

"Among patients virologically suppressed 3 to 9 months after starting ART and with a median follow-up of 2.3 to 3.1 years, persistent low-level viremia between 200 and 499 copies/mL was strongly associated with virological failure, but not with AIDS events or death," the researchers concluded.

"The lack of association of persistent low-level viremia between 50 and 199 copies/mL with virological failure or clinical outcomes supports current U.S. guidelines, which define virological failure as a confirmed viral load > 200 copies/mL," they added.

Cumulative Viral Load

In the second study, Michael Mugavero of the University of Alabama at Birmingham and colleagues looked at a different measure -- time-updated viremia copy-years (VCY), or sum of detectable viral loads -- among participants in the ART Cohort Collaboration.

While the ability of cross-sectional (single time-point) viral load to predict clinical events following ART initiation is well established, it is unclear how best to quantify the relationship between viral load and clinical events or death over time. VCY estimates the "cumulative plasma viral burden" based on the area under the viral load curve from the beginning to the end of the observation period. Higher VCY indicates more time off ART or on failing therapy. A VCY of 10,000, for example, could indicate either having 10,000 copies/mL for 1 year or 1000 copies/mL for 10 years.

The researchers calculated VCY for more than 33,500 cohort participants who started ART between January 2000 and December 2009. Again, most (about 70%) were men, the mean age was 39 years, and the mean baseline CD4 count was approximately 200 cells/mm3. The viral load lower limit of detection was assumed to be 500 copies/mL, and any value below this level was set to 250 copies/mL for the calculation.

During a mean follow-up period of 3.2 years after ART initiation, participants had an average of 13 viral load measurements available. A total of 1341 patients died during follow-up, including 503 AIDS-related deaths (38%), 614 non-AIDS-related deaths (46%), and 224 that were of unclassifiable or unknown cause (17%).

All 3 viral load measures considered -- baseline viral load, time-updated or recent viral load, and time-updated VCY -- predicted all-cause mortality in a single-factor analysis, but time-updated VCY remained strongly predictive after adjusting for the other 2 measures. People with VCY in the 10,000-100,000 range, 100,000-350,000 range, and >350,000 had adjusted HRs of 1.14, 1.56, and 1.74, respectively, relative to those with VCY <10,000.

Separating out causes of death, even stronger associations were observed for AIDS-related mortality (adjusted HRs of 1.45, 2.18, and 3.64, respectively, for the ascending VCY ranges). But associations for non-AIDS mortality were modest.

Overall, viremia copy years predicted all-cause mortality better than current viral load, which in turn did better than baseline or pre-ART viral load.

"Cumulative viral load measures such viremia copy years may have independent and enhanced prognostic value for mortality relative to cross-sectional measures," the researchers concluded.

Viral Load Copy Years Among Seroconverters

Finally, Matthias an der Heiden of the Robert Koch-Institute in Berlin and colleagues looked at the clinical impact of viral load copy years among seroconverters, or newly infected individuals, in the CASCADE Collaboration cohort who had not yet started ART. They hypothesized that ongoing viral replication may contribute to immune activation and its detrimental effects before the CD4 count is low enough to necessitate treatment.

This analysis included 9461 people who seroconverted after 1997 and had available viral load and CD4 cell measured taken within 4-12 months after seroconversion. Most (86%) were men and the median age at seroconversion was approximately 34 years. They were followed starting at 4 months after seroconversion and continuing until they initiated ART, their CD4 count fell below 200 cells/mm3, or they had no available viral load or CD4 count data for more than 12 months, accruing a total of 15,553 years of follow-up data.

During follow-up 161 participants progressed to AIDS or death, for an overall incidence rate of 10.4 per 1000 person-years. In a multivariate analysis, older age, having a CD4 count of 200-350 or 350-500 cells/mm3 (vs 500-1000 cells/mm3), and average viral load >100,000 copies/mL were significantly associated with a higher risk of clinical AIDS or death. In addition, participants with VCY >100,000 or in the 20,000-100,000 range had a higher rate of AIDS or death than those with VCY <20,000.

"Higher average viral load is associated with a higher risk of clinical AIDS/death," the investigators concluded. "Accumulation of intermediate viral load levels over time leads to higher risks comparable to those associated with high average viral load," they added, suggesting that, "Choice [of] when to start ART may also be influenced by the duration and extent of previous HIV replication."



MA Vandenhende, S Ingle, M May, et al (Antiretroviral Therapy Cohort Collaboration). Impact of Low-Level Viremia On Clinical and Virological Outcomes in Treated HIV Infected Patients. 21st Conference on Retroviruses and Opportunistic Infections (CROI 2014). Boston, March 3-6, 2014. Abstract 1014.

M Mugavero, AO Westfall, MJ Gill, et al (Antiretroviral Therapy Cohort Collaboration). Cumulative Viral Load Predicts All-Cause and AIDS-Related Mortality After Initiation of ART.21st Conference on Retroviruses and Opportunistic Infections (CROI 2014). Boston, March 3-6, 2014. Abstract 565.

M an der Heiden, A Zoufaly, C Sabin, et al (CASCADE Collaboration in EuroCoord).The Clinical Impact of Viral Load Copy Years in Antiretroviral-Naive HIV Seroconverters.21st Conference on Retroviruses and Opportunistic Infections (CROI 2014). Boston, March 3-6, 2014. Abstract 291.