Higher risk with persistent low-level viremia

Organisation: Position: Deadline Date: Location:

People living with HIV who have a detectable but low viral load – in the range of 50 to 500 or 1000 copies/ml – may continue to have a higher risk of Aids-related events, but their likelihood of experiencing serious non-Aids events including heart, liver and kidney disease did not appear to increase, according to a pair of Italian studies presented at the recent 15th European AIDS Conference held in Barcelona, Spain.

It is well known that effective combination antiretroviral therapy (ART) that suppresses HIV replication dramatically reduces disease progression and improves survival. Recent findings from the large START trial showed that people who initiate treatment soon after diagnosis had a lower risk of both Aids-related illnesses and non-Aids events and death compared to those who waited until their CD4 cell count fell.

A small proportion of people with HIV have persistently detectable but low viral load despite treatment, and the effects of low-level viraemia are not fully understood. Even a low level of ongoing viral replication can lead to increased immune activation and inflammation, which may contribute to non-Aids-related co-morbidities such as heart disease and cancer. However, some previous large observational studies have not seen an association between low-level virus and increased risk of Aids, non-Aids illness or death.

Andrea Antinori and fellow investigators with the ICONA Foundation Study Cohort estimated three-month incidence of Aids-related events, severe non-Aids events or death among people with current low-level viraemia compared to those with viral loads below 50 copies/ml.

The analysis included 7,277 participants in the Italian ICONA cohort who had at least one person-year of follow-up spent with a viral load in the range of zero to 1000 copies/ml at least six months after starting first-line combination ART. Three-quarters were men, most were white, the median age was 37 years and 18% had hepatitis C co-infection. The median CD4 count at ART initiation was approximately 300 cells/mm3 and 13% had been diagnosed with Aids. The most common ART regimen was efavirenz (Sustiva) plus tenofovir/emtricitabine (the drugs in Truvada), but a wide range of other drugs were also used.

The researchers classified participants into four groups according to viral load six months after starting ART: 0-50 copies/ml (n = 3919); 51-200 copies/ml (n = 1811); 201-500 copies/ml (n = 1117); and 500-1000 copies/ml (n = 430).

They compared two composite endpoints across the viral load groups: Aids or death due to any cause; and serious non-Aids events according to the definition used in the SMART study (major cardiovascular, kidney or liver disease) and death due to any cause, over the course of three months.

A total of 204 Aids-related events or deaths occurred during 28,429 person-years of follow-up. Incidence rates of Aids-related events or death were 0.52, 1.25, 1.91 and 1.06 per 100 person-years for participants with viral loads of 0-50, 51-200, 201-500 and 500-1000 copies/ml, respectively.

The adjusted risks of Aids-related events or death were 1.74 (a 74% increase), 2.30 (more than double the risk) and 1.30 (a 30% increase) for people with 51-200, 201-500 and 500-1000 copies/ml, respectively, compared to those with 0-50 copies/ml; these differences were statistically significant for the 51-200 and 201-500 groups, but not for the 500-1000 group.

A total of 438 serious non-Aids events or deaths occurred during follow-up. Incidence rates of serious non-Aids events or death were 1.46, 1.77, 1.69 and 2.56 for the respective viral load groups. The adjusted risks of serious non-AIDS events or death were 1.09 (little change), 1.00 (no difference) and 1.54 (a 54% increase) for people with 51-200, 201-500 and 500-1000 copies/ml, respectively, none of which were statistically significant.

“Low-level viremia, in the range of 51-500 copies/ml, was associated with an independent risk of developing a new Aids diagnosis, which was up to 2.3-fold higher than that observed in the presence of suppressed viral load,” the researchers concluded. “Conversely, low-level viremia did seem to predict a more elevated risk of serious non-Aids events.”

They noted, however, that the viral load categories in this study were based on current levels and could not account for sustained periods of low-level viral replication. Occasional viral ‘blips’ are not uncommon during treatment, but persistent viraemia is a greater concern. They also said the short follow-up period may have underestimated long-term effects on clinical endpoints.

These results, they added, “may be useful to plan analyses aimed at better defining threshold of virological failure valid for clinical purposes and to eventually optimise treatment strategies.”

In the second study, Silvia Costarelli from San Gerardo Hospital in Monza, Italy, and colleagues with the Italian MASTER Cohort looked at the association between persistent low-level viraemia (viral load between 50 and 400 copies/m) and increased risk of cardiovascular disease events.

This analysis included 4,393 people who achieved viral load below 50 copies/ml for at least six months after ART initiation. About 70% were men, most were Italian and the median age at ART initiation was approximately 40 years. They had relatively advanced HIV disease with a median CD4 count below 300 cells/mm3. Underlying cardiovascular risk was generally low, with only about 1% having prior cardiovascular events.

A majority of 3576 people maintained full viral suppression, but 574 experienced treatment failure with viral rebound to more than 400 copies/ml and 243 had persistent low-level viraemia on two consecutive tests.

During follow-up there were 45 cardiovascular events, 57 Aids events and 93 deaths among people with ongoing viral suppression; 18, 53 and 37, respectively, among people with viral rebound; and 6, 5 and 4, respectively, among people with persistent low-level viraemia.

The researchers again considered two composite endpoints: first cardiovascular event, Aids-related event or death; and first cardiovascular event or death, ignoring Aids.

For the first endpoint, unadjusted incidence rates were 11.7, 21.6 and 9.3 per 1,000 person-years for people with full viral suppression, viral rebound and persistent low-level viraemia, respectively. For the second endpoint, incidence rates were 8.6, 11.6 and 9.0, respectively. That is, the risk of adverse outcomes was much greater – nearly twofold higher – for people with viral rebound when Aids was considered, but not when considering only cardiovascular events or death.

In a multivariate analysis, people with viral rebound had a significantly higher risk for both the first and second endpoints compared to people with ongoing viral suppression (hazard ratio 2.15 and 1.5, respectively). However, persistent low-level viraemia was not associated with significantly different hazard rates.

“Our results suggest that persistent low-level viremia does not influence cardiovascular disease,” the investigators concluded. “Moreover, in our setting, persistent low-level viremia does not even influence Aids or death.”

In summary, incidence of cardiovascular events, Aids and death were similar for people with ongoing viral suppression and those with persistent low-level viraemia in the MASTER study, suggesting that low-level viral load does not increase the risk of either Aids or non-Aids events. However, this conflicts with the ICONA findings, which did see a higher risk of Aids among people with low but detectable viral load.

Abstract 1
Objectives: To estimate 3-month incidence of AIDS, severe non-AIDS event (SNAE) and death in people with current LLV and to compare incidence to that of people with current viral load (VL) ≤50 copies/mL.
Methods: Patients enrolled in Icona Foundation Cohort with at least one person-year follow-up (PYFU) spent with a VL in the range of 0-1,000 copies/mL ≥6 months after first initiation of cART.
Endpoints were: AIDS, SNAE (SMART definition) or death for any cause. Incidence rates (95%CI) of endpoints were calculated assuming a Poisson distribution, by most recent (within 3 months) VL stratified according to the following exposure groups: suppressed VL (0-50 copies/mL) and three more strata in the range of LLV of 51-1,000 copies/mL. Poisson regression was fitted to estimate unadjusted and adjusted rate ratios (RR 95%CI) after controlling for potential confounders in Table.
Results: We identified 7,277 patients (male 75%; median age [IQR] 37 [31-43] yrs; IDU 20%; previous AIDS 13%; HCV+ 18%; median [IQR] CD4 count at cART 302 [171-427] cell/mm3). Distribution of patients by VL levels 6 months after starting cART: 0-50 copies/mL (n=3,919), 51-200 (n=1,811), 201-500 (n=1,117), 501-1,000 (n=430). During a total of 28,429 PYFU we observed 204 AIDS and 438 SNAEs, for an overall incident rate (IR per 100 PYFU) of AIDS/death and SNAE/death of 0.72 and 1.54, respectively. Number of events, PYFU, IR and rate ratios of AIDS/death and SNAE/death by LLV strata as time updated variable are reported in Table.
Conclusions: LLV, in the range of 51-500 copies/mL, was associated with a risk of developing a new AIDS diagnosis, which was up to 2.3 fold higher than suppressed VL. Conversely, LLV did not predict a more elevated risk of SNAEs. These results may be useful to better define threshold of virological failure valid for clinical purposes and to optimise treatment strategies.

Aidsmap material
Abstract 1
Abstract 2


Receive Medical Brief's free weekly e-newsletter



Related Posts

Thank you for subscribing to MedicalBrief


MedicalBrief is Africa’s premier medical news and research weekly newsletter. MedicalBrief is published every Thursday and delivered free of charge by email to over 33 000 health professionals.

Please consider completing the form below. The information you supply is optional and will only be used to compile a demographic profile of our subscribers. Your personal details will never be shared with a third party.


Thank you for taking the time to complete the form.