Number of viral load blips during antiretroviral therapy (ART) begun during acute HIV infection predicted viral rebound when ART stopped in a 228-person international analysis. Longer time between HIV seroconversion and starting ART also boosted the off-treatment rebound risk, while longer time on ART during acute infection trimmed rebound risk.
A few people who start ART for acute HIV infection can later suspend therapy without viral rebound, perhaps because HIV reservoirs are smaller during acute infection, the immune system is healthier and ART-induced immunologic recovery is better. Transient rebounds in viral load during sustained therapy -- "blips" -- occur during treatment of acute HIV infection. Researchers who conducted the new study suggested these blips may predict confirmed rebounds after ART stops because blips may reflect the size of the latent HIV reservoir and intermittent immune activation.
To explore these issues, the researchers analyzed data from HIV seroconverters in the CASCADE database embracing Europe, Australia, Canada and sub-Saharan Africa. They focused on people older than 16 years who began ART within six months of their estimated seroconversion dates and had at least three viral load measurements while on ART.
Related: Starting HIV Treatment Soon After Diagnosis Trims Time to Viral Suppression
For each participant, the researchers counted blips, defined as a viral load above 400 copies/mL in a previously suppressed person and a subsequent load below 400 copies/mL without a regimen change. Among participants who later stopped ART, the CASCADE team measured time to loss of viral control, defined as two consecutive viral loads above 1000 copies/mL. They used Cox proportional hazards models to assess factors associated with loss of viral control or with post-treatment control (viral load below 50 copies/mL for at least 24 months after stopping ART).
The analysis focused on 778 adults who started ART within six months of HIV infection and had at least three HIV RNA measures. That group included 228 people (29%) who later stopped ART, of whom 119 (52%) had a rebound and 22 (10%) achieved post-treatment control. Among the 288 people who stopped ART, 90% were men and 73% men who have sex with men. Median time from seroconversion to starting ART measured 2.2 months and median time on ART was 11 months. In the group that stopped ART, 9% had at least one blip above 400 copies/mL and 89% of that group had only one blip.
Among the 119 people who stopped ART and rebounded, median time to rebound was 10.3 months. Proportional hazards modeling identified three factors independently associated with rebound: (1) each blip above 400 copies/mL (hazard ratio [HR] 1.71, 95% confidence interval [CI] 0.94 to 3.10), (2) longer time between HIV seroconversion and starting ART (HR 1.16 per additional month, 95% CI 1.04 to 1.28) and (3) more frequent HIV RNA measurements while on ART (HR 1.10 per mean additional measurement/year increase, 95% CI 1.02, 1.17). Two factors independently predicted lower rebound risk: (1) longer time on ART (HR 0.84 per six-month increase, 95% CI 0.76 to 0.92) and (2) later year of starting ART (HR 0.91, 95% CI 0.84 to 0.98). Factors not linked to rebound risk were CD4 count when ART began, antiretroviral class begun, age at seroconversion, sex and HIV transmission risk.
The 22 post-treatment controllers differed from the 206 people who stopped ART and did not achieve control in four ways: (1) blips above 50 copies/mL: one controller versus 14 noncontrollers, (2) longer time on ART: median 17.4 versus 10.9 months, (3) lower first HIV RNA after HIV diagnosis: median 4.9 versus 5.3 log₁₀ copies/mL and (4) higher CD4 count when ART began: median 562 versus 493 cells/mm³.
The CASCADE researchers see their findings as "the first evidence that frequency and magnitude of viral blips while on ART initiated in PHI [primary HIV infection] is associated with viral rebound among individuals interrupting ART started in PHI." Because they reported linear associations between how soon ART starts, treatment duration and post-treatment control, the investigators say they cannot determine when ART initiation is too late to achieve control.