| |
"Blips & Viral Failure"- Understanding viral blips in people with HIV on antiretroviral therapy: a nationwide Swedish cohort study of 4,210 individuals
|
| |
| |
16 October 2026
Download the PDF here
Conclusion
Individuals on INSTI- and NNRTI-based ART had fewer viral blips than those on PI-based ART. In this cohort, transient viremia below 500 HIV RNA copies/mL was not associated with a higher risk of subsequent virological failure.
The results showed that viral blips are far from rare. In total, 853 individuals, or 20.3 percent of the cohort, had between one and six registered blips during follow-up. The overall blip incidence was 4.35 per 100 person-years, with a 95 percent confidence interval of 4.02 to 4.69. In a sensitivity analysis using a narrower definition of 50 to 199 copies per milliliter, the incidence was slightly lower at 3.80 per 100 person-years. These figures suggest that roughly one in five people on suppressive therapy will experience at least one blip over a five-year period.
Results
In total, 4,210 individuals were included, of whom 853 (20.3%) had 1-6 registered blips. Median observation time was 5.5 years (range 0.5-13.7). Blip incidence was 4.35 (95% CI 4.02-4.69) per 100 person-years. In a sensitivity analysis applying the narrower definition of 50-199 copies/mL, the incidence was 3.80 (95% CI 3.49-4.11) per 100 person-years. Individuals on INSTIs and NNRTIs had similar risk, while those on PIs had a significantly higher risk of blips (hazard ratio [HR] 1.22, 95% CI 1.03-1.45; p = 0.02). Transient viremia was not associated with an increased risk of subsequent virological failure; individuals with blips had a lower risk than those with continuously undetectable HIV RNA (HR 0.54, 95% CI 0.41-0.71; p < 0.0001).
The median blip magnitude was 77 copies/mL (IQR 58-117). Of the 1,180 episodes classified as blips, 799 (67.7%) had a peak below 100 copies/mL and 1,044 (88.5%) below 200 copies/mL, while 136 (11.5%), occurring in 129 individuals, were between 200 and 500 copies/mL.
In the sensitivity analysis, restricting the definition to 50-199 copies/mL reduced the number of qualifying episodes to 1,044 and the number of individuals with at least one blip from 853 (20.3%) to 769 (18.3%), with a blip incidence of 3.80 (95% CI 3.49-4.11) per 100 person-years compared with 4.35 (95% CI 4.02-4.69) for the primary definition. With an upper limit of 399 copies/mL the incidence was 4.20 (95% CI 3.87-4.52) per 100 person-years. The mean number of blips per affected individual was essentially unchanged (1.38 and 1.36 for the primary and the narrower definition, respectively).
Grouped by the core agent class of the initial regimen, secondary virological failure occurred in 145 of 1,738 individuals on NNRTI-based, 187 of 1,222 on PI-based and 55 of 1,250 on INSTI-based regimens (1.18, 2.24 and 1.33 events per 100 person-years, respectively; chi-square, p < 0.0001). Individuals with transient viremia of 20-500 HIV RNA copies/mL had a decreased risk of secondary virological failure compared with individuals with sustained suppression. One possible explanation for the difference observed between core agents is confounding
by indication.
In this retrospective study covering the vast majority of treatment-naïve PWH starting different ART regimens in Sweden between 2007 and 2020 we found that individuals receiving INSTI-based regimens had a decreased risk of viral blips compared with those on PI-based regimens and a similar risk compared with those on NNRTI-based regimens.
For clinicians managing people with HIV, the study provides evidence-based reassurance that viral blips are common, largely benign events that do not predict treatment failure. For researchers, the observation that PI-based regimens carry a higher blip risk than INSTI- or NNRTI-based regimens opens new avenues for investigating how drug class influences residual viral production from latent reservoirs.
Perhaps the most clinically reassuring result concerns the relationship between blips and subsequent virological failure. Contrary to the hypothesis that blips might serve as early harbingers of treatment failure, the study found that individuals with blips had a lower risk of virological failure than those with continuously undetectable HIV RNA. The hazard ratio was 0.54 (95 percent confidence interval 0.41 to 0.71; p < 0.0001). This counterintuitive finding suggests that transient viremia below 500 copies per milliliter is not associated with a higher risk of subsequent failure in this cohort.
The lower failure risk among individuals with blips may reflect surveillance bias rather than a protective biological effect........Patients who experience a blip receive closer monitoring, and any developing resistance or adherence problem may be detected and addressed earlier. Alternatively, the immune systems of individuals who mount transient viral production may be more active in controlling the virus. The study does not distinguish between these possibilities, but the practical message is clear: a single low-level positive result in an otherwise suppressed patient should not automatically trigger regimen changes or alarm.
The team compared integrase strand transfer inhibitor (INSTI)-based regimens with those built around non-nucleoside reverse transcriptase inhibitors (NNRTIs) and protease inhibitors (PIs). INSTIs, which block the viral enzyme that inserts HIV DNA into the host genome, have become the backbone of first-line therapy in many countries due to their potency and tolerability. The analysis revealed that individuals on INSTIs and NNRTIs had similar blip risk, while those on PI-based regimens had a significantly higher risk, with a hazard ratio of 1.22 (95 percent confidence interval 1.03 to 1.45; p = 0.02).
|
|
| |
| |
|
|
|
