Early Treatment Failure Signals: One in 15 HIV Patients in Hubei Miss Viral Suppression at 24 Weeks

When someone living with HIV begins antiretroviral therapy, the first six months are a diagnostic window of enormous clinical consequence. By 24 weeks, the plasma concentration of HIV RNA should have fallen below the threshold that defines virological suppression, and failure to reach that milestone signals a suboptimal early response that can foreshadow drug resistance, sustained viral replication, and continued transmission risk. A retrospective cohort study conducted in central China has now quantified how often this early milestone is missed and, crucially, which patients are most likely to miss it.

The 24-week mark is not an arbitrary checkpoint. It reflects the pharmacological expectation that modern antiretroviral combinations, when taken consistently, drive plasma HIV RNA below the limits of detection within roughly three to six months of initiation. Clinical guidelines in many countries, including China’s national free ART program, embed viral load testing around this time point precisely because it separates patients on track for durable suppression from those whose regimens may need adjustment, adherence interventions, or intensified follow-up. The Hubei cohort, drawn from the provincial CDC’s electronic HIV database, offers a rare real-world window into how this milestone performs outside the carefully controlled conditions of clinical trials, where participant selection and support structures tend to inflate success rates.

A non-suppression rate of 6.81 percent across more than four and a half thousand treatment-naive patients may appear modest, but it represents a meaningful fraction of individuals whose early treatment course has already diverged from the expected trajectory. Each of these patients carries a higher probability of accumulating resistance mutations, experiencing longer periods of infectious viremia, and requiring more complex second-line therapy. From a public health perspective, the figure also matters at the population level: China’s strategy of expanding treatment coverage as a pillar of HIV control depends on the assumption that treated individuals achieve and maintain suppression. Early non-suppression, if left unaddressed, erodes that assumption.

The protective association with integrase strand transfer inhibitor-based initial regimens is among the most consequential findings. INSTIs have transformed HIV care globally because of their potency, tolerability, and high barrier to resistance, and their inclusion in first-line therapy has expanded steadily in China in recent years. The roughly two-thirds reduction in the odds of 24-week non-suppression observed here aligns with the pharmacological profile of this drug class, which achieves rapid declines in viral RNA and is generally forgiving of occasional missed doses compared with older agents. The finding carries practical weight for program planners weighing the cost of newer agents against the downstream costs of virological failure, resistance testing, and regimen switching.

By contrast, the elevated risk associated with non-nucleoside reverse transcriptase inhibitor-based regimens reflects both the biology and the history of HIV treatment in China. NNRTI-based combinations, often built around efavirenz, have long formed the backbone of the national free treatment program because they are inexpensive and domestically produced. Yet efavirenz is associated with neuropsychiatric side effects that can undermine adherence, and NNRTIs have a lower genetic barrier to resistance, meaning that even modest lapses in adherence can permit the emergence of resistant variants. The study’s identification of NNRTI-based starts as a risk factor for early non-suppression therefore echoes a broader international shift toward INSTI-containing first-line therapy, while also highlighting the equity question of whether patients in resource-constrained settings can access the more effective options.

The dose-response relationship with baseline viral load is biologically coherent. Patients entering treatment with higher plasma HIV RNA have a larger viral population to suppress, a greater probability of pre-existing minority resistant variants, and a longer expected time to reach undetectable levels. At a fixed 24-week checkpoint, those starting from higher baselines are structurally more likely to remain above the suppression threshold even when their treatment is working. This does not mean such patients are destined to fail; many will achieve suppression by 48 weeks. But it does mean that the 24-week measurement, interpreted in isolation, can flag individuals whose pharmacokinetic and virological starting conditions demand closer surveillance rather than immediate regimen change.

The mirror-image finding for baseline CD4 count tells a complementary story. Advanced immunosuppression at treatment initiation, reflected in counts below 200 cells per microliter, is associated with poorer early virological outcomes for reasons that include larger viral reservoirs, immune activation, possible subclinical opportunistic infections, and in some cases malabsorption or comorbidities that affect drug exposure. The observation that patients with counts between 200 and 499 cells per microliter fared better than those below 200 underscores a principle long recognized in HIV medicine: earlier diagnosis and treatment initiation improve outcomes across virtually every metric. As China expands routine testing and same-day or rapid ART initiation, the proportion of patients presenting with very low CD4 counts should decline, and with it, presumably, the burden of early non-suppression.

The nearly tenfold increase in odds associated with HIV acquisition through illicit drug use stands out as the strongest single risk factor in the analysis. This association likely operates through several converging pathways. People who use drugs face well-documented barriers to adherence, including unstable housing, incarceration, stigma within healthcare settings, and the logistical demands of daily medication amid active substance use. Opioid agonist therapy, peer support, and integrated addiction-HIV care have all been shown to improve suppression in this population, but access to such services varies widely. In China, where methadone maintenance programs exist but coverage and integration with HIV care remain uneven, the finding points to a specific, actionable gap: patients with drug-related acquisition routes need adherence support structures designed for their circumstances, not merely the same quarterly clinic visits offered to everyone else.

The lower non-suppression rate among women, with odds reduced by about a third, invites careful interpretation. Sex differences in antiretroviral pharmacokinetics are documented for several agents, and body composition, hormonal factors, and adherence patterns have all been proposed as contributors to sex-based outcome differences in HIV care. However, the direction of such associations varies across studies and settings, and unmeasured confounders, including differences in care engagement, disclosure, and social support, may play a role. The authors’ finding should therefore be read as a signal warranting further investigation rather than a settled clinical rule.

Methodologically, the study’s reliance on a provincial CDC electronic database confers both strengths and limitations. The database captures a large, geographically defined population under a standardized national treatment program, which strengthens external validity for similar settings in China. Yet retrospective designs inherit the data quality of routine care: viral load measurements may be missing or mistimed, adherence is typically not measured directly, and the 24-week window itself may be recorded with variable precision. The use of logistic regression with adjustment for demographic, clinical, and regimen factors mitigates some confounding, and the mention of inverse probability weighting in the abbreviations suggests the authors attended to issues such as loss to follow-up or informative censoring. Still, associations observed in observational data cannot fully exclude residual confounding by indication, since clinicians may preferentially assign certain regimens to patients perceived as more adherent or more robust.

The choice of a 200 copies per milliliter threshold for defining non-suppression also deserves attention. Guidelines differ on whether suppression should be defined below 200, below 50, or below the assay’s detection limit, and the proportion of patients classified as non-suppressed rises as the threshold tightens. A 200 copies per milliliter cutoff is clinically meaningful because persistent low-level viremia above this level is associated with increased risk of subsequent virological failure and resistance emergence, but it is more permissive than thresholds used in some high-income settings. Comparisons with studies using stricter definitions should therefore be made cautiously.

The study period, spanning February 2022 to June 2025, coincides with a period of significant evolution in China’s treatment guidelines and drug supply. The gradual expansion of INSTI-based and single-tablet regimens within the national program, alongside the continued role of domestically produced generic formulations, means the cohort likely contains a mixture of treatment eras. If regimen selection shifted over time, secular trends in non-suppression rates could partly reflect changing drug availability rather than changing patient behavior. The multivariable adjustment for initial regimen type addresses this partially, but temporal confounding remains a consideration for readers interpreting the pooled 2022 to 2025 estimate.

Beyond the specific associations, the study reinforces a model of HIV care in which the early treatment period functions as a triage window. Patients entering therapy with high viral loads, low CD4 counts, drug-related acquisition routes, or NNRTI-based regimens are identifiable at the moment of treatment initiation, before any viral load result is available. Risk-stratified follow-up, in which such patients receive earlier and more frequent viral load monitoring, structured adherence counseling, and prompt regimen review, could convert the 24-week measurement from a retrospective audit into a prospective intervention point. This aligns with a growing international consensus that early detection of non-response, including low-level viremia, enables timely switches that preserve future treatment options.

The findings also speak to the broader goal of ending the HIV epidemic as a public health threat. Treatment as prevention rests on the premise that suppressed individuals do not transmit the virus. Every patient who remains viremic at 24 weeks represents a window of extended transmission potential, particularly if unaware of their non-suppressed status. Reducing early non-suppression is therefore not only a clinical quality measure but a transmission-control strategy. In a province like Hubei, which combines large urban centers with extensive rural populations, the operational challenge is ensuring that the patients flagged as high risk, who are often the least connected to care, actually receive the intensified support the data suggest they need.

Finally, the study contributes to a still-thin evidence base on early virological outcomes in China. Much of the published literature on treatment failure in low- and middle-income countries focuses on 12-month outcomes or on pediatric and adolescent populations, while data on the 24-week checkpoint among adults initiating therapy under routine program conditions remain comparatively scarce. By quantifying both the prevalence and the modifiable determinants of early non-suppression in a large provincial cohort, the authors provide a benchmark against which other Chinese provinces, and programs in comparable settings, can measure their own performance. As INSTI access broadens and testing initiatives pull patients into care at higher CD4 counts, repeating this analysis in future cohorts will reveal whether the early treatment window is becoming more forgiving, and whether the disparities identified here are narrowing or persisting.

Subject of Research: Virological non-suppression at 24 weeks after antiretroviral therapy initiation among people living with HIV in Hubei, China

Article Title: Prevalence and determinants of virological non-suppression at 24 weeks among people living with HIV in Hubei, China

Article References: Hong, K., Zhao, D., Zheng, J., Liu, Y., Chen, L., Liu, L., Qi, S., Hu, Q., Tang, H., & Liang, K. (2026). Prevalence and determinants of virological non-suppression at 24 weeks among people living with HIV in Hubei, China. BMC Public Health. https://doi.org/10.1186/s12889-026-29870-0

Image Credits: AI Generated

DOI: 10.1186/s12889-026-29870-0

Keywords: HIV, antiretroviral therapy, virological non-suppression, viral load, CD4 count, INSTI, NNRTI, China, cohort study, public health, treatment adherence, viral suppression