The investigation was designed as a cross-sectional study drawing participants from the provincial HIV-positive surveillance repository in Jiangxi, a province in southeastern China with a substantial population under long-term antiretroviral care. The researchers applied standardized sample submission criteria to assemble two distinct cohorts. The first comprised 104 treatment-naïve individuals whose baseline viral load had been measured before starting therapy or within 30 days of initiating antiretroviral treatment, allowing assessment of pretreatment drug resistance. The second comprised 307 treatment-experienced individuals who had received standardized antiretroviral therapy for at least six months and who met the definition of virological failure, operationalized as a viral load exceeding 400 copies per milliliter of plasma. By anchoring enrollment in the surveillance system rather than in convenience sampling from single clinics, the study sought to capture a province-wide snapshot of resistance rather than the picture at selected treatment centers alone.

Genotypic resistance testing focused on the pol gene, which encodes the viral enzymes reverse transcriptase and protease, and on the integrase gene, which encodes the enzyme that inserts viral DNA into the host genome. Resistance profiles were interpreted using the Stanford HIV Drug Resistance Database, a widely used algorithm that translates sequence mutations into predicted levels of susceptibility for individual antiretroviral drugs. This genotypic approach detects resistance-associated mutations in the viral genome, the molecular footprints left behind when imperfectly suppressed virus replicates under drug pressure. Because HIV-1 lacks proofreading capability in its reverse transcriptase and replicates prolifically, every untreated infection already carries a swarm of viral variants, and drug exposure selectively amplifies those variants fortunate enough to carry mutations that blunt drug binding or activity.

The headline numbers concern multidrug resistance. In the treatment-naïve group, representing the largely transmitted or pretreatment component of the resistance landscape, multidrug resistance stood at 10.58 percent. Among patients who had failed therapy, the figure reached 28.01 percent, a statistically significant increase reported at a p value below 0.001. This near tripling quantifies what the authors describe as a profound shift: once antiretroviral pressure is applied and virological control is lost, the virus does not merely accumulate resistance to one agent but frequently acquires mutations that compromise entire drug classes simultaneously. For clinicians, multidrug resistance is the form of resistance that matters most, because it constrains the options available for constructing an effective salvage regimen and foreshadows cumulative treatment failure at the population level.

Class-by-class analysis showed that mutations associated with non-nucleoside reverse transcriptase inhibitors predominated. The K103N and K103R mutations, classic signatures of resistance to nevirapine and efavirenz, were detected in 28.66 percent of the virological failure cohort, compared with 9.61 percent among treatment-naïve individuals. These substitutions alter the hydrophobic pocket to which non-nucleoside inhibitors bind, reducing drug affinity by orders of magnitude while leaving viral replicative capacity largely intact, which is precisely why they outcompete wild-type virus under non-nucleoside-based therapy. Alongside this, the M184V and M184I mutations were markedly enriched in the failure group at 25.73 percent. M184V/I confers resistance to lamivudine and emtricitabine, the nucleoside analogues that typically anchor first-line regimens, and its prevalence in failed therapy severely compromises the nucleoside reverse transcriptase inhibitor backbone on which those regimens depend.

Against this troubling backdrop, two drug classes remained notably durable. Susceptibility to protease inhibitors and to integrase strand transfer inhibitors exceeded 93 percent in the study population, indicating that these components of combination therapy retained activity in the overwhelming majority of patients even after treatment failure. This durability reflects both the genetic barrier and the usage patterns of these drugs in the region: protease inhibitors require multiple sequential mutations to lose potency, and integrase inhibitors, particularly the second-generation agents dolutegravir and bictegravir, impose a high fitness cost on resistant variants. The findings therefore support a clinical pivot toward robust salvage regimens built around integrase strand transfer inhibitors, which the authors advocate urgently, especially as Jiangxi and comparable high-burden settings transition toward dolutegravir-based first-line therapy as the new standard of care.

Yet the integrase picture was not entirely reassuring. The study detected the polymorphic accessory mutations L74I or L74M in 2.93 percent of the virological failure cohort. Accessory mutations of this kind do not usually cause outright resistance to first-generation integrase inhibitors such as raltegravir or elvitegravir on their own, but they can contribute to reduced susceptibility and may serve as stepping stones toward higher-level resistance when combined with other substitutions. Their presence at baseline, before patients have accumulated extensive integrase inhibitor exposure, underscores a gap in current surveillance practice: because integrase inhibitors have only recently been scaled up in many Chinese treatment programs, baseline integrase gene sequences have not been routinely collected. The authors argue that establishing integrase surveillance now, before widespread dolutegravir and bictegravir exposure reshapes the viral population, is essential for detecting emerging resistance pathways early.

The distinction between pretreatment drug resistance and acquired drug resistance frames the public health interpretation of these findings. Pretreatment resistance, at roughly one in ten treatment-naïve individuals, reflects resistance transmitted from person to person or acquired during prior, sometimes undocumented, drug exposure. Acquired resistance, by contrast, is generated de novo within patients whose replication is incompletely suppressed, often through adherence gaps, pharmacokinetic variability, or inadequate regimens. The Jiangxi data show that the second process is the dominant engine of the province’s multidrug resistance burden, nearly tripling its prevalence. Each case of virological failure left unmanaged does not merely harm the individual patient; it can seed transmitted resistance into the wider epidemic, so intensified molecular surveillance and prompt regimen switching upon failure are the twin strategies the findings support.

Operationally, the study argues for concrete changes in the region’s treatment architecture. Regular viral load monitoring must be strengthened so that failure, defined here as a viral load above 400 copies per milliliter, is detected and acted upon before extensive resistance accumulates. Resistance genotyping should be deployed at the point of confirmed failure rather than reserved for exceptional cases, and the tested gene panel should expand to include integrase as second-generation integrase inhibitors become first-line. Salvage regimens should be anchored in drugs retaining high susceptibility, with the >93 percent susceptibility to protease and integrase inhibitors in this cohort suggesting these classes are the most reliable foundations. As China consolidates dolutegravir-based therapy nationally, the Jiangxi experience offers a quantitative baseline: a population carrying 10.58 percent pretreatment multidrug resistance that, under prior regimens, climbed to 28.01 percent after treatment exposure. Sustaining the durability that integrase inhibitors currently provide will depend on surveillance that moves as fast as the virus does.

Subject of Research: Acquired multi-class HIV-1 drug resistance emerging after antiretroviral therapy exposure in Jiangxi Province, China

Article Title: Treatment exposure drives a profound shift toward complex multi-class HIV-1 drug resistance

Article References: Mao, Z. Q., Mao, J., Zheng, H., Hu, Y., Tang, X. M., & Zhang, Q. L. (2026). Treatment exposure drives a profound shift toward complex multi-class HIV-1 drug resistance. Virology Journal. https://doi.org/10.1186/s12985-026-03308-w

Image Credits: AI Generated

DOI: 10.1186/s12985-026-03308-w

Keywords: HIV-1, drug resistance, antiretroviral therapy, virological failure, multidrug resistance, NNRTI, NRTI, integrase inhibitors, protease inhibitors, pretreatment drug resistance, acquired drug resistance, genotypic resistance testing