Large-scale evaluation of HIV-1 DNA drug resistance testing as a robust tool for clinical decision-making: A...

Human immunodeficiency virus type 1 (HIV-1) drug resistance remains a major challenge in HIV/AIDS management, particularly in individuals with low-level viremia (LLV) where RNA-based drug resistance testing (DRT) often fails. Although HIV-1 DNA DRT represents a promising alternative, its clinical utility has been constrained by insufficient evidence. This nationwide study in China enrolled 9,428 people living with HIV (PLWH), analyzing 10,903 samples spanning a wide viral load (VL) spectrum. To improve RNA detection, an optimized primer design combined with an extracellular particle (EP)-HIV co-isolation technique was developed. We then evaluated the reproducibility of drug resistance mutation (DRM) profiles between paired RNA and DNA DRTs using Sanger sequencing (SS), with single-molecule sequencing employed to establish a dominant sequence threshold. Our findings demonstrated that primer optimization and EP-co-isolation significantly enhanced RNA amplification success. DRMs were prevalent across all VL strata. The combined concordance and degeneracy rates (C/D rates) (where multiple DNA DRMs included all RNA-derived DRMs) between RNA and DNA DRTs ranged from 90.4% to 100% in different gene regions, with higher discordance rates observed in the nucleoside reverse transcriptase inhibitor (NRTI) and non-NRTI (NNRTI) regions. Based on Stanford penalty scores across 25 antiretroviral drugs, the degeneracy group showed a 98.3% ± 1.7% interpretation agreement. Even within the discordance group, mean agreement remained high (89.5% ± 5.0%), with only four NNRTIs exhibiting agreement below 85%. The dominant sequence proportion threshold for HIV-1 DNA was determined to be 24.6%. This study provides strong evidence supporting the integration of HIV-1 DNA DRT into clinical practice for reliable drug resistance surveillance and treatment monitoring.