Resistance mechanisms for INSTIs and NRTIs in bictegravir/FTC/TAF therapy

Abstract

Ten studies encompassing 19,608 participants demonstrated that bictegravir/emtricitabine/tenofovir alafenamide (B/F/TAF) exhibits distinct resistance patterns depending on treatment history. Pre-existing NRTI resistance was common in treatment-experienced populations, ranging from 2.7% in treatment-naive individuals to 25% in real-world cohorts, with M184V/I mutations present in 10-16% of participants. Despite this, viral suppression rates remained 96-100% among individuals with archived NRTI resistance, indicating that these mutations do not compromise B/F/TAF efficacy. Pre-existing INSTI resistance was rare (0.6-4%), and treatment-emergent resistance to any B/F/TAF component was absent in randomized trials through 144 weeks but occurred in 3-4% of real-world treatment-experienced populations. The critical resistance mechanism identified was prior INSTI virologic failure, which increased viral rebound risk 2.68-fold, while major INSTI resistance mutations increased risk 4.21-fold. Phenotypic data revealed that Q148H+G140S, which confers high-level resistance to first-generation INSTIs, maintained bictegravir sensitivity at <2.5-fold change, demonstrating bictegravir’s activity against some INSTI-resistant variants. B/F/TAF demonstrates an exceptionally high barrier to de novo resistance development in INSTI-naive populations but faces challenges in individuals with established INSTI resistance from prior treatment failures, while archived NRTI resistance does not predict virologic failure.

Methods

We analyzed 10 sources from an initial pool of 200, using 8 screening criteria. Each paper was reviewed for 7 key aspects that mattered most to the research question. More on methods

Paper search

We performed a semantic search across over 138 million academic papers from the Elicit search engine, which includes all of Semantic Scholar and OpenAlex.

We ran this query: “Resistance mechanisms for INSTIs and NRTIs in bictegravir/FTC/TAF therapy”

The search returned 200 total results from Elicit.

Screening

We screened in sources based on their abstracts that met these criteria:

• Drug Combination: Does this study involve patients treated with bictegravir/FTC/TAF combination therapy?
• Resistance Outcomes: Does this study report resistance mechanisms, mutations, or genotypic resistance patterns for INSTIs and/or NRTIs?
• Study Population: Does this study include HIV-positive patients of any age?
• Resistance Evidence: Does this study report laboratory or clinical evidence of drug resistance?
• Study Design: Is this study a randomized controlled trial, cohort study, case-control study, cross-sectional study, case series with ≥5 patients, systematic review, or meta-analysis?
• Relevant Drug Focus: Does this study include bictegravir/FTC/TAF (rather than focusing solely on other HIV drug combinations)?
• Beyond PK/PD Only: Does this study report resistance outcomes (rather than only pharmacokinetic or pharmacodynamic data without resistance outcomes)?
• Resistance Analysis Included: Does this study include resistance analysis (rather than involving only drug-susceptible isolates without any resistance analysis)?

Results

Characteristics of Included Studies

Ten studies evaluating resistance mechanisms in bictegravir/emtricitabine/tenofovir alafenamide (B/F/TAF) therapy were included, encompassing 19,608 participants across clinical trials and real-world cohorts.

The studies comprised both treatment-naive populations and treatment-experienced individuals switching to B/F/TAF.

Pre-existing Resistance Mutations

Pre-existing NRTI resistance mutations were common across studies, while INSTI resistance mutations were rare.

Pre-existing NRTI resistance prevalence ranged from 2.7% in treatment-naive populations to 25% in a real-world cohort of treatment-experienced individuals. M184V/I mutations, which confer resistance to emtricitabine and lamivudine, were the most commonly detected NRTI mutations, present in 10-14% of participants across multiple studies. Thymidine analog mutations (TAMs) were identified in 7% of participants in the BRAAVE study.

Pre-existing INSTI resistance mutations were substantially less common, with prevalence rates of 0.6-4% across studies that reported this data. Among the specific mutations identified, the Q148H substitution, particularly in combination with G140S, was detected in multiple studies. Other documented INSTI resistance mutations included E92G, Y143C/H, S147G, N155S, and R263K.

Treatment-Emergent Resistance

Treatment-emergent resistance to B/F/TAF components was rare across all studies.

The majority of clinical trials reported zero treatment-emergent resistance to any B/F/TAF component through 48-144 weeks of follow-up. In the pivotal 144-week studies 1489 and 1490 involving treatment-naive participants, no emergent resistance was observed despite 21 participants qualifying for resistance testing. Similarly, study 4030 involving treatment-experienced participants switching to B/F/TAF reported no treatment-emergent resistance among three participants meeting criteria for resistance analysis.

Real-world cohort data revealed slightly higher rates of emergent resistance. The Virostar-1 study reported 4% emergent INSTI or NRTI resistance-associated mutations (RAMs) among virologic failures, though notably documented a complete absence of dual resistance to both NRTIs and INSTIs. More recently, the 2025 Marcelin study reported 3% treatment-emergent RAMs with B/F/TAF, with dual INSTI and NRTI RAMs observed in some cases.

Virologic Outcomes

High rates of viral suppression were maintained across studies, including among participants with pre-existing resistance mutations.

Viral suppression rates at 48 weeks ranged from 97-100% across clinical trials. Notably, the presence of pre-existing NRTI resistance, including M184V/I mutations, did not diminish treatment efficacy. In the BRAAVE study, 100% of participants with NRTI resistance and 100% with INSTI resistance achieved HIV-1 RNA <50 copies/mL at week 48. Similarly, 96% of participants with archived M184V/I maintained suppression through 48 weeks.

Treatment-naive participants achieved rapid viral suppression. In the subset with pre-existing INSTI resistance, one treatment-naive participant with Q148H+G140S achieved viral suppression by week 4 and maintained suppression through week 48. The 144-week follow-up data confirmed durability, with 98% of participants with transmitted drug resistance substitutions achieving HIV RNA <50 copies/mL.

Virologic failure rates remained low across studies. Real-world cohorts reported failure rates of 5.3-6.8%, while clinical trials documented failure rates of approximately 1% or less. When viral blips occurred, they were uncommon, observed in only 2.7% of participants in one study.

Risk Factors for Virologic Failure

Prior INSTI virologic failure and the presence of major INSTI resistance mutations emerged as the primary risk factors for viral rebound on B/F/TAF therapy.

The Italian ARCA cohort identified specific risk factors with quantified associations. After controlling for confounding, a history of INSTI virological failure was associated with a 2.68-fold increased risk of viral rebound (aRH 2.68, 95% CI: 1.40-5.12). The presence of major INSTI-DRMs conferred an even higher risk, with an adjusted hazard ratio of 4.21 (95% CI: 1.18-15.02). Notably, NRTI resistance was not associated with viral rebound in this cohort.

Adherence factors played a role in virologic outcomes. In the treatment-naive population, low adherence (<95%) and missed doses were identified as likely contributors to virologic rebound, though preexisting resistance substitutions did not reach statistical significance as predictors of treatment failure (P >0.05).

The presence of pre-existing resistance mutations varied widely but generally did not predict treatment failure in the absence of prior INSTI failure. Pre-existing NRTI resistance was detected in 14-24% of participants across studies, with M184V/I mutations present in 10-16%, yet these mutations did not compromise virologic outcomes in most cohorts.

Resistance Mechanisms

Limited mechanistic data on resistance to B/F/TAF components were available across the included studies. One study with full-text availability provided phenotypic resistance data for specific mutations. The Q148H+G140S combination in integrase showed high-level resistance to raltegravir and elvitegravir but remained sensitive to bictegravir, with a fold-change of 2.14. This pattern demonstrated differential cross-resistance between INSTIs, with mutations conferring resistance to first-generation INSTIs not necessarily affecting bictegravir sensitivity.

Potential resistance mechanisms for bictegravir were identified, including S153F/Y or R263K substitutions, with or without M50I, which conferred low-level reduced susceptibility. Secondary INSTI resistance substitutions such as M50I, S119P/R/T, and E157K/Q were noted but did not confer resistance independently.

Cross-resistance patterns indicated that bictegravir maintained activity against some INSTI resistance mutations. In one case, virus with Q148H+G140S demonstrated full sensitivity to bictegravir but only partial sensitivity to dolutegravir (phenotype <2.5-fold change and >4-fold change, respectively). Both bictegravir and dolutegravir were described as having a high barrier to resistance in vitro and in clinical studies.

Most studies did not report detailed mechanistic data on phenotypic resistance levels, compensatory mutations, or fitness costs associated with resistance to B/F/TAF components.

Synthesis

The evidence demonstrates consistently low rates of treatment-emergent resistance to B/F/TAF across diverse populations and settings, yet reveals important distinctions in resistance patterns between treatment-naive and treatment-experienced populations, particularly those with prior INSTI exposure.

Context-Specific Resistance Risk

Studies of treatment-naive populations universally reported zero emergent resistance through 48-144 weeks of follow-up, while real-world cohorts of treatment-experienced individuals showed emergent resistance rates of 3-4%. This differential likely reflects the impact of prior ART exposure rather than inherent differences in resistance barriers. The Italian ARCA cohort provided mechanistic insight: among treatment-experienced participants, those with prior INSTI virologic failure had a 2.68-fold increased risk of viral rebound, and those with major INSTI-DRMs had a 4.21-fold increased risk. Critically, NRTI resistance—despite being highly prevalent at 25% in this cohort—was not associated with virologic failure. Both findings may be correct within their respective margins: B/F/TAF demonstrates a high barrier to de novo resistance development in INSTI-naive patients while maintaining efficacy despite archived NRTI resistance, but faces challenges in populations with established INSTI resistance from prior treatment failures.

NRTI Resistance Paradox

A striking pattern emerged regarding M184V/I mutations. Pre-existing M184V/I was detected in 10-16% of participants across switching studies, yet 96-100% of individuals with these mutations achieved and maintained viral suppression. This apparent paradox may be explained by emtricitabine’s dual role in the regimen. While M184V/I confers high-level resistance to emtricitabine and lamivudine, these mutations also reduce viral fitness and increase susceptibility to tenofovir. The continued presence of bictegravir, to which participants had no prior exposure in most studies, likely provides sufficient antiviral activity to maintain suppression even when emtricitabine efficacy is compromised. This interpretation is supported by the complete absence of dual NRTI and INSTI resistance development in the Virostar-1 study, suggesting that as long as bictegravir maintains full activity, archived NRTI mutations do not drive virologic failure.

Study Quality Hierarchy

The evidence base includes both high-quality randomized trials with rigorous virologic monitoring and real-world observational studies. Among the 21 participants across randomized trials who qualified for resistance testing due to virologic failure, zero developed emergent resistance to study drugs. The largest and longest randomized trial (N=1274, 144 weeks) found no emergent resistance despite comprehensive deep-sequencing surveillance at 15% detection thresholds. In contrast, real-world French cohorts reported 3-6% emergent resistance or virologic failure rates. However, these real-world studies included participants with more complex treatment histories—up to second-line therapy or multiple prior regimens—and may have had less stringent adherence monitoring than controlled trials. The randomized trial evidence thus suggests an extremely high intrinsic barrier to resistance for B/F/TAF, while real-world data reveal that this barrier can be overcome in populations with prior INSTI failure.

Mechanistic Explanations

The limited phenotypic data available support the clinical observations. The Q148H+G140S pattern, while conferring high-level resistance to raltegravir and elvitegravir, maintained bictegravir sensitivity at <2.5-fold change. This demonstrates that bictegravir’s structural properties allow it to maintain binding affinity to integrase even in the presence of mutations that abolish first-generation INSTI activity. The high genetic barrier to bictegravir resistance is further evidenced by the requirement for multiple mutations (e.g., S153F/Y plus R263K plus M50I) to achieve even low-level reduced susceptibility. Secondary mutations like M50I, S119P/R/T, and E157K/Q do not independently confer resistance, suggesting that resistance pathways require coordinated multi-step evolution that rarely occurs under selective pressure from B/F/TAF.

Population and Adherence Factors

While pre-existing resistance patterns did not predict treatment failure in most cohorts, adherence emerged as a critical factor. Low adherence (<95%) and missed doses contributed to virologic rebound in treatment-naive populations, yet even in these cases, no treatment-emergent resistance developed. This suggests that suboptimal adherence leads to viral rebound through insufficient drug exposure rather than through selection of resistant variants. The high forgiveness of B/F/TAF for adherence lapses likely reflects the long half-lives of tenofovir alafenamide’s active metabolite and the sustained intracellular bictegravir concentrations, which maintain suppressive drug levels even with intermittent dosing.

Temporal Patterns

Durability of response extended through 144 weeks in treatment-naive populations without emergence of resistance, and through 36 months in a real-world switching cohort with only 5.3% viral rebound. The stability of suppression over these extended periods, even among participants with archived resistance, indicates that B/F/TAF does not select for resistance over time in the absence of pre-existing INSTI resistance. The early achievement of suppression—within 4 weeks in one participant with Q148H+G140S—and maintenance without viral blips in the INSTI-resistant subset further supports the potency of the regimen even against partially resistant virus.

For treatment-naive individuals or those without prior INSTI exposure, B/F/TAF demonstrates an exceptionally high barrier to resistance regardless of archived NRTI mutations. For treatment-experienced individuals with prior INSTI virologic failure or documented major INSTI mutations, alternative regimens may be preferable, as this population faces a 2.7-4.2-fold increased risk of virologic failure. The absence of dual NRTI-INSTI resistance development suggests that when resistance does emerge in experienced populations, it typically affects either the INSTI or NRTI component rather than both simultaneously, preserving some treatment options.