Highlights
– In a 13-site prospective cohort (ACTG A5381–Hakim) across six PEPFAR-supported countries, initiation or switching to tenofovir–lamivudine–dolutegravir (TLD) yielded high viral suppression for people suppressed at switch and for ART-naive individuals.
– Participants who switched from failing protease-inhibitor regimens (VL >1000 copies/mL) had lower suppression and were the only group with emergent dolutegravir-associated mutations (G118R, R263K) in three individuals.
– Objective adherence assessment (tenofovir diphosphate in dried blood spots) was significantly lower among those with virological failure, implicating incomplete adherence as a major driver of suboptimal outcomes and resistance emergence.
Background
Dolutegravir-based regimens, commonly combined as tenofovir disoproxil fumarate, lamivudine, and dolutegravir (TLD), have become the backbone of first- and many second-line antiretroviral therapy (ART) programs globally because of potency, tolerability, lower cost, and a high genetic barrier to resistance. International programmes supported by the US President’s Emergency Plan for AIDS Relief (PEPFAR) have widely adopted TLD. Nonetheless, real-world data from low- and middle-income settings are needed to define long-term virological outcomes, the frequency and context of dolutegravir-resistance emergence, and the role of adherence when patients are switched from other regimens or start ART de novo.
Study design
The Advancing Clinical Therapeutics Globally (ACTG) A5381–Hakim Study is a prospective cohort conducted at 13 PEPFAR-supported sites in Haiti, Kenya, Malawi, South Africa, Uganda, and Zimbabwe between Oct 28, 2019, and Sept 27, 2022. Participants were aged ≥10 years and either initiated or switched to TLD. Participants were grouped as follows:
– Group 1 (NNRTI → TLD): 1A had HIV-1 RNA >1000 copies/mL at switch; 1B had ≤1000 copies/mL at switch.
– Group 2 (PI → TLD): 2A had HIV-1 RNA >1000 copies/mL at switch; 2B had ≤1000 copies/mL at switch.
– Group 3: ART-naive individuals initiating TLD.
Primary outcomes included viral suppression (HIV-1 RNA ≤1000 copies/mL) over 36 months, emergence of dolutegravir-associated resistance mutations, and adverse events. Adherence was objectively assessed in a nested case–control substudy using tenofovir diphosphate (TFV‑DP) concentrations in dried blood spots (DBS).
Key findings
Enrollment and population:
1241 participants were enrolled; after exclusions, 1237 were analysed: group 1A n=44, 1B n=425, 2A n=173, 2B n=416, and group 3 n=179. Overall, 65% were female and 35% male. Twelve participants (≈1%) discontinued TLD because of adverse events.
Viral suppression (HIV-1 RNA ≤1000 copies/mL):
– Group 1A (NNRTI switch, failing at switch): 88% (37/42) suppressed at 6 months; 76% (16/21) at 24 months. Interpretation: many achieved suppression after switching despite initial failure, though numbers are small and drop-out reduces precision over time.
– Group 1B (NNRTI switch, suppressed at switch): 99% (380/384) at 6 months; 98% (368/375) at 24 months. Interpretation: switching suppressed patients from NNRTI-based regimens to TLD preserved high suppression rates.
– Group 2A (PI switch, failing at switch): 72% (118/165) at 6 months; 70% (45/64) at 24 months. Interpretation: lower suppression in this subgroup compared with suppressed switchers or ART-naive individuals.
– Group 2B (PI switch, suppressed at switch): 95% (376/395) at 6 months; 93% (190/204) at 24 months. Interpretation: similar to group 1B, switching suppressed PI-treated patients maintained suppression in most participants.
– Group 3 (ART‑naive initiating TLD): 90% (136/151) at 6 months; 90% (128/143) at 24 months. Interpretation: high efficacy for treatment initiation in routine programme settings.
Resistance outcomes:
Mutations associated with reduced dolutegravir susceptibility were identified in three participants, all in group 2A: the G118R and R263K mutations (these mutations are known to decrease integrase inhibitor susceptibility and arise under selective pressure). No dolutegravir-resistance mutations were detected in other groups during follow-up.
Adherence substudy (TFV‑DP in DBS):
In a nested case–control analysis (87 case-control pairs at 6 months), TFV‑DP concentrations were significantly lower among participants with HIV-1 RNA >1000 copies/mL than among those with ≤1000 copies/mL (p<0·0001), indicating that incomplete cumulative adherence was strongly associated with virological non-suppression and, by implication, with the risk of resistance emergence.
Safety:
Twelve participants discontinued TLD owing to adverse events (≈1%); overall the regimen was well tolerated in this large, programme-based cohort.
Interpretation and clinical implications
Preserved suppression after switching supports current international recommendations to transition patients who are virologically suppressed on older regimens (NNRTI- or PI-based) to TLD. The very high suppression rates in groups 1B and 2B demonstrate that the switch is safe and effective at the programme level across diverse low- and middle-income country settings.
However, the combination of lower suppression in individuals switched from failing PI regimens (group 2A), objective evidence of poor adherence among those who failed, and the detection of dolutegravir-associated mutations in three such individuals highlights an important caveat. While dolutegravir has a higher genetic barrier to resistance than earlier integrase inhibitors and NNRTIs, resistance can still emerge when selective pressure is applied in the setting of ongoing viremia and suboptimal drug exposure. Pre-existing nucleoside reverse transcriptase inhibitor (NRTI) resistance or insufficient adherence to the entire regimen may create the conditions for integrase resistance to develop after switching.
Operationally, these findings suggest a pragmatic pathway for programmes:
– Continue to switch virologically suppressed patients to TLD without delay, supported by routine viral load monitoring.
– For patients with confirmed virological failure at time of planned switch, prioritise targeted interventions before or at switch: intensified adherence support, repeat viral load measurement, and, where available, resistance testing to guide regimen choice.
– Monitor closely after switching any patient with recent unsuppressed viral load; early detection of persistent viremia allows timely adherence interventions or regimen adjustment and limits the risk of emergent resistance.
Mechanistic considerations
The integrase mutations observed (G118R, R263K) have been reported to reduce susceptibility to dolutegravir and are associated with fitness costs to the virus — which may partly explain why dolutegravir resistance remains relatively uncommon. Nonetheless, under the selective pressure of ongoing replication with partial drug exposure (e.g., subtherapeutic TFV‑DP levels), these mutations can be selected. In addition, if the NRTI backbone has compromised activity due to pre-existing resistance, the effective regimen becomes closer to dolutegravir monotherapy, further increasing selection risk.
Study strengths and limitations
Strengths include a large, multisite prospective design in routine programme settings across six countries, objective adherence assessment using TFV‑DP in DBS, and systematic viral load follow-up up to 36 months. These features increase generalisability to public-health ART programmes in similar settings.
Limitations include smaller subgroup sizes for some comparisons (notably group 1A), loss to follow-up and fewer participants with on-treatment viral load measures at later timepoints which reduced precision, and observational (non-randomised) design which limits causal inferences. The resistance testing strategy and timing were not described in full detail here; thus, detection of resistance may depend on sampling thresholds and testing practices. Finally, implementation contexts vary across sites, and resources for routine resistance testing are limited in many settings, which constrains direct application of genotype-guided approaches.
Conclusions
The ACTG A5381–Hakim prospective cohort provides reassuring evidence that TLD is highly effective when used to replace older regimens in virologically suppressed people living with HIV and in ART-naive individuals in PEPFAR-supported programmes. However, clinicians and programmes should be alert to the risk of reduced suppression and rare emergence of dolutegravir resistance when switching patients who have ongoing viremia, particularly in the context of poor adherence. Targeted adherence interventions, vigilant post-switch viral load monitoring, and access to resistance testing where feasible remain important to preserve the long-term efficacy of dolutegravir-based ART at the population level.
Funding and trial registry
Funding: National Institutes of Health and PEPFAR (as reported by the study). ClinicalTrials.gov: not specified in the manuscript citation provided.
Selected references
1) Marc JB, McCarthy C, Wallis CL, et al.; ACTG A5381–Hakim Study Team. Virological and drug-resistance outcomes for people living with HIV initiating or switching to tenofovir, lamivudine, and dolutegravir in six PEPFAR-supported countries: a prospective cohort study. Lancet HIV. 2025;12(12):e836–e849. doi:10.1016/S2352-3018(25)00162-6.
2) World Health Organization. Consolidated guidelines on HIV prevention, testing, treatment, service delivery and monitoring: recommendations for a public health approach (2021 update). World Health Organization; 2021. (Provides contemporary guidance on dolutegravir use in resource-limited settings.)
Note: Readers seeking operational detail and site-level implementation data should consult the full Lancet HIV manuscript and supplementary appendices for methods, genotyping procedures, and per-site outcomes.
