Non-INSTI-based ART combination on the horizon for broad HIV population

The investigational combination of doravirine plus islatravir (DOR/ISL)—a fixed-dose, once-daily oral antiretroviral therapy (ART) regimen that does not rely on integrase strand-transfer inhibitors (INSTIs)—is shaping up to be a promising alternative in the management of HIV, showing noninferiority to bictegravir/emtricitabine/tenofovir alafenamide (BIC/FTC/TAF) in individuals initiating treatment for the first time and maintaining virologic suppression in those switching from various ART regimens, as shown in three phase III studies presented at CROI 2026.

First-line treatment option

The MK-8591A-053 trial, which involved treatment-naïve adults with HIV-1 RNA of ≥500 copies/mL at baseline, met its primary noninferiority endpoint. [CROI 2026, abstract 177]

By week 48, the percentage of participants who achieved virologic suppression (HIV-1 RNA <50 copies/mL) was very high—91.8 percent with DOR/ISL vs 90.6 percent with BIC/FTC/TAF—with a treatment difference of only 1.2 percent (95 percent confidence interval [CI], –3.7 to 6.2), reported first study author Prof Jürgen Rockstroh from the University Hospital Bonn in Bonn, Germany.

CD4+ T-cell count increased by a mean of 218 cells/mm³ in the DOR/ISL arm and by 226 cells/mm³ in the BIC/FTC/TAF arm (difference, –6.8 cells/mm³, 95 percent CI, –35.8 to 22.2).

Discontinuation due to lack of efficacy was very low, according to Rockstroh, with four participants in the DOR/ISL arm and two in the BIC/FTC/TAF arm. Treatment-emergent resistance developed in two participants in the DOR/ISL arm.

In terms of safety, drug-related adverse events (AEs) occurred in 14.1 percent of participants in the DOR/ISL arm vs 18 percent in the BIC/FTC/TAF arm. The most common AEs were weight gain (2.6 percent vs 3.4 percent, respectively), headache (2.2 percent vs 3.4 percent), decreased estimated glomerular filtration rate (0.7 percent vs 3 percent), and abdominal distension (0.4 percent vs 2.2 percent).

Two participants in the DOR/ISL arm experienced drug-related serious AEs, including drug reaction with eosinophilia and systemic symptoms in one and liver injury not meeting Hy’s Law in the other. Two participants in the BIC/FTC/TAF arm died: one had alcoholic cirrhosis and the other had dilated cardiomyopathy.

Rockstroh noted that DOR/ISL was not a hepatitis B virus (HBV)-protective combination. During the trial, HBV infections occurred in three participants in the DOR/ISL arm. Two of them had hepatitis B core antibody (HBcAb) detected after ALT elevations and continued DOR/ISL. The third participant had HBV DNA and HBsAg positivity and discontinued DOR/ISL; HBV resolved on BIC/FTC/TAF. “This emphasizes the importance of vaccination in people with HIV as per current guidelines,” he said.

Finally, participants in both treatment arms gained weight during the 48-week trial. The mean increase in weight was 3.6 kg with DOR/ISL and 3.9 kg with BIC/FTC/TAF, with the difference not statistically significant.

Switch treatment option

Updated 96-week data from two phase III trials also demonstrated the efficacy and safety of DOR/ISL as a switch treatment option in virologically suppressed adults.

MK-8591A-052

In the blinded MK-8591A-052 trial, virologic suppression was maintained through week 96 in 88.9 percent of participants who switched to DOR/ISL and in 90.1 percent of those who remained on BIC/FTC/TAF (difference, –1.2 percent, 95 percent CI, –6.5 to 5.0). [CROI 2026, poster 514]

“A switch to DOR/ISL had no adverse effects on total lymphocyte and CD4+ T-cell counts through week 96,” said lead investigator Dr Amy Colson from Cambridge Health Alliance, Cambridge, Massachusetts, US.

Mean changes in CD4+ T-cell counts were 21.9 cells/mm³ in the DOR/ISL vs 26.7 cells/mm³ in the BIC/FTC/TAF arm (difference, 1.6 cells/mm³, 95 percent CI, –31.5 to 34.7). Mean changes in total lymphocyte counts were –0.02 vs 0.01 x 109 cells/L, respectively (difference, –0.03 x 109 cells/L, 95 percent CI, –0.10 to 0.05).

Five participants in the DOR/ISL arm discontinued medication due to lack of efficacy in two, due to protocol-specified decreases in CD4+ T-cell and/or total lymphocyte count in the other two, and due to transient low-level HBV viremia (DNA ≤36 IU/mL) in the fifth who was HBcAb-positive at baseline. Treatment-emergent resistance developed in one participant in the DOR/ISL group.

“DOR/ISL was generally well tolerated over 96 weeks. The AE profiles of DOR/ISL and BIC/FTC/TAF were comparable,” noted Colson.

Drug-related AEs occurred in 11.1 percent of participants in the DOR/ISL arm (including a serious drug-related AE in one [0.3 percent]) vs 10.5 percent in the BIC/FTC/TAF arm. The most common AEs were diarrhoea (1.5 percent), flatulence, decreased lymphocyte count, and pruritus (1.2 percent each) with DOR/ISL; and sleep disorder, osteopenia, decreased weight, and decreased appetite (1.2 percent each) with BIC/FTC/TAF. Drug-related AEs led to discontinuation in six and four participants in the respective treatment arms. No deaths occurred over 96 weeks.

New HBV infections occurred in four DOR/ISL participants, all of whom remained asymptomatic. Two of these were HBV surface antigen positive with DNA >200 000 IU/mL and discontinued DOR/ISL; HBV viremia subsequently resolved after initiation of an ART regimen with anti-HBV activity. There were no cases of clinical HBV reactivation.

Changes in weight over 96 weeks were 0.34 kg in the DOR/ISL arm and 0.43 in the BIC/FTC/TAF arm.

“These findings are consistent with the study’s primary week-48 results,” Colson said.

MK-8591A-051

In the open-label MK-8591A-051 trial, switching to DOR/ISL maintained high rates of virologic suppression through 96 weeks of treatment, according to lead investigator Prof Chloe Orkin from Queen Mary University of London, London, UK. [CROI 2026, poster 515]

Virologic suppression rates at week 48 were 95.6 percent among participants who switched to DOR/ISL and 91.9 percent among those who continued with their baseline ART (bART), including those based on INSTIs, non-nucleoside reverse transcriptase inhibitors (non-NRTIs), and boosted protease inhibitors. At this point, participants in the bART arm were then switched to DOR/ISL, while those in the DOR/ISL arm continued with their regimen, so that all participants were on DOR/ISL between weeks 48 and 96. Virologic suppression rate at week 96 remained high at 92.6 percent.

At week 96, two participants in the bART arm had HIV-1 RNA ≥50 copies/mL. One of them discontinued due to lack of efficacy, while the other was resuppressed on DOR/ISL at week 108 and remains in the study.

“There was no evidence of viral resistance to either DOR or ISL in any participants,” Orkin said.

During the first 48 weeks, mean changes in total lymphocyte count and CD4+ T-cell count were comparable between the DOR/ISL and bART arms. At week 96, participants in the DOR/ISL arm had a mean increase of 3.6 percent in total lymphocyte count and 4.5 percent in CD4+ T-cell count. None of the participants met the discontinuation criteria for decrease in total lymphocyte count or CD4+ T-cell count.

Between weeks 48 and 96, drug-related AEs were documented in six participants in the DOR/ISL arm and 20 in the bART arm. The most common AEs were diarrhoea (3.3 percent) and weight gain (2.2 percent) in the DOR/ISL arm, and weight gain (1.7 percent) and abnormal dreams (1.7 percent) in the bART arm. One participant in the DOR/ISL arm had a drug-related serious AE (previously reported multiple-drug overdose). There was one death in the DOR/ISL arm, due to myocardial infarction that was not related to treatment.

New HBV infection occurred in one participant in the bART arm, for whom BIC/FTC/TAF was initiated following discontinuation of DOR/ISL. There were no cases of clinical HBV reactivation.

Changes in weight from week 48 to 96 were –0.01 kg in the DOR/ISL arm and 0.89 kg in the bART arm. Weight change was greater in participants who switched from weight-suppressive bART, such as those containing efavirenz (EFV) and/or tenofovir disoproxil fumarate (TDF). Meanwhile, participants who switched from a regimen that did not contain EFV or TDF had minimal weight change.

A non-INSTI-based alternative

Overall, the findings from these three phase III trials “support DOR/ISL as a promising non-INSTI-based two-drug regimen for the treatment of HIV-1 in a broad population,” Rockstroh said.

Doravirine is a next-generation non-NRTI, while islatravir is an investigational nucleoside reverse transcriptase translocation inhibitor. These drugs, according to Rockstroh, are well-suited for a two-drug oral ART, given their complementary mechanisms of action and activity against common NRTI and NNRTI resistance-associated variants.

“In light of concerns of emerging resistance to INSTIs, which may affect their efficacy in the future, or for people who may not tolerate INSTIs, a non-INSTI-based regimen, such as DOR/ISL, with comparable efficacy to an INSTI-based regimen may provide an alternative option for people with HIV,” he said.

DOR/ISL trials

Conducted at 116 sites across 20 countries, MK-8591A-053 included 536 adults (median age 32 years, 25 percent female at birth, 42 percent White, median baseline CD4+ T-cell count 370 cells/mm³) with HIV-1 RNA of ≥500 copies/mL who were initiating treatment for the first time. These participants were randomly assigned to receive either doravirine 100 mg/islatravir 0.25 mg (n=269) or bictegravir 50 mg/emtricitabine 200 mg/tenofovir alafenamide 25 mg (n=267) orally once daily.

MK-8591A-052 involved 513 adults (median age 47 years, 21.4 percent female at birth, 60.8 percent White, median baseline CD4+ T-cell count 691 cells/mm³) living with HIV-1 who had undetectable viral load (<50 copies/mL) for ≥3 months on BIC/FTC/TAF. They were randomly assigned to either switch to DOR/ISL (n=342) or continue BIC/FTC/TAF (n=171).

In MK-8591A-051, a total of 551 adults (median age 51 years, 39.7 percent female at birth, 39 percent White, median baseline CD4+ T-cell count 706 cells/mm³) with HIV-1 RNA <50 copies/mL for ≥3 months on stable, oral 2- or 3-drug ART regimen participated. They were randomly assigned to switch to DOR/ISL (n=366) or to continue their baseline ART regimen for 48 weeks then switch to DOR/ISL thereafter (n=185).

The baseline ART regimen was INSTI-based in 64.2 percent of participants, non-NRTI-based in 30.3 percent, and protease inhibitor-based in 5.4 percent.

All trials excluded individuals with active hepatitis B virus infection. Additionally, none of the participants in the two switch trials had previous virologic failure and documented resistance doravirine. For MK-8591A-053, Rockstroh pointed out that study participants who were not immune to HBV were encouraged to go for HBV vaccination.

However, in a question-and-answer session, a member of the audience argued that for any HIV studies involving dual therapy without tenofovir, HBV vaccination should be a mandated protocol requirement rather than an optional recommendation.

The three trials are ongoing, with data being evaluated through week 144.