Novel Long-Acting Anti-HIV Compounds

At the 2026 Conference on Retroviruses and Opportunistic Infections (CROI), there was a strong emphasis on long-acting (LA) antiretroviral therapy (ART) formulations. Moreover, several compounds showed promise for twice-yearly dosing, suggesting that a complete twice-yearly ART regimen is viable in the near future. Gupta and colleagues (Abstract 174) presented data on GS-3242, an investigational LA integrase strand transfer inhibitor (InSTI). They investigated the pharmacokinetic (PK) parameters associated with 3 different doses of intramuscular administration of GS-3242 in people without HIV. The results supported a possible dosing schedule of every 4 months, potentially as infrequently as every 6 months. They investigated the antiviral activity of GS-3242 monotherapy in people with HIV (PWH) not on ART using an oral formulation dosed on day 1 and day 2. They found a mean plasma HIV-1 RNA level reduction of 2.31 log₁₀ copies/mL through day 11, which is comparable to other InSTIs. Future studies will investigate this compound in novel combinations of LA regimens.

Back and colleagues (Abstract 176) presented data on LA formulations of VH-184, an investigational InSTI with antiviral activity against HIV-1 isolates resistant to second-generation InSTIs such as dolutegravir (DTG) and bictegravir (BIC). They evaluated several formulations with subcutaneous and intramuscular administration. Only mild injection-site reactions were noted. They concluded based on available PK data that twice-yearly dosing is feasible, which is being pursued in ongoing studies.

Thakkar and colleagues (Abstract 175) presented PK data on LA formulations of VH499, an investigational HIV-1 capsid inhibitor, in people without HIV. They investigated various formulations including subcutaneous and intramuscular administration. Injection-site reactions were common and occurred more frequently with subcutaneous formulations; however, these adverse events were generally mild or moderate. Based on the population PK models generated from these data, future studies will investigate twice-daily dosing of VH-499.

Rolle and colleagues (Abstract 178) presented data on monthly intramuscular long-acting cabotegravir (LA CAB) given with every 6-month N6LS, an investigational broadly neutralizing antibody (bNAb) targeting HIV-1, in people with stable virologic suppression with documented susceptibility to N6LS. A total of 125 participants were randomly assigned 2:2:1 to intravenous N6LS plus LA CAB, N6LS subcutaneously plus LA CAB, or continued oral ART, respectively. The 12-month results presented were consistent with prior 6-month results, ie, the intravenous formulation was better tolerated than the subcutaneous formulation, and virologic suppression was generally sustained from month 6, with 94% achieving plasma HIV-1 RNA level below 50 copies/mL in the intravenous group. No additional participants in the intravenous group experienced confirmed virologic failure (VF) from month 6 to month 12.

Single-Tablet Oral Regimens

Single-tablet regimens have generally been studied in those initiating ART or in those who are stably suppressed on ART without a history of VF. Orim and colleagues (Abstract 181) presented data on the use of a BIC/LEN single-tablet regimen in those virally suppressed on a complex regimen and with a history of VF. Participants with stable virologic suppression on a complex regimen (median, 3 tablets/d) were randomly assigned 2:1 to switch to BIC/LEN or remain on a complex regimen. Both arms did well through 48 weeks, with high rates of virologic suppression, very low rates of confirmed VF, and no emergence of resistance. Participants switching to BIC/LEN reported greater treatment satisfaction than those remaining on a complex regimen.

Messner and colleagues (Abstract 513) presented data on switching to BIC/LEN in people who were stably suppressed on BIC/emtricitabine (FTC)/tenofovir alafenamide (TAF) without history of VF. They conducted a randomized, double-blinded clinical trial and found that switching to BIC/LEN resulted in virologic outcomes that were noninferior to staying on BIC/FTC/TAF.

Rockstroh and colleagues (Abstract 177) compared doravirine (DOR)/islatravir (ISL) with BIC/FTC/TAF in those initiating ART for the first time. This randomized, double-blind trial enrolled 536 participants. The rates of viral suppression were high in each group, and DOR/ISL achieved noninferiority vs BIC/FTC/TAF (91.8% vs 90.6%, respectively; difference, 1.2%; 95% CI, −3.7 to 6.2). The rates of confirmed VF were similar in both arms; 2 participants in the DOR/ISL arm and zero participants in the BIC/FTC/TAF arm had treatment-emergent resistance mutations. Suppression rates in the DOR/ISL arm were similar among those with a plasma HIV-1 RNA level above 100,000 copies/mL and those above 500,000 copies/mL. Weight gain did not differ between arms. These data suggest that DOR/ISL is an additional viable 2-drug regimen for those initiating ART.

Two presentations reported on 96-week outcomes in clinical trials that compared switching those with virologic suppression on BIC/FTC/TAF (Abstract 514) or other oral ART (Abstract 515) to DOR/ISL. The studies found that switching to DOR/ISL resulted in noninferior virologic outcomes vs continuing the baseline ART regimen. Post and colleagues (Abstract 519) performed a subgroup analysis of these trials to describe the renal outcomes of those with mild or moderate renal impairment at the study baseline. They found similar virologic outcomes in those with and without renal impairment and found that DOR/ISL did not adversely affect renal outcomes.

Real-World Data on LA CAB/RPV

There were several presentations on the use of longacting cabotegravir/rilpivirine (LA CAB/RPV) in real-world settings. Rolle and colleagues (Abstract 525) investigated an early switch to LA CAB/RPV among people initiating ART with oral DTG/lamivudine (3TC). Participants were offered the option of switching to LA CAB/RPV as soon as their plasma HIV-1 RNA level was below 50 copies/mL. They found low rates of viral nonsuppression through month 11 (approximately 5%); 1 of 129 (<1%) experienced emergence of resistance to LA CAB/RPV. This suggests that switching from oral induction ART to LA CAB/RPV as soon as plasma HIV-1 RNA suppression is achieved is a reasonable strategy.

Abstracts 527 to 529 compared virologic outcomes among those initiating LA CAB/RPV without viral suppression and those with viral suppression. These studies were consistent in finding that low-level detection of HIV-1 RNA was more common among people initiating LA CAB/RPV with viremia; however, these rates were low in both groups. Similarly, confirmed VF was very low in both groups. Data on median CD4+ counts of the study populations were limited and only mentioned in Abstract 527, where the median CD4+ count was above 500 cells/μL. Baseline drug resistance mutations (DRMs) among the study participants were not mentioned. These data support the use of LA CAB/RPV in those without virologic suppression, but caution remains in its use in those with preexisting DRMs or low CD4+ cell counts, as these groups were not studied.

These results were supported by Chen and colleagues (Abstract 530), who presented data from a small clinical trial that randomly assigned 45 individuals with ongoing viremia despite oral ART with no DRMs associated with CAB or RPV. Participants were randomly assigned to continue oral ART or change to LA CAB/RPV. Median CD4+ count was 319 cells/μL. They found more superior virologic suppression among those randomly assigned to LA CAB/RPV than in those who continued oral ART (19/25 [76%] vs 8/20 [40%], respectively; P = .03).

Dietrich and colleagues (Abstract 534) examined the relationship between body mass index (BMI) at the time of LA CAB/RPV initiation and subsequent virologic outcomes in the OPERA (Observational Pharmaco-Epidemiology Research and Analysis) cohort. The rates of viral suppression among BMI strata (<30 kg/m², 30 to <40 kg/m², 40 kg/m² and higher) were similar at approximately 95%. The rates of confirmed VF through a median of 16 months of follow up were low in all groups. These data support the use of LA CAB/RPV in PWH with obesity.

Spinelli and colleagues (Abstract 484) evaluated the relationship between low-level viremia and plasma trough concentrations of CAB and RPV. They found that low-level viremia occurred more commonly among those receiving every 2-month dosing of CAB/RPV than in those receiving monthly dosing, and that low-level viremia was related to lower CAB concentrations, suggesting that dose adjustment based on drug concentrations may be needed in a subset of individuals.

Resistance to InSTIs

A themed discussion focusing on noncanonical mutations leading to InSTI resistance highlighted Abstracts 550, 551, 553, and 585, which showed that envelope (env) mutations were associated with InSTI resistance. Kengni and colleagues (Abstract 550) examined mutations in PWH in Cameroon who had VF on DTG-based ART. They observed 17 new noncanonical mutations, the majority in env and a few in group-specific antigen (gag) that conferred InSTI resistance. They found mutations frequently occurring outside integrase, with 100% of these mutations occurring in env.

Hikichi and colleagues (Abstract 551) presented bench data on HIV-1 mutations acquired in the presence of raltegravir or DTG. HIV-1, in the presence of raltegravir, caused the emergence of mostly integrase resistance mutations known to confer resistance to InSTIs. By contrast, HIV-1 exposed to DTG first showed the emergence of env mutations that permitted ongoing viral replication in the presence of DTG. Subsequently, the virus acquired mutations in the nucleocapsid (NC) and integrase. These experiments demonstrated that HIV-1 can use noncanonical resistance pathways in env to continue replicating, which thereby allows it to acquire mutations in integrase and NC.

Coetzee and colleagues (Abstract 553) presented real-world data on these noncanonical mutations in PWH with VF on DTG in South Africa, where they observed mutations in gag and env leading to InSTI resistance. Charpentier presented data (Abstract 585) from a cohort of 145 PWH who had VF while taking an InSTI regimen. In the cohort for whom they had NC sequence data, 10.8% had mutations in the NC that confer resistance to InSTIs. However, these individuals did not have mutations in integrase.

Swaine and colleagues (Abstract 566) presented data from the United Kingdom about the efficacy of BIC/FTC/ TAF in a cohort of 238 individuals who had prior nucleoside reverse transcriptase inhibitor (nRTI) resistance-associated mutations (RAMs) but no InSTI RAMs. None of the individuals with K65R/N had VF on BIC/FTC/TAF. VF or low-level viremia was rare in their cohort, occurring in 3.4% of their participants, with nonadherence contributing to half of those cases. No major InSTI emergent resistance was observed in those with VF. These data are similar to previously reported results such as the NADIA (Nucleosides and Darunavir/Dolutegravir In Africa) trial, where the use of DTG paired with an nRTI backbone containing tenofovir was effective in achieving virologic suppression despite high background nRTI resistance.¹

Santoro and colleagues (Abstract 568) presented resistance data from an Italian cohort on LA CAB/RPV. They defined VF as 2 consecutive viral loads above 50 copies/mL or 1 viral load above 50 copies/mL with a change in ART. In all, 2.1% of participants on LA CAB/RPV met their endpoint of VF, with a median time of 6.6 months to VF. They found that 74.4% of those with VF had nonnucleoside reverse transcriptase inhibitor (NNRTI) or InSTI resistance, and 43.9% had substantiial resistance to second-generation InSTIs. Reassuringly, 92.9% of their participants with VF were able to suppress on an alternative regimen.

Jeffrey presented data (Abstract 554) from experiments evaluating the resistance profile of VH-184 (previously mentioned above), a new third-generation InSTI that is being developed as an LA injectable. When testing the activity of VH-184 against pseudotyped viruses from participants who had VF in prior trials, they found that VH-184 was able to retain more potent activity in the presence of InSTI mutations than BIC. This holds promise for VH-184 as a treatment for individuals in whom second-generation InSTIs have failed.

Resistance to Lenacapavir

Bekerman and colleagues (Abstract 586) presented resistance data from a phase II, randomized, open-label trial examining the efficacy of once-weekly oral ISL and oral LEN (ISL/LEN) as switch therapy. From baseline genotyping, 5 participants who had reverse transcriptase (RT) RAMs (including M184V) maintained virologic suppression on ISL/LEN at week 96. In 2 participants who had viremia on ISL/LEN and had posttreatment genotyping performed, no treatment-emergent resistance was seen.

Gistand (Abstract 582) presented real-world data from a cohort of 50 PWH with baseline resistance in San Francisco, California, who started on LA LEN. A total of 92% of these individuals were on injectable therapy only, 70% of the individuals had baseline NNRTI resistance, and 10% had baseline InSTI resistance. A total of 44% of individuals were virally suppressed before starting LEN, but this increased to 100% by week 15 after LEN was added. At week 140, 95.7% of the cohort had sustained viral suppression. These data support using LEN to achieve viral suppression even in individuals with background resistance and demonstrate that it can be done in combination with a fully injectable regimen.

Two abstracts from Begovic and colleagues (Abstract 579) and Margot and colleagues (Abstract 581) showed that capsid polymorphisms at T107 do not confer resistance to LEN except for the T107N mutation.

VanderVeen and colleagues (Abstract 567) presented resistance data from a phase II, open-label study for a combination of 2 bNAbs, teropavimab and zinlirvimab, administered with LEN twice yearly for the treatment of HIV-1. Emergent resistance of LEN was rare, occurring in only 1 of 80 participants.

Empiric Treatment Strategies for Pneumonia in Infants With HIV

Globally, infants with HIV face a high mortality rate, with pneumonia as the leading cause of death.² In Abstract 151, the EMPIRICAL (Empirical Treatment Against Cytomegalovirus and Tuberculosis in HIV-infected Infants With Severe Pneumonia) trial evaluated whether empiric cytomegalovirus treatment with valganciclovir improved survival in 558 infants with HIV aged 28 to 365 days who were hospitalized for severe pneumonia. EMPIRICAL was a phase II-III, randomized, open-label, 2×2 factorial study conducted in 19 hospitals across 6 African countries.³ Participants were randomly assigned to receive standard of care alone (empiric antibiotics plus Pneumocystis jirovecii treatment), or in combination with valganciclovir, tuberculosis treatment, or both, with ART initiated on day 15 (±7 days) if the child was not already taking prescribed ART. The groups were recategorized and analyzed according to the receipt of valganciclovir. The primary outcome was all-cause mortality at day 15 and through 1 year of follow up. The median age was 4.4 months, 49% were female, and 25% had severe malnutrition. A total of 71% of the children were newly diagnosed with HIV; median CD4+ cell percentage was 15.3% and median HIV-1 RNA level was 6.3log10 copies/mL. Analysis revealed that 51% of infants with available results had cytomegalovirus viremia attributable to pneumonia. All-cause mortality at day 15 was significantly lower in the valganciclovir group than in the no-valganciclovir group (23.2% vs 27.0%, respectively; hazard ratio [HR], 0.60; 95% CI, 0.41–0.87). Although the survival benefit was most pronounced in the first month of follow up, the overall 1-year mortality rate was 43.1% with valganciclovir and 47.5% without (adjusted HR [aHR], 0.79; 95% CI, 0.62–1.02). No evidence of interaction between empiric valganciclovir and empiric tuberculosis treatment was noted. Valganciclovir was not associated with more adverse events, neutropenia, or anemia during the 1-year follow-up period. The authors concluded that empiric valganciclovir improves early survival in infants with HIV with severe pneumonia, suggesting that its inclusion in care guidelines may be warranted.

Advancing ART-Free Remission in Children

One strategy under investigation for advancing ART-free remission in children is the use of bNAbs to limit the HIV reservoir size and sustain viral suppression, with the longterm goal of achieving ART-free remission. Earlier results from the Tatelo study suggested that this approach could potentially work in children.⁴ The IMPAACT (International Maternal Pediatric Adolescent AIDS Clinical Trials Network) 2042 Tatelo Plus study (Abstract 152) extends this work by testing 3 broader and more potent LA bNAbs (VRC07-523LS, PGDM1400LS, and PGT121.414.LS) for treatment. Tatelo Plus is a single-arm, open-label, phase I/II trial in Botswana that enrolled children who started ART near birth. In step 1, the children received combination bNAbs and ART for 32 weeks. In step 2, children who remained virally suppressed and met prespecified eligibility criteria for ART interruption then stopped ART for a 24-week bNAb-only period. Study sizes were small: 12 children completed Step 1, and all maintained viral suppression (<40 HIV RNA copies/mL) during the 32-week period. Ten children entered step 2 to receive bNAb-only treatment. During the ART interruption phase of step 2, seven of the 10 children (70%) received all 3 bNAbs, and 3 (30%) received 2 bNAbs upon the detection of resistance to the third antibody. Among the 10 children who entered step 2, 80% were female, median age at ART initiation was 3 days, and median age at step 2 entry was 9 years. All 10 children had negative HIV DNA polymerase chain reaction testing, and 9 children (90%) had negative HIV enzyme immunoassay testing at step 2 entry. All 10 children maintained viral suppression, with HIV-1 RNA level below 40 copies/mL throughout the entire 24-week bNAb-only period. The bNAb regimen was well tolerated, with no grade 3 or 4 adverse events reported. The study findings support the use of combination bNAbs, guided by phenotypic susceptibility testing and biomarker criteria, as a strategy for ART interruption. These results represent an initial step toward future bNAb trials aimed at achieving ART-free remission in children.

In the nonrandomized, open-label, IMPAACT P1115 trial, infants with in utero HIV-1 infection were initiated on ART (2 nRTIs + nevirapine + raltegravir) within 48 hours of birth. A subset of these children received the monoclonal antibody VRC01 within 72 hours. Abstract 839 reported on 22 infants with confirmed in utero HIV-1 infection: 11 received ART alone and 11 received a combination of ART and VRC01. By week 96, 81% of these infants achieved virologic control, with a cumulative incidence of first confirmed viral suppression increasing from 52% at week 24 to 81% at week 96. Furthermore, 27% of these infants had sustained virologic control through week 96. In addition to these findings, Abstract 840 demonstrated that VRC01 was safe and well tolerated among 222 neonates in the IMPAACT P1115 trial, with 98% of doses administered within 48 hours of birth. Local injection-site reactions were rare (1.1%) and mild, and no systemic injection reactions were reported. Median VRC01 levels at 1-week postbirth dose were similar for infants with HIV and without HIV. However, for repeated doses, VRC01 clearance was faster among infants with HIV than predicted from data for HIV-exposed but uninfected children, particularly among those with concurrent viremia after 10 weeks. These data highlight the importance of accounting for HIV status and viremia when determining the optimal dosing of bNAbs for infants.

Very early initiation of ART in infants may limit the formation of the latent HIV reservoir and increase the likelihood of ART-free remission. Cromhout and colleagues (Abstract 153) characterized the variation in time to viral rebound after stopping ART in children in a prospective analytic treatment interruption (ATI) study in KwaZulu-Natal, South Africa. Nineteen infants were enrolled in this study from an initial birth-treatment cohort of 330 mother-child pairs with HIV, where all infants were started on ART at birth and initiated on 3-drug ART within 21 days of birth. Children eligible to enroll were older than 36 months of age and virally suppressed (plasma viral load <30 copies/mL) for more than 24 months and met stringent immune and reservoir criteria for ATI. The children stopped ART and underwent close safety monitoring, and ART was restarted if plasma viral load increased to more than 30 copies/mL. A total of 42% of the study children were male and median age was 59 months. Thirteen of 19 children (68%) experienced viral rebound within 6 weeks and 6/19 (32%) remained virally suppressed beyond 12 weeks. Among these 6 children, 3 later had viral rebound at 4.8, 18, and 24 months, and 3 remained off ART with sustained viral suppression for 52, 30, and 10 months. The authors noted that males undergoing ATI were more likely to remain virally suppressed longer than females in the cohort, suggesting that innate immune sex differences may contribute to this disparity in ART-free remission dynamics. This was suggested by prior research from the investigators, who reported on 5 children, all male, who had maintained ARTfree aviremia following unscheduled ATI.⁵ The authors noted that the proportion of very-early treated children (32%) who maintained viral suppression beyond 12 weeks post ATI in this study was higher than has been previously reported in adults. The findings suggest greater potential for ART-free remission in children who initiate ART early, underscoring the need for additional studies combining ATI with bNAbs and other cure strategies.

Advances in Pediatric ART

Two-drug ART is increasingly under investigation as an alternative to 3-drug regimens for HIV treatment, but evidence in children remains limited. D3/Penta 21 (DTG/3TC Fixed-Dose Formulations For the Maintenance of Virological Suppression in Children With HIV Infection Aged 2 to <15 Years Old) (Abstract 154) was a 96-week, open-label, non-inferiority trial that evaluated the efficacy and safety of a 2-drug regimen (DTG/3TC) compared with a 3-drug control regimen (DTG-based + 2 nRTIs) in children. The study enrolled children with HIV who were virally suppressed for at least 6 months, had no prior treatment failure, were aged 2 years to less than 15 years, and weighed at least 10 kg. A total of 386 children were enrolled globally and randomly assigned to DTG/3TC or DTG-based 3-drug ART. The median age was 8.3 years and 54% were female. The median CD4+ cell percentage was 38%. In the intention-to-treat analysis, confirmed viral rebound (viral load, ≥50 copies/mL) by week 96 occurred in 11 of 193 (5.9%) in the DTG/3TC arm and 13 of 193 (6.6%) in the 3-drug ART arm, meeting criteria for noninferiority. Most children with confirmed viral rebound of 50 copies/mL or higher were resuppressed by week 96. Safety was similar between groups, with similar serious adverse events and grade 3 or higher events; there were no ART-modifying adverse events. The proportion of children with emergent InSTI (DTG) or nRTI DRMs at confirmed viral load of 50 copies/mL or higher was similar between the 2 groups. Across 96 weeks, health care resource utilization was comparable between the 2 arms; however, drug costs were $145 less per child per year in the DTG/3TC arm than in the DTG-based 3-drug ART arm. These findings support DTG/3TC as a simpler maintenance option for children, with efficacy and safety comparable to DTG-based 3-drug ART.

IMPAACT 2017 (MOCHA; More Options for Children and Adolescents) was a phase I/II, noncomparative study that evaluated the safety, tolerability, and PKs of LA CAB/RPV among adolescents. Results through week 96 (and end-of-study follow up) were reported in Abstract 155. The study enrolled 144 virally suppressed adolescents with HIV who were aged 12 years to younger than 18 years and weighed 35 kg or more, across 18 sites in 5 countries. The median age was 15 years; 51% were female, and 92% had perinatally acquired HIV. Participants switched from oral ART and received a 4-week oral CAB/RPV lead in, followed by injectable CAB/RPV. A total of 95% of participants completed follow up through week 96, and 137 of 144 (95%) received at least 13 injections. Drug-related adverse events through week 96 occurred in 60 of 144 (42%) of participants. Injection-site reactions were the most common (37%) and were typically mild and short in duration (≤7 days), with pain reported more often after LA RPV than after LA CAB injections (31% vs 16%, respectively). There were 2 grade 3 drug-related adverse events (abscess and pain/abscess) and 1 grade 4 anaphylaxis-like postinjection reaction that resolved but led to discontinuation of the study drug. At week 96, a total of 94.4% of participants maintained viral suppression (HIV-1 RNA level, <50 copies/mL). No confirmed VF events occurred. At week 96, RPV and CAB trough concentrations were similar to those observed in adults and remained above efficacy targets, supporting an acceptable PK profile. Among those assessed at week 96, all participants preferred injections over daily pills.

Abstract 854 presented safety and PK data on CAB/RPV from the IMPAACT 2036 CRAYON (Cabotegravir and Rilpivirine Long-Acting Injections in Young Children) study. The trial evaluated 61 children with HIV aged 2 years to 12 years and weighing between 10 and below 40 kg. Following a 4-week oral lead-in period, participants transitioned to monthly intramuscular injections of LA CAB/RPV using weight-band dosing. Children achieved plasma concentrations and drug exposure levels comparable to those observed in adults and adolescents. The most common adverse event was grade 1 or 2 injection-site pain. No children discontinued treatment due to safety concerns.

Children who are on DTG-containing ART regimens and tuberculosis prophylaxis with 12 weeks of once-weekly rifapentine and isoniazid (3HP) present a unique challenge, as children clear DTG more rapidly and rifapentine may further decrease drug exposure. Salazar-Austin and colleagues (Abstract 150) presented results from the DOLPHIN-KIDS (Rifapentine and Isoniazid TB Preventive Therapy [3HP] for Children Taking Dolutegravir-Based Antiretroviral Treatment) trial, a 2-stage, phase I/II study evaluating DTG PK, safety, and virologic suppression in children with HIV receiving DTG with 3HP. In stage 1, the children received twice-daily DTG with 3HP for 12 weeks. Subsequent PK modeling predicted that once-daily DTG would produce therapeutic concentrations in children weighing at least 10 kg, which was tested in stage 2. In this second stage, children with HIV weighing at least 10 kg with established viral suppression (<50 copies/mL) received ART consisting of once-daily DTG plus 2 nRTIs, which was initiated at least 4 weeks before starting 3HP.

The participants completed 3HP for 12 weeks concurrent with their ART, followed by an 8-week follow-up period. The children underwent semiintensive and sparse DTG PK sampling in addition to HIV-1 RNA viral load monitoring. DTG PK sampling results were considered acceptable if the lower 95% CI of the observed trough concentration was more than 158 ng/mL and all trough values remained above the protein-adjusted (PA) IC90 (64 ng/mL). Of the 41 enrolled children (median age, 5.5 years; range, 1.5–16.9 years; 46% female, all Black African), 22% had less than 90% adherence by pill count during 3HP treatment. The lower 95% CI for DTG trough concentrations with once-daily DTG was 200 ng/mL overall and 75 ng/mL at 3 days post dose. However, 2 participants did not meet the PK criteria and had predose trough levels below 64 ng/mL at the 3-day postdose timepoint, although none fell below the limit of quantification. No treatment-related grade 3 or higher adverse events occurred. The regimen was well tolerated, with no participants discontinuing once-daily DTG-based ART while on 3HP. Most participants (95%) remained virally suppressed at study completion and 2 participants developed transient HIV-1 RNA blips that later became undetectable following adherence counseling. The authors noted that although the minimum effective DTG trough concentration is not yet firmly established, the PA IC90 of 64 ng/mL serves as the current target. The authors raised the question of whether infrequent, time-limited decreases in DTG concentration are clinically relevant, given that all children on daily DTG-based ART achieved viral suppression at study completion. They concluded that 3HP as tuberculosis prophylaxis may be coadministered with once-daily DTG and 2 nRTIs in children with HIV who weigh at least 10 kg without the need for DTG dose adjustment, provided it is combined with effective adherence counseling.

In Abstract 850, the authors showed that children weighing 3 to 20 kg with HIV and tuberculosis coinfection and being treated with ART and rifampicin-based TB treatment achieved effective drug levels with twice-daily DTG that were comparable to standard once-daily DTG without rifampicin, with the dosing strategy considered safe.

Abstract 851 presented findings from the UNIVERSAL1 (Pharmacokinetic Study of a Novel DTG/FTC/TAF Dose Ratio for Children) study, which evaluated a new, simplified DTG/FTC/TAF dose ratio designed to improve ease of use and treatment adherence for children in African settings. The study showed that once-daily administration of 5 mg/15 mg/1.88 mg dispersible tablets (for children weighing 3 kg to <20 kg) and 50 mg/200 mg/25 mg film-coated tablets (for children weighing 20 to <25 kg) achieved DTG and FTC exposure levels similar to those observed in adults. Although TAF and tenofovir disoproxil fumarate (TDF) exposures varied across weight bands, the levels remained within established ranges that were considered safe and effective.

Pregnancy, Maternal Health, and Infant Exposure

The DoraDO (Doravirine Dose Optimization in Pregnancy) study (Abstract 797) is an ongoing, randomized, open-label trial investigating the safety and PKs of DOR, a second-generation NNRTI, among 76 pregnant South African women initiating ART between 8 and 26 weeks of gestation. Participants were randomly assigned to receive DOR (100 mg once daily) in combination with TDF/3TC or the standard of care ART (TDF/3TC/DTG). The study found that although maternal DOR exposure was reduced during the second and third trimesters, trough concentrations remained consistently above the therapeutic target (>3 ng/mL). Safety data from 46 women showed no significant differences in serious adverse events between the DOR and standard-of-care arms, with no HIV transmissions or birth defects reported. Furthermore, despite a moderate transfer of DOR into breast milk (∼30%), infant drug levels remained low postpartum. These results suggest that DOR is a promising option for use during pregnancy; however, follow-up studies are needed to inform clinical guidelines for DOR use during pregnancy and breastfeeding.

In an ongoing observational cohort of 261 pregnant women with HIV in Kenya receiving DTG-based ART (Abstract 787), Marwa and colleagues evaluated the association between timing of isoniazid initiation by trimester for tuberculosis prevention and adverse pregnancy outcomes. The study found that those who initiated isoniazid in the first trimester experienced a significantly higher frequency of any adverse outcome than those starting in the second trimester (43.2% vs 27.6%, respectively; risk ratio, 1.56; 95% CI, 1.06–2.31; P = .024). Specifically, first-trimester initiation was associated with a higher incidence of stillbirth (7.7% vs 4.0%; RR, 2.50; 95% CI, 1.02–6.08; P = .043) and preterm birth (21.2% vs 7.7%; RR, 2.73; 95% CI, 1.36–5.00; P = .005). Although these results suggest that the timing of isoniazid initiation is an important safety consideration, the authors noted that larger studies are needed to confirm these findings.

Findings from the Pediatric HIV/AIDS Cohort Study (Abstract 830), which enrolled children in the US and Puerto Rico, showed that children who are HIV exposed but uninfected have a substantial risk of developing cardiometabolic complications. In an analysis of 1669 participants, the incidence of meeting a metabolic trigger (defined as BMI >95th percentile) was 98.5 cases per 1000 person-years (421 participants over 4273 person-years of follow up). Among the subset of children undergoing fasting metabolic testing, the weighted prevalence was 18.6% for dyslipidemia and 16.8% for insulin resistance. These data suggest that HIV-exposed but uninfected children who exhibit a high BMI have substantial cardiometabolic risk factors. In addition, the findings underscore the need for further investigation into how in utero HIV and ART exposure and early life factors influence cardiometabolic health in the long term.

Supporting Adolescent Transitions in HIV Care

Adolescents with perinatally acquired HIV transitioning to adult services are a vulnerable group, with loss of viral suppression and decreases in retention in care frequently observed during this transition period. InT-SHA-VIP (Interactive Transition Support for Adolescents With HIV Comparing Virtual and In-Person Delivery) (Abstract 156) was a hybrid implementation-effectiveness, stepped-wedge, cluster-randomized trial evaluating an adolescent-friendly transition support intervention conducted at 16 clinics in KwaZulu-Natal, South Africa. Eligibility criteria included adolescents aged 15 to 19 years with perinatally acquired HIV who had been receiving ART for at least 6 months, who were aware of their HIV status, and had scored low to intermediate on a transition readiness assessment. The intervention consisted of facilitated peer discussion groups, delivered virtually or via monthly in-person sessions. Of the 372 adolescents enrolled, 178 were assigned to the immediate-intervention arm and 146 to delayed standard of care. Within the immediate-intervention arm, 53% attended in-person sessions and 47% attended virtual sessions. At 9 months, the in-person delivery significantly increased viral suppression (HIV-1 RNA <200 copies/mL) compared with delayed standard of care (93% vs 84%, respectively; adjusted odds ratio [aOR], 2.9; P = .035). In contrast, virtual delivery did not significantly impact viral suppression (87% vs 84%, respectively; aOR, 0.9; P = .9). Retention in care (a composite outcome based on pharmacy refills and clinic attendance) improved relative to delayed care for in-person delivery (27% vs 13%; aOR, 2.7; P = .004) and virtual delivery (57% vs 13%; aOR, 7.7; P <.001). Overall, the InTSHA-VIP model improved transition outcomes by boosting viral suppression through in-person delivery and enhancing retention with in-person and virtual delivery modes, showing that structured, adolescent-centered support can improve transition to adult HIV care.