A number of innovative approaches to treating children were presented at last week’s Conference on Retroviruses and Opportunistic Infections (CROI 2026) in Denver, US. Several were based on the fact that infants have good prospects for long-term viral suppression without antiretroviral therapy (ART). Failing that, children and adolescents, who may have particular problems with adherence to daily pills, are prime candidates for long-acting injectable formulations.
Globally, 1.4 million children under 14 have HIV, with 130,000 newly infected every year. Of them, only 57% are on ART and 46% virally suppressed.
Viral control off ART in early-treated children
Dr Gabriela Cromhout of the University of KwaZulu Natal in South Africa presented a continuation of a series of studies of a cohort of children given ART immediately after birth. In 2023 she presented the cases, now published in Nature Medicine, of five young boys out of a cohort of 281 children born with HIV who maintained persistently low viral loads off treatment (viral remission) despite their ART being stopped accidentally. At Denver she presented a study where ART was stopped intentionally, as an analytical treatment interruption (ATI), in 19 children. Of these, three have continued to stay in viral remission off ART, and another three had delayed viral rebounds.
The Ucwaningo Lwabantwana (‘Learning from Children’) cohort now consists of more than 330 children given two-drug ART from birth and a third antiretroviral from an average of ten days after it. From this cohort, nine boys and ten girls were selected for the Azaphile (‘Good Life’) study. They were aged four to six years and had been virally suppressed (below 30 copies) since they were two on average. An important selection criterion was that the reservoir of integrated HIV DNA in their lymphocyte cells, which was on average 142 copies per million cells when they started ART, now had to be zero or nearly zero (average, two per million cells).
Their ART was stopped and they were closely monitored, weekly in the first month, then every two weeks and then monthly, with any viral load over 30 copies defined as a ‘rebound’. Thirteen of the 19 experienced a rebound within six weeks, all but one within a month.
Of the other six, three experienced delayed rebounds, maintaining viral loads below 30 for five, 18 and 24 months before having to restart ART. Three more have not restarted ART yet. One has been off ART for ten months (the only girl with a delayed rebound so far), but two have been off ART without rebound for 30 and 52 months. Cromhout pointed out that in studies of adults taken off ART without any other intervention, 96% rebound within six weeks.
In the whole cohort, including the previous five boys, there are now eight children still in viral remission off ART. All of these have viruses with low replicative capacity.
“We hypothesise that greater numbers of very-early treated children would achieve cure/remission following broadly neutralising antibody (bnAb) interventions,” Cromhout added, hinting that such studies were underway.
Three bnAbs in experimental children’s trial in Botswana
Immediately before her presentation there was one from Botswana, which presented preliminary results from a study giving three bnAbs to children and taking them off antiretrovirals.
Dr Gbolahan Ajibola of the Botswana–Harvard AIDS Institute Partnership told the conference about the follow-on of the Tatelo study (which means ‘The next thing’ in Tetswana). In the original study, presented at CROI 2022, 10 out of 25 children given two bnAbs (VRC01-LS and 10-1074) maintained a viral load below 40 for 24 weeks.
In Tatelo Plus the aim is to see if bnAb treatment can lead to post-treatment control. Twelve children (eight girls, four boys, nine of whom had been in the original Tatelo study) were given three bnAbs, initially in addition to ART. They had an average age of nine, had started ART within a few days of birth and had had a viral load below 40 on dolutegravir-based ART for more than 24 weeks.
The three bnAbs were VRC07-523-LS, PGDM1400-LS and PGT121.141-LS. In a previous study of the same bnAb combination given to adults not on ART, the strongest predictor of virological control was whether the participants had pre-existing resistance to one or more of the bnAbs. So in Tatelo Plus, children with resistance to more than one of the three bnAbs were excluded from the second part of the study, where participants were taken off ART and received the bnAbs only. This excluded two children.
Of the other ten, four had virus susceptible to all three bnAbs and continued on the three. Two were fully resistant to one of the bnAbs and received the other two.
In the other four children, resistance to the bnAbs could not be determined; assays testing susceptibility to bnAbs are difficult to perform and results often indeterminate. But the trial protocol said that children could be continued on the three bnAbs without ART as long as they had no detectable integrated HIV DNA in their cells.
Not only was this the case in all ten, but nine of the children were also HIV-antibody negative on a standard ELISA HIV test. Seroreversion – losing your antibodies to HIV – is extremely rare in adults, but more common in children who are treated very early. These four children were therefore maintained on the triple bnAb combo without ART, with the aim of seeing if HIV DNA, possibly including antibody loss, is as good a guide to antibody efficacy as resistance tests.
The result, so far, is that all ten children have maintained undetectable viral loads below 40 on the bnAbs alone – as the children in the original Tatelo study did. But the next phase will be to see if they continue to control their HIV when they stop taking the bnAbs. If so, it will be the first study to show that bnAbs can promote post-treatment control in children.
Long-acting injectables in teenagers
In the meantime, however, tolerable and effective ART regimens for children with HIV remain limited and adherence often becomes harder as they enter adolescence. So they are a group who may particularly benefit from long-acting injectable ART regimens.
Week 24 results from the IMPAACT 2017 (MOCHA) study of long-acting, injectable cabotegravir and rilpivirine in adolescents were reported at CROI 2024. At this year’s conference the final, week 96, results were presented by Dr Aditya Gaur of St Jude Children’s Research Hospital in Memphis, Tennessee.
The study recruited 144 adolescents aged 12 to 17 across 18 sites on three continents, all switching to cabotegravir/rilpivirine. After 96 weeks, none had experienced a protocol-defined virological failure (two consecutive viral loads over 200). Indeed, 94% had not had a single viral load over 50 after 96 weeks – the remaining 11 participants (7.6%) had had a single viral ‘blip’ over 50, of whom just two (1.4%) had a blip over 200.
Retention was excellent too, with just over 95% of participants staying in the study and receiving the 13 injections of each drug (two injections, every eight weeks).
Forty-four participants (31%) had a grade three adverse event, the most common a rise in creatine phosphokinase levels, but these were temporary and thought not to be related to the study drugs.
Forty-two per cent did report injection site reactions, such as redness, swelling and nodules. Almost all were grade 1 (mild) or 2 (moderate), and they became less common over time. After week 48, less than 10% of rilpivirine injections and less than 5% of cabotegravir injections were associated with injection site reactions.
Two participants experienced grade 3 (severe) reactions, both abscesses, and one had a grade 4 (life-threatening) reaction – anaphylaxis, a severe allergic reaction. Although this person fully recovered, they did not resume the injections.
The researchers also described ‘post-injection reactions’ which they had not anticipated when planning the study. These only occurred with 1% of injections, and initially looked like anaphylaxis, with shortness of breath, agitation, bronchospasm, dizziness and numbness, but only lasted for a few minutes. They are thought to be a short-lived autonomic reaction, similar to ones seen with other types of injection.
Despite the injection site reactions, and 50-60% of participants reporting that the rilpivirine injections hurt ‘more than a bit’ (compared with about 20% for the cabotegravir injections), more than 97% of participants said they’d prefer the injections to daily pills.
