Novel antibody-based regime keeps HIV at bay for six months
Higher dose chosen for ongoing study
Gus Cairns, reporting from HIV Glasgow, 19 November 2024.
A long-lasting combination therapy regimen consisting of one antiretroviral drug and two broadly neutralising antibodies (bnAbs) kept HIV suppressed in 26 out of 32 people who changed from daily oral therapy. Three people withdrew from the study before it ended, while another three developed a detectable viral load (over 20 copies) on the lower dose of one of the bnABs. However, all 15 people who stayed on the higher dose maintained viral suppression for six months without taking further anti-HIV medication.
The open-label study, run by the pharmaceutical company Gilead Sciences, recruited 32 people who had had undetectable viral loads on standard oral antiretroviral therapy (ART) for at least 18 months. In place of their existing ART it gave them one infusion each of Gilead’s two proprietary bnAbs teropavimab and zinlirvimab, plus their capsid inhibitor drug lenacapavir (Sunlenca). Lenacapavir was given as a single subcutaneous injection plus a ‘loading dose’ of one lenacapavir pill on the same day, plus a second pill taken the following day. Viral load was measured every four weeks. At week 26 the participants resumed taking a standard ART regimen (unless they had already developed a detectable viral load). After re-starting ART, viral load was measured at weeks 38 and 52.
Teropavimab (TAB) and zinlirvimab (ZAB) are long-acting versions of the bnAbs 3BNC117 and 10-1074, which block the activity of the HIV envelope protein gp120 in two different places, and stop the virus from entering cells. Lenacapavir (LEN) meanwhile prevents the assembly of a vital component of mature viral particles at a later stage of HIV’s life cycle. Results from a phase 1 study of a LEN/TAB/ZAB regimen were reported at the Conference on Retroviruses and Opportunisitic Infections (CROI) last year, but the study at Glasgow was designed to find the optimal dose of ZAB. This will now be taken forward into a phase 2 study which will randomise people to standard ART or to six-monthly LEN/TAB/ZAB.
While bnAbs can be highly effective and non-toxic when used to either treat or prevent HIV, their Achilles heel is that their activity, despite their name, is often not that broad: many are not active against all strains of HIV. Because of this, candidates for the LEN/TAB/ZAB trial were screened in advance to see how sensitive their strain of HIV was to either bnAb. Studies had already shown that roughly 90% of strains of HIV were sensitive to either TAB or ZAB, though only 50% of strains were sensitive to both.
Testing for viral susceptibility to antibodies, which are large and complex proteins, is technically difficult and yields imprecise results compared to resistance testing for small-molecule antivirals and antibiotics, so the main purpose of the screening was to avoid recruiting people who would be very unlikely to benefit from the regimen. Roughly 90 people were screened for bnAb sensitivity before recruitment. Of these 12 were not susceptible to either of the bnAbs and were excluded. In the end, 32 people were enrolled: 21 people susceptible to both bnAbs, five susceptible only to TAB and six susceptible only to ZAB.
Presenting the study at HIV Drug Therapy Glasgow this week, Dr Paul Cook of East Carolina University emphasised that the ideally effective concentrations of either antibody had not been firmly established. Because of this, while a 10mcg/ml dose of TAB was used for all participants, two different doses of ZAB were used, 10mcg/ml and 30 mcg/ml. These doses were not allocated on the basis of participants’ bnAb sensitivity – instead, 16 participants each were randomly allocated to receive the higher or lower dose. The 18 participants were aged 46 on average and six of them were women. Eighteen participants were White, seven Black and three Asian; ten were Latino. They all had medium-to-high CD4 counts (ranging from 449 to 1916). Two participants (one in each dosing group) withdrew from the study before the 26-week ending date, and one was excluded because they were found to have active hepatitis B and should not have been enrolled.
Another three, all of them in the low-dose ZAB group, experienced virologic rebound. The first was found to have a viral load of about 750 when tested at week 16 – despite being sensitive to both TAB and ZAB – and restarted oral ART early, which re-suppressed their HIV. They were the only study participant who developed a drug resistance mutation to lenacapavir subsequent to viral rebound. The second, who was not sensitive to TAB, was found to have a viral load of about 70 at week 26 when they were due to restart their ART anyway; they also re-suppressed their HIV.
The third participant was interesting. On the day they took their LEN/TAB/ZAB they had a viral load of about 35 – not high enough to exclude them from the trial, but higher than the lower limit of the viral load test used, which could detect down to 20. They caught COVID at week 13 and their viral load went up to over 100 at week 20. After that, they never completely re-suppressed their HIV, even after restarting ART, and their viral load was about 120 at week 52, six months after restarting ART. They have not changed their ART regimen and have not acquired drug resistance. Dr Cook speculated their viremia might be connected to immune activation consequent to COVID, but there needs to be more investigation of their responses both to COVID and to the bnAbs. The COVID was counted as an adverse event that was not related to the study medication. Those that were were almost entirely lenacapavir injection site reactions. One participant in the low-dose ZAB group developed fever after infusion of their bnAbs, but this quickly resolved. Six participants in each of the ZAB dose groups developed anti-drug-antibodies, which have limited the effectiveness of bnAbs in some other studies, but in this case were at low levels and did not affect efficacy, safety or drug levels.
Speaking of drug levels, LEN levels fell after peaking at week 12, but on average were four times the average inhibitory level at week 26, though below that level in a few cases. TAB and ZAB levels fell slowly from week one but were still 30-50 times higher than a reference level of 2mcg/ml set as the minimum effective dose. However, this limit was set on the basis of pharmacokinetic modelling alone and the true therapeutic level may be higher. For this reason, the higher dose of ZAB has been taken forward into the phase 2 trial, as there were no virological rebounds in that arm.
Reference
Little SJ (presenter Cook PP) et al. Efficacy and safety analysis of lenacapavir with broadly neutralising antibodies, teropavimab and zinlirvimab, in people with HIV-1 highly sensitive to one or both broadly neutralising antibodies. HIV Glasgow 2024, abstract O23.
Congress delegates can view the presentation here.
All abstracts from the congress are available here (open access).
Reporting from HIV Glasgow 2024 by Keith Alcorn, Gus Cairns and Roger Pebody has been financially supported by the Congress. The writers are editorially independent of the Congress and the presenting speakers.
Photo credit: Alan Donaldson Photography