Injectable HIV treatment works for people with poor HIV control in Africa

Injectable treatment with long-acting cabotegravir and rilpivirine is as effective as dolutegravir-based oral treatment in adults with HIV in sub-Saharan Africa who have a history of treatment non-adherence, a three-country study reported at the 13th International AIDS Society Conference on HIV Science (IAS 2025) in Kigali this week.

Presenting the results, Dr Fiona Cresswell of University Hospital Sussex said that long-acting treatment has an important role to play in the management of people who struggle with daily pill-taking in Africa.

Long-acting injectable cabotegravir and rilpivirine (CAB/RPV) is licensed for use in virally suppressed people. Last year, CARES, a large study conducted in Africa, reported that CAB/RPV was just as effective at maintaining viral suppression as the standard form of oral antiretroviral treatment offered through public health services.

But questions remained over the suitability of injectable CAB/RPV for use in people who are not virally suppressed and have a history of non-adherence. UNAIDS says that up to one million people in eastern and southern Africa may have unsuppressed HIV despite taking antiretroviral treatment and might benefit from a long-acting injectable form of treatment.

The IMPALA study was designed to compare injectable treatment every two months with long-acting CAB/RPV to a three-drug, dolutegravir-based oral regimen. The study enrolled people with HIV who had suboptimal HIV control, defined as either a viral load measurement above 1000 copies/ml in the past two years, a history of being lost to follow-up for at least four weeks, or who were unlinked to care for at least three months after HIV diagnosis. However, participants needed to maintain a viral load below 200 copies/ml on a combination of tenofovir, lamivudine and dolutegravir during the three-month screening phase before being randomised to continue oral treatment or switch to injectable treatment.

The study excluded people on second-line treatment, pregnant or breastfeeding women and people with active hepatitis B (HBV surface antigen positive).

The study randomised 540 people in Kenya, Uganda and South Africa; 60% of participants were female, had been taking antiretroviral treatment for a median of 7.8 years, and 78% had taken a non-nucleoside reverse transcriptase inhibitor in the past.

The primary study outcome was the proportion of participants in each study arm with viral load below 50 copies/ml at week 48. Injectable CAB/RPV proved non-inferior to oral treatment (91% vs 89%) at week 48, including in people with a body mass index of 30 or above (94% vs 90%), where efficacy might have been compromised due to higher body mass.

There were five confirmed virologic failures (two viral loads above 200 copies) in the study, all in the CAB/RPV arm. Viral sequencing for drug resistance was possible in four of the five cases, all showing high-level cabotegravir resistance and three showing high-level rilpivirine resistance. All were able to resuppress viral load on an oral regimen of tenofovir, lamivudine and dolutegravir (n=4) or tenofovir, emtricitabine and lopinavir/ritonavir (n=1).

Four participants discontinued study drug due to drug-related adverse events, none of them grade 3 or above, or injection site reactions. Participants overwhelmingly preferred injectable treatment because it offered privacy and consequently reduced stigma.

Although these results are encouraging for low- and middle-income countries, the findings may not change treatment practice quickly.

The World Health Organization this year recommended long-acting CAB/RPV as an alternative switching option for adults who are stable on antiretroviral treatment and without active hepatitis B infection, meaning that long-acting treatment will soon be incorporated into national guidelines in low- and middle-income countries.

And this week, the Medicines Patent Pool announced that ViiV Healthcare had agreed to expand the voluntary licensing agreement for cabotegravir to enable its use in long-acting treatment. Rilpivirine is already off patent in most countries, or will be within two years.

But it’s likely to take at least two years, if not more, for generic versions of injectable CAB/RPV to become available in low- and middle-income countries. In the medium term, an injectable regimen of two-monthly cabotegravir and six-monthly lenacapavir is the most promising option for widespread implementation in low- and middle-income countries, Dr Cissy Kityo of Uganda’s Joint Clinical Research Centre told a conference plenary session. A clinical trial is needed to evaluate the regimen.

Hepatitis B in IMPALA

The IMPALA study also evaluated hepatitis B acquisition and reactivation among participants. Whereas regimens containing tenofovir and either emtricitabine or lamivudine provide protection against hepatitis B acquisition, neither cabotegravir nor rilpivirine do so.

Ten per cent of potential study participants were excluded because they had active hepatitis B (hepatitis B surface antigen positive) or lacked immunity after previous infection (hepatitis B surface antibody titre < 10 IU/L). Although vaccination against hepatitis B is recommended for people with HIV in the three countries where the study took place, only 7% of participants had evidence of vaccine-induced immunity.

Liver enzymes (ALT, ALP and bilirubin) were tested at baseline and every six months during follow-up; ALT elevations were observed in 2.6% in the tenofovir/lamivudine/dolutegravir arm and 0.7% in the CAB/RPV arm. While there were no cases of hepatitis B reactivation, one participant in the CAB/RPV arm acquired hepatitis B, diagnosed after a significant liver enzyme elevation was detected at month 12. As a result, the study Data and Safety Monitoring Board recommended that all participants in the CAB/RPV arm should receive hepatitis B vaccination according to national guidelines.

Testing for active hepatitis B before switching to CAB/RPV may prove challenging in some settings where diagnostic capacity is limited.

“The risk of clinically significant hepatitis B reactivation is low, so the benefits of switching to CAB/RPV are likely to outweigh the risks of reactivation,” Dr Ubaldo Bahemuka concluded.

Long-acting CAB/RPV in people with unsuppressed HIV in the United States

Dr Ricky Hsu reported on the outcomes of people with HIV who initiated injectable CAB/RPV outside clinical trials at clinics in the United States participating in the OPERA cohort study between 2021 and the end of 2023. During this period, 3304 people began treatment with CAB/RPV, including 368 (11%) who had detectable viral load. The median level of virus in those with unsuppressed HIV who switched to injectable CAB/RPV was 120 copies/ml.

Those with unsuppressed HIV had a median age of 41 years, 30% were female, 57% were Black and 18% were Hispanic. The majority (63%) were receiving care in the southern United States. Before switching to CAB/RPV, 69% were taking an integrase inhibitor. The median time on the previous regimen was 17 months.

Completed treatment initiation was defined as receiving the first two injections no more than 67 days apart; 90% (331) received their first injections within this window, and of 293 people who received further injections, 59% received all injections on time, 33% had at least one delayed injection and 13% missed at least one injection.  

Of those with unsuppressed HIV who started injectable treatment, 258 (78%) were still receiving it at the end of the analysis period in February 2024. Among those who completed initiation and had a viral load test within six months of initiation, 85% had a viral load below 50 copies/ml.

Thirty-six people (12%) who had at least one viral load test after initiation never recorded a viral load below 50 copies/ml after switching. Three people experienced virologic failure, defined as two consecutive viral load measurements above 200 copies/ml or one viral load above 200 copies/ml followed by discontinuation within two months. In two of the three cases of virologic failure, major resistance to integrase inhibitors and non-nucleoside reverse transcriptase inhibitors was detected, in one case after missing two doses, but in the other cases, adherence to the injection schedule was good.