Two-monthly injectable HIV therapy kept 91% suppressed in people who struggled with daily pills

Design

Phase 3b, open-label, randomised non-inferiority trial (IMPALA)

Population

540 adults in Kenya, South Africa and Uganda with suppressed HIV-1 and adherence or engagement problems

Primary outcome

Viral load below 50 copies/mL at week 48

Effect

91.0% vs 89.2%, difference 1.9 points (95% CI −3.1 to 6.9); virological failure 1.9% vs 0%

IMPALA randomised 540 adults at seven African sites. All had viral loads below 200 copies/mL but a record of suboptimal adherence or clinic engagement. Half switched to intramuscular cabotegravir plus rilpivirine every eight weeks; half stayed on daily dolutegravir-based therapy.

At 48 weeks, 91% on injections and 89% on oral therapy had viral loads below 50 copies/mL, which met the trial's non-inferiority margin. Five people on injections had confirmed virological failure, against none on oral therapy. Serious adverse events were similar.

The finding matters because long-acting regimens were first tested in highly adherent participants. Here they held up in a group defined by difficulty with daily treatment. Rilpivirine resistance is the risk to watch, and baseline resistance testing was not done, which mirrors many real settings.

• Participants were already suppressed below 200 copies/mL at entry; this is a switch strategy, not first-line treatment.
• Injections were every eight weeks after two loading doses four weeks apart.
• Confirmed virological failure occurred in 1.9% on injections and none on oral therapy.
• Missed injection visits matter as much as missed tablets, so tracing systems are part of the treatment.
• Availability of long-acting cabotegravir plus rilpivirine in India is limited; check local supply before raising it.

Why it matters

It challenges the assumption that long-acting therapy is only for the most adherent patients; this trial enrolled the people who miss doses and clinic visits.

Don't overread it

Five virological failures occurred only on injections, and baseline resistance was not tested, so the trade-off is real.

The statistics, in plain English

Non-inferior means injections were not meaningfully worse, here within 10 percentage points. The failure difference of 1.9 points has an interval from 0.3 to 3.5, so it is unlikely to be chance, though the numbers are small.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.