On 7 September 2026, the South African government reported that 55,123 people had started lenacapavir — the HIV prevention injection given once every six months — across 360 public health facilities in 24 districts [source: SAnews, 2026]. Six weeks earlier, health officials from nine African countries gave a different measure at the International AIDS Conference in Rio de Janeiro: the 191,620 doses administered by mid-July were roughly 14% of what those countries had planned to deliver during 2026 [source: aidsmap, 2026].
Both figures are accurate, and together they describe this drug better than either does alone. Lenacapavir has cleared every scientific and regulatory bar a new prevention tool normally faces. What is unfinished is everything that comes after approval: manufacturing, pricing, tendering, logistics, clinic staffing, and the arithmetic of guaranteeing a second injection to whoever received a first. Confused coverage of this drug mixes those two layers together. This article keeps them apart.
What the trials measured — and what came after
The efficacy figures worth using come from the US Centers for Disease Control and Prevention, which reviewed both pivotal Phase 3 trials to issue its clinical recommendation. In PURPOSE 1, conducted among cisgender women in sub-Saharan Africa, efficacy was 100%; in PURPOSE 2, which enrolled cisgender men and transgender and non-binary participants, it was 96% [source: CDC MMWR, 2025]. Both numbers are measured against background HIV incidence — the rate expected among comparable people using no pre-exposure prophylaxis (PrEP) at all — and not against oral PrEP.
That denominator matters more than it sounds. Gilead Sciences, the manufacturer, announced in June 2025 that there had been zero infections among the 2,134 participants in the PURPOSE 1 lenacapavir arm and two among the 2,179 in PURPOSE 2, a company announcement rather than an independent audit [source: Gilead Sciences, 2025]. The CDC's later review records what a press release could not: after the primary analyses were locked, two further infections appeared in PURPOSE 1 and one more in PURPOSE 2 [source: CDC MMWR, 2025]. The drug did not become less effective; the counting window moved. A third figure, 89%, circulates in commentary and compares the PURPOSE 2 lenacapavir arm with that trial's oral control arm instead of with background incidence [source: Daily Maverick, 2026] — a different denominator, not a different finding.
The regimen is also more specific than "one shot, twice a year." Starting it involves an injection plus oral doses on the first two days, after which injections fall every 26 weeks, and the CDC graded the recommendation as strong with high certainty of evidence [source: CDC MMWR, 2025]. The oral lead-in is easy to drop from a summary and impossible to drop from a clinic workflow.
The regulatory work is finished. That was the faster part.
The US Food and Drug Administration approved lenacapavir for HIV prevention on 18 June 2025, marketed as Yeztugo [source: Gilead Sciences, 2025]. The World Health Organization followed on 14 July 2025 with a strong recommendation graded at moderate to high certainty of evidence, and its wording is deliberate: an additional prevention choice within combination prevention, not a replacement for oral PrEP. The same guidelines recommended HIV rapid diagnostic tests for starting and continuing injectable PrEP, so that testing requirements would not themselves become the barrier [source: WHO, 2025].
Europe followed within weeks. The European Medicines Agency's human medicines committee issued a positive opinion on 24 July 2025, and the European Commission granted marketing authorisation on 26 August 2025 under the name Yeytuo, covering the EU-27 plus Norway, Iceland and Liechtenstein [source: European Medicines Agency / European Commission, 2025].
Then came the step most coverage skips, and the one that actually governs supply. On 6 October 2025, WHO added lenacapavir to its prequalification list through an abridged pathway that took 36 days [source: WHO Prequalification, 2025]. Prequalification is the precondition for procurement by UN agencies and the Global Fund, which makes "approved somewhere" and "purchasable at scale by a donor-funded programme" two different states. South Africa's regulator SAHPRA registered the drug in October 2025, the first national approval in Africa according to a specialist regulatory tracker [source: PrEPWatch, 2026].
One drug, three product names — and one approval that is not for prevention
The same molecule carries different names in different markets: Yeztugo in the United States, Yeytuo in the European Union. In South Korea, the Ministry of Food and Drug Safety approved lenacapavir on 7 April 2026 as Sunlenca injection and tablets — but for treating multidrug-resistant HIV-1 infection, with orphan drug designation, and not for prevention. Korean pharmaceutical trade publications reported this consistently, although the regulator's own notice could not be retrieved for this article [source: Korean pharmaceutical trade press, 2026]. A Korean headline about "the HIV shot" therefore refers to treatment, not to the preventive use discussed here.
WHO wrote the gap into its own recommendation
The most useful sentence about this rollout was written by the agency recommending it, on the day it recommended it: while access to lenacapavir outside clinical trials remained limited, WHO urged governments, donors and partners to begin rolling it out immediately. The guidelines themselves were blunter, recording that availability had so far been restricted to trial settings and that follow-up time was limited [source: WHO, 2025]. That described July 2025, and access outside trials has since begun — but as a diagnosis of where the bottleneck sits, it has aged well.
The interval between recommendation and first real-world dose is the simplest measure of implementation. The first injections outside a clinical trial in Africa were given on 1 December 2025 in Eswatini, where 23 people were dosed across five clinics that day, with Zambia starting the same afternoon; both countries began on initial allocations of roughly 500 doses each, supported by the Global Fund and the Children's Investment Fund Foundation [source: Positively Aware, 2025]. Five months after a global recommendation, 500 doses per country.
The price ladder: US$28,218, US$40 and US$75
On the day of FDA approval, UNAIDS put the US price at US$28,218 per person per year and called it an obstacle to expansion in low- and middle-income countries [source: UNAIDS, 2025]; the figure is UNAIDS's, and the company's own release published no list price. A generic pathway exists on unusually specific terms: under an agreement among Unitaid, the Clinton Health Access Initiative, Wits RHI and India's Dr. Reddy's, announced on 24 September 2025, generic injectable lenacapavir is to be supplied at US$40 per person per year across 120 low- and middle-income countries [source: Clinton Health Access Initiative, 2025], which UNAIDS welcomed the same day [source: UNAIDS, 2025]. The date to hold on to is 2027: US$40 is a contracted price for a product that has not yet shipped.
The licensing groundwork came earlier. Gilead announced in October 2024 that it had granted royalty-free voluntary licences to six generic manufacturers covering 120 high-incidence, resource-limited countries, with technology transfer beginning that December [source: Gilead Sciences, 2024]. That is a company announcement, and a licence is not a shipment. Announcing a PEPFAR partnership in September 2025, chief executive Daniel O'Day said the company was providing the medicine at no profit within that partnership [source: Gilead Sciences, 2025] — a corporate statement, not an independently audited one.
The 120-country list has edges, and Brazil sits outside them. Brazil contributed participants to the PURPOSE trials but was not included among the licensed countries; Brazilian estimates put domestic production at around US$75 per patient per year, and the civil-society organisation ABIA has criticised the exclusion [source: UN News, 2026]. UNAIDS Executive Director Winnie Byanyima put the principle sharply: "Innovation without access is not innovation; it is injustice" [source: UN News, 2026]. All three prices are annual per-person figures, and the roughly 705-fold span between the top and bottom rungs is the access story compressed into one ratio.
What has actually happened in the field
The flagship number is a target, and it has already moved once. On 14 April 2026, the Global Fund, the US government and the Children's Investment Fund Foundation announced support for an additional one million people over three years, raising the goal to three million cumulatively by 2028 [source: The Global Fund, 2026]; the original access agreement of 9 July 2025 had set that figure at two million [source: The Global Fund, 2025]. The same announcement sorted countries into three stages — active rollout, initial volumes delivered, and approved for future introduction — which is the clearest available picture of how uneven this is [source: The Global Fund, 2026].
The delivery figures are early but no longer trivial. UNAIDS recorded more than 6,000 people reached with long-acting prevention across five sub-Saharan countries as of March 2026 [source: UNAIDS, 2026]. Four months later, officials told AIDS 2026 that nine countries had administered 191,620 doses and initiated roughly 66,000 people by mid-July 2026, across 900 sites and with more than 10,000 providers trained [source: aidsmap, 2026]. The two counts differ by four months and by the number of countries included, not by disagreement, and they measure different things: 191,620 is doses, 66,000 is people. Reading two dated snapshots as a contradiction is the commonest error in coverage of this drug.
South Africa shows how long the last mile takes even where the system works. A first batch of 37,920 doses arrived on 7 April 2026 [source: South African Government, 2026], the national programme launched in June 2026, and by 7 September 2026 some 55,123 people had been initiated across six provinces, 71% of them women and largely pregnant and breastfeeding women [source: SAnews, 2026]. That is five months from dock to five-figure uptake, in the country running the world's largest HIV programme.
There is early evidence that the injection reaches people oral PrEP did not. A CDC survey across two sites in Zambia and twenty in Eswatini between December 2025 and January 2026 found that 71% of lenacapavir recipients in Zambia and 54% in Eswatini were using PrEP for the first time [source: CDC EIS Conference abstract, 2026]. Malawi reported the same direction among its 303 recipients [source: aidsmap, 2026]. These are two-month samples from 22 sites and are not nationally representative — and they are also the only field measurements yet available on the question that matters most for a new option.
Eswatini reserved 10,000 doses — and that is the whole problem in one number
The most informative detail from AIDS 2026 is a piece of inventory management. Eswatini had initiated about 1,000 adolescent girls and young women across 32 sites, and had set aside 10,000 doses for patients returning for their second injection, which limited how many new people it could start [source: aidsmap, 2026]. A six-month injection creates an obligation the moment it is given: the next dose falls due in 26 weeks, and a programme that cannot guarantee it has to throttle new initiations to protect those already enrolled. That is not a limitation of the drug. It is arithmetic imposed by finite supply.
The same session recorded the other face of that arithmetic. Mozambique had initiated about 10,000 people across 55 facilities but reported declining uptake, which officials attributed to limited drug availability; South Africa had planned 345,000 shots for 2026 [source: aidsmap, 2026]. The presenters were health ministry and national AIDS programme officials, and the figures reach us through a specialist HIV publication rather than a published dataset.
The countercurrent: prevention shrank in the same year it expanded
This rollout is happening during the sharpest contraction in HIV funding in two decades. International HIV funding fell from US$8.8 billion in 2024 to US$7.3 billion in 2025, an 18% drop that UNAIDS described as the lowest level in nearly twenty years [source: UNAIDS, 2026].
Service data moved with the money. UNAIDS reported PrEP use down 38% across 62 countries and HIV testing down 22% in high-burden settings between 2024 and 2025, with prevention accounting for just 11% of all HIV spending in 2024 [source: UNAIDS, 2026]. The year a twice-yearly injection began reaching clinics is also the year existing prevention lost ground — including the oral PrEP programmes the injection is meant to complement rather than replace.
The epidemiology has not cooperated either. UNAIDS reported 1.2 million new HIV infections and 570,000 AIDS-related deaths in 2025, with new infections rising in three regions and 21 countries; of the 41 million people living with HIV, about 9 million are not on treatment [source: UNAIDS, 2026]. In sub-Saharan Africa, roughly 3,000 adolescent girls and young women acquire HIV every week. UNAIDS also modelled a scenario in which the lost funding never returns: an estimated 6 million additional infections and 4 million additional deaths by 2029 — a conditional projection, not a forecast [source: UNAIDS, 2026].
No source reviewed here establishes that aid-budget changes caused any specific delay in this particular rollout, and this article does not claim that. What the record supports is narrower: a new tool is expanding inside a system that has less money, fewer tests and fewer people on oral PrEP than it did a year earlier.
What to watch
Three numbers describe the distance still to cover. UNAIDS estimates that 20 million people will need antiretroviral-based prevention by 2030 [source: UNAIDS, 2026]; the Global Fund's goal is three million people reached by 2028 [source: The Global Fund, 2026]; the number actually initiated on lenacapavir across nine countries as of mid-July 2026 is roughly 66,000 [source: aidsmap, 2026]. Need, target and delivery are three different categories, not three points on one scale.
Four things are worth watching from here. Whether generic supply actually begins in 2027 at or near the contracted US$40, since every projection of scale rests on that. Whether the twelve countries approved for future introduction convert into active rollouts, and how quickly. Whether the licensing map widens to countries such as Brazil that contributed trial participants but sit outside the agreement. And whether people come back for the second dose — data on that do not exist yet, which is itself a finding, because Eswatini's reserved 10,000 doses show programmes already planning around a question nobody can answer.
The clinical question about lenacapavir has largely been settled by regulators on both sides of the Atlantic. The open questions are about prices, licences, tenders and follow-up visits — a less dramatic list, and the one that will decide whether a drug that works becomes a drug that matters.