Optimising scale-up of injectable lenacapavir for HIV pre-exposure prophylaxis in South Africa: A modelling study and economic evaluation.

backgroundSouth Africa accounts for 20% of global HIV infections. Six-monthly injectable lenacapavir (LEN) for HIV pre-exposure prophylaxis (PrEP) has superior efficacy to oral tenofovir disoproxil fumarate/emtricitabine (TDF/FTC), and similar efficacy to 2-monthly injectable cabotegravir (CAB). With LEN's regulatory approval, South Africa faces critical implementation decisions amid constrained domestic resources and reduced international funding. We evaluated the epidemiological impact, cost-effectiveness, and optimal populations for LEN roll-out in South Africa. METHODS AND

findingsUsing Thembisa v4.8, a deterministic compartmental HIV transmission model, we simulated the impact of LEN scale-up, expanded oral TDF/FTC, and CAB scale-up, compared to a baseline of current TDF/FTC provision over 20 years (2026-2045) in South Africa. We modelled PrEP among women, including adolescent girls and young women (AGYW), female sex workers (FSW), pregnant and breastfeeding women (PBFW), men who have sex with men (MSM), and heterosexual men. For TDF/FTC scale-up, we doubled baseline initiation rates. For LEN and CAB, conservative and optimistic scenarios assumed initiation rates similar to or double those under TDF/FTC scale-up, respectively. Duration of use varied by subpopulation under TDF/FTC (3-6 months), conservative (LEN: 6-12 months, CAB: 4-8 months), and optimistic (LEN: 12-24 months, CAB: 8-16 months) scenarios. We modelled strategies to maximise impact of ~500,000 LEN doses allocated for 2026-2027, and national roll-out by subpopulations. We compared LEN cost-effectiveness to other HIV interventions, including antiretroviral treatment (ART). Costs are presented from the South African government's perspective in undiscounted 2025 United States Dollars (USD). Providing LEN to 1.7-2.9 million South Africans annually averted 19%-31% of infections and saved 3%-5% of life years lost to HIV, reaching incidence <0.1% in 2039-2043, 10-14 years earlier than baseline. TDF/FTC and conservative LEN scale-up increased HIV programme costs by 3%; however, LEN was more cost-effective, costing $2,301-$3,567/life year saved (LYS) versus $8,143/LYS (TDF/FTC scale-up) and $11,114-$16,118/LYS (CAB). While promising for HIV prevention, scaling up condoms, ART, HIV testing, and medical male circumcision may be more cost-effective than LEN; notably ART at 95% coverage, saved 4.8-8-fold more life years compared to LEN, costing $577/LYS. Prioritising AGYW, MSM, and FSW for the initial allocation maximised infections averted, while national roll-out prioritising FSW and MSM were most cost-effective, and even cost-saving under scale-up to FSW. Study limitations include uncertainty in achieving modelled uptake, risk-differentiated uptake of long-acting products, and real-world implementation costs.

conclusionsDelivering LEN to persons with elevated HIV risk in South Africa is more cost-effective than existing oral PrEP and can speed up HIV incidence reduction. Prioritising uptake among groups at highest risk is essential to maximise impact and cost-effectiveness.