BACKGROUND: WHO recommends a boost COVID-19 vaccine dose for people with immunosuppression, including people living with HIV. Data from randomised controlled trials assessing approaches to homologous and heterologous COVID-19 vaccine administration in people with HIV is scarce. We aimed to evaluate safety and immunogenicity of vaccine strategies in people with and without HIV.
METHODS: This phase 2a, observer-blind, randomised trial enrolled adults with a CD4 count of at least 100 cells/μL in five clinical research sites in South Africa. People with HIV (on antiretroviral therapy and a HIV-1 viral load of 1000 copies per mL or less) or without HIV who previously received one-dose of Ad26.COV2.S or two-doses of BNT162b2; and unvaccinated people with HIV with previous SARS-CoV-2 infection were randomly assigned (1:1:1) using a computer-generated randomisation list with block randomisation to receive a single-dose of intramuscular Ad26.COV2.S, BNT162b2, or SARS-CoV-2-rS-protein-nanoparticle (SARS-CoV-2-rS-PN). Participants, investigators, and site staff (except for site pharmacists) were masked to participant treatment assignment. The primary outcome was humoral immune responses at day 15 in people with HIV, evaluated by the geometric mean titres (GMT) and geometric mean fold change (GMFC) of SARS-CoV-2-specific spike protein receptor-binding domain antibody and of SARS-CoV-2-specific neutralising antibody, which was assessed in a modified intention-to-treat population that excluded participants with SARS-CoV-2 infection within 7 days post-vaccination. Safety was reported up to 12 months post-study vaccination. No imputation methods were applied, and analyses were based on all available observations. The amount and reasons for missing data were documented and examined to assess potential bias. The trial was registered with ClinicalTrials.gov (NCT05515042) and South African National Clinical Trial Registry (DOH-27-062022-4961) and is complete.
FINDINGS: Between July 25, 2022, and Feb 3, 2023, 694 participants (599 people with HIV and 95 people without HIV) were randomly assigned to Ad26.COV2.S (201 with HIV, 32 without HIV), BNT162b2 (203 with HIV and 31 without HIV), or SARS-CoV-2-rS-PN (195 with HIV and 32 without HIV). In 694 total participants, the mean age was 42 years (SD 11), 424 (61%) were female, and 269 (39%) were male. Safety analyses included 694 participants and modified intention-to-treat included 687 participants. Among 588 people with HIV, there was a significant increase in GMT of both binding (p<0·0001) and neutralising antibodies in all three vaccine groups (p<0·0001) from baseline to Day 15. The GMFC for binding antibodies at day 15 was 2·2 (95% CI 1·9-2·5) for Ad26.COV2.S, 7·1 (6·3-8·1) for BNT162b2, and 3·6 (3·2-4·1) for SARS-CoV-2-rS-PN. The GMFC for neutralising antibodies at day 15 was 2·6 (2·2-3·1) for Ad26.COV2.S, 10·0 (8·6-11·7) for BNT162b2, and 4·5 (3·8-5·3) for SARS-CoV-2-rS-PN. The BNT162b2 boost induced the highest peak humoral response. Local and systemic reactions in people with HIV were similar to those observed in people without HIV and were mostly mild or transient. No deaths or safety issues were reported.
INTERPRETATION: A single dose of ancestral Ad26.COV2.S, BNT162b2, or SARS-CoV-2-rS-PN vaccines was safe and induced robust and durable immune responses, regardless of CD4 count and viral suppression. This vaccine strategy should be considered in low-income and middle-income countries.
FUNDING: Coalition for Epidemic Preparedness Innovations.
