AI-generated research brief — verify at source
BNT162b2 Boost Elicits Strongest Humoral Response in People Living with HIV
Among people living with HIV who received a single booster dose of one of three COVID-19 vaccines, BNT162b2 produced the largest increases in both binding and neutralising antibodies at day 15, with neutralising antibody titres rising approximately tenfold from baseline (geometric mean fold change 10.0, 95% CI 8.6–11.7). All three vaccine strategies were safe and immunogenic regardless of CD4 count or viral suppression status, supporting their potential use in immunocompromised populations in resource-limited settings.
What Was Studied
This trial investigated whether a single booster dose of one of three COVID-19 vaccines — the adenoviral vector Ad26.COV2.S, the mRNA vaccine BNT162b2, or the recombinant spike-protein nanoparticle SARS-CoV-2-rS-PN — could safely induce meaningful immune responses in people living with HIV, a population for whom WHO recommends additional vaccine doses due to immunosuppression but for whom randomised trial data comparing homologous and heterologous approaches have been sparse.
How It Was Studied
This was a phase 2a, observer-blind, randomised controlled trial conducted at five clinical research sites in South Africa between July 2022 and February 2023. A total of 694 adults were enrolled, comprising 599 people living with HIV (on antiretroviral therapy with a viral load of 1,000 copies per mL or less and a CD4 count of at least 100 cells/µL) and 95 HIV-negative adults. Participants had either previously received one dose of Ad26.COV2.S, two doses of BNT162b2, or were unvaccinated people living with HIV with prior SARS-CoV-2 infection; they were randomly assigned 1:1:1 to receive a single intramuscular dose of Ad26.COV2.S, BNT162b2, or SARS-CoV-2-rS-PN. The primary immunogenicity endpoint was humoral response at day 15, assessed via geometric mean titres and geometric mean fold changes in spike receptor-binding domain binding antibodies and neutralising antibodies, with safety follow-up extending to 12 months post-vaccination.
What Was Observed
- BNT162b2 produced the highest antibody increases across both endpoints. For binding antibodies, the geometric mean fold change at day 15 was 7.1 (95% CI 6.3–8.1), and for neutralising antibodies it was 10.0 (95% CI 8.6–11.7) — roughly three to four times higher than the fold changes seen with Ad26.COV2.S (2.2 and 2.6, respectively) and approximately twice those seen with SARS-CoV-2-rS-PN (3.6 and 4.5, respectively).
- All three vaccines induced statistically significant increases in both binding and neutralising antibody titres from baseline to day 15 among people with HIV (p<0.0001 for each comparison), meaning the immune responses were unlikely to be due to chance and were consistent across vaccine groups despite differences in magnitude.
- Safety profiles were comparable between people living with HIV and HIV-negative participants. Local and systemic reactions were mostly mild or transient across all groups, and no deaths or serious safety signals were identified over the 12-month follow-up period.
- Immune responses were robust regardless of CD4 count or degree of viral suppression, indicating that even participants with more advanced immunosuppression — within the CD4 threshold of at least 100 cells/µL — mounted meaningful humoral responses to all three vaccine strategies.
Why This Matters
Randomised trial evidence directly comparing homologous and heterologous COVID-19 booster strategies in people living with HIV has been limited, leaving uncertainty about which approach best supports this immunologically vulnerable group. These findings establish that mRNA boosting with BNT162b2 yields substantially greater antibody fold rises than either the adenoviral vector or protein subunit alternatives in this population. For low- and middle-income countries where HIV burden is high and vaccine access is varied, this evidence base is particularly relevant to booster policy design.
How to Read This Result
While the immunogenicity findings are internally consistent and statistically robust, the trial measured antibody responses as a surrogate endpoint rather than clinical protection against COVID-19 disease, and the ancestral-strain-based vaccines studied may have limited direct applicability to current variant-adapted immunisation programmes.
Limitations
The abstract does not explicitly report study limitations beyond noting that no imputation methods were applied for missing data and that the analyses relied on available observations, with potential bias from missing data examined but not fully characterised.