Immunogenicity and reactogenicity of SARS-CoV-2 vaccines in people living with HIV in the Netherlands: A nationwide prospective cohort study

Abstract

Background: Vaccines can be less immunogenic in people living with HIV (PLWH), but for SARS-CoV-2 vaccinations this is unknown. In this study we set out to investigate, for the vaccines currently approved in the Netherlands, the immunogenicity and reactogenicity of SARS-CoV-2 vaccinations in PLWH.

Methods and findings: We conducted a prospective cohort study to examine the immunogenicity of BNT162b2, mRNA-1273, ChAdOx1-S, and Ad26.COV2.S vaccines in adult PLWH without prior COVID-19, and compared to HIV-negative controls. The primary endpoint was the anti-spike SARS-CoV-2 IgG response after mRNA vaccination. Secondary endpoints included the serological response after vector vaccination, anti-SARS-CoV-2 T-cell response, and reactogenicity. Between 14 February and 7 September 2021, 1,154 PLWH (median age 53 [IQR 44-60] years, 85.5% male) and 440 controls (median age 43 [IQR 33-53] years, 28.6% male) were included in the final analysis. Of the PLWH, 884 received BNT162b2, 100 received mRNA-1273, 150 received ChAdOx1-S, and 20 received Ad26.COV2.S. In the group of PLWH, 99% were on antiretroviral therapy, 97.7% were virally suppressed, and the median CD4+ T-cell count was 710 cells/μL (IQR 520-913). Of the controls, 247 received mRNA-1273, 94 received BNT162b2, 26 received ChAdOx1-S, and 73 received Ad26.COV2.S. After mRNA vaccination, geometric mean antibody concentration was 1,418 BAU/mL in PLWH (95% CI 1322-1523), and after adjustment for age, sex, and vaccine type, HIV status remained associated with a decreased response (0.607, 95% CI 0.508-0.725, p < 0.001). All controls receiving an mRNA vaccine had an adequate response, defined as >300 BAU/mL, whilst in PLWH this response rate was 93.6%. In PLWH vaccinated with mRNA-based vaccines, higher antibody responses were predicted by CD4+ T-cell count 250-500 cells/μL (2.845, 95% CI 1.876-4.314, p < 0.001) or >500 cells/μL (2.936, 95% CI 1.961-4.394, p < 0.001), whilst a viral load > 50 copies/mL was associated with a reduced response (0.454, 95% CI 0.286-0.720, p = 0.001). Increased IFN-γ, CD4+ T-cell, and CD8+ T-cell responses were observed after stimulation with SARS-CoV-2 spike peptides in ELISpot and activation-induced marker assays, comparable to controls. Reactogenicity was generally mild, without vaccine-related serious adverse events. Due to the control of vaccine provision by the Dutch National Institute for Public Health and the Environment, there were some differences between vaccine groups in the age, sex, and CD4+ T-cell counts of recipients.

Conclusions: After vaccination with BNT162b2 or mRNA-1273, anti-spike SARS-CoV-2 antibody levels were reduced in PLWH compared to HIV-negative controls. To reach and maintain the same serological responses as HIV-negative controls, additional vaccinations are probably required.

Trial registration: The trial was registered in the Netherlands Trial Register (NL9214). https://www.trialregister.nl/trial/9214.

Fig 1. Flow chart of included PLWH.

*Participants who tested positive for SARS-CoV-2 antibodies in serum at baseline measurement were excluded from further articipation ^†Samples were not stored adequately or were not sent to be analysed in the central laboratory at the Erasmus University Medical Centre. ^††One participant received a combination of ChAdOx1-S and BNT162b2 vaccines and was therefore not included in the per protocol analysis. PLWH, people living with HIV.

Fig 2. Antibody concentration in PLWH and controls after vaccination.

Anti-spike SARS-CoV-2 binding antibodies after BNT162b2 (HIV-negative, n = 94; PLWH, n = 884), mRNA-1273 (HIV-negative, n = 247; PLWH, n = 100), ChAdOx1-S (HIV-negative, n = 26; PLWH, n = 150), or Ad26.COV2.S (HIV-negative, n = 73; PLWH, n = 20) vaccination in PLWH and the HIV-negative control group. The thick horizontal bar shows the geometric mean concentration, also indicated in the numbers above the graphs, with error bars showing geometric standard deviation. The dotted lines show the lower limit of detection of the performed test (4.81 BAU/mL), the positivity cutoff (33.8 BAU/mL), and the hyporesponse cutoff (300 BAU/mL). BAU, binding antibody units; LLoD, lower limit of detection; PLWH, people living with HIV; S, spike.

Fig 3. Cellular immune responses against SARS-CoV-2 in subgroup participants (PLWH).

(A) Cellular immune response to wild-type spike by ELISpot assay (Pre, n = 23; Post, n = 45): IFN-γ SFCs after subtraction of MOG. Statistics performed using Mann–Whitney U test: p = 0.002. Negative responses: Pre, 8; Post, 5. (B) Cellular immune response to wild-type and Delta spike in AIM assay (n = 14): percentage of CD4+ CD137+ OX40+ T-cells after subtraction of DMSO. Dotted line shows the LLoD at 0.001%. Statistics between pre- and post-vaccination for WT in PLWH performed using Wilcoxon matched-pairs signed rank test (p = 0.005); pre- and post-vaccination for Delta: ns. Statistics between PLWH and controls with Mann–Whitney U test: not significant. (C) Cellular immune response to wild-type and Delta spike in AIM assay (n = 14): percentage of CD8+ CD137+ CD69+ T-cells after subtraction of DMSO. Dotted line shows the LLoD at 0.001%. Statistics performed using Wilcoxon matched-pairs signed rank test (p = 0.008); pre- and post-vaccination for Delta: ns. Statistics between PLWH and controls with Mann–Whitney U test: not significant. Green circles: mRNA vaccines; blue squares: vector-based vaccines. Pre: before vaccination; Post: 4–6 weeks after second vaccination. AIM, activation-induced marker; BAU, binding antibody units; cART, combination antiretroviral therapy; DMSO, dimethyl sulfoxide; ELISpot, enzyme-linked immune absorbent spot; GMC, geometric mean concentration; INF-γ, interferon-γ; LLoD, lower limit of detection; MOG, myelin oligodendrocyte glycoprotein; PBMC, peripheral blood mononuclear cell; PLWH, people living with HIV; SFC, spot-forming cell; WT, wild type.

Fig 4. Severity of adverse events.

Severity of adverse events present in the 7 days after vaccination in people living with HIV, comparing mRNA (BNT162b2 and mRNA-1273) and vector (ChAdOx1-S and Ad26.COV2.S) vaccines, shown as percentage (%) of participants.