Failure to seroconvert after two doses of BNT162b2 SARS-CoV-2 vaccine in a patient with uncontrolled HIV.

SARS-CoV-2 infection elicits similar antibody responses in HIV-negative individuals and people living with HIV that is well controlled by antiretroviral therapy (ART). SARS-CoV-2 vaccines, including the mRNA vaccine BNT162b2, are efficacious in clinical trials, inducing antibodies and T cells specific to the spike protein.2, 3, 4 However, vaccine responses in people living with HIV have been assessed only in those stable on ART. Immune function in people with HIV and uncontrolled HIV replication is impaired due to destruction of the CD4 T cells that help B cells during an antibody response. Immune responses to vaccines (eg, against influenza and hepatitis B) in people living with HIV are inferior to those in the general population. We present a case of one individual with uncontrolled HIV replication who did not respond to two doses of the BNT162b2 SARS-CoV-2 vaccine.

The index patient, with advanced HIV, was recruited to a cohort study, along with 13 people living with HIV suppressed on ART (median CD4 count 590 cells per μL [SD 202; range 310–940]) and 43 HIV-negative controls, all of whom had received one or two doses of BNT162b2. The index patient had no history of SARS-CoV-2 infection or AIDS-defining conditions and had already received two doses of BNT162b2 24 days apart. Blood samples were obtained 16 days and 44 days after the second dose (appendix p 2). At day 16, the patient's HIV viral load was 831 764 copies per mL and CD4 count was 20 cells per μL (CD4% 4·6%; CD4/CD8 0·05). ART with bictegravir, emtricitabine, and tenofovir alafenamide and prophylaxis for opportunistic infections were initiated at day 16. At day 44 (29 days after ART initiation), HIV viral load was undetectable and CD4 count was 70 cells per μL.

Post-vaccine samples from the index patient showed no IgG reactivity against the S1 subunit of the spike protein by an in-house ELISA (appendix pp 1–2). By contrast, an HIV-negative, SARS-CoV-2-naive participant had an S1-specific IgG titre of 43·4 μg/mL 44 days after the second dose of BNT162b2, consistent with the binding titres across the wider cohort, in which all participants produced S1-specific IgG, even after only one dose (appendix pp 1–2). No SARS-CoV-2-specific neutralisation was observed at either timepoint for the index patient. By contrast, the HIV-negative, SARS-CoV-2-naive vaccine recipient had a neutralisation titre of 1/656 after the second dose. No quantifiable spike protein-specific T cells, evaluated via ELISpot (MAIPN4550; Merck Millipore, Darmstadt, Germany), were observed in the index patient compared with the HIV-negative control participant, sampled 44 days after the second dose, and a subsample of people living with well controlled, stable HIV after one dose of BNT162b2 (appendix p 2). Although no spike protein-specific T cells were detected via intracellular cytokine staining in the index patient, responses against cytomegalovirus pp65 were dominated by CD8 T cells, strikingly so after ART (appendix p 2). The profound peripheral immune cell perturbations in the index patient were accompanied by increased frequencies of CD8 T cells with a terminally differentiated effector memory phenotype (CCR-7− and CD45RA+) and an inverted CD4/CD8 ratio, which could influence the size of T-cell responses to SARS-CoV-2.

In conclusion, an individual with profound HIV-related immune dysfunction did not seroconvert to a SARS-CoV-2 vaccine. We suggest monitoring SARS-CoV-2 seroconversion in people with advanced HIV and considering repeat vaccination upon HIV suppression and CD4 count improvement with ART. This case highlights an urgent need to establish correlates of vaccine efficacy in people living with HIV, particularly in those with suboptimal viral suppression or ongoing perturbed immune function, to better inform clinical management and guidelines.

LEM has received honorarium for lectures from the London School of Hygiene and Tropical Medicine. SR-J has received honorarium for lectures from the London School of Hygiene and Tropical Medicine, declares additional grant support from UK Research and Innovation, EDCTP, Kumamoto University, the Rosetrees Trust, and the Global Challenges Research Fund, and is a trustee and past president of the Royal Society of Tropical Medicine and Hygiene. RKG has received consulting fees from UMOVIS Lab and Janssen. FB has received payment for teaching research methods from Gilead Sciences. LW has received honorarium for participation in advisory boards, speaker fees, or both from Gilead Sciences, ViiV, Janssen, MSD, Theratechnologies, Mylan, and Cipla, and has also received conference attendance support from ViiV. All other authors declare no competing interests. ET and AA contributed equally to this Correspondence. This Correspondence was supported by a Medical Research Council (MRC) grant to LEM (MR/R008698/1), a MRC studentship (MR/N013867/1), and a MRC grant (MR/M008614) and a National Institutes of Health grant (R01AI55182) to DP. We thank Peter Cherepanov of the Francis Crick Institute (London, UK) for supplying recombinant S1 antigen. We are grateful to Rebecca Matthews, Paulina Prymas, Marzia Fiorino, Thomas Fernandez, Nnenna Ngwu, the clinical research teams at the Mortimer Market Centre (London, UK) and the Ian Charleson Day Centre (London, UK), and all the clinic staff and participants.