Low agreement and frequent invalid controls in two SARS-CoV-2 T-cell assays in people with compromised immune function

Abstract

T-cell response plays an important role in SARS-CoV-2 immunogenicity. For people living with HIV (PWH) and solid organ transplant (SOT) recipients there is limited evidence on the reliability of commercially available T-cell tests. We assessed 173 blood samples from 81 participants (62 samples from 35 PWH; 111 samples from 46 SOT recipients [lung and kidney]) with two commercial SARS-CoV-2 Interferon-γ (IFN-γ) release assays (IGRA; SARSCoV- 2 IGRA by Euroimmun, and IGRA SARS-CoV-2 by Roche). The reliability between the tests was judged as low (Cohen's kappa [Κ] = 0.20; overall percent agreement [OPA] = 66%). A high proportion of tests were invalid (22% Euroimmun; 8% Roche). When excluding these invalid tests, the agreement was higher (Κ = 0.43; OPA = 90%). The low reliability between the two T-cell tests indicates that results should be interpreted with caution in SOT recipients and PWH and that SARS-CoV-2 T-cell tests need to be optimized and further validated for use in vulnerable patient populations.