Retrovirus-based pseudotyped virus neutralisation assays overestimate neutralising activity in sera from participants receiving integrase inhibitors

Retroviral pseudotype-based virus neutralisation assays are widely used to estimate functional immunity, but may be unsuitable for testing human immunodeficiency virus (HIV)-infected individuals receiving integrase inhibitor treatment. We evaluated these assays for measuring severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and hepatitis C virus (HCV) neutralisation in people living with HIV. SARS-CoV-2 neutralisation was assessed using HIV-based SARS-CoV-2 pseudoviruses in sera from a longitudinal Malawian cohort (n = 1,876), detecting neutralisation across timepoints in 10.5–54.5% of HIV-uninfected vs. 85.5–93.9% of HIV-infected participants (n = 96). HIV-infected sera were re-tested using vesicular stomatitis virus (VSV)-based SARS-CoV-2 pseudoviruses, estimating seroprevalence at 5.6–65.2%, suggesting HIV-based assays overestimate neutralisation. HIV-based VSV-glycoprotein(G) pseudoviruses confirmed non-specific inhibition in 75.0-87.9% of HIV-infected participants. HCV neutralisation was assessed in UK-based HCV patients (n = 100, n = 90 HIV-infected) using murine leukaemia virus (MLV)-based HCV pseudoviruses. Non-HCV-specific inhibition was detected in integrase inhibitor recipients. Median neutralisation of MLV(HCV) pseudoviruses was higher in integrase inhibitor recipients (71.8% vs. 21.3%). Testing HIV(SARS-CoV-2) and MLV(HCV) pseudoviruses against antiretroviral drugs showed interference by integrase inhibitors (specifically, dolutegravir). Isolating IgG from serum removed the interference by residual drugs. Retrovirus-based assays are therefore unsuitable for testing individuals receiving integrase inhibitors. Protective immunity to viruses has likely been over-reported in HIV prevalent populations.