Longitudinal analysis of antibody responses to the Pfizer BNT162b2 vaccine in Patients Undergoing Maintenance Hemodialysis

ABSTRACT Background and objectives Patients undergoing hemodialysis are at higher risk of developing severe complications upon SARS-CoV-2 infection and were prioritized in the Portuguese vaccination campaign. Immunogenicity of COVID-19 vaccines in hemodialyzed patients was not addressed by the phase 3 clinical trials leading to their emergency approval. Design, setting, participants, and measurements We performed a prospective, longitudinal, cohort analysis of 156 hemodialyzed patients and 143 age-matched controls scheduled for BTN162b2 vaccine. Excluded from analysis were five patients previously diagnosed with SARS-CoV-2, three sero-positive for anti-SARS-CoV-2 N, two dropouts and two deaths. ELISA was used to quantify anti-full-length Spike IgG, IgM and IgA levels in sera collected at day of the first vaccine dose (t0); 3 weeks later (day of the second dose, t1); and 3 weeks after the second inoculation (t2). Results Seroconversion after the first vaccine dose (t1) was remarkably low in patients, with positivity for anti-spike IgG, IgM and IgA antibodies of 29.4%, 12% and 41%, respectively. The second vaccine dose raised seroconversion to 90.9% and 83.9% for IgG and IgA, respectively, while IgM positivity remained unchanged. At t1 the anti-Spike IgG level was significantly lower in patients with ages below 70 years when compared to age-matched controls, showing a profile similar to aged individuals (above 70 years). Immunosuppression was associated with lower antibody responses along the vaccine schedule (p=0.005 at t1; p=0.008 at t2). Noteworthy, previous unresponsiveness to hepatitis B vaccination (75/129, 58% of patients negative for anti-HBs antibodies) did not correlate with humoral unresponsiveness to BTN162b2. Other clinical and laboratory parameters had marginal correlations with response to vaccination. Conclusions The large majority of hemodialyzed patients showed IgG seroconversion upon BNT162b2 mRNA vaccination but a sizable proportion of patients presented poor responses. These results support further investigation into the relationship between vaccination, serologic response and host protection.