Assessment of Serology after Sars-Cov-2 Vaccine in Allogeneic HCT Recipients

Introduction: Randomized trials demonstrated ~95% efficacy of SARS-CoV-2 spike messenger RNA (mRNA) vaccines. Patients (pts) after allogeneic hematopoietic cell transplant (HCT) have a variable period of immune deficiency and the impact of diagnosis, treatment regimens, GvHD, and immunosuppression on vaccine (vacc) immunogenicity is unknown. Methods: We performed a retrospective analysis of 149 consecutive pts (Table) who received a SARS-CoV-2 vacc between 12/17/2020, and 5/21/21, and were tested for anti-SARS-CoV-2 S1/S2 antibodies. Serology testing was performed with the Liaison® SARS-CoV-2 S1/S2 IgG assay (DiaSorin) with ≥15 AU/mL defined as a positive result. Pts with prior COVID-19 infection were excluded. Pts received mRNA-1273/Moderna (n= 46), BNT162b2/Pfizer-BioNTech (n= 100) or Jannsen vacc (n= 3). Reactogenicity was not investigated. Demographic and treatment variables were tested for prediction of vaccine response using Chi-square test or Fisher's exact test for categorical variables and two-sample t test for continuous variables. Univariate and multivariable logistic regression analyses with backward selection using Akaike information criterion (AIC) were used to examine interdependence of those variables and odds ratios (OR) with 95% confidence intervals (CI). Results: Pts underwent HCT from a related HLA matched sibling (n=36), related haploidentical (n= 23), or matched/mismatched unrelated donor (n= 89). 93% received fludarabine in the HCT conditioning regimen (data not shown). All pts received a calcineurin inhibitor (CNI) and 76 pts received ATG for GvHD prophylaxis. All pts achieved at least mixed donor lymphoid engraftment (data not shown). Median time from HCT to 1st vaccination was 26 months (range, 3-258 months). Median age at time of vaccination was 61 years (range, 24-78) and 75 (50%) were female. Serology was tested at a median of 37 days (range, 6-119 days) after the second vacc dose. Serology was tested <14 days in 3 pts; all were seropositive. No pt developed COVID-19 during the period of observation. 101 pts (67%) tested positive for anti-SARS-CoV-2 S1/S2 antibodies (vacc responders). Of the responders, the median time from HCT to 1st vacc was 45.5 months (range, 3-258, SD 39.78). Among the 23 pts between 3-9 months after HCT, 26% (n=6) had a positive antibody response, but all were receiving ongoing immunosuppression at time of vaccination. 29% (n=29) vacc responders were receiving prednisone (pred) in the management of cGvHD at the time of vaccination. 48 pts did not mount an antibody response (vacc non-responders). Of the non-responders, 30 pts were receiving cGvHD treatment at the time of vacc, 31 pts were taking pred, and 20 pts were taking CNIs. In univariate analysis, we found a history of prolonged use of pred (>8 weeks) and/or CNIs, on current treatment for cGvHD at time of vacc, and receipt of rituximab in the preceding 12 months predicted for lack of response (Table). Active use of pred and treatment with pred >8 weeks in the preceding 12 months prior to vacc predicted vacc non-response [OR 0.221; 95% CI (0.106 - 0.456); p<0.001] and [OR 0.408; 95% CI (0.197 - 0.844); p=0.016] in univariate analysis, respectively, however, active use of pred was predictive [OR 0.07; 95% CI (0.016-0.304; p<0.001] while pred treatment >8 weeks was not [OR 2.00; 95% CI (0.55-7.298; p=0.293] in multivariable analysis. Other significant predictors for non-response in the multivariable analysis include pt use of ruxolitinib [OR, 0.233, 95% CI, (0.067-0.808); p=0.022], and rituximab within 1 year [OR, 0.026, 95% CI, (0.007-0.099); p<0.001]. Discussion: In this study, we found that 67% allogeneic HCT pts developed anti-SARS-CoV-2 S1/S2 antibodies after SARS-CoV-2 vaccination. Predictors of non-response after adjustment for potential confounders, were factors that are expected to suppress immune response including active use of immunosuppressive medications. Consistent with prior studies, anti-CD20 therapy likely impairs humoral response to vaccination. Ruxolitinib also appears to impair response. However, a proportion of pts being actively treated for cGvHD responded to vaccination and these pts should still be encouraged to receive vaccination in consideration of the COVID-19 mortality risk. Many questions remain including the protective benefit of immune response, the duration of response, and the potential value of booster vaccinations in non-responders. [Figure: see text] DISCLOSURES: Rowley: ReAlta Life Sciences: Consultancy.