Background:

Hematopoietic stem cell transplant (HSCT) recipients are uniquely vulnerable to adverse outcomes of SARS-CoV-2 infection. Small, mostly observational studies suggest that some HSCT recipients may not generate protective antibody responses following SARS-CoV-2 vaccination. We conducted a meta-analysis to estimate the prevalence and identify predictors of vaccine non-response. Methods A comprehensive search of electronic databases, including MEDLINE (Ovid), Embase (Elsevier), Web of Science Core Collection (Clarivate), the Cochrane Central Register of Controlled Trials (Wiley), and the Cochrane COVID-19 Study Register was conducted on January 20, 2023. We defined a non-response as not achieving a seroconversion (positive anti-S IgG titer) after receiving at least two vaccine doses, indicated by study-specific assay cut-off value. Only studies assessing COVID-19 vaccine induced antibody (anti-S IgG) responses in adult (≥ 18 years) HSCT recipients were included. With 95% confidence intervals (CI) across all studies, a random-effects model was used to combine the pooled effect sizes. Quality and risk of bias assessment were determined using the Newcastle-Ottawa scale and ROBINS-I tool, respectively. Results Out of 903 unique articles identified and 439 screened, 45 were included in this analysis comprising 4568 participants. Pooled absent sero-conversion was 20% (95% CI: 17% − 24%) with significant heterogeneity ( I 2  = 95.10%) among included studies (1 clinical trial, 1 cross-sectional study, 1 case-control study, and 42 observational cohort studies). Subgroup analyses showed no difference between autologous [0.21 (95%CI 0.12–0.31)] and allogeneic [0.20 (95%CI 0.17–0.24)] transplant recipients. Identified predictors of non-response included time interval between transplantation and vaccination (< 12 months), concurrent anti-CD20 therapy, and specific treatments (high-dose glucocorticosteroid, calcineurin inhibitor, and anti-thymocyte globulin) for graft versus host disease. No publication bias was observed but the Galbraith’s plot asymmetry showed evidence of small-study effects. Conclusion Our findings emphasize the significant prevalence of non-responsiveness to SARS-CoV-2 vaccination in HSCT recipients and underscore need for close monitoring and aggressive risk factor management in this immunocompromised population.