COVID-19 Vaccine Response in Patients with Hematologic Malignancy: A Systematic Review and Meta-Analysis

Introduction: Emerging data suggests that seroresponse (SR) in patients with hematologic malignancy following COVID-19 vaccination is likely lower than in patients without blood cancer. The objective of this study was to perform a systematic review and meta-analysis on SR in patients with hematologic malignancy who received COVID-19 vaccination (submitted to PROSPERO for registration). Methods: We searched PubMed and EMBASE from December 1, 2020, to July 22, 2021, to identify studies of SR following COVID-19 vaccine in adult patients with hematologic malignancy (including studies in which patients with hematologic malignancy represented a subset of a broader population). Patients with positive serologic response at baseline (prior to vaccination) or known COVID-19 infection were excluded. The primary outcomes were pooled SR estimates following COVID-19 vaccination in patients with hematologic malignancy, and pooled SR estimates of subgroups based on hematologic malignancy type. Secondary outcomes were pooled relative risk ratio (RR; compared to non-cancer controls) based on dichotomous-effect SR in all patients, and subgroups based on hematologic malignancy type, treatment status, and use of anti-CD20 therapy. Pooled estimates and RR with its associated 95% confidence intervals (CIs) were calculated using MetaXL (EpiGear), and Reference Manager (Cochrane) using random effects model. Results A total of 17 studies comprising 2834 patients with hematologic malignancy from Europe, United Kingdom and North America were included (Figure 1). The pooled estimate for SR was 58% (95% CI 48-67%, I (2) 95%), with a RR of 0.53 (95% 0.42-0.66, I (2) 94%) when compared to controls (10 studies with comparison group, 1092 hematologic malignancy patients, 830 controls; Figure 2). The pooled estimate for SR varied by type of hematologic malignancy: lymphomas SR 52% (95% CI 36-68%, 7 studies, 832 patients, I (2) 94%); chronic lymphocytic leukemia (CLL) SR 42% (95% CI 25-60%, 6 studies, 921 patients, I (2) 93%); plasma cell dyscrasias SR 66% (95% CI 47-83%, 8 studies, 611 patients, I (2) 95%); myeloproliferative neoplasms (MPNs, including chronic myelogenous leukemia) SR 83% (95% CI 68-95%, 6 studies, 227 patients, I (2) 58%); acute leukemia SR 86% (95% CI 77-94%, 2 studies, 67 patients, 46 acute myelogenous leukemia [AML] and 15 acute lymphocytic leukemia], I (2) 0%). The RR for SR also varied by type of hematologic malignancy: lymphomas (excluding CLL) RR 0.48 (95% CI 0.34-0.68, 4 studies, 337 patients, I (2) 89%); CLL RR 0.37 (95% CI 0.25-0.53, 3 studies, 194 patients, I (2) 54%); plasma cell dyscrasias RR 0.73 (95% CI 0.62-0.86, 5 studies, 323 patients, I (2) 70%); RR MPN 0.78 (95% CI 0.62-0.99, 3 studies, 199 patients, I (2) 90%). The pooled estimate for SR in those receiving treatment was 42% (95% CI 26-58%, 9 studies, 683 patients, I (2) 94%). The pooled estimates for SR for those receiving anti-CD20, bruton tyrosine kinase inhibitor (BTKi), or venetoclax were 13% (95% CI 1-32%, 6 studies, 367 patients, I (2) 88%), 42% (95% CI 17-71%, 3 studies, 319 patients, I (2) 75%), and 20% (95% CI 0-54%, 3 studies, 39 patients, I (2) 66%), respectively. The RR for those receiving treatment for their hematologic malignancy compared to those who were not receiving treatment was 0.51 (95% CI 0.37-0.71, 8 studies, 579 patients, I (2) 89%; Figure 3). . The RR of patients receiving anti-CD20 therapy compared to non-cancer controls was 0.13 (95% CI 0.02-0.93, 102 patients, I (2) 73%). For patients treated with anti-CD20 therapy, the RR of those receiving vaccination within 9-12 months compared to beyond 9-12 months was 0.12 (95% CI 0.06-0.25, 2 studies, 74 patients, I (2) 0%; Figure 4). Conclusion: Our systematic review and meta-analysis suggests that patients with hematologic malignancy have a lower SR rate following vaccination compared to controls. Furthermore, SR is variable across different types of hematologic malignancy, with very good response rates seen in patients with myeloid diseases (MPN and AML) and poor response rates seen in lymphoma and CLL. Active treatment, particularly anti-CD20 therapy within 12 months of vaccination, is associated with a particularly low SR following vaccination. Additional studies are needed to understand non-humoral responses to vaccination, and to guide decisions regarding how to optimize vaccine response in patients with blood cancer. We plan to update the systematic review and meta-analysis as more data become available. [Figure: see text] DISCLOSURES: No relevant conflicts of interest to declare.