Objective: To systematically evaluate the association between human leukocyte antigen (HLA) genotypes and the risk of hematological malignancies (HM) and explore the correlation between HM-related HLA and protein levels. Methods: Based on data from 438 464 participants in the prospective cohort UK Biobank (UKB), the Cox proportional hazards regression model was used to assess the associations between HLA genotypes and the risk of incident HM, and the multiple linear regression model was used to evaluate the correlation between HLA and plasma protein levels. All models were adjusted for age, sex, genotyping batch, and the top five gene primary components (PCs). Results: Ten HLA genotypes were identified as significantly associated with the risk of HM. The development of lymphoid malignancies (Hodgkin lymphoma, non-Hodgkin lymphoma, and lymphocytic leukemia) was jointly influenced by both HLA-I and HLA-II loci. Specifically, B*08:01, C*07:01, DQA1*05:01, DQB1*02:01, DRB1*03:01, and DRB3*01:01 consistently increased the incidence risk of lymphoid malignancies. In contrast, the risk of multiple myeloma was exclusively associated with HLA-I locus B*07:02 (HR=1.30, 95% CI: 1.14~1.47, Padj=8.518×10-3) and C*07:02 (HR=1.30, 95% CI: 1.15~1.47, Padj=8.518×10-3). Enrichment analysis demonstrated that HLA-related proteins were significantly enriched in cell adhesion molecule interactions, as well as core immune signaling pathways including PI3K-Akt, MAPK, and NF-κB. Furthermore, HM-susceptible HLA genotypes significantly influenced the expression levels of four immune checkpoint-related protein families: killer cell immunoglobulin-like receptors, leukocyte immunoglobulin-like receptors, programmed cell death proteins, and TIGIT. Conclusion: The HLA genotype is closely related to the susceptibility to HM, and its potential biological mechanism may be realized by regulating immune checkpoint proteins and related signaling pathways. It not only confirms the value of HLA genotype as a key biomarker for assessing high-risk populations for HM but also provides multi-omics clues for understanding the tumor immune escape mechanism affected by HLA genetic variations.