SNRPB and CEP290, predicting the prognosis of diffuse large B cell lymphoma and associated with tumour immune microenvironment.

BACKGROUND

Diffuse large B-cell lymphoma (DLBCL), the most prevalent type of non-Hodgkin's lymphoma, exhibits significant correlations with efferocytosis-related molecules (ERMs) concerning invasion, metastasis, and clinical outcomes. This study aims to establish an efferocytosis-related gene signature specifically linked to DLBCL.

METHODS

Key module genes linked to DLBCL were identified via weighted gene co-expression network analysis (WGCNA) in GSE32018. Univariate Cox analysis of GSE31312 revealed ERMs associated with DLBCL survival. Differential expression analysis identified differentially expressed genes (DEGs) between DLBCL subtypes and normal samples. Venn diagram analysis identified common DEGs and key module genes. A DLBCL gene signature was built by using univariate Cox and least absolute shrinkage and selection operator (LASSO) analysis. Gene functional enrichment, immune microenvironment, and immunotherapy analyses compared two risk subgroups. Prognostic gene expression was validated at the single-cell level.

RESULTS

In the GSE32018 dataset, 1760 key module genes related to DLBCL were identified. Using GSE31312, 14 ERMs associated with DLBCL prognosis were determined.Then, an ERMs-related prognostic signature, including small nuclear ribonucleoprotein polypeptides B (SNRPB) and centrosomal protein 290 (CEP290), was established. Independent prognostic analysis showed that the RiskScore derived from this signature was a prognostic factor. Significant immune microenvironment differences were observed between two risk subgroups. Additionally, chemotherapeutic drug sensitivity results indicated the signature could predict therapeutic response. Eventually, expression of SNRPB and CEP290 was confirmed in B cells.

CONCLUSION

The prognostic signature comprised of SNRPB and CEP290 based on ERMs-DEGs was established, providing a theoretical basis and reference value for DLBCL research.