Serum microRNA Prognostic Model and Underlying Immune Alterations in Diffuse Large B Cell Lymphoma

Serum microRNA prognostic model and underlying immune alterations in diffuse large B cell lymphoma

Abstract

Background: MicroRNAs (miRs) have been investigated as prognostic biomarkers and therapeutic targets in diffuse large B-cell lymphoma (DLBCL). Translating these findings into clinical approaches should be validated by independent patient cohorts and a better understanding of underlying biology mechanism remains great interests.

Methods: MiRNA microarray was performed on serum samples of 20 DLBCL patients at diagnosis, remission and at relapse. Subsequent real-time PCR validation of 504 newly diagnosed DLBCL patients was used to generate a serum prognostic model in predicting clinical outcomes. The mechanism of action of miR signature on tumor progression was further analyzed by whole exome sequencing and RNA sequencing.

Results: Four serum miRs (miR-21, miR-155, miR-130b and miR-28) were identified to predict relapse in a training cohort of 225 DLBCL patients and validated in a validation cohort of 279 DLBCL patients. Multivariate Cox regression models indicated that when the International Prognostic Index was controlled, this serum miR prognostic model was an independent prognostic factor in DLBCL. Moreover, mutations of transcription factors (TFs) were significantly correlated with these four miRs. Twenty lymphoma-associated TFs regulated at least one miR by transmiR v2.0 database and different genetic mutation burden was present between low-index and high-index of the model. Furthermore, RNA sequencing data manifested that the four-miR signature correlated with immune status of the DLBCL patients.

Conclusions: An easy-to-use miR prognostic model was utilized to predict disease relapse and prognosis of DLBCL patients. The tumor-specific mutation burden and immune dysregulation potentially contribute to tumor progression in DLBCL.

Key words: MicroRNAs, Diffuse Large B-cell Lymphoma, Prognosis, Signaling pathways.