Identification of biomarkers for predicting central nervous system involvement in patients with diffuse large B‐cell lymphoma

Introduction: Central Nervous System (CNS) involvement in Diffuse Large B Cell Lymphoma (DLBCL) is a rare complication associated with bad prognosis. The CNS Prognostic Index identifies patients in high risk of CNS relapse; however, its specificity is limited and not accurate. Biomarkers that can predict CNS relapse are still missing. For that reason, we aimed to characterize the transcriptomic, genomic and epigenomic hallmarks of systemic tumors at diagnosis of patients that will relapse in the CNS. Methods: We conducted a retrospective study including 38 patients with nodal DLBCL that relapsed within the first two years after diagnosis: 12 with CNS relapse and 26 with systemic relapse. As controls, we included 21 patients that did not relapse. DNA and RNA were extracted from tumors at diagnosis. Immune infiltration was evaluated using the nCounter PanCancer Immune Profiling Panel (Nanostring); COO was determined using the Lymph2Cx assay. Transcriptomic profiles were generated by bulk RNA sequencing; mutational analysis was performed using a custom targeted NGS panel. Methylation profiles were described using the Illumina Infinium MethylationEPIC Array 850K (Figure 1A). Results: Immune cell composition was similar between groups; however, several genes were overexpressed at diagnosis in patients with CNS relapse. Genes related to chemokine signaling as CCL14, IL7R, and CCR6; adhesion like VCAM1 and ITGA4; and immune checkpoints as BTLA were upregulated in the CNS relapse group in comparison to patients without relapse (Figure 1B). Among these, the ITGA4 gene was also found to be overexpressed in the CNS relapse group when compared to both systemic and no relapse groups together (Figure 1C). No differences in COO were observed between groups. Whole transcriptomic data showed enrichment in genesets related to proliferation and survival in CNS relapse group in comparison to all others (Figure 1D). Patients with CNS relapse had increased mutation load and frequency in PIM1, IGLL5 and KMT2D genes. A total of 343 CpG sites were differentially methylated in CNS relapse group in comparison to No CNS relapse, involving 211 genes (Figure 1E). Keywords: Diagnostic and Prognostic Biomarkers, Extranodal non-Hodgkin lymphoma, Genomics, Epigenomics, and Other -Omics No conflicts of interests pertinent to the abstract.