Closing the gap: Genetic profiling of B-cell lymphomas using integrative whole-exome and whole-transcriptome sequencing

Abstract Background: The treatment of B cell lymphomas has evolved significantly; however, a portion of patients still experience inferior outcomes, indicating an ongoing need for better ways to predict relapse and individualize therapy. Lymphomas are primarily classified based on lineage and cell of origin, and as a result, current systems do not fully capture the disease's molecular complexity. Recent studies have identified distinct B cell lymphoma genetic subtypes with high reproducibility and direct clinical relevance. Consequently, we are establishing a comprehensive lymphoma biobank with integrated genomic, transcriptional, and clinical data to directly inform the design of molecularly-stratified therapeutic trials. Our initial lymphoma panel results, presented here, highlight the clinical utility of this approach. Methods: DNA samples isolated from B cell lymphoma specimens derived from either formalin-fixed, paraffin-embedded (FFPE) tissue or bone marrow (BM) were collected. DNA was extracted using the Ionic Purification System (Bionano, Germany). Using the DHCancerseq assay, library preparation was performed using Agilent SureSelect Human Exome v8 on the Agilent Magnis NGS Prep System (Agilent Technologies, USA). The RNA library was prepared using the Illumina TruSeq RNA Exome sequencing assays. RNA sequencing was not performed on BM specimens. Raw next-generation sequencing data was processed using the AUGMET analysis platform to identify single nucleotide variants, small deletions, small insertions, amplifications, fusions, and splice variants. Additionally, a dynamic whole genome copy number plot was generated for the identification of large gains, losses, and loss of heterozygosity. We retained variants assessed to meet classification criteria tiers 1A-2D using a consensus recommendation from the Association for Molecular Pathology. Results: Thirty-eight specimens (n=26 FFPE; n=12 BM), including 12 diffuse large B cell lymphomas (DLBCL), 10 mantle cell lymphomas (MCL), 2 follicular lymphomas (FL), and 14 unspecified B cell lymphomas, were analyzed. Clinically relevant variants were identified in 79% of cases (n=30/38), totaling 56 tier 1A-2D pathogenic variants. All DLBCL cases (n=12/12) harbored at least one pathogenic variant. The most frequently altered genes were TP53 (n=10/56, 18%), MYD88 (n=4/56, 7%), NOTCH2 (n=4/56, 7%), ATM (n=3/56, 5%), CARD11 (n=3/56, 5%), KMT2D (n=3/56, 5%), EZH2 (n=3/56, 5%), and CREBBP (n=3/56, 5%). Biallelic TP53 loss was identified in 50% of MCL cases (n=5/10). In particular, a loss of heterozygosity in TP53 was identified in 1 MCL sample which could not have been identified using traditional fluorescence in situ hybridization-based methodologies. Additionally, transcriptomic analysis reclassified a high-grade B cell lymphoma to a Philadelphia chromosome-positive B cell acute lymphoblastic leukemia transformed from FL after identification of a 9;22 translocation. Whole-genome visualization identified complex karyotypes, key deletions, and loss of heterozygosity events affecting TP53 and CDKN2A across multiple B cell lymphoma subtypes.Conclusion: Whole-exome and whole-transcriptome sequencing of B cell lymphomas identified a high percentage of clinically relevant variants, confirming its potential for use in clinical practice. This infrastructure provides a framework for developing personalized therapeutic strategies and molecularly-stratified clinical trials across B cell lymphomas. Future work will pair these data with clinical outcomes to identify biomarkers for targeted therapy, stratify patients for clinical trials, and validate new molecular classifications.