Coding and non-coding drivers of mantle cell lymphoma identified through exome and genome sequencing

Objectives Mantle cell lymphoma (MCL) is an uncommon B-cell non-Hodgkin lymphoma that is incurable with standard therapies. The genetic drivers of this cancer have not been firmly established and the features known to contribute to differences in clinical course remain limited. We sought to extend our understanding of the molecular etiology of this malignancy using an integrative genomic analysis of diagnostic biopsies. Methods We performed exome sequencing on 51 frozen MCL tumors and analyzed these alongside previously published exome cohorts. We sequenced tumour genomes and matched constitutional DNA from 34 frozen MCLs, along with matched constitutional DNA, to more broadly identify the pattern of non-coding mutations. Based on mutations identified in this discovery cohort, we re-sequenced 18 recurrently-mutated genes in 212 archival MCLs, each having clinical follow-up data. We also performed RNA-seq on 110 of these cases and analyzed these data for alternative splicing and differential expression, including the differential splicing of HNRNPH1 in the context of recurrent intronic mutations. We investigated the functional and phenotypic effect of mutations and deregulated HNRNPH1 protein through ectopic expression of full-length HNRNPH1 and a mini-gene containing the exons and introns affected by mutations. Using custom droplet digital PCR (ddPCR) assays, we validated alternative splicing patterns in HNRNPH1 itself and other targets identified through re-analysis of available CLIP-seq data. Results In addition to confirming the prognostic association of TP53 and NOTCH1 mutations in MCL, we identified two additional genes associated with outcome: EWSR1 with poor outcome (HR = 5.6) and MEF2B with good outcome (HR = 0.2). By comparing mutation patterns to diffuse large B-cell lymphoma (DLBCL), we identified an MCL-specific missense hot spot in MEF2B, non-specific truncating mutations in EWSR1, and truncating mutations affecting the DAZAP1 C-terminus in both MCL and DLBCL. The DAZAP1 mutations are predicted to alter protein sub-cellular localization and disrupt protein-protein interactions. We also identified the focal recurrence of non-coding mutations surrounding a single exon of the HNRNPH1 gene that were largely restricted to MCL. These mutations affected a region bound by HNRNPH1 protein and disrupted the preferred binding motif of this protein. Intronic mutations were significantly associated with alternative splicing of the HNRNPH1 mRNA and appear to disrupt a negative regulatory loop that normally limits the level of HNRNPH1. Using cell-based assays, we have evaluated the role of HNRNPH1 in cell survival and proliferation. Our interrogation of alternative splicing events in downstream targets implicate HNRNPH1 as a master splicing regulator which may broadly perturb the transcriptome and proteome to favor lymphomagenesis in MCL. Conclusions We discovered three novel MCL-related genes with roles in RNA trafficking or splicing, namely EWSR1, DAZAP1, and HNRNPH1. Mutations in these RNA-binding proteins were identified in 49 of 291 (17%) samples analyzed. Our results improve the current understanding of the MCL mutational landscape, highlight the similarities and differences between MCL and DLBCL, and strongly implicate a role for aberrant regulation of RNA metabolism in MCL pathobiology. We elucidated a functional role for recurrent non-coding HNRNPH1 mutations specific to MCL and identified multiple downstream targets. We continue to explore putative trans targets of HNRNPH1, a novel oncoprotein in MCL. Disclosures Steidl: Seattle Genetics: Consultancy; Roche: Consultancy; Bristol-Myers Squibb: Research Funding; Bayer: Consultancy; Nanostring: Patents & Royalties: Filed patent on behalf of BC Cancer; Juno Therapeutics: Consultancy; Tioma: Research Funding. Connors:Bristol-Myers Squibb: Consultancy; Seattle Genetics: Honoraria, Research Funding; Takeda Pharmaceuticals: Honoraria. Villa:Roche, Abbvie, Celgene, Seattle Genetics, Lundbeck, AstraZeneca, Nanostring, Janssen, Gilead: Consultancy, Honoraria. Johnson:Roche: Consultancy, Employment, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Travel fees, gifts, and others, Research Funding; Abbvie: Consultancy, Employment, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Merck: Consultancy, Honoraria; BMS: Consultancy, Honoraria; BD Biosciences: Other: Provided a significant proportion of the antibodies used in this project free of cost.; Seattle Genetics: Honoraria; Lundbeck: Employment, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Travel fees, gifts, and others, Research Funding. Scott:Janssen: Consultancy, Research Funding; NanoString: Patents & Royalties: Named inventor on a patent licensed to NanoSting [Institution], Research Funding; Celgene: Consultancy; Roche/Genentech: Research Funding.

Download Full-text

Perturbations in HNRNPH1 Splicing and Abundance Affect Global Splicing and Proliferation in Mantle Cell Lymphoma

Blood ◽

10.1182/blood-2020-141389 ◽

2020 ◽

Vol 136 (Supplement 1) ◽

pp. 23-24

Author(s):

Krysta M Coyle ◽

Quratulain Qureshi ◽

Prasath Pararajalingam ◽

Nicole Thomas ◽

Timothy E Audas ◽

...

Keyword(s):

Alternative Splicing ◽

Mantle Cell Lymphoma ◽

Cell Lymphoma ◽

Intron Retention ◽

Gene List ◽

Protein Abundance ◽

Mantle Cell ◽

Sirna Knockdown ◽

Retained Introns

Objectives Mantle cell lymphoma (MCL) is an uncommon B-cell non-Hodgkin lymphoma that is incurable with standard therapies. The genetic drivers of this cancer have not been firmly established and the features known to contribute to differences in clinical course remain limited. We previously discovered non-coding and silent mutations in HNRNPH1 that affect its splicing and contribute to poor outcomes for patients with MCL. We sought to extend our understanding of the mechanisms by which HNRNPH1 contributes to MCL pathology using a combination of in vitro models and integrative analysis of RNA sequencing from MCL tumors. Methods We previously sequenced ribosomal RNA-depleted RNA from 130 MCL tumors. Based on our earlier identification of mutations in HNRNPH1 and altered splicing of this gene, we performed differential splicing analyses using rMATS and leafcutter. We investigated the functional and phenotypic effect of deregulated hnRNP H1 protein through siRNA knockdown. Results Our previous work identified that splicing of HNRNPH1, and not total mRNA expression, correlated with protein abundance in MCL tumors. As a result, our analysis of alternative splicing focused on events associated with altered splicing of HNRNPH1. We identified 155 unique alternative splicing events (ΔPSI > 0.1, FDR < 0.1). Gene ontology analysis identified various aspects of RNA processing which are significantly enriched within this gene list, including mRNA splicing, transport, and metabolic process. This nominates HNRNPH1 as part of the complex network controlling alternative splicing within MCL. Available CLIP-seq in HeLa cells provides evidence for direct interactions between hnRNP H1 and transcripts identified by our analysis (e.g. RBM25, EIF4A1, HNRNPA2B1). Of the 155 events we identified, more than half involved retained introns. Generally, retained introns result in non-productive RNA species, which indicates that this program of intron retention in MCL is a mechanism by which protein abundance can be regulated by hnRNP H1. For all cases with available Mantle Cell Lymphoma International Prognostic Indicator (MIPI) classification, we determined the splicing ratio for HNRNPH1 and observed a general association between high MIPI scores and a lower ratio of non-productive HNRNPH1 transcripts. This suggested that the increased hnRNP H1 abundance we observed in HNRNPH1-mutant tumors contributes to increased proliferation of MCL cells. We verified this in vitro with siRNA knockdown of HNRNPH1 in HEK cells, which resulted in a significant decrease in cell proliferation. Conclusions We have described a pattern of alternative splicing in MCL that is associated with alterations in HNRNPH1 splicing and related protein abundance. The prevalence of retained introns suggests that hnRNP H1 regulates the abundance of protein-coding transcripts via alternative splicing coupled to nonsense-mediated decay. We continue to explore targets of hnRNP H1, a novel oncoprotein in MCL. Disclosures Morin: Celgene: Consultancy.

Download Full-text

Abstract PO-04: Noncoding mutations in mantle cell lymphoma disrupt regulation of HNRNPH1 by alternative splicing

10.1158/2643-3249.lymphoma20-po-04 ◽

2020 ◽

Author(s):

Krysta M. Coyle ◽

Quratulain Qureshi ◽

Prasath Pararajalingam ◽

Nicole Thomas ◽

Timothy E. Audas ◽

...

Keyword(s):

Alternative Splicing ◽

Mantle Cell Lymphoma ◽

Cell Lymphoma ◽

Mantle Cell

Download Full-text

Regulation of HNRNPH1 via alternative splicing is disrupted by non-coding mutations in mantle cell lymphoma

10.26226/morressier.5ebd45acffea6f735881aeb8 ◽

2020 ◽

Author(s):

Krysta Mila Coyle

Keyword(s):

Alternative Splicing ◽

Mantle Cell Lymphoma ◽

Cell Lymphoma ◽

Mantle Cell

Download Full-text

Unraveling Actionable Target Mutations in Formalin Fixed Tissue in Mantle Cell Lymphoma

Blood ◽

10.1182/blood-2020-136944 ◽

2020 ◽

Vol 136 (Supplement 1) ◽

pp. 25-26

Author(s):

Joana M. Rodrigues ◽

May Hassan ◽

Anna Porwit ◽

Sara Ek ◽

Mats Jerkeman

Keyword(s):

Cell Cycle ◽

Mantle Cell Lymphoma ◽

Research Funding ◽

Cell Lymphoma ◽

Population Based ◽

Fixed Tissue ◽

Formalin Fixed Tissue ◽

Recurrent Mutations ◽

Mantle Cell ◽

Formalin Fixed

Introduction: Mantle Cell Lymphoma (MCL) is a B-cell neoplasm with poor prognosis and high relapse frequency. It follows an heterogenous clinical course and, despite clear improvements in survival among young and fit patients, most patients eventually relapse. Recurrent mutations have been previously described, pinpointing alterations in DNA damage pathways, cell survival and cell-cycle regulation. Nevertheless, the current genomic knowledge of the disease cannot fully explain its high clinical variability. The present study aimed at revealing the mutation profile for the development of personalized treatment approaches in a population-based cohort of patients diagnosed with MCL. Material & Methods: We performed sequencing across ~230 lymphomagenesis-related genes on 75 formalin-fixed paraffin-embedded (FFPE) samples from symptomatic patients diagnosed with MCL in Southern Sweden. The panel covered the exonic parts of the chosen genes with capture probes provided by TWIST Bioscience. Sequence alignment was done against the human genome version 37 and variant calling was performed resorting to Sention TNscope. IGV was used for manual validation of each identified mutation. Samples were pre-selected according to DNA quality. Data on patient characteristics, prognostic factors and treatment was extracted from the Swedish Lymphoma Register. Results: MCL tumors were highly heterogenous with respect to the mutational status of the analyzed genes, ranging between 0 and 45 mutated genes in this cohort. ATM was the most commonly mutated gene (49%), followed by KMT2D, NOTCH2, TP53, FAT4, BIRC3, CCND1, UBR5, MEF2B, NOTCH1,SMARCA4, DNAH5,SAMHD1 and SP140 (Figure 1). Our results are in accordance with the previously published analyses, although the frequency of samples with mutated NOTCH2 was higher (18.7%, compared to 3-12.5% of affected tumors in previous studies), and with less samples with mutated NOTCH3 (6.7% of the studied tumors were mutated, vs 12.5% reported by Yang et al., (2018)). We found mutations in DNAH5, PCLO, FAT3 and BRCA1 genes, which have not been previously reported in MCL. Additionally, FAT4 (14.7%) showed recurrent mutations in our cohort, but has only been referred as commonly mutated in MCL before by Zhang et al. (2014). Main affected pathways were NOTCH-related and TP53-related pathways, followed by chromatin remodeling and cell cycle pathways. Due to low numbers of patients with non-classic morphology, no mutations were significantly different comparing classical and non-classical histology, although there was an enrichment of TP53 and NOTCH1 mutations in non-classic cases. It seems that cases with non-classic morphology have stronger driver mutations, which may reflect their aggressiveness. Cases with high proliferation (Ki67 >30%) showed a higher frequency of SMARCA4 (30%, vs 7.7% for low proliferation tumors) and TP53 mutations (36% in highly proliferative tumors vs. 17% in low proliferative tumors). No differences in mutation patterns were found in cases with low vs. high SOX11 expression. Unsurprisingly, TP53 mutations were the strongest independent factor for prognosis among the studied genes. Conclusions: Using the results from a population-based cohort of MCL, we report the feasibility of applying a targeted panel and sequencing technologies on DNA from formalin fixed tissue, as a method applicable in clinical routine in tailoring treatment approaches for MCL. Figure 1. Top 15 genes mutated in the cohort studied. Frequency of mutations on each gene is presented on the right panel and in the upper panel the proportion of affected genes by sample can be seen. Darker blue identifies a missense mutation, light blue corresponds to a frameshift deletion, pink color matches frameshift insertions, orange represent in frame deletions, whereas red in frame insertions. Light green identifies nonsense mutations and dark green stands for different types of mutation affecting that gene in that sample. Figure Disclosures Jerkeman: Roche: Research Funding; Abbvie: Research Funding; Celgene: Research Funding; Gilead: Research Funding; Janssen: Research Funding.

Download Full-text