Abstract Diffuse large B cell lymphoma (DLBCL) is the most common hematologic malignancy and is characterized by a striking degree of clinical and molecular heterogeneity. Gene expression profiling has long been used to define subtypes and understand the heterogeneity within DLBCL. Microarray-based gene expression studies have identified cell of origin subgroups (Alizadeh et al., 2000), as well as gene expression signatures derived from host inflammatory response (Monti et al., 2005), stromal tissues (Lenz et al., 2008) as well as other biological processes. However, the connections between these gene expression signatures and genetic alterations is largely unknown. In this study, we sought to define connections between gene expression signatures and genetic alterations to better understand the biology of DLBCL. We enrolled a total of 1001 DLBCL patients and comprehensively defined the landscape of genetic mutations, copy number alterations and expression through whole-exome and transcriptome sequencing. We identified 150 genetic drivers of DLBCL including many novel, clinically relevant genes (e.g. SPEN, SETD1B and KLHL14). We began the gene expression analysis with a comprehensive universe of nearly 9500 annotated gene sets from widely used gene set databases (Kegg, Reactome, MSigDB), several lymphoma-specific gene signature sources (Lenz et al. 2008, Monti et al. 2005, and Shaffer et al.) and other published sources. From this large universe, we identified 1228 gene sets that comprised genes with high intra gene set correlation. The vast majority of these 1228 gene sets also exhibited high inter-gene set correlation indicating a high degree of redundancy between these gene sets. We developed an approach to singular value decomposition to define 31 exemplar gene sets that essentially replicated the patterns of the 1228 described gene sets. Our results recapitulated a number of described patterns underlying DLBCL gene expression including distinct gene sets reflecting cell of origin, host response, BCR and Oxidative Phosphorylation (Monti et al., 2005), as well as the “Stromal 1” and “Stromal 2” signatures (Lenz et al., 2008) signatures. There was no association between the expression of stromal signatures and overall mutational burden (p=0.4). Our data identified the genetic underpinnings of each of these distinct gene sets that provide a better understanding of the underlying biology affected by different genetic mutations. Interestingly, genetic alterations in RHOA were strongly associated with the proliferation gene signature and poor survival in DLBCL. We sought to better define the role of RHOA mutations in DLBCL which are mostly unknown. We knocked out the expression of RHOA in vitro through the CRISPR-Cas9 approach and found that RHOA deletion was lethal in both ABC DLBCL (LY3, TMD8, HBL1) and GCB DLBCLs (Pfeiffer and SUDHL4), results were validated with shRNAs. Conversely, cells overexpressing R5Q mutant proliferated 30% faster than vector control cells and more than 10% faster than wildtype overexpressing cells, directly implicating these mutations and RHOA overexpression in the growth of DLBCL cells. We further investigated the effects of RHOA knockout in vivo using two complementary mouse models. We bred Rhoa conditional knockout mice with Mb1-Cre transgenic mice, in which Cre is restricted to the B cell lineage as well as AID-Cre transgenic mice in which Cre is restricted to germinal center (GC) B cells. While B cells represented the vast majority of cells in the lymphoid organs of Rhoa wildtype mice, we found a dramatic reduction in B cell population and a resulting increase in T cell percentages in Rhoa deleted mice (p Collectively, our data define the genetic basis for observed changes in gene expression in DLBCL. Our experimental data define RHOA as an oncogene with roles in B cell development, germinal center maintenance, and DLCBL growth. Disclosures Leppa: Roche: Consultancy, Honoraria, Research Funding; Bayer: Research Funding; Merck: Consultancy, Honoraria; Takeda: Consultancy, Honoraria, Research Funding; Janssen Cilag: Consultancy, Research Funding. Flowers: Abbvie: Consultancy, Research Funding; Infinity: Research Funding; Clinical Care Options: Research Funding; National Cancer Institute: Research Funding; Spectrum: Consultancy; Onyx: Research Funding; OptumRx: Consultancy; Prime Oncology: Research Funding; Janssen Pharmaceutical: Research Funding; Acerta: Research Funding; Gilead: Consultancy; Seattle Genetics: Consultancy; Eastern Cooperative Oncology Group: Research Funding; Research to Practice: Research Funding; Burroughs Welcome Fund: Research Funding; Celgene: Consultancy, Research Funding; V Foundation: Research Funding; Bayer: Consultancy; Educational Concepts: Research Funding; Pharmacyclics LLC, an AbbVie Company: Research Funding; Genentech/Roche: Consultancy, Research Funding; Millennium/Takeda: Research Funding; National Institutes Of Health: Research Funding; TG Therapeutics: Research Funding. Gribben: TG Therapeutics: Honoraria; Kite: Honoraria; Janssen: Honoraria; Celgene: Honoraria; Pharmacyclics: Honoraria; Abbvie: Honoraria; Acerta: Honoraria; Karyopharm: Honoraria; Genentech/Roche: Honoraria. Hsi: Cellerant Therapeutics: Research Funding; Abbvie: Research Funding; Seattle Genetics: Consultancy, Honoraria, Speakers Bureau; Eli Lilly and Co.: Research Funding. Evens: Merck: Consultancy; Amgen: Consultancy; Seattle Genetics: Consultancy; Spectrum Pharmaceuticals: Consultancy; Novartis: Consultancy; Millennium: Consultancy; AbbVie: Consultancy; Affimed: Consultancy; Pharmacyclics: Consultancy; Celgene: Consultancy; Kite Pharma: Consultancy. Reddy: Abbvie: Consultancy; BMS: Consultancy; Celgene: Consultancy; Gilead: Speakers Bureau. Gordon: Janssen: Other: Data Monitoring Committee.
Read full abstract