Biological Features of a High-Risk Transcriptional Molecular Subtype in Diffuse Large B-Cell Lymphoma

Blood(2022)

Cited 0|Views51
No score
Abstract
Background: We previously identified a high-risk subset of newly diagnosed Diffuse Large B-cell Lymphoma (DLBCL) through unsupervised clustering of transcriptomic data. This cluster, called A7 (Aggressive Lymphoma Subtype 7), is characterized by poor event-free survival on R-CHOP therapy, having roughly double the hazard rate of non-A7 patients. The cluster stratifies outcome in a manner independent of known risk-associated factors, including clinically defined ones like the international prognostic index, as well as molecularly defined ones like cell-of-origin (COO), or tumor microenvironment infiltration (TME26). Here we characterize the biological features of the A7 subtype. Methods: Samples from the ROBUST clinical trial (N=1016) were characterized in terms of gene expression, single nucleotide variants (SNV), copy number aberrations (CNA), and immunohistochemistry (IHC) protein features. Transcriptomic and genomic patterns were replicated in an independent cohort of DLBCL patients (MER, N=343). We performed differential gene expression analysis between A7 and all other subtypes, and aggregated differential expression results by biological pathways. We also applied classifiers to score patients along COO and TME26 dimensions. We quantified differential cell type abundance by IHC, and also explored mutational/copy number features associated with A7. Results: A7 is enriched for ABC COO but is not defined by it, with 80-85% of A7 patients classified as ABC, but only 25-35% of ABC patients classified as A7. The cluster is characterized by upregulation of MYC targets, E2F targets, G2M checkpoint, oxidative phosphorylation, and ABC-associated signatures, as well as downregulation of the p53 pathway and GCB-associated signatures. From an immune perspective, A7 is associated with "cold" tumors with little immune activity. The cluster is associated with significant downregulation of immune and inflammatory signatures. IHC data showed that A7 patients exhibit reduced immune populations including T cells and macrophages. We examined genetic features of A7 patients, but found no significantly enriched nonsynonymous somatic mutations (FDR<0.05). Several mutations associated with ABC such as such as ETV6, and PIM1 are nominally enriched in A7, reflecting the mostly ABC nature of A7. We found significantly enriched CNAs in A7 (FDR<0.05), suggesting copy number changes involving multiple genes may be genetic drivers of A7 rather than point mutations. In A7, roughly 60% of patients exhibit arm-level copy number gains on chromosomes 3p, 3q, and 18q, and 44% exhibit focal deletions on chromosome 9p. MYC dysregulation in A7 was observed from multiple perspectives, including upregulation of both MYC gene expression and MYC target pathways in RNAseq data, as well as increased protein expression measured by IHC. The A7-specific upregulation of MYC was not found to be associated with tumor cellularity, nor was it associated with MYC translocation or amplification. These common mechanisms of MYC regulation observed previously in the GCB subtype of DLBCL were not present in A7, suggesting other mechanisms of MYC regulation. Conclusion: A7 is a high-risk molecular subtype of DLBCL associated with poor response to R-CHOP therapy and is characterized by enrichment of ABC COO, low immune infiltration in the tumor microenvironment, downregulation of immune and inflammatory pathways, and upregulation of MYC, MYC target, and metabolism signatures. The subtype can be identified by its homogeneous molecular characteristics, potentially making it amenable to novel therapies targeting specific biology such as MYC and low immune infiltration, in contrast to clinically defined high-risk populations which may be biologically heterogeneous and with mixed outcome. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal
More
Translated text
Key words
high-risk,b-cell
AI Read Science
Must-Reading Tree
Example
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined