Researchers Identify New Subtype of DLBCL, Develop Assay to Stratify Risk

Researchers have defined a clinically and biologically distinct subgroup of tumors within germinal center B-cell&ndash;like diffuse large B-cell lymphoma characterized by a gene expression signature of high-grade B-cell lymphoma with&nbsp;<em>MYC&nbsp;</em>and&nbsp;<em>BCL2&nbsp;</em>and/or&nbsp;<em>BCL6&nbsp;</em>rearrangements, according to a study published in the&nbsp;<em>Journal of Clinical Oncology</em>.

Researchers have defined a clinically and biologically distinct subgroup of tumors within germinal center B-cell—like diffuse large B-cell lymphoma (GCB-DLBCL) characterized by a gene expression signature of high-grade B-cell lymphoma (HGBL) withMYCandBCL2and/orBCL6rearrangements (HGBL-DH/TH), according to a study published in theJournal of Clinical Oncology.1The researchers also created an assay that uses routinely available biopsy samples to identify such patients who would benefit from dose-intensive regimens or novel treatment strategies.

“We sought to identify gene expression features that distinguish HGBL-DH/TH-BCL2[HGBL-DH/TH withBCL2rearrangements] from the remainder of GCB-DLBCLs. In doing so, we discovered a distinct molecular subgroup that comprises 27% of GCB-DLBCLs, only one-half of which were HGBL-DH/TH-BCL2,” wrote the authors, led by Daisuke Ennishi, PhD, of the University of British Columbia. “Relative to the other GCB-DLBCLs, these tumors have significantly lower expression of genes associated with the light zone (LZ) of the germinal center and have mutations and gene expression features that imply potentially targetable vulnerabilities.”

Ennishi et al analyzed RNA sequencing data from 157 samples of de novo GCB DLBCLs to define gene expression differences between HGBL-DH/TH-BCL2and other GCB-DLBCLs. Of this discovery cohort, 25 were HGBL-DH/TH-BCL2.

Discovery cohort samples were GCB-DLBCLs with availableMYCandBCL2fluorescence in situ hybridization (FISH) results from a cohort of 347 diagnostic biopsy samples of patients with de novo DLBCL treated with rituximab (Rituxan) plus cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) treatment.2These cases were selected from the British Columbia Cancer population-based registry.

In the recent study, Ennishi et al analyzed targeted resequencing, whole-exome sequencing, RNA sequencing, and immunohistochemistry data in order to assess the genetic, molecular, and phenotypic features associated with the double-hit signature.

They then analyzed 2 external cohorts with RNA sequencing data available from a total of 440 patients to explore the prognostic significance and molecular features associated with double-hit signature (DHITsig) DLBCL.

To validate the novelDLBCL90 NanoString assay, the authors analyzed biopsy samples from 322 of their 347-patient dataset. They also analyzed 88 transformed follicular lymphomas with DLBCL morphology and 26 HGBLs from patients treated at their institution.

Ennishi et al identified 104 genes that were differentiated between HGBL-DH/TH-BCL2and other GCB-DLBCLs. They used the expression of these genes to identify 2 groups of GCB-DLBCL tumors. The DHITsig-positive group consisted of 42 tumors, or 27% of the group. This positive group included 22 of the 25 FISH-determined HGBL-DH/TH-BCL2tumors.

The DHITsig-negative group consisted of 115 GCB tumors, or 73% of the total. The negative group included 3 HGBL-DH/TH-BCL2tumors.

The DHITsig had been developed previously without knowing patient outcomes, so the investigators then used data from the 157 de novo GCG-DLBCL R-CHOP—treated group to determine its prognostic impact.MYCandBCL2translocations and protein expression of MYC and BCL2 were observed more frequently in DHITsig-positive patients (allP<.001), which was an expected finding.

They also found that DHITsig-positive patients had a significantly shorter time to progression (TTP;P<.001). Disease-specific survival (DSS) and overall survival (OS) were also significantly shorter compared with DHITsig-negative GCB patients (log rankP<.001, .012, respectively). These patients experienced outcomes comparable to those with activated B-cell—like (ABC) DLBCL from the cohort of 347 patients.

An independent data set of 262 patients was also tested with the gene expression model, and the investigators again found that patients who were identified as DHITsig-positive had a shorter OS compared with other GCB-DLBCLs (P<.001).

Looking at the gene expression and biological nature of DHITsig-positive tumors, the investigators discovered that these tumors exhibited lower expression of immune and inflammation signatures and more frequent loss of MHC class I and II protein expression compared with other GCB tumors. In addition, DHITsig-positive tumors demonstrated frequent mutations inCREBBP,EZH2Y646,DDX3X,TP53, andKMT2D(false discovery rate [FDR] <.10).On the other hand, mutations inTNFAIP3,KLHL6,NFKBIE,TET2,CD58, andSTAT3were more frequently found in DHITsig-negative tumors (FDR <.10).

The study authors wanted to investigate how to translate DHITsig into a clinically relevant assay, so they reduced the 104-gene RNA sequencing model into a 30-gene module. They applied the module to 171 GCB-DLBCL tumors from the 347-patient cohort (including 156 from the discovery cohort), which showed 26% of the samples were DHITsig-positive, 64% were DHITsig-negative, and 10% were DHITsig-indeterminate. The misclassification rate was 3% against the RNA sequencing comparator.

They then applied the DLBCL90 assay to the 322 available biopsy samples from the 347-patient de novo DLBCL cohort. They learned that the DHITsig was not seen in ABC-DLBCL, although 4% of 102 were classified as indeterminate by the assay. According to the authors, the prognostic significance for TTP, DSS, and OS of DHITsig was maintained (allP<.001).

Ennishi et al found that 11 of 25 cases of DHITsig-positive transformed follicular lymphoma were classified as HGBL-DH/TH-BCL2.None of the 50 in the DHITsig-negative group received that classification. They also found that the DHITsig was shared with the majority of B-cell lymphomas with high-grade morphology.

Of note, the assay showed that patients with DHITsig-negative GCB-DLBCL had a good prognosis, with a DSS rate of 90% at 5 years. Even in advanced-stage disease, the DSS rate of the DHITsig-negative patients with GCB-DLBCL was 87% at 5 years. The authors believe that these data provide strong evidence that R-CHOP is sufficient for these patients, avoiding the need to escalate therapy.

Ennishi et al also identified the DHITsig in other tumors that had high-grade morphology. According to the revised 2017 World Health Organization (WHO) classification, these tumors fall into the HGBL not otherwise specifiedand HGBL-DH/TH with high-grade morphology groups. The authors proposed that the new signature be used to identify tumors with high-grade molecular features.

They believe that the DHITsig could form the basis of a new, more inclusive WHO category that includes DLBCLs that do and do not have these rearrangements as genomic testing becomes more widespread in clinical practice.

“This system would classify tumors with shared biology together while significantly expanding a group of patients with an established need for dose-intensive regimens or alternative therapies and may drive acceleration of clinical trials aimed at improving outcomes,” Ennishi et al wrote.


  1. Ennishi D, Jiang A, Boyle M, et al. Double-hit gene expression signature defines a distinct subgroup of germinal center b-cell-like diffuse large b-cell lymphoma.J Clin Oncol.2018;37(3):190-201. doi: 10.1200/JCO.18.01583.
  2. Ennishi D, Mottok A, Ben-Neriah S, et al. Genetic profiling of MYC and BCL2 in diffuse large B-cell lymphoma determines cell-of-origin-specific clinical impact.Blood.2017;129(20):2760-2770. doi: 10.1182/blood-2016-11-747022.