New research suggests a gene expression signature can distinguish high-grade diffuse large B-cell lymphomas (DLBCLs) from other germinal center B-cell–like (GCB) DLBCLs.
Researchers identified GCB DLBCL patients with this 104-gene signature who had a “distinct mutational landscape” and inferior treatment outcomes. David W. Scott, MBChB, PhD, of the British Columbia Cancer Research Centre in Vancouver, and his colleagues described these patients in the Journal of Clinical Oncology.
Dr. Scott and his colleagues began their study by analyzing data from 157 patients with de novo GCB DLBCL. Twenty-five of these patients had double- or triple-hit high-grade B-cell lymphoma with BCL2 translocations (HGBL-DH/TH-BCL2).
The researchers identified 104 genes that were the “most significantly differentially expressed between HGBL-DH/TH-BCL2 and other GCB DLBCLs” to create their double-hit gene signature (DHITsig).
The signature divided the patients into two groups — 42 patients (27%) whose tumors were positive for the DHITsig and 115 (73%) whose tumors were negative. Notably, 22 of the 25 HGBL-DH/TH-BCL2 tumors were DHITsig-positive and 3 were negative.
The DHITsig was not associated with clinical variables such as tumor volume, but it was associated with prognosis. Treatment outcomes were inferior in patients who were DHITsig-positive.
The 5-year time to progression rate was 81% in patients who were DHITsig-negative and 57% in those who were positive (P less than .001). The 5-year overall survival rate was 81% and 60%, respectively (P = .001).
The researchers observed similar results in a validation cohort of 262 patients with GCB-DLBCL who received rituximab-based therapy. The 5-year overall survival rate was 76% in patients who were DHITsig-negative and 49% in those who were positive (P less than .001).
Dr. Scott and his colleagues also evaluated the DHITsig in a second validation cohort of 162 patients with GCB DLBCL.
In analyzing data from all three cohorts, the researchers found that mutations in MYC, BCL2, CREBBP, EZH2Y646, DDX3X, TP53, and KMT2D were more frequent in DHITsig-positive patients and mutations in TNFAIP3, KLHL6, NFKBIE, TET2, CD58, and STAT3 were more common in DHITsig-negative patients.
Additional analyses suggested the cell of origin for DHITsig-positive tumors comes from the intermediate zone or dark zone of the germinal center.
Finally, the researchers found they could use a “clinically relevant assay” to detect the DHITsig. They added a 30-gene module to the Lymph3Cx assay to create a NanoString-based assay called DLBCL90.
The team tested DLBCL90 in 171 GCB DLBCL patients. In this group, 26% of patients were DHITsig-positive, 64% were negative, and 10% were indeterminate. The prognostic significance of the signature was maintained with the assay results, according to the researchers.
Dr. Scott and his colleagues also wanted to validate the association between the DHITsig and HGBL-DH/TH-BCL2, so they tested the DLBCL90 assay in two additional groups of patients.
First, the assay was used in 88 patients who had transformed follicular lymphoma with DLBCL morphology. Eleven of the 25 DHITsig-positive tumors and 4 of the 13 DHITsig-indeterminate tumors were HGBL-DH/TH-BCL2. However, none of the 50 DHITsig-negative tumors were HGBL-DH/TH-BCL2.
The researchers then used the DLBCL90 assay on 26 HGBL tumors. Twenty-three of these were DHITsig-positive and 3 were indeterminate.
This research was supported by the Canadian Cancer Society Research Institute and other organizations. The researchers reported relationships with Seattle Genetics, Roche, Janssen, Celgene, and various other companies.