Based on molecular studies, DLBCL is now further divided in three subtypes with distinct pathogenesis and clinical outcomes. Fluorescence in situ hybridization (FISH) studies identified a subgroup of DLBCL with a poor clinical outcome harboring concurrent gene rearrangements of the MYC, BCL2 and/or BCL6 proto-oncogenes leading to the over-expression of c-Myc, Bcl-2 and Bcl-6, inferior response rates to rituximab-based chemotherapy, and a shorter progression-free survival (PFS)/overall survival (OS). This group of patients is now categorized as double-hit (DHL) or triple-hit (THL) DLBCL. Immunohistochemistry (IHC) studies suggested that DLBCL patients with over-expression of Bcl-2 and c-Myc proteins exhibit a similar clinical course than those patients with DHL/THL. Alternative transcriptional (i.e. gene amplification or chromosomal gain) or post-translational regulatory mechanisms (yet to be defined) may be responsible for the over-expression of c-Myc, Bcl-2, or Bcl-6 in some patients with DHL/THL phenotype. The appropriate therapy for DHL/THL remains to be defined, but retrospective studies had demonstrated that at standard doses of R+CHOP are suboptimal. In order to study the impact of more aggressive therapeutic approaches in the management of DHL/THL, we retrospectively evaluated our single institution outcome experience over the last 14 years. Using the lymphoma translational database that includes 611 DLBCL patients, we identified 24 patients (M=13/F=11) (4%) with FISH-confirmed DHL/THL DLBCL. Demographic characteristics, clinical data, treatment history in the front line (including the use of CNS prophylaxis, high-dose chemotherapy and autologous stem cell support [HDC-ASCS] or allogeneic bone marrow transplant [allo-BMT]) were collected. In addition, response rate, PFS and OS were calculated. Since the first case-reports were published, we observed a steady increase in the number of DHL/THL patients at our Institution (7 vs. 18 cases before or after 2011) which represents an increasing medical awareness of this new clinical entity. The mean age at diagnosis was 62 years (25 to 85 years), most of them Caucasians (N=22). The median Ki67 proliferation index was 90%. Using the Han’s algorithm, 11 of the DHL/THL were categorized as germinal center B-cell like (GCB), 6 patients as non-GCB, and 7 patients could not be classified. By FISH studies, 58% were DHL (involvement of MYC and BCL2 [N=22] or BCL6 [N=10]) and 42% THL (involvement of MYC, BCL2 and BCL6, N=10). Gene rearrangements involving MYC, BCL2, or BCL6 were observed in 18 patients and MYC, BCL2 or BCL6 gain in 6. Clinically, 95% of the patients presented with stage III/IV, 67% with High-intermediate/High IPI score, and 79% of the patients had extranodal disease. Front-line chemo-immunotherapy included 1) standard doxorubicin-containing regimens: R+CHOP (N=10) or R-EPOCH (N=7), or 2) intensified regimens: R-DHAC (N=2) and R+HyperCVAD/R+HDMTXCytarabine (N=4). Prophylaxis IT chemotherapy was administered to 16 (67%) patients. The complete remission (CR) rate was 62.5% and 29% of the patients underwent HDC-ASCS (N=4)/allo-BMT (N=2) in first (N=2) or second-remission (N=4). In this group of patients, the use of intensified regimens was associated with a non-statistically significant improvement is OS when compared to R+CHOP/R+DA-EPOCH (OS at 3 years of 85.7% vs. 58%). Similarly, an improvement in OS was observed among patients receiving HDC-ASCT/allo-BMT as consolidation (71% vs. 60%, P=0.27). Of interest, CNS prophylaxis was associated with an improvement in OS at 3 years (81% vs. 37%, P=0.032). No differences in clinical outcomes were observed between DHL/THL harboring gene-rearrangements vs. gene gain. In summary, our single institution experience suggests that DHL/THL are increasingly being recognized as more aggressive sub-types of DLBCL with a dismal outcome with conventional therapeutic approaches. A high-index of suspicion should be raised and testing for DHL/THL in those DLBCL presenting with advance stage, multiple extra-nodal site of disease, and an elevated Ki67 index should be done. The accurate identification of DHL/THL patients is necessary, as they appear to benefit from rituximab-based intensified cytarabine-containing regimens, CNS prophylaxis and HDC-ASCT/allo-BMT consolidation.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution