Activated B-Cell (ABC) Profile Does Not Have An Inferior Outcome Compared to Germinal Center B-Cell (GCB) Profile Determined by Immunohistochemistry Tissue Microarray in Poor Prognosis Diffuse Large B-Cell Lymphoma (DLBCL) Treated with Rituximab and Dose- Dense and High-Dose Chemotherapy (HDC) Plus Autologous Stem Cell Transplantation (ASCT).

Introduction: A specific molecular signature divides DLBCL into GCB and ABC profiles; in pre-Rituximab era, GCB group had a better prognosis. The addition of Rituximab (R) to chemotherapy improves the outcome in DLBCL. It remains uncertain the prognostic significance of GCB and ABC subgroups in Rituximab era.

Patients and methods: from August 2000 to September 2006, 120 previously untreated patients <61 years affected by DLBCL, at aa-IPI intermediate/high or high risk, were enrolled into a phase II GIMURELL trial: 4 courses of bi-weekly R-MegaCEOP14 (375 mg/m² R on day 1, 1200 mg/m² CTX + 110 mg/m² EPI + 1.4 mg/m² VCR on day 3 and 40 mg/m² PDN days 3 to 7), 2 R-MAD (8 mg/m² Mitoxantrone + 2 g/m²/12h Cytarabine + 4 mg/ m²/12h Dexamethasone days 1 to 3 and 375 mg/m² R on day 4 and before peripheral blood stem cell harvest as in vivo purging) followed by ASCT with BEAM as conditioning regimen. Thus far, tissue microarray (TMA) blocks were created from 92 cases and they were studied in immunohistochemistry; sections were stained with antibodies to CD20, CD3, CD10, CD138, Ki67, Bcl6, Bcl2, MUM1 and cyclinD1. Patients were classified as GCB or non-GCB (ABC) according to the Hans’ algorithm. The samples were scored positive for CD10, Bcl-6 and MUM-1 if 30% or more of tumor cells were stained. Cases with the coexpression of CD10 and MUM-1 were classified as GC-activated according to Chang and grouped with GCB for analysis purpose. Crude Kaplan-Meier overall (OS) and event-free (EFS) survival curves were estimated both overall and stratified by TMA features. Differences between curves were tested using the 2-tailed log-rank test..

Results: 92 samples were centrally reviewed and studied in immunoistochemistry. Overall 69 were evaluable; 23 were excluded due to inadequate pathological materials in 16 and incorrect inclusion criteria in 7. Thirty-seven patients were classified as GCB, 24 as ABC and 8 as GC-activated. Patient characteristics were well balanced between GCB and ABC subtypes. With a median follow-up of 53 months, OS in all 92 patients was 81% (95% CI: 71% - 88%); at the same timeframe, OS in GCB was 77% vs 86% for ABC (p=.5). With a median follow-up of 46 months, EFS in all 92 patients was 74% (95% CI: 64% - 82%); at that time, EFS in GCB was 65% vs 87% for ABC (p=.2).

Conclusions: Rituximab as adjuvant to dose-dense and HDC with ASCT support is effective in high risk DLBCL in both GCB and ABC subgroups. In this R-HDC+ASCT study ABC profile as determined by TMA does not represent a poor prognostic feature. However 25% of patients still experienced progression. Further and more detailed biological analysis are needed to better identify these patients.