Molecular profiles have been shown to correlate with survival in patients with diffuse large B-cell lymphoma (DLBCL) receiving conventional chemotherapy. However, most patients now receive monoclonal anti-CD20 antibodies in addition to chemotherapy, which might erase the poor prognosis associated with some markers (

Mounier et al, Blood 2003;101:4279
) and reveal new prognostic factors related to rituximab mechanisms of action. This prompted us to undertake a gene expression study on the samples of patients (pts) included in the GELA LNH98-5 study that compared CHOP and rituximab-CHOP (R-CHOP) regimen. We collected all available frozen tissue samples from patients included into the LNH98-5 trial and completed the series with patients treated with the same regimens in participating centers during the same period. All those patients were between 60 and 80 years old, had histologically reviewed DLBCL, Ann Arbor stage II to IV disease and an ECOG Performance Status ≤3. High quality RNA from 53 patient (43 from the study itself and 10 from the additional cohort) treated with CHOP (30 pts) or R-CHOP (23 pts) were analyzed with Affymetrix HU133A micro-arrays.

The main clinical characteristics of this subset of patients were similar to that of the remaining group of 346 patients involved in the LNH98-5 study. Two parameters were strongly correlated with disease progression: a high International Prognostic Index (IPI = 4 or 5) (relative risk (RR) 3.9 when compared to pts with an IPI = 2 or 3, p=0.002) and treatment regimen (CHOP: RR 4.2 when compared to R-CHOP, p=0.001). In both treatment arms, patients with a germinal center (GC) DLBCL profile had a better outcome than patients with an activated B-cell (ABC) one (ABC versus GC, RR of progression 2.6, p=0.06) and several transcripts previously associated with prognosis, such as BCL6, showed a strong correlation with survival (BCL6 low versus BCL6 high, RR of progression 3.1, p=0.004), further corroborating the clinical relevance of this series of patients. We then used Cox models adjusted for IPI and treatment regimen to test the association with outcome of each Affymetrix microarray probeset 1) in the entire population, 2) or with a differential effect according to the treatment regimen.This strategy allowed us to identify several transcripts that were highly correlated with survival, either in both groups or specifically in the R-CHOP group. Interestingly, some of these transcripts are related to apoptosis regulation, immune cell infiltration or B-cell signal transduction. We are currently investigating the prognostic significance of this series of transcripts, using Quantitative RT-PCR, in an independent set of 43 DLBCL patients who presented the same clinical characteristics than the initial test group and received R-CHOP in GELA centers after the closure of the LNH98-5 trial.

Topics:
chemotherapy regimen, diffuse large b-cell lymphoma, gene expression profiling, rituximab, r-chop, brachial plexus neuritis, cyclophosphamide, doxorubicin, prednisone, and vincristine, antibodies, cd20 antigens, disease progression

Author notes

Corresponding author

2005, The American Society of Hematology
2005
Sign in via your Institution