Survival Impact of Time Repeated bcl2/IgH Rearrangement Measurements in Follicular Lymphoma Patients Treated with Front-Line Autotransplantation in the GELF-94 Trial by the GELA.

Abstract In the autotranplantation framework, an undetectable MRD before ASCT seems to protect for disease relapse and to increase survival. Using data from the GELF94 trial, which randomized consolidation treatment after achieving clinical response between front line ASCT and chemotherapy, we aimed to assess the impact of MRD on survival controlling for consolidation treatment and time repeated measurement. Of 401 patients included from 07/94 to 03/01, 209 received 12 cycles of CHVP (cyclophosphamide, doxorubicin, teniposide, and prednisone) plus interferon alfa in 18 months (CHVP-I arm) and 192 received 4 cycles of CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) then high-dose therapy with total body irradiation and ASCT (CHOP-HDT arm). Bcl2/IgH rearrangement was prospectively assayed by PCR in bone marrow (BM) and in blood at diagnosis and repeated every 6 months during the first year then annually. Time repeated measurement was stopped at clinical relapse or instigation of a new treatment. At diagnosis, 225 patients had material available for bcl2/IgH rearrangement analysis: BM for 182 (45 %) patients and blood for 199 (50%). In the latter, 105 of the 199 patients (53%) were found to have a bcl2/IgH rearrangement in the MBR breakpoint, whereas bcl2/IgH rearrangement in mcr was observed in 5 patients (3%) and no rearrangement at MBR or mcr in the remaining 94 patients (44%). No differences were found according to bcl2/IgH rearrangement in terms of response rate (82% for bc2- vs 80% for bcl2+) and.5 yrs survival (85% for bcl2- and 79% for bcl2+). Time repeated bcl2/IgH rearrangement measurements were available for 142 patients (ASCT n=75, chemo n=67): in BM for 79 patients and in blood for 85 patients. The clinical characteristics were well balanced between patients with/without time repeated measurement. By time dependant Cox’s model, the significant prognostic factors for survival were age below 40 yrs (RR= 21, p=0,005), complete clinical response (RR=5, p=0.02) and bcl2/IgH rearrangement negativity (RR=4, p=0.03). There was no treatment impact. These findings confirm the importance of molecular response in addition to the clinical response as a critical factor for prognosis. The results of this study suggest that in FL patients, the eradication of cells bearing the Bcl2/IgH rearrangement is highly desirable early in the treatment and should be maintained as long as possible.