Search for a command to run...
Abstract Background: The course of follicular lymphoma (FL) is characterized by iterative relapses. Several reports have demonstrated that intensive chemotherapy followed by autologous stem cell transplantation (auto-SCT) is an attractive treatment. Because a subgroup of patients obtains prolonged complete remission after auto-SCT, some authors have postulated that auto-SCT may cure FL patients. In order to address this issue and to characterize long-term responders, we retrospectively studied the outcome of 80 FL patients who underwent auto-SCT in our institution. Patients characteristics: Between 1989 and 2004, 80 FL patients underwent auto-SCT in the Haematology Department of Nantes Medical University, France. All patients had a FL according to the WHO classification. Median age at diagnosis was 45.7 (range 27–62) and median age at the time of transplantation was 48.1 (range 27–69). Auto-SCT was performed upfront in 36 cases, at first relapse in 36 cases and at a subsequent relapse in 8 cases. Median time from diagnosis to auto-SCT was 53 months. At diagnosis, 25 patients presented with a FLIPI <2 and ≥ 2 in 34 cases (data missing =21). Prior to auto-SCT, 71 patients received a CHOP or CHOP-like regimen and 24 patients received a cytarabine-containing regimen. The conditioning regimens prior auto-SCT were TBI-cyclophosphamide (63 patients) or BEAM (17 patients). The source of stem cell was either bone marrow (19 patients) or peripheral blood stem cell (59 patients); 2 patients received both. In note, the graft was purged using a CD34+ selection in 30 cases or a B-depletion in 19 cases. In the course of their disease (prior or after transplantation), 35 patients received Rituximab. Results: Median follow-up (calculated from the time of auto-SCT) was 6.2 years (range 0.2–16.6) (data missing =1). No patient died during the auto-SCT procedure. After auto-SCT, 71 patients reached CR/CRu. Thirty-five patients relapsed and 10 patients experienced histological transformation (HT)(including 8 cases with HT at first relapse). Myelodysplasia/AML occurred in 5 cases. At the time of the analysis, 68 patients were still alive. Causes of death were: HT in 6 cases, AML in 3 cases, follicular lymphoma related in 2 cases and not related to lymphoma in one case. The 3-, 5- and 7-year progression free survival (PFS) estimates were 74.7%, 66.5% and 58.6%, respectively. In note, a plateau occurred on the PFS Kaplan-Meier curve after 8 years. The 3-, 5- and 7-year overall survival (OS) estimates were 92.9%, 86% and 79.7%, respectively. In univariate analysis, five variables (age at the time of auto-SCT < 50y, a purged graft, a Rituximab-containing therapy, an extra-nodal excluding bone marrow involvement at diagnosis and FLIPI <2 ) were found to be statistically significant at a p<0.15 level and were included in the multivariate analysis. In multivariate analysis, the PFS was significantly better for patients with a FLIPI<2 (p=0.043) and patients<50 years at the time of auto-SCT (p=0.03). A trend for a better PFS was observed for patients who received Rituximab in the course of the disease (p= 0.084). Conclusion: By multivariate analysis, we demonstrate that the FLIPI is a strong predictive index to predict PFS after auto-SCT. Moreover, our results confirm the major interest of auto-SCT in the treatment of FL and highlight that the opportunity to perform auto-SCT should not be jeopardize during the course of the disease. Finally, the plateau on the PFS curve suggests that some FL patients could be cured by auto-STC.