Long-term outcome of high-dose sequential chemotherapy with autografting (i-HDS) in follicular lymphoma at diagnosis: An update of the prospective multicenter consecutive trial of the "Gruppo Italiano Trapianto di Midollo Osseo" (GITMO).

Introduction. i-HDS is a promising though experimental treatment for FL patients at diagnosis. We have published that i-HDS is feasible also in the context of a multicenter trial with limited toxicity and promising results (Ladetto et al, Blood, 2002). This analysis is an update at 42 months from the previous closing date. Particular attention has been devoted to late toxicities and outcome according to molecular remission (MR). Patients and methods. 92 untreated patients aged 18–60 years with stage III-IV FL requiring treatment were enrolled by 20 Italian centers between 1997 and 1999 and evaluated on an intention-to-treat basis. Inclusion criteria have been previosly described. Clinical features at diagnosis were: Ann Arbor stage IV: 84%; BM involvement: 80%; extranodal and extra-BM disease: 55%; bulky mass: 51%; elevated serum LDH: 37%; "B" symptoms: 30%; leukemic disease: 12%; aaIPI 2: 37%. The i-HDS schedule included 2APO, 2DHAP, Etoposide 2g/sqm, Methotrexate 8g/sqm and Cyclophosphamide 7g/sqm with PBPC collection. The myeloablative regimen was Mitoxantrone 60mg/sqm + Melphalan 180mg/sqm followed by 5–8x106 CD34+ cells/kg. Minimal residual disease analysis with the bcl-2/IgH rearrangement was available in 47% of patients. Results: 87% of patients completed the planned treatment. CR rate was 88%. Currently the median follow-up is 62 months. Late toxic effects included five myelodysplastic syndromes and /or secondary leukemias (MDS/2AL) all occurring within two years from the end of treatment. In 4 of 5 cases MDS/2AML occurred in subjects already re-treated for disease relapse. Two fatal solid tumors (one gastric cancer and one lung cancer) were also recorded. Long-term hematopoietic fuction was satisfactory in all patients not developing MDS/2AML. Heart failure occurred in three patients, always manageable with oral medical treatment. The projected overall survival and progression-free survival (PFS) at 7,5 years are 74% and 56%. Notably, following i-HDS, patients with aaIPI2 had an outcome comparable to those with aaIPI 1 (p=NS). Attainement of MR in the first year following i-HDS was highly predictive even for the long term outcome. Among patients achieving MR the relapse rate (RR) was 13% while in those failing to achieve MR was 81% (p<0001). Conclusions. This long-term analysis indicates the following: a) i-HDS allows long-term PFS in 56% of advanced FL patients regardless of the aaIPI. This result compares favorably with those achieved with Rituximab-free non-intensified therapy at least for high-risk patients; b) MR achievement is a powerful indicator for prolonged PFS; c) although severe side effects occurred in this series, relapse remains the major cause of treatment failure. Thus, treatment implementation is required. Rituximab inclusion is probably one of the most feasible approaches, suitable for multicenter trials. The new GITMO trial, comparing CHOP vs i-HDS (both supplemented with Rituximab), currently ongoing for high-risk (aaIPI2) FL patients, will help clarifying wheter dose-intensification has still a role in the monoclonal antibody era.