Incidence and Risk-Factors of Secondary Myelodysplastic Syndrome/Acute Leukemia Following High-Dose Chemotherapy and Autograft: A Long-Term Analysis on 307 Lymphoma Patients.

INTRODUCTION High-dose chemotherapy (HDT) with autologous stem cell transplantation is an effective treatment option for both non-Hodgkin’s Lymphoma (NHL) and Hodgkin’s Lymphoma (HL). Its clinical applicability has been considerably widened by peripheral blood progenitor cells (PBPC). Indeed, HDT with autograft is now the most frequently employed treatment for patients <65 y.o. at relapse and there is also evidence supporting its use as first-line approach in high-risk patients. Secondary myelodysplastic syndrome/acute leukemias (s-MDS/AL) is a well known late toxic effect in long-term survivors following HDT and autograft. However, the reported actuarial incidence rates are variable, ranging from as low as 1% up to 24% and risk factors associated with the occurrence of sMDS/AL have not been fully established. Aim of this study was to evaluate frequency, actuarial projection, and risk factors of sMDS/AL in a large and homogeneous series of lymphoma patients who received HDT and autograft PATIENTS AND METHODS The study has evaluated 307 lymphoma patients treated at our Institution, between 1988 and 2003, with high-dose sequential (HDS) chemotherapy and autograft. The series included 38 patients with HL, and 269 with NHL (153 high-grade and 116 low-grade NHL). Median age was 46 yrs., 180 patients were male. Overall, 207 patients received HDS as first-line therapy, and 100 patients as salvage treatment following one or more lines of conventional chemo-radiotherapy. Among 307 patients entering the HDS program, 240 completed the whole protocol with the final autograft. Most patients were autografted with PBPC and only a few received either BM cells alone or BM cells + PBPC. RESULTS At a median follow-up of 5.5 yrs., 134 (65%) of 207 patients receiving front-line HDS and 44 of 100 treated for refractory/recurrent lymphoma are alive. s-MDS/AL occurred in 14 (4.5%) patients (10 after PBPC and 3 after BM autograft, 1 after HDS without the autograft procedure). The cumulative probability of developing sMDS/AL is 4.8% at 5 yrs, with a median time to sMDS/AL occurrence of 45 mos. since autograft. Refractory/relapsed status at HDS was the only factor associated with the development of sMDS/AL (p=0.014 vs. HDS first-line). None of the other clinical characteristics, including age, sex, histology, type of graft and number of CD34+ re-infused appeared to be of relevance. The association between disease status at HDS and sMDS/AL development was still observed if the analysis was limited to 240 autografted patients. CONCLUSIONS Overall, the incidence of sMDS/AL is one of the lowest so far reported in lymphoma patients treated with HDT and ASCT. Thus, the use of single agents at high doses does not increase the risk of sMDS/AL. In addition, the study shows a strong association between sMDS/AL and the use of HDT as salvage therapy. This supports an early use of HDT and autograft in those high-risk lymphoma patients with low chances of cure if treated with a conventional approach.