Autologous and Allogeneic Stem Cell Transplantation for T-Cell Lymphoma: The M.D. Anderson Cancer Center Experience,

Abstract

Abstract Abstract 4118 BACKGROUND: Despite the advent of novel agents, outcomes of T cell Lymphoma (TCL) with conventional chemotherapy are poor. We have previously reported (Rodriguez J, JCO 2001) encouraging results with autologous stem cell transplantation (ASCT). Others have advocated allogeneic transplants (Allo-SCT). Herein, we report our long-term experience in 196 TCL patients (pts) who had either ASCT (n=119) or Allo-SCT (n=77). METHOD and PATIENTS: This is a retrospective analysis of TCL pts treated with SCT at our institution between 1986 and 2009. We adopted WHO 2008 classification for TCL and divided pts accordingly into Nodal TCL (N-TCL), extra nodal TCL (EN-TCL), and T cell lymphoblastic Lymphoma (T-LBL). Sixty one pts (31%) had peripheral t-cell not otherwise specified (PTCL), 50 (26%) had anaplastic large cell (ALCL), 19 (10%) angioimmunoblatic (AITL), 34 (17%) EN-TCL, and 32 (16%) had T-LBL. ASCT and Allo-SCT pts were balanced for Ann-Arbor stage, gender distribution and time from initial diagnosis. However, Allo-SCT pts were younger (median age 41 vs. 49 yrs, respectively, (p=0.002), more heavily pretreated (p=0.01), were more likely to have extranodal (p<0.001) or bone marrow involvement (p=0.001), and refractory disease (p=0.01) at the time of transplant. RESULTS: A. [ASCT]: Most pts (82%)received BEAM-like conditioning. With a median follow-up of 39 months (range, 3–208), the OS and PFS rates were 39% and 30%, respectively. Risk factors for OS and PFS were evaluated among the largest histologic subgroups, i.e., PTCL, ALCL and AITL: among those 94 pts who received ASCT, having a transplant > CR1 (p=0.001), Prognostic Index for TCL(PIT) >0 (p=0.009), IPI>1 (p=0.03), refractory disease (p=0.001) were associated with worse OS and PFS (Figure 1). The 3-year OS and PFS for CR1 pts (n+33) were 90% and 70%, respectively. A subset of pts (n= 11) within the CR1 group received induction chemotherapy with Hyper-CVAD before ASCT. The 3-year OS and PFS rates for these pts were both 100%. We were not able to detect any difference in outcomes between the different histologies studied, including ALK(+) and ALK (−) ALCL. B. [Allo-SCT] Most pts (75%) received myeloablative conditioning. There was a trend for better outcomes for transplants > year 2000. With a median follow-up of 65 months (range, 5–235), the OS and PFS rates were 43% and 30%, respectively. OS and PFS rates for EN-TCL(n=18) were 49% and 36%, respectively. Among the N-TCL (n=33) group, unlike ASCT, most pts (87%) were >CR1. The 3-year OS and PFS rates for these pts were 38% and 23%, respectively. Both IPI and PIT >0 were determinants for worse PFS, whereas IPI >0 and marrow involvement were predictors for worse OS. No plateau however was observed in these pts, unlike those with T-LBL, where a clear plateau of 40 % was observed. CONCLUSION: This is the largest single institution analysis to investigate the role of SCT for TCL. Our review demonstrates that ASCT can induce long-term remissions in CR1 pts with N-TCL. Prospective studies comparing ASCT in this setting to other conventional treatments are warranted. Although allo-SCT was found to be effective in T-LBL, its role in other forms of TCL is yet to be determined. Innovative strategies remain needed for pts with relapsed disease. Disclosures: No relevant conflicts of interest to declare.