Title: Myeloablative allogeneic stem cell transplantation for lymphoblastic lymphoma in Sweden: A retrospective study
Abstract: The World Health Organization from 2008 classifies B- or T-cell lymphoblastic leukemia/lymphoma as a precursor lymphoid neoplasm [1]. Conventionally, when a lymph node mass is present and 25% or more of the nucleated cells in the bone marrow are lymphoblasts, a diagnosis of lymphoblastic leukemia is preferred to lymphoblastic lymphoma. Data about the patients (≥16 years of age), who had undergone allogeneic stem cell transplantation (ASCT) due to T- and B-cell lymphoblastic lymphoma (LBL) in Sweden, were collected from patient files, local registries, and the The European Group for blood and marrow transplantation (EBMT) data base in all participating centers (Stockholm, Gothenburg, Lund, Linköping, Uppsala, and Umeå). Nineteen patients were transplanted in four Swedish centers (Stockholm, Uppsala, Lund, and Gothenburg) between August 1987 and February 2005. A patient who underwent syngeneic stem cell transplantation and two others who were transplanted with reduced intensity conditioning regimens were excluded from this study. All the patients had a lymph node biopsy-proven histologic diagnosis of a precursor B or T lymphoblastic lymphoma. The cut-off value of 30% for blast cells in the bone marrow was used for distinguishing between leukemia and lymphoma, because the majority of patients were diagnosed before 2004. The study protocol was approved by the Regional Ethical Review Board in Gothenburg and the patients gave their informed consent before transplantation. They were treated according to different chemotherapy protocols (Wollner protocol, NOPHO, HD Ara-C and Mitoxantrone, Riehm, HyperCVAD, DexaBEAM, Swedish National therapy program for adult ALL, and Norwegian national program for ALL). Decisions about the transplantations were made, respectively, at the transplantation centers. Among the patients in complete remission (CR), only one patient was transplanted in CR2. human leurocyte antigen (HLA)-identical related donors were used in 14 patients (73%), HLA-A, B, and Dr locus (DR) identical unrelated donors were used in four patients (23%) and mismatched unrelated in one patient (4%). Other initial patient characteristics are shown in detail in Table I. Probabilities of overall survival and relapse-free survival were estimated using Kaplan–Meier method and compared with the log-rank test [2]. The incidence of graft versus hoct disease (GVHD), transplantation related mortality (TRM), and relapse were estimated using a cumulative incidence curves estimator [3]. All the tests were two sided. Analyses were performed using a cmprsk package (developed by Gray, June 2001), Splus 6.2 software (S-plus 6.2, Insightful, Seattle, WA) and Statistica software (Statistica, StatSoft, Tulsa, OK). Transplantation outcome of the 19 patients is presented in Table I. Acute GVHD Grades II–IV developed in six patients with a cumulative incidence of 32%. The incidence of acute GVHD was higher in patients who received peripheral-blood stem cells (PBSC) than bone marrow (BM) (67% vs. 15%; P = 0.02). Chronic GVHD developed in 4/16 patients with a cumulative incidence of 25% within 2 years. The cumulative incidence of TRM at 1 and 3 years was 32%. The incidence of TRM was higher in the unrelated donor transplants than in the sibling transplants (80% vs. 14%, P = 0.002) and in T- compared with B-LBL (42% vs. 0%; P = 0.06, ns). The cumulative incidence of relapse was 42%. The incidence of relapse was higher in the patients with >1 year between diagnosis and ASCT than in the patients with ASCT <1 year after diagnosis (71% vs. 25%, P = 0.06, ns). The 5-year overall survival (OS) and the progression-free survival (PFS) were 30% and 32%, respectively. The OS and PFS were longer in the patients who were transplanted <1 year after the diagnosis versus patients transplanted >1 year after diagnosis (P = 0.06, ns). There are not many reports of the results of allogeneic stem cell transplantation for lymphoblastic lymphoma in the literature. We report the results of the patients who were transplanted at the time when lymphoblastic lymphoma was treated predominantly as high-grade lymphoma rather than ALL. All the patients were transplanted in chemosensitive disease, most in CR1 and with myeloablative conditioning regimen, preferentially with cyclophosphamide and TBI. The incidence of acute and chronic GVHD does not differ significantly from that of other studies done in similar circumstances. The high incidence of TRM, especially in the matched unrelated donor (MUD) transplants, and the high relapse risk (42%) contribute to low PFS and OS. The results of our study are comparable to previous studies in lymphoblastic lymphoma [4-6]. The PFS and OS were almost identical to the previous report of the Swedish ALL group [7]. However, our study shows better results for the patients transplanted <1 year after the diagnosis compared with those transplanted >1 year after diagnosis. A year after transplant, the survival curve reaches plateau. To analyze whether there was a time related outcome improvement, we divided the patients into two groups: those transplanted between 1987 and 1998 and those transplanted between 1998 and 2005. Surprisingly, the group transplanted between 1987 and 1998 had a significantly superior PFSand OS compared with later transplants (P = 0.01) (Fig. 1). Upon a closer look at the two groups we noticed that most of the patients (91%) transplanted before 1998 were transplanted <1 year after the diagnosis. Only 25% of the patients transplanted after 1998 were transplanted a year or longer after the diagnosis (Table I). The patients transplanted between 1987 and 1998 had received less chemotherapy before the transplantation. Another contributor to the better results of this group may be the fact that more patients were in CR at ASCT and that fewer received PBSC than in the other group. The use of PBSC led to more acute GVHD and higher TRM. Furthermore, among those transplanted between 1998 and 2005, more patients had unrelated donors. In this group, there were significantly more acute GVHD and TRM compared with that of the matched related donors. Although this group is profoundly heterogeneous, this retrospective analysis confirms that a selection of patients and donors plays a crucial role for the outcome. We have shown that myeloablative allogeneic stem cell transplantation for lymphoblastic lymphoma led to a long-term survival in a third of the patients. Finally, when the transplantation was done <1 year after the diagnosis in CR, using matched sibling donors and BM as a stem cell source, with fewer chemotherapy cycles before the transplant, better results for ASCT were achieved. Overall survival after myeloablative conditioning and ASCT in 19 patients with lymphoblastic lymphoma transplanted (1987–1998) and (1999–2005). Vladimir Lj Lazarevic*, Mats Remberger*, Hans Hägglund*, Helene Hallböök*, Gunnar Juliusson*, Eva Kimby*, Claes Malm*, Anders Wahlin*, Hamdy Omar*, Jan-Erik Johansson*, * Department of HematologyUniversity Hospital LundLund University, Lund, Sweden.