Prognostic Impact of Different High Risk Features in Allogeneic Stem Cell Transplantation for Acute Lymphoblastic Leukemia in First Complete Remission.

In acute lymphoblastic leukemia (ALL) specific clinical and biological features confer high relapse risk and inferior overall survival (OS) after treatment with conventional chemotherapy alone. The differential prognostic impact of these high risk features after treatment with allogeneic hematopoietic stem cell transplantation (HCT) has not been well studied.

79 adult ALL patients in first complete remission (CR) received allogeneic HCT at our center between 1995 and 2008. All patients were high or very high risk according to German Multicenter Study Group for Adult ALL (GMALL) criteria. Median age was 36 years (range: 17-68). Patients received high-dose conditioning consisting of 12 Gy total body irradiation ± etoposide ± cyclophosphamide (n=69, 87%) or reduced intensity conditioning (RIC) consisting of fludarabine/busulfan/ATG (n=10, 13%) and HSCT from related (n=34, 43%) or unrelated (n=45, 57%) donors. Bone marrow (n=17, 22%) or peripheral blood stem cells (n=62, 78%) were given. Graft-versus-host-disease prophylaxis was CSA/MTX for high-dose conditioning or CSA/MMF for RIC.

Patients were classified as high risk or very high risk due to Philadelphia chromosome-positive disease (Ph+) (n=30, 38%), leukocytosis>30/nl at diagnosis in B-ALL (n=25, 23%), late response to induction therapy in B-ALL (>week 4) (n=13, 16%), early or mature T-ALL (n=13, 16%), pro-B-ALL/t(4;11) (n=8, 10%), persistence of minimal residual disease (MRD) (>week 16) (n=8, 10%) or complex aberrant karyotype (n=6, 8%). 57 patients (72%) presented with one high risk feature, whereas 20 patients (25%) and 2 patients (3%) presented with two or three features, respectively. Currently, after a median follow-up of 56 months (7-169) 49 patients (62%) remain alive. Projected OS of the whole cohort at 1, 2 and 5 years was 78%, 70% and 55% and leukemia-free survival was 77%, 66% and 55%. Cumulative incidence of non-relapse mortality (NRM) and relapse mortality (RM) at 5 years was 23% and 18%, respectively. In multivariate Cox regression analysis, a non-significant trend for inferior OS was seen for patients with early or mature T-ALL (hazard ratio (HR): 2.03 (95%CI: 0.92-4.52), p=0.082), whereas no differential effect on OS, NRM or RM was seen for any other high risk feature (Table 1). In additional analyses, inferior OS (HR 1.81 (95%CI: 1.02-3.29), p=0.043) and increased RM (HR 2.17 (95%CI 1.16-4.05), p=0.015) was observed for patients with more than one high risk feature.

In summary, this single center study on allogeneic HCT in high risk ALL found a negative prognostic trend for early or mature T cell immunophenotype. No differential prognostic impact on OS, NRM and RM was seen for other high risk features as defined by GMALL criteria, however this conclusion is limited by the low patient number in some of the subgroups. Overall survival for the whole cohort was 55% at 5 years, with inferior OS and higher RM being observed in patients with more than one high risk feature.

No relevant conflicts of interest to declare.