Prognostic Analyses for Adult Relapsed / Refractory Acute Lymphoblastic Leukemia Patients Who Received First Allogenic Hematopoietic Stem Cell Transplantation

Background

Patients with acute lymphoblastic leukemia (ALL) refractory to chemotherapy have poor prognoses. The role and the appropriate protocol of allogenic hematopoietic stem cell transplantation (allo-HSCT) on these patients have not yet been completely established. Therefore, we performed analyses using the Transplant Registry Unified Management Program (TRUMP) database in Japan for adult ALL patients who received allo-HSCT at non-CR status.

Methods

Patients of ages 16 years or older diagnosed with ALL who received their first allo-HSCT between 2008 and 2018 were included. The median follow-up of survivors was 44.6 months. We first compared their pre-HSCT characteristic and post-HSCT outcomes between CR and non-CR patients. In addition, we performed multivariate analysis of pretransplantation variables and developed a predictive risk stratification system for survival within this cohort.

Results In total, 4430 patients (CR at allo-HSCT, 3627; non-CR, 803) with median age of 42 years (range, 16-76) were included in the study. The 2-year overall survival (OS) after allo-HSCT was 29 % with non-CR patients, while 72 % for CR patients (p < 0.01). The 2-year event free survival (EFS) was 27% with non-CR patients, while 69% for CR patients (p < 0.01). The 2-year non relapse mortality (NRM) was 23% with non-CR patients, while 17% for CR patients (p < 0.01). Non-CR patients were featured with younger age (p < 0.01), worse performance status (p < 0.01), higher hematopoietic cell transplantation-specific comorbidity index (HCT-CI) (p < 0.01), lower ratio of existence of Philadelphia chromosome (p < 0.01), higher ratio of cord blood transplantation (p < 0.01), and longer period from diagnosis to transplantation (p < 0.01) compared with CR patients. Multivariate analysis associated with shorter OS for non-CR revealed poor PS (2-4; hazard ratio (HR), 1.90; p < 0.01), high HCT-CI (3-; HR, 1.40; p < 0.01), higher marrow blasts at HSCT (10-49%; HR, 1.70; p<0.01, and 50% or more; HR, 1.94; p<0.01), and older patients (50 years or older; HR, 1.42; p < 0.01). On the other hand, longer OS for non-CR was observed in the sub-cohort of primary induction failure (HR, 0.60; p < 0.01), and longer period from diagnose to HSCT (1 year-; HR, 0.78; p=0.02). Regarding post-HSCT factors, GVHD, both acute and chronic, had positive impacts on prognosis (aGVHD, HR, 0.79; p < 0.01, and cGVHD, HR, 0.78; p < 0.04) when treated as time-dependent covariates in the non-CR cohort.According to the results of multivariate analysis, a prognostic scoring system was established. Patients with good prognosis group featured by younger age, good PS, low HCT-CI, T-lineage, non-Ph karyotype, less than 10 % bone marrow blasts, primary induction failure, myeloablative conditioning, and HLA-matched donor had 55.9% 2-year OS, whereas patients with poor prognosis group had 13.9% 2-year-OS. Non-CR patients with less than 10 % bone marrow blasts had favorable outcome irrespective of age, sex, Ph karyotype , disease status, graft type, HLA mismatch, and period from diagnosis to transplantation. But non-CR patients with poor PS, high HCT-CI, T-lineage, and reduced intensity conditioning regimen had poor prognosis even though they had low bone marrow blasts.

Conclusions

Prognosis after allo-HSCT for non-CR ALL patients was not yet satisfactory. However, longer survival can be expected in some subgroups shown in this study such as low blast counts at HSCT These results can suggest the importance of blast control before HSCT even in case of non-CR status. Novel molecular targeting drugs such as Blinatumomab and Inotuzumab ozogamicin can be good candidates for pre-HSCT therapies in the relapse/refractory ALL patients.