Outcomes of Patients with Acute Lymphoblastic Leukemia (ALL) with Central Nervous System (CNS) Involvement Undergoing Allogeneic Hematopoietic Cell Transplantation (AlloHCT): A Single Center Retrospective Analysis

Background:

Central nervous system (CNS) is the most common extramedullary site involved in acute lymphoblastic leukemia (ALL) at diagnosis and at the time of relapse. Although some studies have shown an adverse impact of CNS involvement on ALL outcomes when treated with chemoradiotherapy, limited data is available on the prognostic implication of CNS involvement in patients who undergo AlloHCT. The objective of our study was to analyze characteristics and outcomes of patients with ALL and history of CNS involvement who had undergone AlloHCT.

Methods:

We reviewed all consecutive patients with ALL who had undergone AlloHCT between October 2004 and December 2013 at City of Hope and identified cases with history of CNS involvement either at diagnosis or CNS relapse [isolated or combined CNS and bone marrow (BM)] prior to AlloHCT. Patient characteristics and outcomes were collected and analyzed.

Results:

Among 436 patients transplanted during the selected period, 56 (13%) had history of CNS involvement prior to AlloHCT. Of those patients with prior CNS involvement, 11 (20%) had CNS disease at presentation, 25 (44%) had isolated CNS relapse, and 20 (36%) had combined CNS/BM relapse. Median age for eligible patients was 25.5 (range=2.4-61.4). Poor risk cytogenetics were identified in 22 (40%) of patients.  Donor was sibling in 24 patients (43%) and graft source was peripheral blood in 34 (61%) patients. Disease status at transplant was CR1 in 11 patients (20%), CR2 in 31 patients (55%), and advanced/refractory disease in 14 patients (25%). The conditioning regimen was myeloablative in 42 (75%), and TBI-based in 40 (71%) patients.

With overall median follow up of 13.5 months (range=0.3-112.8), 1- and 2-year EFS were 49.8% and 42.3%, while 1- and 2-year OS were 57 and 45%, respectively. Day 100 and 1-year non-relapse mortality were 10.7% and 23%, and 1- and 2-year relapse rate were 27% and 30%, respectively. Only 6 patients (10%) relapsed post AlloHCT with CNS disease. Patients with CNS disease at presentation as well as combined CNS/BM relapse had inferior outcomes compared to those with isolated CNS relapse prior to transplant (2-year OS 32.7 and 33.7 vs. 72%, respectively. P=0.0123). In univariate analysis, among all factors examined, age was the only significant factor for OS [pediatric vs. adult: HR= 2.44 (CI: 1.06-5.62); p = 0.04]. In multivariate analysis, age was independent factor for both OS [pediatric vs. adult: HR = 2.67 (CI: 1.03-6.95), p = 0.04] and relapse [pediatric vs. adult: HR = 4.57 (CI: 1.04-20.07), p = 0.05]. WBC on presentation was independent factor for only relapse [more vs. less than 30K: HR = 3.53 (CI: 1.02-12.21), p = 0.04].

Conclusions:

AlloHCT is effective in treating ALL with CNS involvement and post-transplant CNS relapse is uncommon. Patients who undergo AlloHCT with isolated CNS relapse have superior outcome compared to those with combined CNS and BM relapse or CNS involvement at diagnosis.