Abstract
Introduction
Relapse, graft-versus-host disease (GvHD) and GvHD-associated mortality are major obstacles to success of transplantation from unrelated and haploidentical donors in children with acute lymphoblastic leukemia. Future directions will focus on optimizing conditioning regimens and enhancing graft-versus-leukemia effect.
Negative depletion of α/β(+) T cells and CD19+ B lymphocytes, which permits to maintain mature donor-derived natural killer cells and γδ(+) T cells in the graft may improve GvHD control, immune reconstitution and prevent the relapse.
Patients and methods:
A total of 67 pediatric patients with acute lymphoblastic leukemia (T-ALL- 26, B-ALL-41, 29 female, 38 male, median age 9,4 years, range 0,15-20) underwent allogeneic HSCT between May 2012 and May 2016. Forty two patients received haploidentical graft, 25 - a graft from matched unrelated donor. Disease status at transplant was CR1 in 19pts., CR2 in 35 pts. and CR>2 13 pts. Transplantation in CR1 was performed according to risk stratification scheme in the current institutional ALL protocol.Fifty patients recieved treosulfan-based myeloablative preparative regimen (74%), TBI-based regimen was used in 17 patients (26%).
Two regimens of GvHD prophylaxis were used: regimen 1 (n=28): ATG (horse, ATGAM) 50mg/kg,
post-grafting immunosuppression consisted of short course Tacro/MTX (n=20) before day +30 or no post-transplant prophylaxis (n=8[ММ1] ); regimen2 (n=39): ATG (rabbit, thymoglobuline) 5mg/kg, rituximab 200mg/m2 (n=30), bortezomib (n=24) and post- transplant bortezomib (n= 19) or Tacro (n=19). All patients with TBI- based regimen received rabbit ATG. TCRαβ+/CD19+ depletion of HSCT with CliniMACS technology was implemented in all cases according to manufacturer's recommendations. The median dose of CD34+ cells in transplant was 10 x106/kg (range 3,9-18,8), TCRα/β - 19x103/kg (range 0,2-300).
Results
Primary engraftment was achieved in 64 of 67 pts. (two patients died before engraftment), the median time to neutrophil and platelet recovery was 13 and 14 days, respectively. All evaluable patients achieved sustained complete donor T cell and myeloid chimerism by day +30. Early (100 day) mortality was pTRM was 7,5% (95% CI: 0,3-17),2-year pTRM - 17% (95%CI: 9-30). The 3 early deaths included bacterial sepsis (n=2) and viral infections (n=1), seven late: viral infection in four pts. (ADV=2, ADV+CMV=1, CMV=1), bacterial sepsis in two pts. and rhinocerebral mucormycosis in 1 pt., all late deaths were associatedwith chronic GvHD and prolonged corticosteroid therapy.
Cumulative incidence of acute graft-versus-host disease (GvHD) grades II - IV and III - IV was 23,9% (95% CI: 16-36), and 7,5% (95% CI: 3-17) respectively. CI of cGvHD was 22,9% (95% CI: 14-36). Regimen 2 was more effective in prevention of aGvHD II-IV: CI at 2 year after HSCT was 12,8% vs 35,7% in regimen 1, p=0,05 and in cGvHD 8,8% vs 35,7%, p=0,028. No correlation between graft composition, donor type in aGvHD and cGvHD was noted
Cumulative incidence of relapse at 2 years was 32% (95%CI: 22-47). Two years pEFS (event=death or relapse) was 49,6% (95%CI: 36-63), 2-year pOS - 50% (95%CI: 40-67). In patients, who received TBI-based conditioning pEFS was 62% (95%CI: 37-86), as compared with treosulfan-based 46,5% (95%CI: 31-62), p=0,65. There was no significant difference in survival and relapse rate according leukemia subtype and donor type. Median time of follow-up for survivors was 2 years (range, 0,3 - 4).
Discussion:
We confirm that the depletion of TCR-alpha/beta and CD19 lymphocytes from the graft ensures high engraftment rate and acceptable transplant-related mortality in pediatric ALL patients. Viral infections and leukemia relapse await further improvement of control. All major outcomes were equivalent between transplantation from unrelated and haploidentical donor.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
