Allogeneic stem cell transplantation (allo-SCT) represents a curative option for intermediate- and high-risk acute leukemias (AL). The number of unmanipulated haploidentical allo-SCT (haplo-SCT) is increasingly used with favorable outcomes. Incidence of graft-versus-host disease (GvHD) in haploidentical bone marrow (BM) transplants using post-transplant cyclophosphamide (PTCy) is low, counterbalanced by an excess in disease recurrence; and acute GvHD using mobilized peripheral blood stem cells (PBSC) ranges between 30% and 40%. The ultimate choice of graft source depends on the design of the full transplantation package based on transplantation center experience.

We conducted a retrospective analysis of 32 patients (59% male), who received an haplo-PTCy with double source of stem cells, G-CSF primed bone marrow plus G-CSF-mobilized PBSC, for high-risk or advanced acute leukemia in two Brazilian centers, Hospital Santa Marcelina (n=23) and Hospital São Paulo (n=9), from 2013 to 2019. The median age patients were 27.5 years (range 17-60 years). Median disease time before haplo-PTCy was 8.9 months (3.6-108). There were 13 acute myeloid leukemia (AML), 17 acute lymphoblastic leukemia (ALL), one mixed phenotype acute leukemia and one dendritic cell leukemia. 6/17 ALL were Ph1 positive. 34% of the patients received 2 treatment protocols to achieve CR and 12.5% had submitted to more than two treatments. So, at the time of transplant 75% (n=24) was in first CR (CR1) although one (3%) patient was minimal residual disease (MRD) positive and six (18.7%) there were no MRD available. The others patients (n=8, 25%) were on second or third CR. The HCT-CI comorbidities was ≥ 3 in two patients, and there were 15 patients (46%) with carbapenem-resistant gram-negative bacilli (CRGNB) colonization before transplant. Panel reactive antibody was positive in two patients. The donor was a sibling in 68.7% (n=22), father, mother and child in two (6%), three (9%) and five (15%) patients, respectively. The conditioning was reduced intensity (RIC) in 87.5% (n=28) patients, with fludarabine, cyclophosphamide and total body irradiation (TBI) 200 cGy. After conditioning, patients received G-CSF primed bone marrow grafts in combination with PBSC no ex-vivo T cells depleted, and cyclophosphamide 50 mg/Kg/day IV on days +3 and +4 post-transplant, as well as GVHD prophylaxis.

Six (18.7%) patients died for sepsis before 60 days (10 to 58 days), all had had CRGNB before transplant, four those ones with no grafting, died from days +10 to +19. Acute GvHD grade III-IV was observed in two patients, who died at +48 and +95 days. High mortality related to transplant (TRM) was observed considering all patients. CRGNB was a determining factor in these early deaths. If we excluded all CRGNB patients of this study the mortality could be 11% (2 patients with GVHD in 17 patients transplanted). Four patients (16%) has severe cGvHD. Nine patients (37%) relapsed in two years. Two years OS and DFS were 40% and 37.5%, respectively.

In conclusion, with a median follow-up of 2 years, haplo-PTCy with double source leads to 40% overall survival in 32 patients with high-risk advanced acute leukemia, with 16% (n=2) being in treatment for cGvHD. The causes of death were relapse 43% (n=7), early sepsis in patients with CRGNB colonization 37.5% (n=6), grade IV acute GvHD in 12,5% (n=2), and one patient died for pneumonia community at D+285. Our data suggests that haplo-PTCY double source is a feasible option in these cases. However, CRGNB colonization in aggressive disease is the main factor that should be considered as exclusion for haplo-SCT in development countries.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution