Key Points• Children with AL given haplo-HSCT after ab T-and B-cell depletion are exposed to a low risk of acute and chronic GVHD and NRM.• The leukemia-free, GVHDfree survival of patients given this type of allograft is comparable to that of HLAmatched donor HSCT recipients.Allogeneic hematopoietic stem cell transplantation (HSCT) from an HLA-haploidentical relative (haplo-HSCT) is a suitable option for children with acute leukemia (AL) either relapsed or at high-risk of treatment failure. We developed a novel method of graft manipulation based on negative depletion of ab T and B cells and conducted a prospective trial evaluating the outcome of children with AL transplanted with this approach. Eighty AL children, transplanted between September 2011 and September 2014, were enrolled in the trial. All children were given a fully myeloablative preparative regimen. Anti-T-lymphocyte globulin from day 25 to 23 was used for preventing graft rejection and graft-versus-host disease (GVHD); no patient received any posttransplantation GVHD prophylaxis. Two children experienced primary graft failure. The cumulative incidence of skin-only, grade 1-2 acute GVHD was 30%; no patient developed extensive chronic GVHD. Four patients died, the cumulative incidence of nonrelapse mortality being 5%, whereas 19 relapsed, resulting in a 24% cumulative incidence of relapse. With a median follow-up of 46 months for surviving patients, the 5-year probability of chronic GVHD-free, relapsefree survival (GRFS) is 71%. Total body irradiation-containing preparative regimen was the only variable favorably influencing relapse incidence and GRFS. The outcomes of these 80 patients are comparable to those of 41 and 51 children given transplantation from an HLA-identical sibling or a 10/10 allelic-matched unrelated donor in the same period. These data indicate that haplo-HSCT after ab T-and B-cell depletion represents a competitive alternative for children with AL in need of urgent allograft. This trial was registered at www.clinicaltrials.gov as #NCT01810120. (Blood. 2017;130(5):677-685)

Abstract Traditionally, hematopoietic stem cell transplantation (HSCT) from both HLA-matched related and unrelated donors (UD) has been used for treating children with acute leukemia (AL) in need of an allograft. Recently, HLA-haploidentical HSCT after αβ T-cell/B-cell depletion (αβhaplo-HSCT) was shown to be effective in single-center studies. Here, we report the first multicenter retrospective analysis of 127 matched UD (MUD), 118 mismatched UD (MMUD), and 98 αβhaplo-HSCT recipients, transplanted between 2010 and 2015, in 13 Italian centers. All these AL children were transplanted in morphological remission after a myeloablative conditioning regimen. Graft failure occurred in 2% each of UD-HSCT and αβhaplo-HSCT groups. In MUD vs MMUD-HSCT recipients, the cumulative incidence of grade II to IV and grade III to IV acute graft-versus-host disease (GVHD) was 35% vs 44% and 6% vs 18%, respectively, compared with 16% and 0% in αβhaplo-HSCT recipients (P < .001). Children treated with αβhaplo-HSCT also had a significantly lower incidence of overall and extensive chronic GVHD (P < .01). Eight (6%) MUD, 32 (28%) MMUD, and 9 (9%) αβhaplo-HSCT patients died of transplant-related complications. With a median follow-up of 3.3 years, the 5-year probability of leukemia-free survival in the 3 groups was 67%, 55%, and 62%, respectively. In the 3 groups, chronic GVHD-free/relapse-free (GRFS) probability of survival was 61%, 34%, and 58%, respectively (P < .001). When compared with patients given MMUD-HSCT, αβhaplo-HSCT recipients had a lower cumulative incidence of nonrelapse mortality and a better GRFS (P < .001). These data indicate that αβhaplo-HSCT is a suitable therapeutic option for children with AL in need of transplantation, especially when an allele-matched UD is not available.

Microenvironmental factors contribute to the immune dysfunction characterizing acute myeloid leukemia (AML). Indoleamine 2,3-dioxygenase 1 (IDO1) is an interferon (IFN)-γ-inducible enzyme that degrades tryptophan into kynurenine, which, in turn, inhibits effector T cells and promotes regulatory T-cell (Treg) differentiation. It is presently unknown whether childhood AML cells express IDO1 and whether IDO1 activity correlates with patient outcome.We investigated IDO1 expression and function in 37 children with newly diagnosed AML other than acute promyelocytic leukemia. Blast cells were cultured with exogenous IFN-γ for 24 hours, followed by the measurement of kynurenine production and tryptophan consumption. No constitutive expression of IDO1 protein was detected in blast cells from the 37 AML samples herein tested. Conversely, 19 out of 37 (51%) AML samples up-regulated functional IDO1 protein in response to IFN-γ. The inability to express IDO1 by the remaining 18 AML samples was not apparently due to a defective IFN-γ signaling circuitry, as suggested by the measurement of signal transducer and activator of transcription 3 (STAT3) phosphorylation. Co-immunoprecipitation assays indicated the occurrence of physical interactions between STAT3 and IDO1 in AML blasts. In line with this finding, STAT3 inhibitors abrogated IDO1 function in AML blasts. Interestingly, levels of IFN-γ were significantly higher in the bone marrow fluid of IDO-expressing compared with IDO-nonexpressing AMLs. In mixed tumor lymphocyte cultures (MTLC), IDO-expressing AML blasts blunted the ability of allogeneic naïve T cells to produce IFN-γ and promoted Treg differentiation. From a clinical perspective, the 8-year event-free survival was significantly worse in IDO-expressing children (16.4%, SE 9.8) as compared with IDO-nonexpressing ones (48.0%, SE 12.1; p=0.035).These data indicate that IDO1 expression by leukemia blasts negatively affects the prognosis of childhood AML. Moreover, they speak in favor of the hypothesis that IDO can be targeted, in adjunct to current chemotherapy approaches, to improve the clinical outcome of children with AML.