Donor Leukocyte Infusion after Stem Cell Transplantation in Patients with Juvenile Myelomonocytic Leukemia.

Juvenile myelomonocytic leukemia (JMML) is a rare clonal disorder of early childhood. Currently, only allogeneic stem cell transplantation (SCT) offers long-term cure. Relapse remains the major cause of treatment failure. Although graft-versus-leukemia (GVL) effect most likely plays an important role in controlling JMML, the benefit of donor leukocyte infusion (DLI) following SCT in JMML is currently unknown.

Patients and methods: Twenty-one patients with JMML who received DLI after SCT, including 4 patients after given a second SCT, were studied (BMT 14, PBSCT 6, CB 1). The median age at SCT was 15 (8–99) months. A normal karyotype, monosomy 7 or other aberrations were observed in 15, 2, and 4 patients, respectively. Six patients were transplanted from a matched sibling and 15 from an alternative donor. Chimerism analyses were performed by microsatellite PCR system or FISH for sex mismatch in all the patients. Response was defined as the achievement of complete chimerism (CC) and no evidence of hematological relapse. DLI was given either for the development of mixed chimerism (MC) in 7 patients (MC group) or for cytogenetic/hematological relapse in 14 patients (relapse group). Prior to DLI, cyclosporin A had been stopped in all patients, and no child had received chemotherapy.

Results of DLI: Five of the 21 patients received a single DLI, 16 patients 2–6 infusions (median 3). The total T cell dose given ranged from 9x10⁴ to 2.4 x10⁸/kg. Six of 21 patients responded: 3 of 7 patients in the MC group and 3 of 14 patients in the relapse group. The infusion of at least 1x10⁷/kg T cells was needed for durable response. Response was observed in all karyotype subgroups. None of the 6 patients receiving DLI from a matched sibling responded. Five patients developed acute GVHD following DLI and 4 of them responded to DLI. On the contrary, only 2 of the 16 children who did not show acute GVHD after DLI responded. Chronic GVHD developed in 2 responders. The outcome of even the responders was unfavorable. Only one of the responders is alive in remission, with severe chronic GVHD, 72 months after DLI. Two patients relapsed 26 days and 54 months after DLI (one as gastric chloroma), and 3 died of complications of DLI (acute GVHD, bone marrow aplasia, and hyper-eosinophilic syndrome). Four non-responders and one responder with subsequent relapse were rescued by a second SCT.

Conclusion: This study shows that DLI can induce a GVL effect in some of the JMML patients. However, the benefit of DLI in our series of patients was limited because of severe complications in responders and lack of a durable effect. Some modification of DLI, with previous cytoreduction by chemotherapy or a novel drug such as E21R and concomitant administration of cytokines such as interferon alpha, can possibly improve the result of DLI.