HLA-identical umbilical cord blood transplantation from a sibling donor in juvenile myelomonocytic leukemia

As recently described by Flotho et al. juvenile myelomonocytic leukemia (JMML) is a rare type of childhood leukemia not only characterized by young age, hepatosplenomegaly, thrombocytopenia and monocytosis, but also by molecular aberrations in the RAS-RAF-MEK-ERK signaling pathway and GM-CSF-hypersensitivity.1 Although progress has been made in unraveling the molecular background of JMML, the only curative treatment option for these children is stem cell transplantation (SCT).2,3 Previous studies have shown that in JMML the graft versus leukemia (GvL) effect of the SCT plays an important role in the prevention of relapse.2,4 Hence, donor and stem cell source selection could play an important role in the outcome of JMML patients. SCT using unrelated, immunological naive, umbilical cord blood (UCB) has shown to be effective for pediatric JMML.3 UCBT has increased the available pool of hematopoietic stem cell transplantation donors, especially for young children.5 This is important for patients with rare HLA-haplotypes and with diseases that may rapidly progress like JMML. However, it is conceivable that the relatively immunological naivety of cord blood stem cells from a newborn HLA identical sibling donor may be a negative factor for outcome.4,6

In this report we describe 5 JMML patients registered in the database of the European Working Group on Childhood MDS (EWOG-MDS), who have received umbilical cord stem cells from an HLA-identical sibling (Table 1). So far only MacMillan et al. has described a JMML patient who received HLA-identical sibling umbilical cord cells.7 More information is available on unrelated UCBT in JMML.3,7,8 Recently, from the combined EWOG-MDS/EBMT registry, Locatelli et al. described 100 JMML patients of whom 7 received an unrelated UCBT. These 7 patients showed a delayed hematologic recovery, but the outcome was comparable to the children treated with other stem cell sources.3 Rocha et al. reported that children receiving a related UCBT for malignancies, bone marrow failure syndromes, hemoglobinopathies, inborn errors of metabolism or immunodeficiencies, had a slower hematopoietic recovery and a lower risk of acute and chronic GVHD than children receiving a related bone marrow transplantation.9 Gluckman et al. have shown that the amount of acute GVHD in related UCBT is lower (18%) than in unrelated UCBT (32%).10 These studies underscore the relative immune naivety of the UCB cells and the feasibility of the use of UCBs especially for childhood transplantation settings. However, especially in JMML, where the GvL-effect is an important contributor to the success of the transplantation procedure, it could be questioned whether these relatively naive umbilical stem cells from sibling donors should be used in all patients, especially when alternative donors would be available.

Figure 1.DLI: donor lymphocyte infusion (week 10 and 12); 6-MP: 6-mercapto-purine; PBMC: peripheral blood mononuclear cells; UCBT: umbilical cord blood transplantation; % chimerism reflects the percentage of donor bone marrow in the patient.

In all cases full donor-chimerism was found after transplantation. In the cases that relapsed, a mixed chimerism was found prior to clinical progression. In addition, in Patient 1 increasing mixed donor-chimerism developed from day 42 onwards without clinical signs of relapse (Figure 1). Mixed donor-chimerism after SCT in JMML has been shown to be an important predictor for relapse, therefore discontinuation of immunosuppressive therapy (IST) is the first step to prevent relapse if this occurs early after transplantation.11 To avoid relapse, a second stem cell transplantation was considered, but the parents were very reluctant to proceed, because of the absence of any clinical signs of JMML and the excellent clinical condition of the child. Therefore, 6-mercaptopurine (6-MP) (average dose 40 mg/m) was started on day 145 after SCT. When, after initial improvement, donor-chimerism increased, 6-MP was replaced by 13-cisretinoic acid (100 mg/m/day every other week) for 2.5 years until the age of five years. Thereafter, the child remained well with a current follow-up of five years after SCT. At present the percentage of donor cells is 15%, without any clinical sign of JMML. At the moment no series are available on the use of 6-MP and 13-cis-retinoic acid in a relapse setting after SCT for JMML. Although Locatelli et al. showed that second SCT was successful in 7/15 cases with a relapse after SCT, the issue whether to proceed to a second SCT in all cases of mixed donor-chimerism without any clinical signs of JMML has not been clarifed.3

Table 1.(A) Patients’ characteristics of the JMML patients. (B) Transplantation characteristics of the JMML patients.

We conclude that in JMML umbilical cord stem cells can be considered a good alternative source for HSCT. However, whether the use of sibling HLA-identical UCB should be advocated needs to be confirmed in larger series of JMML patients.

Acknowledgments

Prof. Dr. P.D. Emanuel for his expert advise on he management of Patient 1; dr. C.P. Kratz for performing the mutation analysis of the JMML patients.

Footnotes

• Funding: the scientific work of ACHdV is supported by a grant from the KIKA Foundation, Amstelveen, The Netherlands.