An unusual case of donor-derived myelodysplastic syndrome following double-unit umbilical cord blood transplantation in acute lymphoblastic leukemia


  • Conflict of interest: Nothing to report.

Umbilical cord-blood transplantation is considered an effective treatment strategy for acute lymphoblastic leukemia (ALL) when a human leukocyte antigen (HLA)-matched donor is unavailable. The use of a second unit helps ensure engraftment in larger adults and those with comorbidities, even though only one unit engrafts in most patients [1, 2]. Herein, we present the clinical and laboratory characteristics of a patient who developed donor-derived myelodysplastic syndrome (ddMDS) after double umbilical cord-blood transplantation (dUCB HSCT). To our knowledge, no cases of ddMDS have been described in a patient with a history of ALL in molecular remission after receiving a dUCB HSCT. Current molecular techniques, including analysis of short tandem repeats (STR) and fluorescence in situ hybridization (FISH) allowed us to firmly establish donor origin.

Our patient is a 25-year-old man, who was previously diagnosed with B-ALL with hyperdiploidy in January 2006. Bone marrow examination revealed a hyperdiploid karyotype, 56,XY,+X,inv(2)(p11.2q13),+4,+6,+9,+14, +14,+17,+18,+21,+21. To determine the patient's constitutive karyotype, peripheral blood was stimulated with phytohemagglutinin (PHA), a T-cell specific mitogen, revealing a 46,XY,inv(2)(p11.2q13)C[15] karyotype. As T-cells are extremely long-lived, this suggests the inv(2) abnormality was constitutive. Cerebrospinal fluid analysis was negative. He was treated with a 5-drug regimen and 20 months of methotrexate, 6-MP, and vincristine maintenance [3]. He attained a morphologically complete remission but had minimal residual disease by FISH-with 5% of cells with +X,+Y,+21,+21,+21,+21 in January 2007. The patient received one cycle of clofarabine, achieving cytogenetic and FISH CR in June 2007.

In July 2007, he received myeloablative conditioning with cyclophosphamide, thiotepa, fludarabine, and total body irradiation, and a 4/6 partially mismatched dUCB HSCT. Graft versus host disease (GVHD) prophylaxis was tacrolimus and methotrexate. Patient and cord blood units were all O+ with a 46XY karyotype. Post-transplant course was complicated by sepsis, acute and chronic GVHD, pulmonary embolism, CMV reactivation, JC virus hemorrhagic cystitis, posterior reversible encephalopathy, steroid induced diabetes, and acute renal failure leading to a living-donor, related renal transplant in April 2009. He received daclizumab induction and mycophenolate mofetil post-transplant until June 2009. He continues to be on prednisone and tacrolimus for suppression of chronic GVHD.

In May 2011, nearly 4 years post-HSCT, a CBC and peripheral blood smear demonstrated new anemia and thrombocytopenia. Bone marrow biopsy and aspirate demonstrated normocellular marrow with relative erythroid hyperplasia, with dysplasia of erythroid and megakaryocytic lineages, and no increase in myeloblasts by flow cytometry or manual differential count. CD19+, CD10+, and immature B-cells were absent by flow cytometry. Cytogenetics revealed a 44, XY,-3,del(4)(q23q33),der(5;17)(p10;q10),-7,t(8;22)(p21;q13) karyotype in 65% of cells. The constitutional inv(2) abnormality seen pretransplant was notably absent, suggesting donor origin of the cell clone (see Fig. 1). In June and August 2011, bone marrow and peripheral blood CD3 (lymphoid origin) and CD33 (myeloid origin) cell chimerism studies were performed. STR analysis of SE33 (ACTBP2) alleles, a robust and highly polymorphic locus, remained 100% Donor 2. Collectively, these results were interpreted as ddMDS. The patient is currently receiving azacitidine, and a search for a second unrelated stem cell donor is underway. The cord blood donor center was notified, and there is currently no evidence of hematologic disease in either donor.

Figure 1.

A: At diagnosis of ALL, in January 2006, cytogenetic analysis of bone marrow revealed a 56,XY,+X,inv(2)(p11.2q13),+4,+6,+9,+14,+14,+17,+18,+21,+21 hyperdiploid karyotype (arrows indicate abnormal chromosomes). B: A bone marrow nucleus from a specimen obtained in January 2007 after FISH showed 5% of cells with Tetrasomy 21 (red signals), two copies of X chromosome (green signals), and one copy of Y chromosome (aqua). C: A partial karyotype of Chromosome 2 from the peripheral blood specimen, PHA stimulated for 72 h, obtained in May 2006, showing a constitutional inversion (2)(p11.2q13). D: At the time of the diagnosis of MDS in May 2011, bone marrow cytogenetic analysis revealed 65% of cells with a 44,XY,-3,del(4)(q23q33),der(5;17)(p10;q10),-7,t(8;22)(p21;q13),+mar karyotype (arrows indicate abnormal chromosomes). E: Five bone marrow nuclei after FISH studies from the June 2011 specimen showing 75% cells with deletion of EGR1 at 5q31 chromosomal location (red) and deletion of P53 at 17p13.1 chromosomal localization (red) as a result of der(5;17); 71% showing a loss of 7q31 locus as a result of Monosomy 7, as well as disomy 21 (red), one X (green), and one Y (aqua) chromosome. F: A partial bone marrow karyotype of Chromosome 2 from May 2011 specimen showing a normal Chromosome 2 from the donor cells and absence of inv(2) observed in the BM and PHA-stimulated PB at the time of diagnosis. [Color figure can be viewed in the online issue, which is available at]

MDS is a heterogeneous family of hematological disorders best characterized by ineffective blood cell production and a risk of transformation to acute leukemia. MDS may occur de novo or after exposure to mutagens, such as chemotherapy or radiation, manifesting between a few months or years afterward [4, 5]. ddMDS is a rare but increasingly recognized complication of HSCT that represents both a diagnostic dilemma and a therapeutic challenge. In the case presented, the recipient received a dUCB HSCT for B-ALL, and the MDS clone was determined to be donor-derived.

The first challenge to determining the origin of MDS after HSCT is correctly distinguishing between donor and recipient cells. In the past, most reported cases of ddMDS or ddAML occurred in the setting of sex-mismatched transplants, whereby cytogenetic analysis could easily distinguish between donor and recipient. As the two UCB units and the recipient had a male chromosomal complement, discriminating cell origin based on cytogenetic analysis of sex chromosomes was not possible. STR testing of an informative sequence (SE33 locus) was consistent with Donor 2 origin. To further confirm donor origin, conventional cytogenetics was used to demonstrate the absence of the previously present constitutional abnormality on Chromosome 2, as shown in Fig. 1.

The occurrence of ddMDS/ddAML after HSCT was first reported in 1971 and subsequently has been documented in various case reports [6, 7]. A total of 51 cases of ddMDS and 13 cases of ddAML have been reported in the literature [8]. The median time from HSCT to the development of ddMDS/ddAML has been reported to be 24 months [9]. The occurrence of ddMDS/ddAML raises the concern for potential development of a hematologic malignancy in the donor. However, there seems to be no report of development of MDS/leukemia in children whose UCB was used in cases of ddMDS/ddAML. A case of ddMDS from a donor with asymptomatic 20q- MDS at time of stem cell donation on retrospective analysis has been reported; this case underscores the capability of pre-existing clonal MDS donor progenitor cells to home to the marrow, successfully engraft, contribute to hematopoiesis, and reconstitute an immune system [10, 11].

UCB allografts are readily available and require low stringency matching. Therefore, they are used for patients without suitable related or unrelated donors. Donor-derived hematologic neoplasms are now recognized as a potential complication of UCB HSCT. Donor-derived EBV+ B-cell PTLD and ddMDS/ddAML has also been previously reported after UCB HSCT [12–20]. Kaplan–Meier analysis of pooled data on 13 cases of ddAML after UCB HSCT seems to show an increased risk of ddAML, when compared with other sources of stem cells [21]. The reason for an increased risk of ddMDS/ddAML in the setting of UCB HSCT is not understood, and the pathogenesis remains elusive. Proposed mechanisms include: (i) sustained host-origin antigenic stimulation, (ii) impaired hematopoietic microenvironment and defective stromal support system, (iii) immune surveillance escape secondary to post-transplant immunosuppressive therapy, (iv) similar genetic susceptibility in cases of related donors, (v) viral driven pathogenesis (CMV, EBV), (vi) delayed effects of conditioning regimen, (vii) transfection of host cell oncogene into donor cells [22, 23]. ddMDS/ddAML is unlikely to be a result of a single mechanism and is probably a combination of the above mechanisms active in a given case, and these combinations may differ from case to case. A very compelling hypothesis for the mechanisms leading to the development of ddMDS/ddAML after SCT is the “2 hit hypothesis” [21, 22]. A donor HSC that has an inherent susceptibility to malignant transformation (Hit 1) is placed within a defective stromal structure elaborating a microenvironment that applies repeated stress signals (Hit 2) inducing additional genetic or even epigenetic mutations promoting malignant transformation. This second hit may also arise from random mutation due to the high proliferative rate needed to regenerate the bone marrow and decreased immune surveillance secondary to immune suppression allowing proliferation of a mutated clone.

ddMDS after UCB HSCT and, in the case presented, dUCB HSCT challenges our understanding of MDS pathogenesis and implicates a potential role of host nonhematopoietic factors in the development of MDS. The numbers of reported cases of ddMDS/ddAML after UCB HSCT are too few to determine actual prognosis and optimal therapeutic approach. Therapeutic options are limited for donor-derived MDS and include supportive care, treatment with hypomethylating agents, induction chemotherapy, and second SCT either with or without the same donor. Cases of ddMDS/ddAML responding to same-donor lymphocyte infusion and retransplant have been described [23, 24]. Current molecular techniques in combination with conventional cytogenetic analysis enabled us to firmly establish the donor-derived origin of MDS in this patient.

R. Cotter*, V. Najfeld* †, L. Isola*, G. Del Toro‡, E. Roman§, B. Petersen†, J. Mascarenhas*, * Division of Hematology-Oncology, The Tisch Cancer Institute, Mount Sinai School of Medicine, New York, New York, † Department of Pathology, Mount Sinai School of Medicine, New York, New York, ‡ Department of Pediatrics, Wyckoff Heights Medical Center, Brooklyn, New York, § Division of Pediatric Hematology-Oncology, NYU Langone Medical Center, New York, New York.