Extensive bone marrow necrosis (BMN) is a relatively rare clinical pathologic entity. Usually, 90% of BMN cases are associated with malignancies (mainly hematological). The morphological features of BMN are disruption of normal marrow architecture, and necrosis of myeloid tissue and medullary stroma.1 The pathogenesis of BMN is still unclear.2 It has been suggested that BMN could be mediated by reactive CD8+ cytotoxic T cells3 or the release of either toxins or soluble mediators.4 BMN might correlate with elevated TNF-α levels.5, 6 Traditionally, there is a high mortality rate in BMN.7, 8 The prognosis of patients with BMN that is secondary to neoplastic disease is extremely poor, with survival not exceeding 6 months from the date of marrow necrosis diagnosis.9 There is no effective treatment for BMN; therefore, it is particularly important to seek an effective treatment.
Umbilical cord-derived MSCs (UC-MSCs) are precursor cells that can differentiate into BM stromal cells, which have a critical role in providing the essential microenvironment for hematopoiesis. On the basis of the hematopoietic support, immunosuppressive properties and low immuno-phenotype of UC-MSCs, we postulated that BMN may be corrected by infusion of allogeneic MSCs. Here, for the first time, we present a BMN case with CML patient that was successfully resolved by UC-MSCs.
A 10-year-old boy was transferred to our hospital on 3 February 2010, because of marked leukocytosis and splenomegaly. The BM was hypercellular with the manifestations of chronic-phase CML. The ratio of BCR-ABL/ABL determined by real-time RT–PCR of peripheral blood metaphases was 64.8%. He was diagnosed with CML chronic phase, and received hydroxyurea 2 g/daily peros for a week, then switched to imatinib mesylate 200 mg/daily peros in March 2010. The BCR-ABL/ABL ratio in the peripheral blood was monitored by real-time RT–PCR every 3 months, and declined gradually to 0.03% 1 year later.
In May 2011, the patient developed increasing bone pain in the shoulders, knees and ankles of both sides, together with fever. He was readmitted to our hospital on 7 June 2011. He had moderate pallor. There was no lymph node enlargement or organomegaly. Hematologic examination revealed 7.6 g/dL hemoglobin, 228 × 109/L platelets, 2.11 × 109/L WBC and 225 × 109/L reticulocytes; in the peripheral blood smear, normoblasts and immature granulocytes were seen; LDH level was increased to 2383 IU/L and alkaline phosphatase level to1013 IU/L. All microbiological tests were negative. The right aspirate from BM was of an altered serosanguinous type. A watery, dark red to clear fluid was aspirated from the left side of the iliac crest. The smears showed a differential, irregular staining with the presence of an amorphous eosinophilic proteinaceous material enmeshed within which were seen ‘ghost-like’ hematopoietic cells with irregular or indistinct cell membranes, and with nuclei showing the nuclear features. BM biopsies revealed extensive foci of gelatinous transformation with necrosis (the necrosis area accounted for 75% of the biopsy). Technetium-99 m sulfur colloid BM imaging showed extensive radioactive anomalies in the skull and chest, ribs, humerus, double side shoulder blades, bilateral iliac crest, femur and tibia and right ankle. Extensive BM necrosis (>50% of BM biopsy showing necrosis) was diagnosed (Figures 1a and c).The ratio of BCR-ABL/ABL was 1.19%. We suspected that the patient might be on the clonal evolution to blast crisis.
Due to the reports about BMN secondary to imatinib usage,6, 10 we stopped imatinib administration to the patient, and gave him allopurinol tablets 100 mg/m2/dose and antibiotics. Supportive therapy and pain control were applied and hematopoietic support was provided to him, when required. Three weeks later, the patient’s symptoms, including fever and bone pain subsequently aggravated, and hematologic examination revealed 8.1 g/dL hemoglobin, 48 × 109/L platelets, 1.78 × 109/L WBC and 285 × 109/L reticulocytes; BM aspiration of both sides posterior superior iliac spine (PSIS) still indicated BMN. After approval of the study by our ethics committee and written informed consent by the patient's guardian, we decided to investigate whether allogeneic UC-MSCs had therapeutic effects on the patient.
Human umbilical cord samples were obtained from a healthy woman with a written informed consent. UC-MSC was prepared as described previously.11
On 6 July 2011, BM aspiration was done from the right and left PSIS, respectively (Figures 2a and b; Table 1); We aspirated the patient's BM from multi-sites, purified and cultured the MSCs, with no success (meanwhile, by the same culture methods, MSCs were harvested successfully from the healthy volunteer). This was followed by administration of a total of 2 × 107 allogeneic UC-MSCs in 20 mL of solution (normal saline) by intra-BM injection via the left PSIS. In all, 20 mL normal saline by intra-BM injection via the right PSIS was used as a control. No UC-MSC infusion-related side effects were noted. Two weeks later, BM aspiration was done again from both side PSIS; besides BM smear, 5 mL of BM specimen was aspirated to perform BM-MSC incubation as described, and sex chromosome detected by FISH for identifying MSC engraftment. The BM smear of the left PSIS showed BMN disappearance and active proliferation of BM cells (Figure 2d, Table 1); however, BMN on the right side was still present (Figure 2c, Table 1). Due to extensive BMN, on 20 July, 2 × 106/kg MSCs were delivered by i.v. infusion without premedication. After 2 weeks, hematologic examination revealed 9.6 g/dL hemoglobin, 167 × 109/L platelets, 12.05 × 109/L WBC and 278 × 109/L reticulocytes; BM smear of both sides recovered to normal (Figures 2e and f, Table 1), and bone pain and fever were relieved.
FISH results showed the patient displayed evidence of mostly donor origin at +14 days after UC-MSC transplantation. This patient was male, and the UC-MSC sample was obtained from a female infant. At +14 days, 197/200 (98.5%) of the MSCs were XX, which represents the most donor origin. The dynamic observation continued at +60 days and 124/200 (62.0%) of the UC-MSCs were XX. At +180 days, the chimerization status disappeared, with 200/200 (100.0%) XY MSCs.
The patient developed B-cell blastic transformation after 2 months, and was treated with chemotherapy with the tyrosine kinase inhibitor imatinib at 400 mg/day; to date, he has been in stable condition for 38 months.
Allogeneic or autologous BM transplantation has been suggested to be a lifesaving treatment strategy.12, 13 But allotransplantation is limited by the availability of histocompatible donors and a variety of potentially lethal complications. Auto-SCT is not suitable for the treatment of acute or chronic leukemia.
We used allogeneic UC-MSC to successfully correct BMN, for the first time. Before the cellular therapy, the BM MSCs were cultivated in vitro, without success. We postulated that this outcome might be due to the patient’s MSC damage, and infusion of allogeneic MSCs might correct the defect. The BMN patient was treated successfully by i.v. injection of allogeneic UC-MSCs, and completed by i.v. infusion of an additional dose of MSCs. This case indicates that BMN might be caused by the damage of BM MSCs, although further research about potential mechanisms involved is still needed. The therapeutic effects of MSC treatment may also be attributed to their immuno-regulatory activities. MSCs have been reported to suppress cytotoxic T cell-mediated cytotoxicity, increase the proportion of CD4+CD25+ FoxP3+ regulatory T cells14 and inhibit cytokine secretion (IL-12, TNF-α and IFN-γ) in activated T cells.15 Allogeneic UC-MSCs might be a better option for BMN treatment.
Lee JL, Lee JH, Kim MK, Cho HS, Bae YK, Cho KH et al. A case of bone marrow necrosis with thrombotic thrombocytopenic purpura as a manifestation of occult colon cancer. Jpn J Clin Oncol 2004; 34: 476–480.
Janssens AM, Offner FC, Van Hove WZ . Bone marrow necrosis. Cancer 2000; 88: 1769–1780.
Markovic SN, Phyliky RL, Li CY . Pancytopenia due to bone marrow necrosis in acute myelogenous leukemia: role of reactive CD8 cells. Am J Hematol 1998; 59: 74–78.
Aboulafia DM, Demirer T . Fatal bone marrow necrosis following fludarabine administration in a patient with indolent lymphoma. Leuk Lymphoma 1995; 19: 181–184.
Seki Y, Koike T, Yano M, Aoki S, Hiratsuka M, Fuse I et al. Bone marrow necrosis with dyspnea in a patient with malignant lymphoma and plasma levels of thrombomodulin, tumor necrosis factor-alpha, and D-dimer. Am J Hematol 2002; 70: 250–253.
Matsue K, Takeuchi M, Koseki M, Uryu H . Bone marrow necrosis associated with the use of imatinib mesylate in a patient with Philadelphia chromosome-positive acute lymphoblastic leukemia. Ann Hematol 2006; 85: 542–544.
Bhasin TS, Sharma S, Chandey M, Bhatia PK, Mannan R . A case of bone marrow necrosis of an idiopathic aetiology: the report of a rare entity with review of the literature. J Clin Diagn Res 2013; 7: 525–528.
Tsitsikas DA, Gallinella G, Patel S, Seligman H, Greaves P, Amos RJ . Bone marrow necrosis and fat embolism syndrome in sickle cell disease: increased susceptibility of patients with non-SS genotypes and a possible association with human parvovirus B19 infection. Blood Rev 2014; 28: 23–30.
Elgamal BM, Rashed RA, Raslan HN . Prevalence of bone marrow necrosis in Egyptian cancer patients referring to the National Cancer Institute. J Egypt Natl Canc Inst 2011; 23: 95–99.
Aras Y, Akcakaya MO, Unal SN, Bilgic B, Unal OF . Bone marrow necrosis secondary to imatinib usage, mimicking spinal metastasis on magnetic resonance imaging and FDG-PET/CT. J Neurosurg Spine 2012; 16: 57–60.
Lu LL, Liu YJ, Yang SG, Zhao QJ, Wang X, Gong W et al. Isolation and characterization of human umbilical cord mesenchymal stem cells with hematopoiesis-supportive function and other potentials. Haematologica 2006; 91: 1017–1026.
Katayama Y, Deguchi S, Shinagawa K, Teshima T, Notohara K, Taguchi K et al. Bone marrow necrosis in a patient with acute myeloblastic leukemia during administration of G-CSF and rapid hematologic recovery after allotransplantation of peripheral blood stem cells. Am J Hematol 1998; 57: 238–240.
Khan AM, Yamase H, Tutschka PJ, Bilgrami S . Autologous peripheral blood progenitor cell transplantation for non-Hodgkin's lymphoma with extensive bone marrow necrosis. Bone Marrow Transplant 1997; 19: 1037–1039.
Selmani Z, Naji A, Zidi I, Favier B, Gaiffe E, Obert L et al. Human leukocyte antigen-G5 secretion by human mesenchymal stem cells is required to suppress T lymphocyte and natural killer function and to induce CD4+CD25highFOXP3+ regulatory T cells. Stem Cells 2008; 26: 212–222.
Kim J, Hematti P . Mesenchymal stem cell-educated macrophages: a novel type of alternatively activated macrophages. Exp Hematol 2009; 37: 1445–1453.
This work was supported by a grant-in-aid from the National Science Foundation of China (No. 81372132).
The authors declare no conflict of interest.