Presently, multiparameter flow cytometry (MFC) and polymerase chain reaction (PCR) have been used for disease load, including measurable residual disease (MRD), monitoring in patients with myelodysplastic syndrome (MDS). MFC is the most commonly method for disease load evaluation. In patients with acute myeloid leukemia, leukemia stem cells (LSCs) determined using MFC for leukemia load and MRD detection is superior to traditional MFC method. In the investigators previous single center study, the investigators demonstrated that detection of disease load, including MRD, by MFC in patients with MDS-EB is superior to predict outcomes after allogeneic stem cell transplantation. Here, the investigators will perform a multi-center, prospective clinical trial to investigate the predictive values of MDS-SC in patients with MDS-EB who received allografting.

Inclusion Criteria:

• Patients with Myelodysplastic syndromes;
• Between 15 and 70 years old;
• Subjects are able to provide written informed consent.

Exclusion Criteria:

• Subjects who cannot comply with the study;
• Patient has severe cardiac (ejection fraction <50%), hepatic (total bilirubin >34μmol/L, ALT, AST >2x upper limit of normal) or renal (blood creatinine >130μmol/L) disease；
• Uncontrolled serious infection;
• Other conditions that do not tolerate transplantation or other therapies.

To determine whether there was any difference in relapse between the pre-transplant MRD-positive and -negative groups, the cumulative incidence approach was used with a test for equivalence of CIF for the difference in the Kaplan-Meier estimate of the 1-year CIR. With a planned sample size of 163 AML/MDS patients, 80% power can be achieved against the hypothesis of CIR as 18.3% and 3.6% for cases in the pre-transplant MRD-positive and -negative groups at a significance level of P = 0.05 in Student's one-tailed t-test.