Risk Factors and Prediction Model of Early Relapse in Acute Myeloid Leukemia after Allogeneic Hematopoietic Cell Transplantation

BACKGROUND: Relapse remains the main cause of death in acute myeloid leukemia (AML) patients after allogeneic hematopoietic stem cell transplantation (allo-HSCT), which limits the success of allo-HSCT for the treatment of AML patients. Identifying the high-risk group of patients who will relapse after allo-HSCT could facilitate early intervention, thereby improving survival.

OBJECTIVE: The objective of this study was to identify prognostic factors associated with post-transplantation early relapse of patients with de novo AML.

METHODS: We enrolled patients aged 14-60 years with AML and received their first allo-HSCT from September 2014 to April 2020 in our center. The primary endpoint was early relapse (relapse within 1 year) after allo-HSCT. The factors associated with early relapse were investigated, and a prognostic model was developed based on multivariate analysis.

RESULTS:Between September 2014 and April 2020, a total of 414 AML patients were enrolled. The overall cumulative incidence of relapse (CIR) after allo-HSCT at 1 year and 4 years were 11.1% (95%CI, 8.1% -14.1%) and 21.1% (95%CI, 17.0%-25.2%), respectively. The 3-year overall survival (OS) of the patients who relapsed within 1-year was 6.2%. The 1-year CIR of patients with the primary resistance was significantly higher than that with non-primary resistance (22.7% vs. 10.3%; P=0.015). The CIR for the patients with DNMT3A mutation group and without DNMT3A mutation group at 1 year was 22.7% (95%CI, 10.9-34.5%) and 10.3% (95%CI, 7.2%-13.4%) (P=0.012). The CIR at 1 year was 7.5% for patients without pre-MRD, 18.6% for patients with pre-MRD, and 18.2% for no remission (NR) patients(P=0.005). The 1-year CIR of patients with WBC ≤20 ×109/L and WBC >20 ×109/L were 10.3% and 22.7% (P= 0.015). For patients who achieved first complete remission (CR1) or second complete remission (CR2) prior to allo-HSCT, the early relapse rate and 4-year CIR were similar between CR1 and CR2. The 1-year CIR for patients transplanted in CR1MRD-, CR2MRD-, CR1MRD+, and CR2MRD+ were 7.9%, 5.9%, 19.2%, and 19.4% (P=0.010). Patients with pre-MRD had a higher relapse rate compared without pre-MRD, no matter in CR1 or in CR2. After multivariable adjustment, primary resistance, pre-MRD, DNMT3A mutation, and white blood cell (WBC) at diagnosis remained statistically significantly associated with early relapse. The predictive model for early relapse based on these features performed well for the early identification of at-risk patients and the AUC was 0.73. The early relapse rate in high-risk and low-risk patients were 23.7% and 7.1%(P<0.001).

CONCLUSION:Primary resistance, pre-MRD, DNMT3A, and WBC count at diagnosis were independent risk factors for early relapse after allo-HSCT. The model might be applied to help identify patients at relapse risk.

No relevant conflicts of interest to declare.