Allogeneic Hematopoietic Stem Cell Transplantation Could Improve Survival of Cytogenetically Normal Adult Acute Myeloid Leukemia Patients with DNMT3A Mutations

Objective:DNMT3A mutations are frequent in cytogenetically normal acute myeloid leukemia (cn-AML) patients and associated with poor survival. The role of allogeneic hematopoietic stem cell transplantation (allo-HSCT) in DNMT3A^mut cn-AML patients remains unclear.

Methods: In this study, we retrospectively analyzed the prognostic impact of DNMT3A mutations and explored the role of allogeneic hematopoietic stem cell transplantation (allo-HSCT) in 308 cn-AML patients who received consolidation of intensive chemotherapy or allo-HSCT in our center from March 2005 to May 2014.

Results: In the whole cohort, the median age was 40 years (range: 16-68 years), and there are 144 males and 164 females. Based on French-American-British (FAB) criteria, there were 5 (1.6%) patients classified as M0, 55 (17.9%) as M1, 74 (24.0%) as M2, 53 (17.2%) as M4, 75 (24.4%) as M5, 16 (5.2%) as M6, 3 (1.0%) as M7, and 27 (8.8%) that were unclassified AML patients. The median blast in BM was 56.12%, and the median white cell counts was 25.22(0.5-355.9)*10⁹/L.

63 patients (20.5%) were identified with DNMT3A exon 23 mutations and R882H was the most frequent variant. The median age of DNMT3A mutated patients was elder than that of the control group (p<0.001), while there were no significant differences in sex, white cell counts, and blast percentage in PB and BM between patients with and without DNMT3A mutations. However, regarding to FAB distributions, more M5 patients (38.1%) were observed in DNMT3A mutated group compared to the controls (20.8%) group (p<0.001). FLT3-ITD and NPM1 mutations were also more often observed in DNMT3A mutated group (p<0.001). DNMT3A^mut patients had shorter overall survival (3-year OS: 31.9% vs. 52.0%, p=0.009) and disease-free survival (3-year DFS: 21.8% vs. 40.1%, p=0.004) compared with DNMT3A^wt patients. Based on FLT3/NPM1/CEBPA mutations, 308 cn-AML patients were divided into favorable/intermediate group (n=262) and unfavorable group (n=46). There were no significant differences in 3-year OS and 3-year DFS between DNMT3A^mut and DNMT3A^wt patients in both favorable/intermediate and unfavorable groups. Additionally, in multivariate analysis age, treatment, FLT3-ITD/NPM1/CEBPA risk classification and DNMT3A mutations were significantly and independently associated with a worse OS and DFS. In the DNMT3A^mut cohort, 23 CR patients received allo-HSCT consolidation and 32 CR patients received chemotherapy consolidation, dramatic differences were observed in 3-year OS (51.7% vs. 28.9%, p=0.048) and 3-year DFS (41.6% vs. 14.9%, p=0.024) between allo-HSCT group and chemotherapy group. Interestingly, when we limited this comparison to the favorable/intermediate risk group only, significant differences were also observed in both 3-year OS (56.0% vs. 34.8%; p=0.036) and 3-year DFS (41.9% vs. 16.7%; p=0.047) between these two groups.

Conclusion:DNMT3A mutation is a poor prognostic factor for cn-AML patients and allo-HSCT could improve survival of cytogenetically normal acute myeloid leukemia patients with DNMT3A mutations.