Next generation sequencing reveals the mutation landscape of Chinese MDS patients and the association between mutations and AML transformations.

Background/objective: Approximately 30% of patients with MDS eventually develop to acute myeloid leukemia (AML). Our study aimed to investigate the mutation landscape of Chinese MDS patients and identify the mutated genes which are closely implicated in the transformation of MDS to AML.

Methods: In total, 412 sequencing data collected from 313 patients were used for analysis. Mutation frequencies between different groups were compared by Fisher's exact. A predictive model for risk of transformation/death of newly diagnosed patients was constructed by logistic regression.

Results: The most frequently mutated genes in newly diagnosed patients were , , , , and and mutations of , , , , and genes were more common in the treated MDS patients. Besides, we found that the mutation frequencies of , , and were significantly higher in MDS patients aged over 60 years. Moreover, two mutation sites, and were detected only at transformation in one patient, while not detected at diagnosis. In addition, the mutation frequencies of and were stable from diagnosis to transformation in two patients. Finally, we constructed a predictive model for risk of transformation/death of newly diagnosed patients combing detected data of 10 genes and the number of to leukocyte, with a sensitivity of 63.3% and a specificity of 84.6% in distinguishing individuals with and without risk of transformation/death.

Conclusion: In summary, our study found several mutations associated with the transformation from MDS to AML, and constructed a predictive model for risk of transformation/death of MDS patients.