Construction and validation of a fatty acid metabolism risk signature for predicting prognosis in acute myeloid leukemia.

Background: Fatty acid metabolism has been reported to play important roles in the development of acute myeloid leukemia (AML), but there are no prognostic signatures composed of fatty acid metabolism-related genes. As the current prognostic evaluation system has limitations due to the heterogeneity of AML patients, it is necessary to develop a new signature based on fatty acid metabolism to better guide prognosis prediction and treatment selection.

Methods: We analyzed the RNA sequencing and clinical data of The Cancer Genome Atlas (TCGA) and Vizome cohorts. The analyses were performed with GraphPad 7, the R language and SPSS.

Results: We selected nine significant genes in the fatty acid metabolism gene set through univariate Cox analysis and the log-rank test. Then, a fatty acid metabolism signature was established based on these genes. We found that the signature was as an independent unfavourable prognostic factor and increased the precision of prediction when combined with classic factors in a nomogram. Gene Ontology (GO) and gene set enrichment analysis (GSEA) showed that the risk signature was closely associated with mitochondrial metabolism and that the high-risk group had an enhanced immune response.

Conclusion: The fatty acid metabolism signature is a new independent factor for predicting the clinical outcomes of AML patients.