Open AccessIdentification of a metabolic gene panel to predict the prognosis of myelodysplastic syndrome
Author(s) -
Hu Fang,
Chen Siliang,
Dai Yujun,
Wang Yun,
Qin Zheyuan,
Li Huan,
Shu Lingling,
Li Jinyuan,
Huang Hanying,
Liang Yang
Publication year - 2020
Publication title -
journal of cellular and molecular medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.44
H-Index - 130
eISSN - 1582-4934
pISSN - 1582-1838
DOI - 10.1111/jcmm.15283
Subject(s) - cohort , medicine , oncology , hazard ratio , proportional hazards model , myelodysplastic syndromes , cohort study , international prognostic scoring system , bone marrow , confidence interval
Myelodysplastic syndrome (MDS) is clonal disease featured by ineffective haematopoiesis and potential progression into acute myeloid leukaemia (AML). At present, the risk stratification and prognosis of MDS need to be further optimized. A prognostic model was constructed by the least absolute shrinkage and selection operator (LASSO) regression analysis for MDS patients based on the identified metabolic gene panel in training cohort, followed by external validation in an independent cohort. The patients with lower risk had better prognosis than patients with higher risk. The constructed model was verified as an independent prognostic factor for MDS patients with hazard ratios of 3.721 (1.814‐7.630) and 2.047 (1.013‐4.138) in the training cohort and validation cohort, respectively. The AUC of 3‐year overall survival was 0.846 and 0.743 in the training cohort and validation cohort, respectively. The high‐risk score was significantly related to other clinical prognostic characteristics, including higher bone marrow blast cells and lower absolute neutrophil count. Moreover, gene set enrichment analyses (GSEA) showed several significantly enriched pathways, with potential indication of the pathogenesis. In this study, we identified a novel stable metabolic panel, which might not only reveal the dysregulated metabolic microenvironment, but can be used to predict the prognosis of MDS.