Identification and validation of a prognostic index based on a metabolic-genomic landscape analysis of ovarian cancer

Purpose: Tumour metabolism has become a novel factor targeted by personalised cancer drugs. This research evaluated the prognostic significance of metabolism-related genes (MRGs) in ovarian serous cystadenocarcinoma (OSC). Methods: MRGs in 379 women surviving OSC were obtained using The Cancer Genome Atlas (TCGA) database. Then, several biomedical computational algorithms were employed to identify eight hub prognostic MRGs that were significantly relevant to OSC survival. These eight genes have important clinical significance and prognostic value in OSC. Subsequently, a prognostic index was constructed. Drug sensitivity analysis was used to screen the key genes in eight MRGs. Immunohistochemistry (IHC) staining confirmed the expression levels of key genes and their correlations with clinical parameters and prognosis for patients. Results: A total of 701 differentially expressed MRGs were confirmed in women with OSC by the TCGA database. The random walking with restart (RWR) algorithm and the univariate Cox and lasso regression analyses indicated a prognostic signature based on eight MRGs (i.e., ENPP1, FH, CYP2E1, HPGDS, ADCY9, NDUFA5, ADH1B and PYGB), which performed moderately well in prognostic predictions. Drug sensitivity analysis indicated that PYGB played a key role in the progression of OSC. Also, IHC staining confirmed that PYGB has a close correlation with clinical parameters and poor prognosis in OSC. Conclusion: The results of the present study may help to establish a foundation for future research attempting to predict the prognosis of OSC patients and to characterise OSC metabolism.