(1) Background: Biomarkers might play a significant role in predicting the clinical outcomes of patients with ovarian cancer. By analyzing lipid metabolism genes, future perspectives may be uncovered; (2) Methods: RNA-seq data for serous ovarian cancer were downloaded from The Cancer Genome Atlas and Gene Expression Omnibus databases. The non-negative matrix factorization package in programming language R was used to classify molecular subtypes of lipid metabolism genes and the limma package in R was performed for functional enrichment analysis. Through lasso regression, we constructed a multi-gene prognosis model; (3) Results: Two molecular subtypes were obtained and an 11-gene signature was constructed (PI3, RGS, ADORA3, CH25H, CCDC80, PTGER3, MATK, KLRB1, CCL19, CXCL9 and CXCL10). Our prognostic model shows a good independent prognostic ability in ovarian cancer. In a nomogram, the predictive efficiency was notably superior to that of traditional clinical features. Related to known models in ovarian cancer with a comparable amount of genes, ours has the highest concordance index; (4) Conclusions: We propose an 11-gene signature prognosis prediction model based on lipid metabolism genes in serous ovarian cancer.
(1) Background: The tumor microenvironment is involved in the growth and proliferation of malignant tumors and in the process of resistance towards systemic and targeted therapies. A correlation between the gene expression profile of the tumor microenvironment and the prognosis of ovarian cancer patients is already known. (2) Methods: Based on data from The Cancer Genome Atlas (379 RNA sequencing samples), we constructed a prognostic 11-gene signature (SNRPA1, CCL19, CXCL11, CDC5L, APCDD1, LPAR2, PI3, PLEKHF1, CCDC80, CPXM1 and CTAG2) for Fédération Internationale de Gynécologie et d’Obstétrique stage III and IV serous ovarian cancer through lasso regression. (3) Results: The established risk score was able to predict the 1-, 3- and 5-year prognoses more accurately than previously known models. (4) Conclusions: We were able to confirm the predictive power of this model when we applied it to cervical and urothelial cancer, supporting its pan-cancer usability. We found that immune checkpoint genes correlate negatively with a higher risk score. Based on this information, we used our risk score to predict the biological response of cancer samples to an anti-programmed death ligand 1 immunotherapy, which could be useful for future clinical studies on immunotherapy in ovarian cancer.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.