A novel disulfidptosis-associated expression pattern in breast cancer based on machine learning

Zhitang Wang; Xianqiang Du; Weibin Lian; Jialin Chen; Chengye Hong; Liangqiang Li; Debo Chen

doi:10.3389/fgene.2023.1193944

A novel disulfidptosis-associated expression pattern in breast cancer based on machine learning

Front Genet. 2023 Jun 29:14:1193944. doi: 10.3389/fgene.2023.1193944. eCollection 2023.

Authors

Zhitang Wang¹, Xianqiang Du¹, Weibin Lian¹, Jialin Chen¹, Chengye Hong¹, Liangqiang Li¹, Debo Chen¹

Affiliation

¹ Department of Breast, The First Hospital of Quanzhou Affiliated to Fujian Medical University, Quanzhou, China.

Abstract

Background: Breast cancer (BC), the leading cause of cancer-related deaths among women, remains a serious threat to human health worldwide. The biological function and prognostic value of disulfidptosis as a novel strategy for BC treatment via induction of cell death remain unknown. Methods: Gene mutations and copy number variations (CNVs) in 10 disulfidptosis genes were evaluated. Differential expression, prognostic, and univariate Cox analyses were then performed for 10 genes, and BC-specific disulfidptosis-related genes (DRGs) were screened. Unsupervised consensus clustering was used to identify different expression clusters. In addition, we screened the differentially expressed genes (DEGs) among different expression clusters and identified hub genes. Moreover, the expression level of DEGs was detected by RT-qPCR in cellular level. Finally, we used the least absolute shrinkage and selection operator (LASSO) regression algorithm to establish a prognostic feature based on DEGs, and verified the accuracy and sensitivity of its prediction through prognostic analysis and subject operating characteristic curve analysis. The correlation of the signature with the tumor immune microenvironment and tumor stemness was analyzed. Results: Disulfidptosis genes showed significant CNVs. Two clusters were identified based on three DRGs (DNUFS1, LRPPRC, SLC7A11). Cluster A was found to be associated with better survival outcomes(p < 0.05) and higher levels of immune cell infiltration(p < 0.05). A prognostic signature of four disulfidptosis-related DEGs (KIF21A, APOD, ALOX15B, ELOVL2) was developed by LASSO regression analysis. The signature showed a good prediction ability. In addition, the prognostic signature in this study were strongly related to the tumor microenvironment (TME), tumor immune cell infiltration, tumor mutation burden (TMB), tumor stemness, and drug sensitivity. Conclusion: The prognostic signature we constructed based on disulfidptosis-DEGs is a good predictor of prognosis in patients with BC. This prognostic signature is closely related to TME, and its potential correlation provides clues for further studies.

Keywords: breast cancer; disulfidptosis; expression pattern; prognostic signature; tumor microenvironment.

Grants and funding

This study was supported by grants from the Science and Technology Project of Quanzhou (No. 2020N030s), the Startup Fund for Scientific Research of Fujian Medical University (No. 2019QH1253), the Youth Project of Health Commission of Fujian Province (No. 2020QNB042), the Natural Science Foundation of Fujian Province, PRC (No. 2020J011275), and the Clinical research Special Fund of Wu Jieping Medical Foundation (320.6750.2021-10-60).