The Comprehensive Analysis of Weighted Gene Co-Expression Network Analysis and Machine Learning Revealed Diagnostic Biomarkers for Breast Implant Illness Complicated with Breast Cancer

Zhenfeng Huang,1,* Huibo Wang,2,* Hui Pang,1,* Mengyao Zeng,3 Guoqiang Zhang,1 Feng Liu1 1Department of Breast Surgery, Harbin Medical University Cancer Hospital, Harbin, Heilongjiang Province, People’s Republic of China; 2Department of Emergency Surgery, The Affiliated Hospital of Qingdao University, Qingdao, Shandong Province, People’s Republic of China; 3Department of Medical Training, Aimiker Technology Development Co., Ltd, Nanjing, Jiangsu Province, People’s Republic of China*These authors contributed equally to this workCorrespondence: Guoqiang Zhang, Department of Breast Surgery, Harbin Medical University Cancer Hospital, No. 150 haping road, Harbin, Heilongjiang, 150081, People’s Republic of China, Email [email protected] Feng Liu, Department of Breast Surgery, Harbin Medical University Cancer Hospital, No. 150 haping road, Harbin, Heilongjiang, 150081, People’s Republic of China, Email [email protected]: An increasing number of breast cancer (BC) patients choose prosthesis implantation after mastectomy, and the occurrence of breast implant illness (BII) has received increasing attention and the underlying molecular mechanisms have not been clearly elucidated. This study aimed to identify the crosstalk genes between BII and BC and explored their clinical value and molecular mechanism initially.Methods: We retrieved the data from Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA), and identified the differentially expressed genes (DEG) as well as module genes using Limma and weighted gene co-expression network analysis (WGCNA). Enrichment analysis, the protein–protein interaction network (PPI), and machine learning algorithms were performed to explore the hub genes. We employed a nomogram and receiver operating characteristic curve to evaluate the diagnostic accuracy. Single-cell analysis disclosed variations in the expression of key genes across distinct cellular populations. The expression levels of the key genes were further confirmed in BC cell lines. Immunohistochemical analysis was utilized to examine protein levels from 25 patients with breast cancer undergoing prosthetic implant surgery. Ultimately, we deployed single-sample Gene Set Enrichment Analysis (ssGSEA) to scrutinize the immunological profiles between the normal and BC cohorts, as well as between the non-BII and BII groups.Results: WGCNA identified 1137 common genes, whereas DEG analysis found 541 overlapping genes in BII and BC. After constructing the PPI network, 17 key genes were selected, and three potential hub genes include KRT14, KIT, ALB were chosen for nomogram creation and diagnostic assessment through machine learning. The validation of these results was conducted by examining gene expression patterns in the validation dataset, breast cancer cell lines, and BII-BC patients. However, ssGSEA uncovered different immune cell infiltration patterns in BII and BC.Conclusion: We pinpointed shared three central genes include KRT14, KIT, ALB and molecular pathways common to BII and BC. Shedding light on the complex mechanisms underlying these conditions and suggesting potential targets for diagnostic and therapeutic strategies.Keywords: breast cancer, breast implant illness, weighted gene co-expression network analysis, hub genes