Transcriptomic and machine learning analyses identify hub genes of metabolism and host immune response that are associated with the progression of breast capsular contracture

Abstract

Capsular contracture is a prevalent and severe complication that affects the postoperative outcomes of patients who receive silicone breast implants. At present, prosthesis replacement is the major treatment for capsular contracture after both breast augmentation procedures and breast reconstruction following breast cancer surgery. However, the mechanism(s) underlying breast capsular contracture remains unclear. This study aimed to identify the biological features of breast capsular contracture and reveal the potential underlying mechanism using RNA sequencing. Sample tissues from 12 female patients (15 breast capsules) were divided into low capsular contracture (LCC) and high capsular contracture (HCC) groups based on the Baker grades. Subsequently, 41 lipid metabolism-related genes were identified through enrichment analysis, and three of these genes were identified as candidate genes by SVM-RFE and LASSO algorithms. We then compared the proportions of the 22 types of immune cells between the LCC and HCC groups using a CIBERSORT analysis and explored the correlation between the candidate hub features and immune cells. Notably, PRKAR2B was positively correlated with the differentially clustered immune cells, which were M1 macrophages and follicular helper T cells (area under the ROC = 0.786). In addition, the expression of PRKAR2B at the mRNA or protein level was lower in the HCC group than in the LCC group. Potential molecular mechanisms were identified based on the expression levels in the high and low PRKAR2B groups. Our findings indicate that PRKAR2B is a novel diagnostic biomarker for breast capsular contracture and might also influence the grade and progression of capsular contracture.

Keywords

• Biomarker
• Breast capsular contracture
• Immune infiltration
• Machine learning algorithms
• PRKAR2B