Crafting a Personalized Prognostic Model for Malignant Prostate Cancer Patients Using Risk Gene Signatures Discovered through TCGA-PRAD Mining, Machine Learning, and Single-Cell RNA-Sequencing

• DOI: 10.3390/diagnostics13121997
• Publication Date: 2023-06-08T05:34:32Z
• Authors: Feng Lyu; Xianshu Gao; Mingwei Ma; Mu Xie; Shiyu Shang; Xueying Ren; Mingzhu Liu; Jiayan Chen
• Keywords: Nomogram; Lasso

Prostate cancer is a significant clinical issue, particularly for high Gleason score (GS) malignancy patients. Our study aimed to engineer and validate risk model based on the profiles of high-GS PCa patients early identification prediction prognosis.We conducted differential gene expression analysis patient samples from The Cancer Genome Atlas (TCGA) enriched our understanding functions. Using least absolute selection shrinkage operator (LASSO) regression, we established validated it using an independent dataset International Consortium (ICGC). Clinical variables were incorporated into nomogram predict overall survival (OS), machine learning was used explore factor characteristics' impact prognosis. prognostic confirmed various databases, including single-cell RNA-sequencing datasets (scRNA-seq), Cell Line Encyclopedia (CCLE), cell lines, tumor tissues.We identified 83 differentially expressed genes (DEGs). Furthermore, WASIR1, KRTAP5-1, TLX1, KIF4A, IQGAP3 determined be factors OS progression-free (PFS). Based these five factors, developed predicting PFS, with C-index 0.823 (95% CI, 0.766-0.881) 10-year area under curve (AUC) value 0.788 0.633-0.943). Additionally, 3-year AUC 0.759 when validating ICGC. KRTAP5-1 WASIR1 found most influential prognosis optimized model. Finally, interrelated immune infiltration, signals in cells scRNA-seq engineered original novel signatures through TCGA learning, providing new insights scarification practice.