Genomic landscape defines peritoneal metastatic pattern and related target of peritoneal metastasis in colorectal cancer

• DOI: 10.1002/ijc.35005
• Publication Date: 2024-05-13T13:33:40Z
• Authors: Minhui Hu; Rui Luo; Keli Yang; Yang Yu; Qiwen Pan; Mingming Yuan; Rongrong Chen; Hui Wang; Qiyuan Qin; Tenghui Ma; Huaiming Wang
• Keywords: Lasso; Primary tumor

The primary objective of this study is to develop a prediction model for peritoneal metastasis (PM) in colorectal cancer by integrating the genomic features cancer, along with clinicopathological features. Concurrently, we aim identify potential target implicated dissemination through bioinformatics exploration and experimental validation. By analyzing landscape from 363 metastatic patients, identified 22 differently distributed variables, which were used subsequent LASSO regression construct PM model. integrated established regression, incorporated two variables seven precisely discriminated cases (AUC 0.899; 95% CI 0.860-0.937) good calibration (Hosmer-Lemeshow test p = .147). Model validation yielded AUCs 0.898 (95% 0.896-0.899) 0.704 0.622-0.787) internally externally, respectively. Additionally, metastasis-related signature (PGS), was composed genes model, has prognostic stratification capability cancer. divergent drives driver PM. Bioinformatic analysis concerning these indicated SERINC1 may be associated Subsequent experiments indicate that knocking down functionally suppresses dissemination, emphasizing its importance CRCPM. In summary, defines pattern reveals