The prognostic model and immune landscape based on cancer-associated fibroblast features for patients with locally advanced rectal cancer

• DOI: 10.1016/j.heliyon.2024.e28673
• Publication Date: 2024-03-26T17:26:34Z
• Authors: Huajun Cai; Yijuan Lin; Yong Wu; Ye Wang; Shoufeng Li; Yiyi Zhang; Jinfu Zhuang; Xing Liu; Guoxian Guan
• Keywords: Nomogram; Gene signature; Infiltration (HVAC)

BackgroundThis study aimed to construct a nomogram based on CAF features predict the cancer-specific survival (CSS) rates of locally advanced rectal cancer (LARC) patients.MethodsThe EPIC algorithm was employed calculate proportion CAFs. differentially expressed genes between high and low subgroups, prognostic were identified via LASSO Cox regression analyses. They then used risk signature. Moreover, GSE39582 GGSE38832 datasets for external validation. Lastly, level immune infiltration evaluated using ssGSEA, ESTIMATE, CIBERSORTx, TIMER.ResultsA higher associated with worse prognosis. Additionally, number metastasized lymph nodes distant metastases, as well in subgroup. Five (SMOC2, TUBAL3, C2CD4A, MAP1B, BMP8A) subsequently incorporated into signature 1-, 3-, 5-year CSS training validation sets. Differences also determined cohort. Furthermore, independent factors, including TNM stage score, combined established nomogram. Notably, our results revealed that proportions macrophages neutrophils levels cytokines secreted by M2 high-risk Finally, significantly cell infiltration.ConclusionHerein, developed rate LARC patients. The model capable reflecting differences infiltration.