Identification Of Novel Molecular Typing Of Colon Cancer And Application Strategies Of Clinical Diagnosis And Treatment Based On The Enrichment Feature Of Secretory Cell Gene Signatures

Background and purposeThe disruption of the balance between cell proliferation,differentiation,and apoptosis in the intestinal mucosa,leading to the dysregulation of the mucosal steady-state environment,is considered an early hallmark of colon cancer.Within the diverse cell types present in the colonic mucosa,secretory cells are a principal population of differentiated cells,primarily comprising goblet cells and enteroendocrine cells.Despite the close association of secretory cells with the mucinous colorectal cancer histological subtype and their interaction with immune cells in the pathogenesis of intestinal inflammatory diseases,whether the enrichment feature of secretory cells is associated with the heterogeneity of the tumor microenvironment(TME)and whether it is helpful for colon cancer patient classification and prognosis prediction have not been thoroughly investigated.Therefore,the main objectives of this study are:(1)to analyze the relationship between the enrichment level of different secretory cell gene signatures and TME,biological pathways,and clinical prognosis in colon cancer tissues;(2)to identify colon cancer secretory cell subtypes based on the enrichment feature of secretory cell gene signatures;(3)to establish a novel scoring model for quantifying secretory cell subtypes and explore its clinical significance;(4)to investigate the molecular mechanisms of chemotherapy resistance in colon cancer patients with specific secretory cell subtypes and their potential therapeutic strategies.Methods and results1)The enrichment levels of different secretory cell gene signatures were obtained in colorectal cancer using PCA algorithm,and their associations with tumor microenvironment,biological pathways,and clinical prognosis were analyzed.There was a significant positive correlation between the enrichment levels of enteroendocrine cells and subtype 2 goblet cells gene signatures and the infiltration of stromal cells,activation of stromal pathways,chemotherapy resistance,and poor prognosis.2)Four robust secretory cell subtypes(SCS1,SCS2,SCS3,and SCS4)were identified in colorectal cancer through clustering analysis.SCS2 subtype was mainly characterized by infiltration of stromal cells and activation of stromal-related pathways,and it had a poor prognosis and could not benefit from adjuvant chemotherapy.3)A new scoring model,the secretory cell subtype(SCS)score,was developed using machine learning algorithms,which could quantify the secretory cell subtypes.SCS score,comprised of 7 key genes,effectively distinguished SCS2 subtype patients and was a powerful tool for predicting prognosis and adjuvant chemotherapy response in patients with colon cancer.4)By using bioinformatics analysis,7 potential genes regulating chemotherapy resistance in SCS2 subtype were identified by screening the fluorouracil-resistant genes selected by CRISPR silencing library and the up-regulated gene sets in SCS2 subtype,and it was found that PI3K inhibitor could be a potential therapeutic drug for SCS2 subtype colorectal cancer patients.ConclusionThis study identified four robust secretory cell subtypes in patients with colon cancer,referred to as secretory cell subtypes(SCS subtypes).Additionally,a quantitative scoring model for secretory cell subtyping(SCS score)was established.A high SCS score could distinguish the SCS2 subtype,which is characterized by stromal cell infiltration and activation of stroma-related pathways,indicating poor prognosis and chemotherapy resistance,but high sensitivity to PI3K inhibitors.Our study suggests that SCS score may become a supplementary method for TNM staging,assisting in prognosis prediction and guiding chemotherapy decisions.