Expression pattern of secretory-cell-related transcriptional signatures in colon adenocarcinomas defines tumor microenvironment characteristics and correlates with clinical outcomes.

Despite the connection of secretory cells to distinct mucus-containing colon cancer histological subtypes and the interaction of secretory cells with immune cells in the pathogenesis of intestinal inflammatory diseases, whether the secretory cell signatures are associated with tumor microenvironment (TME) heterogeneity and can aid in colon cancer patient classification have not been investigated. Here, by performing the principal component analysis and consensus clustering analysis, we identified four distinct expression patterns based on secretory cell signatures which were significantly associated with different clinical behaviors, TME landscape, pathway activation, genomic mutations, and DNA methylation characteristics. Subsequently, a 'SCS score' model was constructed. The high SCS score indicated a pattern of 'secretory cell subtype 2', which was characterized by stromal infiltration and activation, and predicted poor prognosis and low sensitivity to fluorouracil-based chemotherapy and immunotherapy, but high sensitivity to PI3K catalytic subunit inhibitors. In conclusion, our study comprehensively uncovered the tumor heterogeneity related to secretory cell signature expression patterns. Moreover, the SCS score can supplement routine histopathological assessments to guide personalized therapeutic strategies in colon cancer patients.