Abstract
BACKGROUND: SUMOylation is a critical post-translational modification that governs protein stability, subcellular localization, and signal transduction. Accumulating evidence suggests that dysregulated SUMOylation contributes to colorectal cancer (CRC) progression. However, its global impact on tumor microenvironment (TME) remodeling and clinical heterogeneity remains incompletely understood. The objective of this study was to establish a robust SUMOylation-related transcriptional signature to quantify SUMOylation activity in CRC and to elucidate its association with immune infiltration patterns, patient prognosis, and therapeutic responsiveness. METHODS: Using integrated transcriptomic profiles from The Cancer Genome Atlas (TCGA) and multiple Gene Expression Omnibus (GEO) cohorts (total n=1,226), we systematically curated SUMOylation-related genes (SRGs) and developed a SUMOylation-based scoring system (SUMOscore). Patients were stratified into high and low SUMOscore subgroups, followed by comprehensive evaluation of clinical outcomes, tumor staging, stromal components, and immune-cell infiltration landscapes. Potential responses to immune checkpoint blockade (ICB) and 5-fluorouracil (5-FU) chemotherapy were further inferred using established computational frameworks. RESULTS: Five key SRGs were identified to construct a scoring model that categorizes patients into high and low SUMOscore groups. The SUMOscore is an independent prognostic factor for CRC patients. Higher SUMOscore correlates with shorter overall survival (OS), advanced tumor staging, increased stromal infiltration, and lower tumor mutational burden. Further analysis suggests that CRC patients with lower SUMOscore might be more sensitive to immune checkpoint inhibitors and 5-FU chemotherapy. CONCLUSIONS: The SUMOscore captures clinically relevant TME heterogeneity in CRC and may help guide selection of immunotherapy and adjuvant chemotherapy.