Background <p>Endometrial cancer (EC), a leading gynecological malignancy, exhibits heterogeneous progression and limited predictive biomarkers for personalized management. Stratifin (SFN), a cell-cycle regulatory protein, has been implicated in oncogenesis, but its role in EC progression and prognosis remains undefined. This study integrates multi-omics, immunohistochemistry, and survival modeling to establish Stratifin as a novel biomarker for EC.</p> Methods <p>Differentially expressed genes (DEGs) in uterine corpus endometrial cancer (UCEC) were identified by meta-analysis of GSE17025 (<i>n</i> = 147 UCEC vs. 14 normal) and GSE63678 (<i>n</i> = 150 UCEC vs. 10 normal) datasets using limma-voom (FDR &lt; 0.05, |log₂FC| &gt;1). Protein-protein interaction (PPI) networks were constructed via STRING (confidence score &gt; 0.7) and visualized in Cytoscape, with densely connected modules identified using the MCODE plugin (MCODE score &gt; 5). Kaplan-Meier survival curves and log-rank tests were performed on TCGA-UCEC data (<i>n</i> = 586) to evaluate associations between hub gene expression and overall survival (OS). Immunohistochemical (IHC) staining for Stratifin was conducted on tissue microarrays containing 36 normal endometrium, 37 atypical hyperplasia, and 104 endometrioid endometrial carcinoma (EEC) samples, scored by intensity (0–3) and proportion (0–100%), with high expression defined as combined score ≥ 6. The cutoff value was determined based on receiver operating characteristic (ROC) curve analysis to optimize sensitivity and specificity for predicting clinical outcomes. Clinicopathological correlations were analyzed via χ² tests, while univariate/multivariate Cox proportional hazards models assessed independent prognostic value. Immune cell infiltration data from TIMER 2.0 were analyzed using Spearman’s rank correlation.</p> Results <p>A total of 136 common DEGs (58 upregulated, 78 downregulated) were identified, with Stratifin (SFN) localized to a PPI module enriched for cell-cycle regulation (FDR &lt; 10⁻⁵) and mitotic spindle organization. High Stratifin expression in TCGA-UCEC was associated with shorter OS (HR = 1.87, 95% CI = 1.24–2.83, log-rank <i>P</i> = 0.003), consistent with IHC findings showing increased Stratifin positivity from normal endometrium (27.78%) to atypical hyperplasia (40.54%) and EEC (68.27%, <i>P</i> &lt; 0.001). Clinicopathologically, Stratifin expression correlated with early pathological stage (Stage I–II: 85.06% high expression vs. III-IV: 36.36%, χ²=10.217, <i>P</i> &lt; 0.001), but not histological grade (<i>P</i> = 0.639). In the multivariate Cox analysis, high Stratifin expression was coded as the reference group (coded as 0), while low expression was coded as 1, resulting in an HR of 0.043 (95% CI = 0.001–1.128, <i>P</i> = 0.045). This indicates that low Stratifin expression has a protective effect with 95.7% risk reduction compared to high expression, confirming Stratifin as an independent predictor of poor OS when highly expressed, alongside histological grade (HR = 15.682, <i>P</i> = 0.019). Immune profiling revealed negative correlation with CD8 + T-cell infiltration (Spearman’s ρ=-0.251, <i>P</i> = 0.018) and positive correlation with neutrophil infiltration (ρ = 0.273, <i>P</i> = 0.010), suggesting immunosuppressive microenvironment modulation.</p> Conclusion <p>Stratifin is a stage-specific biomarker for EC with dual roles in prognosis prediction and immune microenvironment regulation. Its independent association with OS, combined with immunohistochemical validation, supports integration into risk stratification models.</p>