ObjectiveTo detect the expression of centromere protein F (CENPF) and programmed death-ligand 1 (PD-L1) in colorectal cancer (CRC) tissues, analyze their associations with clinicopathologic characteristics, and assess their correlation and prognostic significance. MethodsThe CRC surgical specimens and their matched adjacent normal tissues, archived at the Department of Pathology, the Fifth Clinical Medical College of Henan University of Chinese Medicine (Zhengzhou People’s Hospital) from January 2019 to January 2023, were retrospectively collected. The expression levels of CENPF and PD-L1 were detected by immunohistochemistry. The positive unit (PU) of CENPF was measured using Image-Pro Plus image analysis software, and the combined positive score (CPS) of PD-L1 was calculated. The associations of CENPF and PD-L1 expression with clinicopathologic parameters were analyzed, and the correlation between CENPF and PD-L1 expression was evaluated using Spearman rank correlation coefficient. Furthermore, disease-free survival curves were plotted using the Kaplan-Meier method. Multivariable Cox proportional hazards regression models were further constructed to evaluate the independent associations of CENPF and PD-L1 expression with disease-free survival in CRC patients. ResultsA total of 122 patients were enrolled in this study. Significantly higher expression levels of both CENPF and PD-L1 were observed in CRC tissues as compared with their matched adjacent normal tissues (11.75±4.83 vs. 3.77±0.85, P<0.001; 13.65±10.58 vs. 0.88±0.60, P<0.001). A positive correlation between CENPF and PD-L1 expression was identified by Spearman analysis (rs=0.534, P<0.001). CENPF and PD-L1 expression were associated with multiple adverse clinicopathologic features of CRC. Both biomarkers were elevated in patients with advanced TNM stage (stage Ⅲ?Ⅳ), lymph node metastasis, and moderate-to-high-grade tumor budding (For CENPF: F=37.067, P<0.001; t=?4.693, P<0.001; t=?2.527, P=0.013, respectively. For PD-L1: H=11.080, P=0.004; t=2.057, P=0.042; t=?3.765, P<0.001, respectively). However, their associations with tumor differentiation and lymphovascular invasion differed: CENPF expression was higher in poorly differentiated CRC and those with lymphovascular invasion (F=8.006, P=0.001; t=2.361, P=0.020), whereas PD-L1 expression showed no significant association with these features (H=5.380, P=0.068; t=1.553, P=0.123). Survival analysis was performed on the 122 CRC cases (73 with high and 49 with low CENPF expression; 66 with high and 56 with low PD-L1 expression). Lower disease-free survival rate was observed in patients with high CENPF or high PD-L1 expression as compared to those with low expression (P<0.001, P=0.016). Further analysis based on combined expression patterns (54 cases with dual-high expression, 37 with dual-low expression, 19 with CENPF-high expression/PD-L1-low expression, and 12 with CENPF-low expression/PD-L1-high expression) revealed that the dual-high expression group had the lower disease-free survival rate (vs. dual-low expression and CENPF-low expression/PD-L1-high expression groups, both P<0.001). Additionally, significantly lower disease-free survival rate was detected in the CENPF-high expression/PD-L1-low expression group than in the CENPF-low expression/PD-L1-high expression and dual-low expression groups (all P<0.001). Finally, multivariate Cox proportional hazards regression analysis demonstrated that poor tumor differentiation, positive lymphovascular invasion, and high CENPF expression were identified as independent risk factors for shortened disease-free survival [HR (95%CI)=2.212 (1.106, 4.423), 3.205 (1.410, 7.285), and 12.295 (2.568, 58.855), respectively], whereas PD-L1 expression was not confirmed as an independent risk factors for shortened disease-free survival [HR (95%CI)=0.879 (0.462, 1.672)]. ConclusionsCENPF and PD-L1 are upregulated and positively correlated in CRC, potentially contributing to tumor progression and immune evasion. Combined detection of CENPF and PD-L1 may aid in risk stratification and immunotherapy strategy selection for CRC.