NHP2 and PRPF4 are hub genes associated with the prognosis of colorectal cancer.

Background: There is an urgent need to identify more accurate and practical prognostic markers for improving the diagnosis and treatment outcomes of colorectal cancer (CRC). This study aimed to identify hub genes of CRC and to evaluate their clinical significance.

Methods: Proteomics analysis was performed on human CRC tissues and adjacent normal tissues, with bioinformatic screening applied to identify hub genes. Differential mRNA/protein expression of the hub genes between CRC tissues and adjacent tissues was validated using public databases combined with RT-qPCR. Potential upstream regulators of hub genes were predicted through integrated analysis of the TF-target database, Starbase, and Comparative Toxicogenomics Database (CTD). Protein expression levels of the hub genes were assessed in 100 tissues by immunohistochemistry (IHC)with semi-quantitative scoring. Prognostic utility of the screened hub genes was evaluated through receiver operating characteristic (ROC) curve analysis and overall survival (OS) stratification.

Results: NHP2 (a small nucleolar ribonucleoprotein) and PRPF4 (pre-mRNA processing factor 4) were identified as two hub genes/proteins in CRC. Both genes exhibited significant upregulation in CRC tissues at transcriptional and translational levels compared with adjacent tissues. ROC analysis demonstrated strong diagnostic potential: NHP2 AUC (area under curve) = 0.819 (95% CI: 0.639-0.958; P < 0.001) and PRPF4 AUC = 0.917 (95% CI: 0.785-1.000; P < 0.001). NHP2 and PRPF4 shared common regulators and exhibited a positive correlation in their expression levels.Clinically, patients with dual high expression of NHP2 and PRPF4 showed markedly reduced prognosis versus single-high or dual-low groups.

Conclusions: Concurrent overexpression of NHP2 and PRPF4 is significantly associated with worse prognosis in CRC. Combined detection of NHP2 and PRPF4 is potentially a valuable prognostic marker of CRC compared to individual assessments. Our findings contribute to the growing body of evidence supporting multi-marker strategies for prognostic evaluation in CRC.