Proteomics analysis identified TPI1 as a novel biomarker for predicting recurrence of intrahepatic cholangiocarcinoma.

BACKGROUND: Intrahepatic cholangiocarcinoma (ICC) is the second most common tumor in primary liver cancer, but the prognostic factors associated with long-term outcomes after surgical resection remain poorly defined. This study aimed to develop a novel prognostic classifier for patients with ICC after surgery. METHODS: Using a proteomics approach, we screened tumor markers that up-regulated in ICC tissues, and narrowed down by bioinformatics analysis, western blot and immunohistochemistry. Prognostic markers were identified using Cox regression analyses in primary training cohort and the predictive models for time to recurrence (TTR) were established. The predictive accuracy of predictive model was validated in external validation cohort and prospective validation cohort. MTT assay, clonal formation assay and trans-well assays were used to verify the effect on the proliferation and migration in ICC cell line. RESULTS: Triosephosphate isomerise (TPI1) was significantly up-regulated in ICC tissues and Kaplan-Meier analysis reveals that higher TPI1 expression was strongly correlated with higher recurrence rate of ICC patients. In the primary training cohort, mean TTR was significantly longer (p < 0.0001) than in the low-risk group (26.9 months for TTR, 95% CI 22.4-31.5) than in the high-risk group (14.5 months for TTR, 95% CI 10.6-18.4). Similar results were observed in two validation cohorts. In addition, a nomogram to predict recurrence was developed. Moreover, Knockdown of TPI1 by shRNA inhibited ICC cell growth, colony information, migration, invasion in vitro. CONCLUSIONS: Current prognostic models were accurate in predicting recurrence for ICC patients after surgical resection.