Development and validation of a prognostic nomogram to predict the recurrence of AFP-negative and DCP-positive hepatocellular carcinoma after curative resection.

PURPOSE

Approximately one-third of hepatocellular carcinoma (HCC) cases are characterized by alpha-fetoprotein (AFP) negativity (AFP-NHCC. Among these patients, around 60% exhibit des-gamma-carboxyprothrombin (DCP) positivity, and DCP-positive patients have a poorer prognosis. As a curative treatment, recurrence after liver resection poses significant challenges to the prognosis of HCC patients. Therefore, it is necessary to determine the relevant risk factors of these patients and provide timely treatment options.

METHODS

This study included 540 patients who underwent resection at Beijing You'an Hospital. 292 patients from 2014 to 2018 constituted the training cohort, while 248 patients from 2018 to 2020 constituted the validation cohort. All patients underwent routine follow-ups until December 2023. Variables were identified through Cox regression, and a nomogram was developed. The nomogram was evaluated using time-dependent receiver operating characteristic curves (ROC), calibration curves, Decision curve analysis (DCA), and Kaplan-Meier (KM) curve analysis.

RESULTS

We found that age, tumor number, tumor size, γ-glutamyl transpeptidase (γ-GT), and prothrombin time (PT) are independent risk factors for HCC recurrence, and a nomogram was developed and validated based on this result to predict recurrence-free survival (RFS) at 1, 2, and 3 years. The performance of the nomogram was further confirmed by the ROC curve, calibration curve, and DCA, all of which showed favorable results. The KM curve analysis clearly distinguishes between two groups of people with different risks in terms of prognosis in both the training and validation sets.

CONCLUSION

In summary, we established and validated a novel nomogram by multivariate Cox regression analysis to predict recurrence in DCP-positive patients with AFP-NHCC after resection. The nomogram, including age, tumor number, tumor size, γ-GT, and PT, demonstrates better predictive ability for AFP-NHCC patients with DCP positive.