Low-grade hepatocellular carcinoma characteristics, a practical nomogram and risk stratification system: a SEER population-based study.

Background: The purpose of this study is to establish a nomogram and risk stratification system to predict OS in patients with low-grade HCC.

Research Design And Methods: Data were extracted from the SEER database. C-index, time-dependent AUCs, and calibration plots were used to evaluate the effective performance of the nomogram. NRI, IDI, and DCA curves were adopted to compare the clinical utility of nomogram with AJCC.

Results: 3415 patients with low-grade HCC were available. The C-indices for the training and validation cohorts were 0.773 and 0.772. The time-dependent AUCs in the training cohort were 0.821, 0.817, and 0.846 at 1, 3 and 5 years. Calibration plots for 1-, 3- and 5-year OS showed good consistency between actual observations and that predicted by the nomogram. The values of NRI at 1, 3, and 5 years were 0.37, 0.66, and 0.64. The IDI values at 1, 3, and 5 years were 0.11, 0.16, and 0.23 (< 0.001). DCA curves demonstrated that the nomogram showed better ability of predicting 1-, 3-, and 5-year OS probabilities than AJCC.

Conclusions: A nomogram and risk stratification system for predicting OS in patients with low-grade HCC were established and validated.