Development and Validation of an α-Fetoprotein Tumor Burden Score Model to Predict Postrecurrence Survival among Patients with Hepatocellular Carcinoma

Abstract

Background: The purpose of this study is to establish a prognostic model to predict postrecurrence survival (PRS) probability after initial resection of hepatocellular carcinoma (HCC).

Study design: Patients with recurrent HCC after curative resection were identified through a multicenter consortium (training cohort, TC); data were from a separate institution were used as validation cohort (VC). The α-fetoprotein (AFP) tumor burden score (ATS) was defined as the distance from the origin on a 3-dimensional Cartesian coordinate system that incorporated 3 variables: largest tumor diameter ( x axis), number of tumors ( y axis), and ln AFP ( z axis). ATS was calculated using the Pythagorean theorem: ATS 2 = (largest tumor diameter) 2 + (number of tumors) 2 + (ln AFP) 2 , where ATS d and ATS r represent ATS at the time of initial diagnosis and at the time of recurrence, respectively. The final model was ATS m = ATS d + 4 × ATS r . Predictive performance and discrimination of the ATS model were evaluated and compared with traditional staging systems.

Results: The ATS model demonstrated strong predictive performance of PRS in both the TC (area under the curve [AUC] 0.70) and VC (AUC 0.71). An ATS-based nomogram was able to stratify patients accurately into low- and high-risk categories relative to PRS (TC: ATS m ≤ 27, 74.9 months vs. ATS m ≥ 28, 23.3 months; VC: ATS m ≤ 27, 59.4 months vs. ATS m ≥ 28, 15.1 months; both p < 0.001). The ATS model predicted PRS among patients undergoing curative or noncurative treatment of HCC recurrence (both p < 0.05). Of note, the ATS model outperformed the Barcelona Clinic Liver Cancer (BCLC), China Liver Cancer (CNLC), and American Joint Committee on Cancer (AJCC) staging systems relative to 1-, 2-, 3-, 4- and 5-year PRS (AUC 0.70, vs. BCLC, AUC 0.50, vs. CNLC, AUC 0.54, vs. AJCC, AUC 0.51).

Conclusions: The ATS model had excellent prognostic discriminatory power to stratify patients relative to PRS.