Assessing prognostic significance of preoperative alpha-fetoprotein in hepatitis B-associated hepatocellular carcinoma: normal is not the new normal.

Background: Hepatitis B (HBV)-associated hepatocellular carcinoma (HCC) is often associated with alpha-fetoprotein (AFP) production. Although serum AFP has been demonstrated to be a prognostic factor for patient survival, optimal cutoff levels remain unclear.

Methods: Patients with HBV-associated HCC treated by primary liver resection were prospectively followed at a single institution between 1995 and 2008. AFP level was categorized into quintiles for Kaplan–Meier analysis and multivariable Cox proportional hazards regression models.

Results: Best 5-year survival after surgery was observed for patients with AFP in the first quintile (1.4-4.1 ng/mL), with progressively worse outcomes for patients in each increasing quintile. AFP was associated with overall survival (HR = 1.61; 95 % CI 1.30-1.98), disease-free survival (HR = 1.26; 95 % CI 1.10-1.44), and 2-year recurrence (HR = 1.30; 95 % CI 1.07-1.57) in multivariate analysis. Noncirrhotic patients (Ishak 1-5) with AFP in quintile 1 had 94 % 5-year survival, compared with 0 % survival for patients with AFP in quintile 5 (2,332.7-327,560.0 ng/mL) and Ishak stage 6 cirrhosis.

Conclusions: Preoperative serum AFP is an independent predictor of prognosis among HBV-HCC patients following surgical resection. Categorizing AFP into quintiles creates the opportunity to observe differences in outcomes even at low serum levels within the normal range. Additionally, combining AFP quintiles and fibrosis staging provides a predictive model of prognosis for HCC. Thus, even small differences in AFP within the normal range may impact prognosis and disease progression for HBV-HCC.