Serum Midkine as a Biomarker for Hepatocellular Carcinoma Treatment Response and Prognostication

BACKGROUND: Up-to-date tumor markers are considered more helpful in the evaluation of treatment response and prognostication than in diagnosis. Identification of a valid biomarker to help clinical management of hepatocellular carcinoma (HCC) is still a challenge, especially in alpha-fetoprotein (AFP) negative HCC. This study evaluated the efficacy of serum Midkine in HCC treatment response and prognostication as well as its association with clinical pathological features. METHODS: The enzyme-linked immunosorbent assay (ELISA) was used to analyze the serum (soluble) Midkine levels in patients with benign and malignant liver disease and healthy subjects. RESULTS: Serum Midkine levels in HCC, hepatic cholangiocarcinoma (HC), chronic hepatitis B (CHB), and decompensated hepatic cirrhosis (DHC) patients were significantly higher than in healthy subjects and patients with benign hepatic tumors, with HCC being the highest. Serum Midkine levels were associated with patients' clinical pathologic features such as tumor size, ascites, high Child-Pugh and BCLC stages, and advanced clinical stages. Serum Midkine was helpful in characterizing the AFP negative HCC patients. A model has been created using univariate and multivariate regression analysis of biomarkers in predicting HCC, which showed that Midkine was one of the key predictive factors in AFP negative HCC. Midkine levels as a parameter for treatment response evaluation and disease progression/prognostication are comparable to computed tomography scan (CT) and magnetic resonance imaging (MRI), especially in non-surgical treatment patients. CONCLUSION: Serum Midkine levels could be used as a parameter for evaluation of treatment efficacy and prognostication of HCC, especially in AFP-negative HCC.