Purpose: Hepatocellular carcinoma (HCC) is the third leading cause of cancer mortality worldwide. A few risk scores have been validated for prognostication of HCC. However, the role of MELD-Na in survival of these particular patient population remains largely unclear. We sought to investigate the role of MELD-Na in survival of patients with HCC. Methods: One hundred and four patients with biopsy or radiographic evidence of HCC (Age: 62+/-8, male 99%;white: 51%, African American 48%, Hispanic 1%, Hepatitis C: 79%; documented cirrhosis: 88%, biopsy proven HCC: 46%, radiographic evidence of HCC 54%). This was a retrospective study of patients at the Memphis VA Medical Center from 2006-2012. There was a mean follow-up of 620+/-534 days. During follow-up, 52% of patients died. Patients were treated with radiofrequency ablation, chemoembolization, resection, sorafenib or liver transplantation. Demographics, radiographic tumor characteristics, and biochemical studies at the time of diagnosis were collected. Different treatment modalities were also investigated in all patients. Student t test and Chi-square tests were used where appropriate for univariate analysis. Cox-regression analysis were performed to determine independent predictors of survival. Results: Univariate tumor characteristics predictive of survival included: number of tumors >2 (p=0.001), size of the largest lesion > 2.7 cm (P=0.04) and bi-lobal involvement (p< 0.001). Biochemical markers MELD-Na (p<0.001), alpha-fetoprotein (P<0.001), alkaline phosphatase (p=0.002) were univariate predictors of survival. Among clinical factors only the presence of portal vein thrombosis (PVT, p<0.001) was a univariate predictor of survival. Ascites approached near significance (p=0.06). Between treatment modalities, liver transplantation was the best predictor of survival (p=0.02). Cox regression analysis revealed MELD-Na, number of tumors, PVT, and transplantation (P values=0.002, 0.004, 0.002, and 0.01 respectively) as independent predictors. Trans-arterial chemoembolization was only predictive of survival in the first year after diagnosis. Conclusion: In this series, several clinical and biochemical factors including MELD-Na, Portal vein thrombosis, number of tumors, and liver transplantation can be used to predict the clinical course of patients with HCC. Chemoembolization predicted survival only during the first year after diagnosis.
Read full abstract