Long-term prospective study of development of hepatocellular carcinoma in compensated cirrhosis

Abstract

Background and Aims: Hepatocellular carcinoma (HCC) is the 5th incident solid tumor in males and the 2nd for mortality. With the aim of identifying clinical/biologic risk factors for HCC development, we started in 2013 a prospective study in patients with liver cirrhosis undergoing hepatic venous pressure gradient (HVPG) measurement (ClinicalTrials.gov:NCT03083002). Method: 445 consecutive patients with liver cirrhosis prospectively enrolled from July 2013 (66.5% CP-A, 23.2% CP-B, 10.3% CP-C) when undergoing HVPG/transjugular liver biopsy, were followed-up every 6 months with US and blood tests. Etiology, portal vein thrombosis, HVPG,esophageal varices, Child-Pugh,MELD Score, time to HCC development were recorded. Incident cases of HCC were biopsied for pathological and transcriptomic characterization. Independent risk factors for HCC were evaluated by Cox regression analysis. Results: Median follow-up was 40 months. 61 patients died during follow-up, 34 developed HCC (incidence 4-5% per year). Preliminary results of analysis of hepatic and circulating biomarkers of angiogenesis, portal hypertension and fibrosis as predictive factors for HCC development indicate marked activation of angiogenesis as related with HCC risk. At univariate analysis HVPG>15 (but not HVPG>10 or >20 mmHg), F2/F3 esophageal varices, viral vs. non-viral etiology,and albumin were statistically associated with HCC development (HVPG/F2-F3 varices collinear). In HVPG>15 multivariate model, none of these factors was significantly associated with HCC development while in F2/F3 esophageal varices model,the latter were independently linked with it (HR 2.258,95% CI:1.135-4.494). Albumin only had borderline significance (.586, CI% .337-1.018). Conclusion: Neither HVPG>10,>15 or >20 was independently associated with HCC development. Previous data indicating HVPG>10 as a significant risk factor for HCC were influenced by the cohort studied, which was free of varices at enrollment. In a cohort of compensated patients with liver cirrhosis not selected for absence of varices, more severe portal hypertension as indicated by F2/F3 varices was the only independent risk factor for HCC development.