Hepatocellular carcinoma: risk groups, surveillance and outcome

Abstract

The burden of hepatocellular carcinoma (HCC) has changed in the past few decades. Although the majority of HCC cases develops in East Asia and Sub-Saharan Africa, HCC has become an increasing problem in Western countries such as the Netherlands. Surveillance for HCC is controversial because of limited high-level evidence for its efficiency. Nevertheless, current guidelines recommend surveillance in high-risk patients, mainly patients with cirrhosis. However, selection of the target population remains a subject of debate. In the first part of this thesis we focused on high-risk patients for HCC development and HCC surveillance. We demonstrated that, even in case of cirrhosis, HCC risk in patients with Wilson disease is low. Our data do therefore not support regular HCC surveillance in patients with Wilson disease in absence of other risk factors. Furthermore, we investigated the contribution of surveillance in HCC patients in “real life” clinical practice in the Netherlands: surveillance was associated with smaller tumor size, earlier tumor stage, with impact on therapeutic strategy. Participation in an HCC surveillance program was an independent predictor of survival. However, there was a profound discrepancy between patients’ expectations of HCC surveillance and the real benefits in daily clinical practice. Absence of cirrhosis in HCC patients is the focus of the second part of this thesis. In the Netherlands, approximately 20% of HCC patients had no underlying cirrhosis. Absence of cirrhosis was strongly associated with female gender and presence of non-alcoholic fatty liver disease (NAFLD) or no risk factors of underlying liver disease. In non-cirrhotic HCC patients, resections were more often performed, with better overall survival despite larger tumor size compared to HCC patients with cirrhosis. Also, we demonstrated that histological steatosis was frequently present in the non-tumoral liver of non-cirrhotic HCC patients, both in patients with and without known causes of underlying liver disease. The high prevalence of steatosis in the non-tumoral liver supports the hypothesis that hepatic steatosis may play a causal role in hepatocarcinogenesis, even in absence of steatohepatitis and may be a cofactor in patients with other well-known risk factors for underlying liver disease. Since overt steatohepatitis was surprisingly rare in non-cirrhotic HCC patients, current paradigmas on HCC pathogenesis in NAFLD may be reconsidered. In the final part of this thesis we discuss presence and treatment of HCC in the Netherlands. In the past decade, there was no trend towards centralization of diagnostic testing and treatments for HCC in the Netherlands, in contrast to the recommendation of the Dutch guideline. Furthermore, significant differences in the use of locoregional therapy for HCC existed between regions. Finally, we demonstrated that in low incidence countries such as the Netherlands, outcome after resection or sorafenib for HCC may differ between various hospital types (university versus non-university hospitals).