Published online Jan 27, 2018. doi: 10.4254/wjh.v10.i1.41
Peer-review started: November 1, 2017
First decision: December 1, 2017
Revised: December 7, 2017
Accepted: January 15, 2018
Article in press: January 15, 2018
Published online: January 27, 2018
Processing time: 86 Days and 9.5 Hours
The incidence of Hepatocellular carcinoma (HCC) has been increasing during the last years and is the second leading cause of cancer related death worldwide. The cause of HCC is closely linked to the prevalence of chronic hepatitis C (HCV) or B virus (HBV) infections as well as the prevalence of alcoholic liver disease. With the advent of the new direct antiviral drugs (DAAs) for hepatitis C (HCV) treatment, epidemiological changes have been already reported in the natural history of liver disease. One of these epidemiological changes in developed countries is due to the stepwise increase in the incidence of non-alcoholic fatty liver disease (NAFLD), an emergent cause of end stage liver disease and HCC.
Indeed, it is estimated that during the next 20 years, non-alcoholic steatohepatitis (NASH), the progressive form of NAFLD, will constitute the most frequent cause of cirrhosis and HCC in developed countries. However, reports regarding epidemiological changes of HCC from developing areas, including South America, have been heterogeneous and none have focused on changing trends in etiologies of HCC over time.
Our aim was to evaluate changes in the etiology of HCC in Argentina during the last seven years, particularly focusing on potential changes associated with NAFLD-HCC.
This cohort study was conducted between January 1 2009 and January 1 2016 in 14 regional hospitals from Argentina. Criteria for inclusion required patients with newly diagnosed HCC as recommended by international guidelines. Etiologies of HCC included viral hepatitis, alcoholic liver disease (alcohol intake exceeding 30 g/d), NAFLD, cryptogenic cirrhosis (CC), cholestatic liver diseases (i.e., primary biliary cholangitis, primary and secondary sclerosing cholangitis), autoimmune hepatitis and other causes including metabolic diseases or miscellaneous causes (e.g., hereditary hemochromatosis, Wilson disease, toxic liver disease). NAFLD diagnosis was established on histological ground or clinically according to international guidelines. Baseline patient and tumor characteristics at HCC diagnosis were recorded. Tumor burden was classified according to Barcelona Clinic Liver Cancer staging (BCLC criteria). Complete patient follow-up was assessed in all included subjects from HCC diagnosis until death or last medical visit.
Categorical data were compared using Fisher’s exact test (2-tailed) or Chi-Square test and continuous variables were compared with Student’s t-test or Mann-Whitney U test according to their distribution, respectively. For survival analysis, Cox regression multivariate analysis estimating hazard ratios (HR) and 95%CI for baseline variables related with 5-year mortality was performed. Kaplan Meier survival curves were compared using the log-rank test.
A total of 708 consecutive adult patients with newly diagnosed HCC from 14 centers were included. Out of 14 hospitals, 6 were LT centers, which contributed with the follow-up of 484 patients (68.4% of the study cohort). Non-cirrhotic patients accounted for 12.6% of the included cohort (n = 89). The prevalence of diabetes mellitus (DBT) was 27.7%. Between 2009 and 2016, HCV related HCC was the most frequent etiology along the whole observation period. No significant changes were observed in the proportion of HCV-HCC. The second most frequent cause of HCC was alcoholic liver disease, remaining stable along the observation period. On the other hand, a striking 6-fold increase in the proportion of HCC-NAFLD cases was observed since 2009 to 2016 (4.3% vs 25.6%; P < 0.0001). NAFLD was the second cause of HCC in 2015 together with alcoholic liver disease. In addition, when patients with cryptogenic cirrhosis/liver disease and DBT (n = 22/68) were considered together as potentially metabolic syndrome/NAFLD, the NAFLD/Cryptogenic + DBT group represented 14.5% (n = 103) of the entire cohort and increased from 8.6% in 2009 to 16.2% in 2014 and 25.6% in 2016, respectively (P = 0.014). There was a higher prevalence of diabetes mellitus (DBT) (61.7% vs non-NAFLD 23.3%; P < 0.0001) in the NAFLD group. The increasing proportion of NAFLD in the overall cohort was not in parallel with an increase of DBT in this population. On the other hand, lower median AFP values at HCC diagnosis were observed between NAFLD-HCC and non-NAFLD groups (6.6 ng/mL vs 26 ng/mL; P = 0.02). Neither NAFLD nor other HCC etiologies were associated with higher mortality.
To the best of our knowledge this is the first multicenter cohort study from a South American country evaluating longitudinal changes in etiologic trends. This changing etiologic scenario documented in developed countries is replicated in developing ones. The documented increase in NAFLD-HCC in our study including a developing country entails in itself a challenging impact on Public Health worth to be considered. HCV infection was the main etiology of HCC throughout the entire observation period, including the last two years, at which time DAA’s therapy was available in Argentina. Second, as reported in other regions of the world, fatty liver disease was observed as an increasing cause of HCC. The prevalence of DBT was significantly higher in NAFLD-HCC when compared to non-NAFLD patients. Finally, NAFLD-HCC was not associated with an increasing risk of mortality, adjusted for the presence HCC surveillance and BCLC stage. NAFLD has been related to an increasing rate of overall cardiovascular morbid-mortality. Consequently, NAFLD will become responsible of an increase in medical resource use in the next years that demands specific health prevention programs focusing in diet and exercise in order to avoid not only liver but also cardiovascular disease development. As previously mentioned, a six-fold increase in NAFLD from 2009 to 2015 was observed in our cohort, mainly in LT centers whereas alcoholic liver disease was a leading cause of HCC in non-LT centers.
NAFLD related HCC has been recognized as a growing burden in the United States and some European regions. This etiologic scenario is not only changing in high-income countries but also it might be happening in developing ones. In Argentina, even though HCV infection is still the main cause of HCC, recent changing trends in etiology of HCC in Argentina suggests that NAFLD might be the leading HCC cause in the next years, becoming an important Public Health issue.