Real impact of liver cirrhosis on the development of hepatocellular carcinoma in various liver diseases—meta‐analytic assessment

Cancer Med. 2019 Feb 21. doi: 10.1002/cam4.1998. [Epub ahead of print]
Real impact of liver cirrhosis on the development of hepatocellular carcinoma in various liver diseases-meta-analytic assessment.
Tarao K1, Nozaki A2, Ikeda T3, Sato A4, Komatsu H5, Komatsu T6, Taguri M7, Tanaka K8.

Abstract BACKGROUND:
It is well known that the incidence of developing hepatocelluler carcinoma (HCC) is increased in liver cirrhosis of different etiologies. However, comparison of HCC incidence in various liver diseases has not yet been estimated. We surveyed this comparison.

METHODS: The PubMed database was examined (1989-2017) for studies published in English language regarding the prospective follow-up results for the development of HCC in various liver diseases. A meta-analysis was performed for each liver disease.

RESULTS: The annual incidence (%) of HCC in the non-cirrhotic stage and cirrhotic stage, and the ratio of HCC incidence in the cirrhotic stage/non-cirrhotic stage were as follows. (a) hepatitis B virus liver disease: 0.37%→3.23% (8.73-fold), (b) hepatitis C virus liver diseases: 0.68%→4.81% (7.07-fold), (c) primary biliary cholangitis (0.26%→1.79%, 6.88-fold), (d) autoimmune hepatitis (0.19%→0.53%, 2.79-fold), and (e) NASH (0.03%→1.35%, 45.00-fold). Regarding primary hemochromatosis and alcoholic liver diseases, only follow-up studies in the cirrhotic stage were presented, 1.20% and 2.06%, respectively.

CONCLUSIONS: When the liver diseases advance to cirrhosis, the incidence of HCC is markedly increased. The development of HCC must be closely monitored by ultrasonography, magnetic resonance imaging, and computed tomography, irrespective of the different kinds of liver diseases.

DISCUSSION
--Download Full-text--
It is known that cirrhosis is present in 80～90% of HCC patients with any underlying liver disease,119 and it is the most important risk factor for HCC. However, comparison of the incidence of HCC in various liver diseases was not accurately and precisely evaluated in previous studies. In this study, we found that the incidence of HCC is highest in HCV LC (4.81%/year) and second highest in HBV LC (3.23%), followed by alcoholic LC (2.06%), PBC LC (1.79%), NASH LC (1.35%), primary hemochromatosis (1.20%), and AIH (0.53%).
        
The incidence of HCC has been wildly studied in patients with HBV LC and HCV LC, and was reported to be 3% and 5%,29, 120 which was almost the same as that in our study.

In this study, we also demonstrated that the incidence of HCC is markedly increased (2.79‐ to 45.00‐fold) in the cirrhotic state compared with non‐cirrhotic state, irrespective of the etiology of liver disease. Why this increase in HCC development occurs in the cirrhotic state must be considered.

First, chronic inflammation may be a key mechanism for HCC development in the cirrhotic state. In this respect, we made clinical observation in the HCV LC patients (Child A) in the past. The LC patients were divided into three groups: Group A: 28 patients whose annual average serum alanine aminotransferase (ALT) level was persistently high (≧80 IU; high‐ALT group); Group B: 28 patients whose annual average serum ALT levels was persistently low (<80 IU; low‐ALT group), and Group C: 13 unclassified patients. The patients had been followed up prospectively. The 5‐year incidence rate of HCC in the high‐ALT group was as high as 53.6% compared with only 7.1% in the low‐ALT group (P < 0.001).120 Thus, this clinical observation demonstrated the importance of chronic inflammation in the development of HCC in HCV LC.
        
The same tendency was demonstrated in HBV LC. Chen et al121 reported that high serum levels of HBV DNA and ALT were independent risk factors for HCC development in HBV infection. Ishikawa et al20 also reported that serum aminotransferase are one of predictive factor for the development of HCC. One important mechanism for high incidence of HCC in HCV LC and HBV LC as compared with the incidence of HCC in LC caused by other etiologies may be the sustained high‐grade inflammation.
        
Furthermore, in autoimmune hepatitis, persistent elevation of serum aminotransaminase was reported to lead to the development of HCC.78 This suggests a role for inflammation in the development of HCC, but this hypothesis has not been confirmed solidly.
         
Second, increases in DNA synthesis in the hepatocytes of cirrhotic patients is suspected as a possible mechanism of HCC development. In our previous study, we demonstrated that in the high DNA synthetic group [BrdU (thymidine analog) labeling index ≧1.5%] 64.3% of patients developed HCC in 5 years, in contrast to 14.3% in the low DNA synthesis (Brd U LI <1.5%) group in HCV LC patients (P < 0.05).122 We further demonstrated that in HCV‐associated cirrhosis patients who showed nodular formation on ultrasound, the cell cycle of hepatocytes was markedly accelerated, as estimated by the bromodeoxyuridine (thymidine analog) uptake into hepatocytes in vitro, and the incidence of HCC was greatly increased.123

Third, accumulated genomic change was also important factor for HCC development. In this respect, Hiatt et al124 reported that C282Y mutation itself may increase the risk of HCC development in hereditary hemochromatosis. In addition, in alcoholic LC, the genetic effect of long‐term alcoholic intake may influence the development of HCC.125

Fourth, obesity, and diabetes are suspected to increase the incidence of developing HCC. In a large epidemiologic study, patients with BMI >35 had an increased risk of cancer, especially HCC, with hazard ratio (HR) of 4.52.126 In addition, diabetes is associated with HCC occurrence, with a HR of 2.39 in a US cohort.127 It is generally accepted that NASH is associated with obesity and diabetes in high percentages.

Notably, in 2014, Flemming et al128 predicted the risk of HCC in patients with cirrhosis, using the ADRESS‐HCC risk model. They examined the l‐year probability of HCC in various liver diseases in 34 932 patients in the national transplantation waitlist database. HCC developed in 1960 patients (5.6%) during a median follow‐up of 1.3 years. Six baseline clinical variables including age, diabetes, race, etiology of cirrhosis, sex, and severity (ADRESS) of liver dysfunction were independently associated with HCC. They settled ADRESS‐HCC risk model from these data and the risk model well‐coincided with the clinical observation. They concluded that the risk model was clinically useful tool for predicting the risk of HCC in patients with diverse etiologies.

In addition, the aging of the patients must be taken into the consideration, as the cirrhotic patients were considered to be older than the non‐cirrhotic patients in almost all liver diseases. In this regard, Asahina et al129 investigated the difference of HCC incidence in aging in HCV‐associated liver disease, and found that the incidence of HCC was higher in the older patients (>65 years old) than the younger patients (<65 years old). The same tendency was observed by Taura et al130 However, the difference in incidence was approximately twofold. So, it is difficult to explain the marked difference in HCC incidence between the cirrhotic state and non‐cirrhotic state found in this meta‐analysis. Regarding other liver diseases, there were very few reports which demonstrated a difference between patients in the non‐cirrhotic and cirrhotic states concerning age.

The next consideration is the prevention of patients not to become cirrhosis state. We demonstrated that the incidence of HCC in the cirrhotic state and that in the non‐cirrhotic state were markedly different in many liver diseases; therefore, efforts to prevent patients with liver diseases from progressing to the cirrhotic state by all means is very important.

In HBV infection, the effort to diminish HBV‐DNA by pegylated interferon (Peg IFN)19 or oral antiviral agents, such as entecavir (ETV), tenofovir, and tenofovir anafenamide, is important. Indeed, ETV and tenofovir were reported to prevent the occurrence of HCC.131, 132In HCV infection, Peg IFN plus ribavirin or direct‐acting antivirals (DAAs) are effective to eliminate HCV, and may be prevent the progression of the disease, resulting in prevention of HCC development.

In primary biliary cholangitis, administration of ursodeoxycholic acid is important. Indeed, the risk of HCC in UDCA‐treated PBC patients was reported to be relatively low.59, 63

In autoimmune hepatitis, well‐designed corticosteroid therapy is important to prevent HCC development, and persistent elevation of serum aminotransaminase is reported to lead the development of HCC.78 It seems that alleviation of serum ALT levels might prevent HCC development, but this hypothesis is not confirmed solidly.
        
In NASH patients, improvement in metabolic syndrome, especially improvement in diabetes mellitus and obesity, is important.

Next, we mentioned whether the treatment of underlying liver diseases substantially modulates the HCC risk or not. We restricted the survey chiefly to the cirrhotic patients because they are at high risk of HCC development.

Continue to full-text article:
https://onlinelibrary.wiley.com/doi/full/10.1002/cam4.1998

© 2019 The Authors. Cancer Medicine published by John Wiley & Sons Ltd.

KEYWORDS: hepatocellular carcinoma; liver cirrhosis; liver diseases; meta-analysis; risk of HCC