Hepatocellular carcinoma in patients with non-alcoholic steatohepatitis – epidemiology, risk factors, clinical implications and treatment

Introduction

Hepatocellular carcinoma (HCC) is the most common primary cancer of the liver occurring worldwide. In 2015 it was the fourth cancer in the world in terms of mortality, and sixth in terms of prevalence. The number of new cases in 2015 increased by 75% in comparison to 1990. The greatest rate of prevalence concerns Asian countries and sub-Saharan region of Africa. The essential etiological factor of HCC development is post-inflammatory cirrhosis induced by viral hepatitis (B or C), HBV or HCV, as well as alcoholic cirrhosis, and any hepatic cirrhosis, regardless of its aetiology [1-3]. In recent years, rapid growth of incidence of metabolic syndrome, obesity and diabetes has been noted worldwide. The majority of authors claim that concurrent non-alcoholic steatohepatitis is becoming a dominant factor of hepatic cirrhosis and HCC [4, 5]. Growing frequency of occurrence, global range, and multimorbidity concomitant with the mentioned pathologies are becoming a true challenge both for practitioners dealing with diagnosis and treatment of HCC, as well as for public health employees in general [6].

Epidemiology

Non-alcoholic fatty liver disease (NAFLD) affects one fourth of the world population. There are various concomitant diseases and pathological states: obesity in 51%, type 2 diabetes in 41%, hyperlipidaemia in 69%, hypertension in 39%, and in 42% of cases all components of the metabolic syndrome [7, 8]. Non-alcoholic steatohepatitis (NASH), a more active and severe form of non-alcoholic fatty liver, affects 1.5% to 6.5% of the general population, which accounts for ca. 10-20% of patients with NAFLD. In papers based on liver biopsy results, transition from NAFLD to NASH was confirmed far more frequently, in as many as 59% of cases [5]. NASH is characterized with a typical histopathological image: ballooning degeneration and inflammation in the hepatocytes, which causes progressing fibrosis and hepatic cirrhosis and hepatocellular carcinoma [5-8]. It is estimated that the mortality caused by liver diseases and general mortality in patients with NAFLD is, respectively, 0.77 and 11.7 per 1000 person-years, while analogical indicators in the case of patients with NASH are considerably higher, respectively 15.44 and 25.56 per 1000 person-years [8]. The main cause of death of patients with NAFLD/NASH is cardiovascular disease (48%); the second, non-liver solid malignant tumours – gastric cancer, pancreatic cancer, bowel cancer, ovarian cancer, lung cancer, breast cancer (22%); and the third, hepatic complications – hepatic cirrhosis and HCC (10%) [5, 9]. It was found that patients with NAFLD incur a 64% higher risk of coronary heart disease, heart attack, stroke and coronary heart disease. Death risk due to heart diseases varies from 1.55 to 1.85 and is considerably higher in comparison to patients with chronic hepatitis C [5, 10-14]. The relation between NAFLD and atherosclerosis, which is the cause of cardiovascular disease (CVD), seems logical. Dyslipidaemia with a high level of triglycerides and decreased level of HDL in combination with insulin resistance in patients with NAFLD accelerates the onset of atherosclerosis [15].

Addressed in numerous papers [5, 6, 16-18], as mentioned above, the unspecified percentage of transition from NAFLD to NASH, slow progression of NASH to hepatic cirrhosis, gravity and significance of global epidemic of obesity and metabolic syndrome undervalued by the majority of doctors, and the consequences stemming from it, all add up to a situation where 38-45% of cases of hepatic cirrhosis in the course of NAFLD are diagnosed only at the stage of advanced liver failure [5, 19]. Moreover, the cases of liver cancers originating from NASH are diagnosed later and are more advanced than hepatocellular carcinomas of different aetiologies [5, 15]. Numerous authors also claim that a large percentage, perhaps even 30-75%, of so-called cryptogenic cirrhosis (CC) may in reality be a form of hepatic cirrhosis induced by NAFLD [5, 20, 21]. It has serious prognostic implications, as the affinity of this kind of cirrhosis to HCC is markedly higher in comparison to cirrhosis of different aetiologies [5, 22, 23].

In the group of patients with hepatic cirrhosis in the course of NASH, hepatocellular carcinoma is becoming the most common cause of death, not only because of cancer itself, but also because of the age of the patients and numerous concomitant diseases [5, 24, 25]. Numerous authors have noted an increase in occurrences of HCC in the course of NAFLD. In the UK in the years 2000-2010 the percentage of cancers induced by NASH increased from 21.5% to 34.8%. In the USA between 2004 and 2009 the annual growth of NASH involvement in the emergence of HCC was 9%. In Asia the percentage of nonviral cases of HCC in the years 1991-2010 increased from 10% to 24.1% [5, 26, 27]. Cumulative annual frequency of HCC incidence in patients with cirrhosis on the background of NAFLD ranges from 2.4% to 12.8%, while the annual risk of transition from NASH to liver cancer is estimated at 0.3% [24, 28, 29]. A permanent increase of the HCC percentage is also observed, reaching 37% of cases, emerged in the course of NASH without concomitant hepatic cirrhosis [24, 29, 30].

Hepatocellular carcinoma risk factors

According to the majority of authors, the most important risk factors of transition from NASH to HCC are: the degree of progression of liver fibrosis, diabetes and insulin resistance, as well as obesity, age and male gender [24, 29]. Diabetes is an independent factor, doubling the risk of HCC, and death risk due to hepatocellular carcinoma in the course of diabetes rises by 1.56 [31-33]. Similarly, obesity is a significant HCC risk factor. Patients with HCC in the course of NASH had higher body mass index (BMI) (27 kg/m²) than patients with HCC with induced HCV (24 kg/m²). With BMI greater than 30 kg/m² the risk of cancer almost doubles, and with BMI greater than 35 kg/m² it increases almost fourfold [24, 29, 31]. The connection of hepatocellular carcinoma with metabolic syndrome, diabetes, obesity and hypercholesterolemia is highly visible. Among patients with HCC in the course of NAFLD, patients with diabetes and obesity prevail. In the group of patients with metabolic syndrome and diabetes, the risk of HCC is five times greater in comparison to patients without these pathologies [31]. Renal failure coexisting frequently with metabolic syndrome is the cause of an increasing number simultaneous liver-kidney transplantations [16]. Not all patients presenting symptoms of metabolic syndrome, such as diabetes, dyslipidaemia or fatty liver disease (FLD), are fat. It affects 10% to 20% of the white population with BMI less than 25 kg/m², where liver fibrosis greater than F2 is diagnosed in ca. 27% of cases. The progression of liver disease in slim patients with metabolic obesity may also lead to hepatic cirrhosis and liver cancer [31]. The aforementioned cryptogenic cirrhosis, in comparison to alcoholic cirrhosis and post-inflammatory cirrhosis, coexists far more frequently with: diabetes (in 56% in comparison to, respectively, 17% and 11%), obesity (in 50% in comparison to, respectively, 17% and 14%), and fatty liver disease (FLD), exceeding 20% (in 61% in comparison to, respectively, 17% and 19%) [24, 31]. This suggests that cryptogenic cirrhosis (CC) in a large percentage is an advanced form of cirrhosis on the background of NAFLD. It has been shown that CC in patients with obesity may be complicated by HCC in 27% of cases [5, 22-24, 31]. In terms of demography and threats stemming from it, numerous studies, included those conducted in Japan, indicated advanced age of the patients with HCC in the course of NASH in comparison to the age of patients with cancer induced with HCV. These regularities concerned both women and men [31, 34]. Among patients with hepatocellular carcinoma males prevail, but such a statistically significant correlation has not been proved in the HCC group in the course of NASH [24, 35]. Among other risk factors in the development of HCC in NAFLD, the increase of intrahepatic iron content and alcohol intake are mentioned. An increased iron concentration in hepatocytes was measured in patients with liver cancer in the course of NASH in comparison to patients with only uncomplicated hepatic steatosis [24, 29]. Chronic alcohol consumption increases the risk of transition from NASH to HCC more than threefold [24, 36].

The mentioned risk factors influencing the increase of oncogenesis in patients with NAFLD gave the basis for numerous studies on investigating its pathomechanisms. The transition from dysplasia to cancer may be induced by numerous carcinogenic mechanisms. Numerous paths and multi-factorial pathogenic mechanisms such as inflammation, cirrhosis, diabetes and insulin resistance, lipid metabolism disorders, intestinal dysbiosis and genetic disorders in the form of polymorphisms play a role, according to many authors, in the development of HCC [24, 29, 31, 37-39].

Following the guidelines of AASLD (the American Association for the Study of Liver Diseases) for patients with hepatic cirrhosis, including those in the course of NASH, they should all be screened for HCC. Ultrasound of the liver every 6 months, together with the evaluation of AFP, is therefore compulsory [40]. It does not (yet?) concern patients with NAFLD/NASH without hepatic cirrhosis. Diagnosing a growing number of HCC cases, reaching even 50%, caused by NAFLD without cirrhosis, and later cancer diagnoses, in comparison to other HCC aetiologies, resulting in delaying the treatment, necessitate the performance of research concerning the rationale behind and the cost efficiency of screening tests in this group of patients [24, 41-44].

Prevention

The proven negative influence of metabolic syndrome, increasing the risk of cancer transition in patients with NASH, paved the way for, initially intuitive, later proven, behaviours, and various therapies potentially helping in cancer prevention. It has been demonstrated that intensive physical exercise decreases the risk of HCC [44, 45]. The treatment of diabetes with metformin reduces the risk of cancer by 7% annually. Similarly, therapy with statins prevents carcinogenesis [46, 47]. BMI reduced through dietary change or bariatric surgery reduces the incidence of diabetes, as well as the risk of HCC occurrence, by decreasing the degree of liver fibrosis [48-50]. In contrast to alcohol, the intake of more than 2 cups of coffee daily decreases the risk of liver cancer by 27% [50, 51]. Papers discuss potentially preventive measure of other medications such as branched-chain amino acids (BCAAs), interferon, non-steroidal anti-inflammatory drugs (NSAIDs) (aspirin) and anticoagulants [50].

Treatment

There is a lack of reports on special methods of HCC treatment dedicated to patients with NAFLD [15, 24]. Patients should be treated in line with BCLC staging [52]. According to Bruix [53], an expert in therapies used in HCC, only patients with a single lump not larger than 5 cm, no feature of decompensated liver fibrosis – ascites, jaundice and portal hypertension, and the number of thrombocytes exceeding 100,000/mm³ – should be qualified as candidates for liver resection surgery. What follows, only patients in whom resection cannot be performed due to liver failure, and who fulfil the Milan criteria (one lump smaller than 5 cm, alternatively up to 3 lumps, each smaller than 3 cm, with no evidence of vascular invasion), should become eligible for liver transplantation (LTx). Thermal tumour ablation is currently the most frequently used strategy in surgical treatment of HCC. Limitations of the application are the patient’s liver failure and unfavourable position of the lesion. In patients with multiple foci of the neoplasm, without extrahepatic spread, transarterial chemoembolization (TACE) is used. Sorafenib, regorafenib and immunomodulators extend the survival of patients in advanced cases of HCC, not eligible for the abovementioned therapies [53]. The basic aim of surgical treatment is radical resection of the tumour and preventing recurrence. The 5-year survival is 60% from the time of resection, with low, lower than 3%, perioperative mortality. Frequent cancer recurrences, reaching 70% in the course of 5 years, pose a major problem [54, 55]. Long-term results of thermal tumour ablation 2 cm in size are comparable to the results of liver resection surgery [54, 55]. After transplantation, due to HCC in patients meeting the Milan criteria, cancer recurrences occur in 10% to 15% of cases. After broadening eligibility criteria, in larger lumps, they occur far more often [55, 56]. In order to improve the results of liver transplantation, many centres adopted various auxiliary criteria, including α-fetoprotein (AFP) concentration in serum, which turned out to be helpful in determining the risk of recurrence, and therefore in the assessment of prognosis. According to many authors, exclusion criteria for supplementation of the Milan criteria with AFP concentration in serum allow for the exclusion of patients with an aggressive form of cancer, resulting in a high degree of recurrences. The level of 1000 ng/ml was adopted as the boundary AFP concentration, above which the risk of recurrence is very high. However, some centres indicate the level of 400 ng/ml [56]. NASH is currently the second indication for liver transplantation in the USA in terms of frequency [57]. Kidney failure, frequently concomitant with metabolic syndrome and diabetes in patients with uncontrolled cirrhosis in the course of NASH, caused an increase of simultaneous kidney liver transplantations. Their percentage increased from 2003 to 2011 in the USA from 6.3% to 19.2% respectively [15]. Moreover, NASH is an increasingly frequent indicator for LTx due to HCC, but it results in numerous clinical implications [55]. Patients with NASH and confirmed liver cancer directed for surgical treatment are usually elderly people, with concomitant diseases, with less developed liver failure and usually larger lumps in comparison to patients with HCC of different aetiologies. Late cancer diagnoses are the results of lack of screening tests in patients with NAFLD, where no cirrhosis was observed yet, as well as difficulties with ultrasound scanning of small focal changes in fatty liver. The period of cancer diagnosis therefore gets extended, which delays the decision to treat the patient. After qualifying the patient for LTx, the waiting time for the surgical treatment of patients with cancer in the course of NAFLD is longer than in patients with hepatic cirrhosis of different aetiologies, taking into account the lower MELD indicator, showing a lower degree of liver failure in these patients. This results in an increased number of cancellations on the waiting list for LTx caused by, in the first place, cancer development exceeding the Milan criteria, and secondly mortality caused by concomitant diseases. An additional factor preventing transplantation is the fact of portal vein thrombosis, occurring more often than in other types of cirrhosis, reaching 10.1%, in patients with NAFLD [15, 24, 58-61]. For these reasons, the majority of patients with HCC in the course of NAFLD have liver resection procedures, and only a small fraction of patients are eligible for LTx [24, 30]. Three- and 5-year survival of patients with NAFLD from the time of LTx does not significantly differ from the results of liver transplantation from other reasons [15, 59, 62-64]. The fact remains that in the first two years from the time of LTx, the main causes of death are IVD, sepsis, and severe obesity [15, 59, 63]. The 5-year survival of the recipient from the time of LTx without obesity in comparison to patients with BMI > 40 kg/m² is 78.8% and 51.3% respectively. 50% of patients with BMI higher than 35 and diabetes die within 1 year from the time of LTx. Some patients with severe obesity cannot, taking into account the high death risk, be qualified for liver transplantation [15, 59, 63]. The mentioned risk factors necessitate a thorough cardiology analysis of patients with NAFLD eligible for surgical treatment. Performing coronary angiography with the use of catheters and coronary artery stenting is recommended, because 70% of cardiac complications occurring in the perioperative period are the cause of 50% of deaths from the time of LTx [15, 59]. Due to numerous risk factors, eligibility of patients with liver cancer in the course of NAFLD for surgical treatment becomes a true challenge for the medical team, which by necessity needs to be interdisciplinary.