Keywords

1 Introduction

Worldwide, primary liver cancer is the fifth most common cancer in men and the ninth most common cancer in women [1, 2]. In 2012, worldwide 782,000 new cancer cases were diagnosed and nearly 746,000 deaths occurred [3]. The prognosis is very poor, with a 5-year survival rate between 5 and 9 %, and thus, primary liver cancer is the second leading cause of cancer-related death worldwide [4]. The predominant form of primary liver cancer is hepatocellular carcinoma (HCC), which accounts for approximately 85–95 % of all primary liver cancer cases, followed by intrahepatic bile duct cancer (IBDC), a cancer that develops in the bile ducts inside the liver [1, 5]. There is a large variation in incidence rates of HCC across geographic regions. More than 80 % of cases with HCC are detected in less developed countries [6]. In general, incidence rates are higher in men than in women [1]. In men, highest incidence rates are detected in Eastern and South-Eastern Asia (age-adjusted incidence rate >20 per 100,000), and lowest rates in Northern Europe and South-Central Asia (age-adjusted incidence rate <5 per 100,000). In women, the highest incidence rates occur in Eastern Asia and Western Africa [age-adjusted incidence rate >8 per 100,000] and the lowest in Northern Europe [age-adjusted incidence rate <2 per 100,000] [1]. However, over the last decades, the incidences of both types of primary liver cancer, HCC and intrahepatic cholangiocarcinoma (ICC) have also increased in the “lower-risk” Western countries such as the USA [7]. Major known risk factors for HCC include chronic infection with hepatitis C virus (HCV) and hepatitis B virus (HBV), exposure to toxins, such as aflatoxin, and excessive alcohol consumption [6]. This could partly explain the geographic variation of HCC occurrence because prevalence of liver cirrhosis in consequence of infection with hepatitis B or C virus, and exposure to toxins is more common in low-income countries compared with high-income countries [5, 8]. The documented increase in HCV- and HBV-related HCC, however, does not fully explain the recent increase in HCC incidence in Western populations, as 20–50 % of HCC remain idiopathic. Different lines of evidence identify non-alcoholic fatty liver disease (NAFLD) as a possible relevant risk factor for occurrence of HCC [9]. NAFLD is the most common form of liver disease in Western countries characterized by accumulation of excessive fat in the liver in the absence of alcohol abuse (12). NAFLD includes a spectrum of liver disorders, ranging from simple steatosis (infiltration of fat in the liver) to the more severe form non-alcoholic steatohepatitis (NASH) [10, 11]. Obesity can alter hepatic pathology, metabolism and promote inflammation, NAFLD and induce pathologic progression and development of NASH. NASH is characterized by prominent steatosis and inflammation and can lead to cirrhosis and ultimately HCC [12]. NAFLD is strongly associated with obesity and its metabolic complications, such as metabolic syndrome and type 2 diabetes [13]. In this context, the increased prevalence of obesity and associated NAFLD could possibly explain rising incidence of primary liver cancer in Western countries over the last decades.

Here, we review the existing evidence on the links between obesity and its metabolic complications—NAFLD, metabolic syndrome and diabetes type 2—and liver cancer incidence and survival. Furthermore, we evaluate current knowledge on potential mechanisms that may possibly explain obesity-associated liver cancer risk and could thereby provide new targets for liver cancer prevention in societies affected by the obesity epidemic.

2 General and Abdominal Obesity, Weight Gain and Risk of Liver Cancer

Recently, an expert review report of the World Cancer Research Fund (WCRF) concluded that there is a sufficient body of evidence to establish higher body fatness as a risk factor for HCC [14]. This evidence comes from studies investigating body fatness based on body mass index measurements (BMI: weight/height2 [kg/m2]), that is considered as an indicator of general obesity. In a dose–response meta-analysis of 12 prospective studies, the risk of HCC was increased by 30 % per each 5 kg/m2 higher BMI [14] (Fig. 1). Parallel lines of evidence have been provided by a number of independently conducted systematic reviews and meta-analyses [15–19]. In those studies, a higher risk of liver cancer was observed in the highest category of BMI compared to the lowest. Using established cut-off values for the BMI, including normal weight (BMI: 18.5 ≤ 25.0 kg/m2), overweight (BMI: 25 ≤ 30 kg/m2) and obesity (BMI: 30 kg/m2) [20], a meta-analysis of 26 prospective studies (including 25,337 participants) observed 18 % higher risk of HCC for individuals with overweight [relative risk and 95 % confidence intervals: 1.18 (1.06–1.31)], and 83 % higher risk in individuals with obesity [relative risk and 95 % confidence intervals: 1.83 (1.59–2.11)] compared to individuals with normal weight [19]. Another meta-analysis of 8 studies including 1,779,471 cohort individuals revealed that the nature of the observed association between BMI and risk of liver cancer was nonlinear (P for nonlinearity <0.001). The relative risks were 1.02 (95 % CI  =  1.02–1.03), 1.35 (95 % CI  =  1.24–1.47) and 2.22 (95 % CI  =  1.74–2.83) for BMI category above 25, 30 and 35 kg/m2 compared with the reference (the median value of the lowest category), respectively [16]. Similar nonlinear association, with the most pronounced increase in risk among persons with a BMI > 32 kg/m2, was reported in another meta-analysis of 21 prospective studies (including 17,624 cases) [17]. In that study, patients with HCV or cirrhosis (but not patients with HBV) with excess weight had a higher risk of liver cancer development than general populations with excess weight. Overall, the conducted meta-analyses reported high heterogeneity of results for the association of obesity with liver cancer that could be mostly accounted for by sex, ethnicity and underlying liver diseases: Stronger associations were seen in men than in women, and in individuals with underlying liver disease or with HCV infection or cirrhosis compared to individuals from the general population. Interestingly, BMI seemed to be more strongly associated with risk in white populations compared with other ethnic groups. A more detailed analysis that evaluated associations according to ethnic groups were recently published within a large sample of the multiethnic cohort study, a population-based prospective cohort study among 482 incident HCC cases identified among 168,476 participants after a median follow-up of 16.6 years [21]. In that study, BMI was strongly associated with liver cancer in Japanese, white, and Latino men, whereas there was no association in black men. Moreover, the study also revealed that BMI strongly correlated with total fat mass, measured by dual-energy X-ray absorptiometry, both in men and women and in all ethnic groups. In contrast, there was a lower correlation value for BMI and visceral or liver fat measured by abdominal magnetic resonance imaging in black men and women [21]. Overall, BMI is strongly correlated with body fat, and thus, it is considered as a good marker for evaluation of total body fatness [22]. However, an important limitation of BMI is that it does not allow accounting for body fat distribution. Therefore, anthropometric measures of abdominal obesity might be more appropriate to reflect differences in body shape and fat distribution, as compared to BMI. However, evidence on the association between measures for abdominal obesity such as waist circumference (WC), waist-to-hip and waist-to height ratio and risk of primary liver cancer remain insufficient [23, 24]. First lines of evidence on the association between abdominal obesity and risk of HCC have been provided by the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort—a large European multi-center cohort study conducted among 359,525 men and women among which during a mean follow-up of 8.6 years 177 cases of HCC have been diagnosed [25]. In that study, abdominal obesity was defined based on established cut-off values provided by the World Health Organization (WHO) (waist circumference ≥102 cm for men and ≥88 cm for women, and waist-to-hip ratio ≥95 can for men and ≥0.80 for women) [20]. The data revealed a twofold higher risk of HCC for individuals above the cut-off values for abdominal obesity compared with individuals below these cut-points after controlling for established liver cancer risk factors, such as age, sex, alcohol intake, smoking, education, infection of hepatitis b and c virus and even after accounting for general obesity (as assessed by BMI). These findings point out that abdominal obesity might be a risk factor for HCC independently from general obesity [25]. Rather than studying markers of total adiposity, studies of obesity and HCC should move beyond BMI and use a better measure for fat-specific depots [26]. When evaluating the role of obesity in liver cancer risk, it is also important to account for the age of onset of obesity—i.e. early life versus later life. So far, only one study reported on the association between early adulthood obesity and risk of developing HCC, suggesting that obesity is associated with an increased risk at a young age in the absence of major HCC risk factors [27]. Furthermore, anthropometric measures such as BMI and WC represent an assessment of a static exposure status and it remains unclear whether dynamic measures of obesity such as weight gain are also associated with a higher risk. Data from the previously mentioned study within the EPIC cohort suggested that weight gain during adulthood (since age 20) was an independent risk factor for HCC reporting a 2.5-fold higher risk of HCC (95 % CI = 1.49–4.13) for the highest versus the lowest tertile of weight gain after taking into account baseline BMI and WC measurements [25]. These results have been further extended with regard to the association between adult weight gain with HCC mortality in a Japanese cohort of 31, 018 men and 41, 455 women aged 40–79 years. In that study, during a median 19-year follow-up, 527 deaths from HCC (338 men, 189 women) were documented. Weight gain since age 20 years was positively associated with liver cancer mortality among women with an underlying liver disease. Thus, women with history of liver disease had an about twofold higher HCC risk for weight gain of 5.0–9.9 kg compared with women with a stable weight (change of −4.9 to 4.9 kg) after controlling for important risk factors [28].

Fig. 1
figure 1

Dose–response meta-analysis of BMI and liver cancer, per 5 kg/m2. Adapted from WCRF/AICR continuous update report for liver cancer [14]

Full size image

3 Metabolic Complications of Obesity in Relation to Liver Cancer

3.1 Non-alcoholic Fatty Liver Disease

Recent studies have suggested that NAFLD and particularly its aggressive form—NASH—are associated with an increased risk of primary liver cancer, mainly HCC [29]. In Western countries, up to 22 % of HCC cases could be attributed to NAFLD [30]. The estimated prevalence of NAFLD is around 20–35 % in developed countries mirroring the observed rates for obesity and the metabolic syndrome. It appears to be more common in men, and it increases with age and after menopause. Some data suggest that Mexican Americans are more likely to have NAFLD and blacks are less likely compared with non-Hispanic whites. More advanced stages of NAFLD are associated with older age, higher BMI, diabetes, hypertension, high triglycerides, and/or insulin resistance. Most NAFLD-related HCCs are believed to develop in the background of a cirrhotic liver [31]. The risk factors for HCC in the setting of NAFLD have not been established [32]. A study from the US indicated one of the most common etiologies of liver disease and cryptogenic cirrhosis (29 %), where half of the patients had histologic or clinical features associated with NAFLD [33]. It has been estimated that in morbidly obese patients that underwent bariatric surgeries, the prevalence of NAFLD can be as high as 98 % [34]. Moreover, this study carried out in a population of young adult, clinically asymptomatic obese patients confirmed the high prevalence of echographically detectable liver steatosis in massive obesity even in young adult patients [34]. Lipid accumulation in NAFLD triggers cancer-related pathways including c-Jun N-terminal kinase (JNK), nuclear factor-kappaB (NF-kβ) and toll-like receptors (TLR) signaling pathway, and overexpression of oncogenic genes [35]. The results from an obesity surgery cohort demonstrated that NAFLD is indeed frequent with over two thirds demonstrating histological presence of NAFLD and 18 % with definitive NASH by liver biopsy [36]. In an experimental study, it has been observed that both genetic and dietary factors related to obesity could promote NASH, liver dysplasia and HCC tumorigenesis in animal models [37]. In livers of obese mice, the occurrence of dysplastic and cancerous lesions showed morphological features of NASH without fully developed cirrhosis [35]. This indicates that liver hyperplasia is evident at the earliest stage of NAFLD and the transformation of malignant liver cells was resultant from the development of NASH instead of cirrhosis [35]. A study by Gutzman et al. [38] suggested also that NAFLD may predispose patients to HCC in the absence of cirrhosis. Finally, NAFLD was suggested to progress to HCC based on the metabolic syndrome development with obesity [39].

3.2 Metabolic Syndrome

Metabolic syndrome is defined as a cluster of metabolic alterations including abdominal obesity, dyslipidemia, hypertension, diabetes and insulin resistance [40]. It has been consistently associated with increased risk of cardiovascular diseases, and it has been also linked to risk of cancer at several sites [41]. NAFLD has been recognized as a hepatic manifestation of metabolic syndrome and its associated complications [42]. NAFLD appears to be most strongly associated with obesity and insulin resistance states including diabetes and with other features of the metabolic syndrome, such as high triglycerides and low high-density lipoprotein cholesterol levels [32]. Individuals with NAFLD/NASH-associated HCC were shown to exhibit a higher prevalence of metabolic features (type 2 diabetes, hypertension, dyslipidemia, coronary artery disease) compared to individuals with non-NAFLD/NASH-HCC. Nevertheless, even in the absence of cirrhosis, the NAFLD/NASH as the hepatic entity of the metabolic syndrome may itself pose an independent risk factor for HCC [43]. Indeed, liver tumors arising in patients with features of metabolic syndrome are with a larger size, well differentiated and mainly occur in the absence of significant fibrosis [44]. In a large pooled European cohort study comprising 578,700 individuals and 266 primary liver cancer cases, a metabolic syndrome score, based on BMI, blood pressure and circulating concentrations of glucose, total cholesterol and triglycerides, was significantly associated with increased risk of primary liver cancer [45]. Further analysis of single metabolic risk factors revealed that particularly BMI and glucose were significantly associated with higher primary liver cancer risk [45]. These findings were confirmed by another large population-based study in the USA that reported a twofold increased risk of HCC in individuals with metabolic syndrome compared to healthy ones [9]. In this context, data from Japanese population also confirmed these findings and reported that most of the patients with NASH who develop HCC were men having high rates of obesity, type 2 diabetes, and hypertension [46]. Additionally, males developed HCC at a less advanced stage of liver fibrosis than females [46]. A meta-analysis of 25 studies indicated the presence of multiple metabolic disorders, including obesity, type 2 diabetes, dyslipidemia and hypertension, as a clinical characteristic of NAFLD-associated HCC [47]. Indeed, almost all NAFLD-associated HCCs (99 %) had at least one type of metabolic disease and 76 % had two or more [47]. Another study showed that the presence of NASH and metabolic syndrome are common metabolic factors in patients with HCC (without infection by HBV and HCV) [48]. In a case–control study, the presence of dyslipidemia (defined by elevated triglycerides and/or lowered high-density lipoprotein) was associated with an increased odds for HCC (Odds ratio: 1.35 (95 % CI = 1.26–2.45) [9, 39]. Moreover, an analysis including cohorts from Austria, Norway and Sweden indicated a twofold increased risk for hypertension regarding the development of HCC [45].

3.3 Type 2 Diabetes

Recent evidence has also pointed to the involvement of more advanced metabolic complications, such as type 2 diabetes in the risk of primary liver cancer. Summary findings of a meta-analysis including 25 prospective studies indicated that diabetes mellitus is associated with twofold higher risk of HCC compared to individuals without diabetes [49]. These data have been supported by a systematic review on the association between anti-diabetic medication use and risk of liver cancer summarizing data from 10 studies including 22,650 cases of HCC in 334,307 patients with type 2 diabetes. The meta-analysis of 8 observational studies showed a 50 % lower HCC incidence with metformin use, 62 and 161 % higher HCC incidence with sulfonylurea or insulin use, respectively. A recent study has confirmed these results [50]. Possible synergistic effects of metabolic factors have been suggested by the results revealing the highest risk of HCC for individuals with both obesity (BMI ≥ 30 kg/m2) and type 2 diabetes [51–53].

In summary, a number of studies have underscored the importance of obesity, NAFLD and related metabolic complications in the development of primary liver cancer. Nevertheless, still broadened researches are needed to better understand the molecular link between the obesity-associated metabolic risk factors and HCC risk.

4 Pathophysiological Mechanisms Linking Obesity and Liver Cancer

The exact pathophysiological mechanisms behind the observed association between obesity, type 2 diabetes and risk of HCC are not completely understood [54]. As described above, one possible explanation for these relations includes the strong association with NAFLD [10, 55]. On the other hand, clinical and epidemiological data have failed to demonstrate hepatic tumor expression in fatty liver tissue [29] leading to the hypothesis that there may not be a single direct link between liver fat accumulation and hepatic carcinogenesis. Parallel lines of evidence have brought the notion on a number of obesity-related pathways during the progression of NASH that could be implicated as potential intermediary risk factors linking obesity and hepatocellular carcinogenesis (Fig. 2) [56]. It has been suspected that an excess fat storage, particularly within the abdomen and around the organs (the visceral fat), is associated with accumulation of fat in the liver, which might be associated with abnormalities in the hepatic metabolism, such as hyperinsulinemia and chronic low-grade inflammation [57–59]. In addition, the adipose tissue itself is defined as an endocrine organ secreting a number of hormones and proteins (growth factors and adipocytokines) known to be involved in altered metabolism and associated disease risk, including some types of cancer [60–62]. Below we review current evidence implicating insulin resistance, chronic inflammation, adipokine secretion, and altered gut microbiota as main intermediate pathways in obesity-liver cancer association.

Fig. 2
figure 2

Pathophysiological mechanisms during the progression to NASH. The development of NASH is initiated by several different risk factors including a high-fat diet, physical inactivity, and genetic predispositions that often lead to obesity and insulin resistance. Exaggerated fat intake and obesity lead to hyperglycemia, hyperlipidemia, and the over expressions of adipocytokines and chemokines and further contribute to insulin resistance in adipose tissue and the liver. Insulin resistance results in hepatic triglyceride (TG) synthesis and the increased delivery of free fatty acids (FFAs) to the liver. Additionally, hepatic steatosis acts as a “first hit” that is followed by a “second hit” in which inflammatory mediators can cause NASH and even cirrhosis. An enhanced storage of TG provokes a series of harmful consequences related to hepatocytes, such as uncontrolled lipid peroxidation, oxidative stress, and endoplasmic reticulum (ER) stress, which can activate hepatic inflammatory pathways. In particular, the recruitment of macrophage/Kupffer cells and an M1-dominant phenotypic shift in macrophages in the liver play a key role in the pathological progression of NASH. Adapted from Hu et al. [56]

Full size image

4.1 Hyperinsulinemia

Hyperinsulinemia exerts coinciding effects with hyperglycemia, type 2 diabetes, and central obesity, thereby suggesting that it may be one of the central mechanisms to explain the obesity-liver cancer link [54]. First lines of evidence in support of this hypothesis came from the Paris Prospective Study cohort, a cohort study of 6237 non-diabetic French working men aged 44–55 years at baseline [63]. In that study, after 23.8 years of follow-up, peripheral hyperinsulinemia—indicative of very high portal insulin concentrations—was associated with a higher risk of fatal liver cancer [63]. Data from the EPIC cohort suggested that elevated concentrations of C-peptide were associated with twofold higher risk of HCC (relative risk: 2.25, 95 % CI = 1.43–3.54; P < 0.0005). These findings could be explained by the fact that the liver, in comparison with other organs, is exposed to high insulin concentrations. Furthermore, hyperinsulinemia is often present in patients with chronic HCV infection and is associated with more advanced hepatic fibrosis. Mechanistic studies demonstrated enhanced hepatic tumor growth in the presence of high insulin concentrations. High insulin levels may directly promote cell proliferation and survival through the phosphoinositide 3-kinase/protein kinase B and Ras/mitogen-activated protein kinase pathways [64]. Furthermore, the insulin-like growth factors I and II (IGF-I and IGF-II), their receptors and their binding proteins play an increasingly role in tumor formation, growth, and metastasis in vivo [65]. Within circulation and tissue compartments, IGF is bound with high affinity to a family of structurally related binding proteins (IGFBP) characterized by different properties [66]. In the rat model of hepatocarcinogenesis, the expression of IGF axis components including IGF-I, IGF-II, IGF-IR, IGFII/M6PR, and individual IGFBP were examined in the sequence of preneoplastic hepatic foci and HCC. Finally, increased expressions of IGF-I and IGF binding protein-4 (IGFBP-4) in altered parenchymal cells, and a decreased expression of IGFBP-1 has been demonstrated. IGF-II was not detected in these pre-neoplastic foci and HCC arising in this model had decreased expressions of IGF-I and IGFBP-4, but IGFBP-1 expression was not significantly altered. Moreover, some HCC showed a more than 100-fold overexpression of IGF-II, whereas other tumors were completely negative for IGF-II expression [67]. In another study, it has been also observed that IGF-1 levels decrease when liver steatosis is worsened showing statistically significant difference between mild-moderate and severe steatosis with no correlation between IGF-1 levels and either homeostasis model assessment (HOMA) or insulin levels [68]. The results from a cross-sectional analysis of data from the Third National Health and Nutrition Examination Survey, 1988–1994 showed that there may still be an important underlying etiological connection between the IGF-1 axis and hepatic steatosis. However, after controlling for important HCC risk factors, this association and trend were extenuated, highlighting the importance of metabolic factors (related to glucose homeostasis and adiposity) in this relation [69].

4.2 Chronic Low-Grade Inflammation

Obesity induces production of pro-inflammatory molecules—chemokines and cytokines—required for the initiation and progression of HCC [70, 71]. Although acute liver inflammation can play a vital and beneficial role in response to liver damage or acute infection, the effects of chronic liver inflammation, including liver fibrosis and cirrhosis, are sufficient in a fraction of individuals to initiate the process of transformation and the development of HCC [72]. Chemokines and their receptors can also contribute to the pathogenesis of HCC, promoting proliferation of cancer cells, the inflammatory microenvironment of the tumor, evasion of the immune response, and angiogenesis [71]. In obese patients, accumulation of lipids in the liver promoted activation of an inflammatory response. At the same time, lipid accumulation increases demand on the endoplasmic reticulum leading to uncontrolled production of reactive oxygen species (ROS). ROS stimulate inflammatory signaling and induce oxidative damage including strand breaks and nucleotide modifications, and DNA damage leading to genomic instability. Thus, sustained hepatic inflammation results in damage to parenchyma, oxidative stress, and compensatory regeneration/proliferation. These inflammation-associated processes could be associated with increased incidence of hepatocellular carcinogenesis; however, evidence remains scarce. In animal models, it was shown that obesity may promote HCC development through elevated production of tumor necrosis factor (TNF) and interleukin 6 (IL-6). In clinical studies, higher levels of IL-6 and C reactive protein (CRP) have been found among patients with HCC, when compared to controls. Recently, data from the EPIC cohort provided first lines of evidence for an independent association between several inflammatory and metabolic biomarkers and HCC risk suggesting their role as intermediate factors in the obesity-liver cancer association [73]. Moreover, a combination of these biomarkers was able to improve risk assessment of HCC beyond established risk factors such as infection with HBV/HBC, smoking, alcohol consumption, etc. (Fig. 3). Notably, these associations were independent of established HCC risk factors and adiposity measures, suggesting that these inflammatory biomarkers may play role as candidate intermediate factors of the association with HCC risk [73]. These data have been confirmed by a case–control study nested in a Japanese cohort with 188 HCC cases and 605 controls which reported that higher concentrations of CRP and Il-6 have been associated with an around twofold and fivefold higher risk of liver cancer, respectively [74]. These associations were independent of hepatitis virus infection, lifestyle-related factors and radiation exposure. Despite these arising data, exact roles of various inflammatory biomarkers as mediators of the association between obesity and HCC have not been evaluated.

Fig. 3
figure 3

Predictive ability of inflammatory and metabolic biomarkers and GLDH beyond the multivariable adjusted model and AFP levels. The biomarkers included in the model have been associated with HCC risk. These include CRP, Il-6, C-peptide, and non-HMW adiponectin. Multivariable model is taking into account matching factors: study center; gender; age (±12 months); date (±2 months); fasting status (<3, 3–6, or >6 h); and time of the day (±3 h) at blood collection. Women were additionally matched according to menopausal status (pre-, peri-[unknown], or postmenopausal) and exogenous hormone use (yes, no, or missing) at blood donation. Further adjusted for education (no school degree or primary school, secondary school, high school, or missing), smoking (never, former, current, or missing), alcohol at baseline, drinking status at baseline (non-drinker or drinker), diabetes (no, yes, or missing), coffee (g/day), HBsAg/anti-HCV (negative, positive, or missing), BMI, and WHtR adjusted for BMI. Adapted from Aleksandrova et al. [72]

Full size image

4.3 Abnormal Adipokine Production

Recently, adipose tissue has been established as an endocrine organ that secretes a variety of biologically active adipokines, such as leptin, adiponectin and resistin. Adipokines play an important role in the physiology of adipose tissue, including food intake and nutrient metabolism, insulin sensitivity, stress, inflammation and bone growth. Several studies reported that adipokine dysregulation contribute to liver fibrosis and influence the pathological state of chronic liver diseases [75–80]. The dysregulated expression of adipokines may therefore provide explanatory mechanisms in the association of obesity with HCC [81]. Among various adipokines, two molecules—leptin and adiponectin—gained much attention in the recent research.

4.3.1 Leptin

Leptin is a well-established adipokine closely linked with the higher BMI and thereby considered as a good proxy measure of general adiposity [82]. Leptin increases with increasing fatty mass as a compensatory mechanism to preserve insulin sensitivity, but persistent hyperleptinemia could be implicated in liver fibrogenesis and carcinogenesis [83, 84]. A recent meta-analysis of 33 studies among 2612 individuals concluded that circulating leptin levels were higher in patients with NAFLD than in controls. Higher levels of circulating leptin were associated with increased severity of NAFLD, and the association remained significant after exclusion of studies involving adolescent populations and morbidly obese individuals [85]. Leptin could play a role in the development of NAFLD through insulin resistance, steatosis, worsening hepatic inflammation and ultimately fibrosis. Leptin has angiogenic properties, promotes cell proliferation and migration, and interacts with growth factors, all of which could promote tumor growth [84]. However, the role of leptin in the development of liver cancer remains controversial with some studies suggesting an important role of leptin in liver fibrosis and carcinogenesis [86], while others demonstrating an inhibitory role of exogenous leptin on tumor size in murine model of HCC [87]. So far, the association of leptin and liver cancer was explored in only one prospective epidemiological study, which suggested a null association [73].

4.3.2 Adiponectin

Adiponectin is one of the most abundantly secreted adipokines in blood circulation, which actions are mainly exerted by the activation of AMP-activated kinase and peroxisome proliferator-activated receptor alpha [88]. Whereas the liver probably is not a source of circulating adiponectin, it is a major target organ of adiponectin metabolism [88]. Adiponectin is implicated in the regulation of steatosis, insulin resistance, inflammation and fibrosis; therefore, it could be expected that that hyperadiponectinemia might suppress liver tumorigenesis and elevated levels of adiponectin would be associated with a reduced risk of HCC [89]. In contrast, experimental studies indicated that adiponectin treatment increased apoptosis of HCC and inhibited its proliferation [89, 90]. Some studies have shown that circulating adiponectin levels are higher in subjects with liver cirrhosis and that they increase in line with fibrosis stage [91]. Paradoxically, several human studies suggested that elevated adiponectin concentrations are associated with higher HCC risk. A hospital-based cohort study in Japan showed that high serum levels of adiponectin were positively associated with the development of HCC in patients with chronic HCV [92]. A nested case–control study conducted in middle-aged Japanese adults with hepatitis virus infection showed that both total and high-molecular weight adiponectin are associated with a higher risk of HCC [93]. A more recent large European cohort study added to this line of evidence suggesting adiponectin and its non-high-molecular weight isoform to contributed substantially to HCC risk [73]. However, null findings have been reported by another cohort study from France, in which serum levels of adiponectin measured in 248 patients with compensated HCV cirrhosis were found to be unassociated with HCC occurrence [91]. Positive associations between adiponectin and HCC risks could be explained by the fact that impaired liver function due to liver disease (including cirrhosis) may lead to hyperadiponectinemia.

4.3.3 Novel Adipokines

Apart from established adipokines, such as leptin and adiponectin, a recent systematic review evaluated the potential link between newly described adipokines and liver histology in biopsy-proven NAFLD patients [76]. Thirty-one cross-sectional studies were included, resulting in a total of seven different investigated adipokines, most of which suggested to be involved in the inflammatory response that develops within the context of NAFLD, either at hepatic or systemic level, and/or hepatic insulin resistance. Based on this literature review clinical studies suggest that chemerin, resistin and adipocyte-fatty-acid-binding protein potentially are involved in NAFLD pathogenesis and/or progression [76]. However, major inconsistency still exists, and there is a high need for larger studies using standardized assays to determine adipokine levels. So far there have not been studies to evaluate potential involvement of inflammation-associated adipokines as potential mediators of the association between obesity and liver cancer risk.

Gut microbiota and bile acid metabolism

Based on animal studies, it was hypothesized that genetic obesity provokes alterations of gut microbiota profile, thereby increasing the levels of deoxycholic acid (DCA), a secondary bile acid produced solely by the 7 alpha-dehydroxylation of primary bile acids carried out by gut bacteria. The enterohepatic circulation of DCA provokes DNA damage and consequent cellular senescence in hepatic stellate cells (HSCs) which, in turn, secrete various inflammatory and tumor-promoting factors in the liver, thus facilitating HCC development in mice [94].

5 Obesity and Liver Cancer Survival

The emerging link between obesity and increased risk of HCC raises the question whether such association could be also observed for prognosis and postoperative complications of HCC. A number of studies have investigated these associations. On the one hand, some studies demonstrated that HCC patients with higher BMI exhibited significantly better prognosis than HCC patients with lower BMI after hepatic resection surgery [95–97]. However, on the other hand, no significant differences in the prognosis were detected between individuals with different levels of BMI in other studies [98–100]. In addition, studies reported that obesity does not influence surgical outcomes in hepatocellular carcinoma patients undergoing curative hepatectomy [101]. A recent systematic review including a total of 14 studies suggested that BMI was not associated with survival (including overall and disease-free survival) in HCC patients. In addition, in these patients, higher BMI was not related to postoperative complications (ascites, bile leaks, and 30-day mortality) [102]. However, HCC patients with higher BMI had increased risk of wound infections. The reason for lack of association between BMI and liver cancer prognosis is not clear. More studies are, therefore, warranted covering large spectrum of anthropometric characteristics of obesity in order to evaluate association between obesity and liver cancer survival.

6 Summary

Accumulating evidence has established an association between higher BMI as an indicator of general obesity and increased risk of primary liver cancer. The associations proved to be stronger in men, in patients with underlying liver disease and in white ethnic groups. Abdominal obesity, weight gain in adult life and metabolic factors related to visceral fat accumulation were also suggested as important risk factors for liver cancer; however, more studies are needed to evaluate these associations. Potential mechanisms that may link obesity and liver cancer include insulin resistance leading to increased levels of insulin and insulin-like growth factors, chronic inflammation due to adipose tissue remodeling, pro-inflammatory cytokine and adipokine secretion, and altered gut microbiota. The association between obesity and metabolic parameters and liver cancer survival remains controversial. More research is warranted in order to evaluate the role of inflammatory and metabolic biomarkers as intermediate risk factors for risk of obesity-associated liver cancer. Better understanding of these associations may help in improving current strategies of liver cancer prevention, particularly in societies with high obesity prevalence.