Etiologic fractions in patients of hepatocellular carcinoma in India with and without a background of cirrhosis: a multi-centric study

Hepatocellular cancer (HCC) typically arises in the background of cirrhosis. The epidemiology of HCC has changed in recent years due to availability of newer antivirals, changing life-styles and greater possibility for early detection. We undertook a multicentric national sentinel surveillance for liver cirrhosis and HCC to assess the attributable risk factors for the development of HCC, both with and without a background of cirrhosis. Data from January 2017 till August 2022 from hospital-based records of eleven participating centers were included. Diagnosed cases of cirrhosis [radiological (multiphase and/or histopathological] and HCC [as per AASLD 2018] were included. History of significant alcohol intake was elicited by AUDIT-C questionnaire. Altogether 5798 enrolled patients were assessed, of which 2664 patients had HCC. The mean age was 58.2 ± 11.7 years and 84.3% (n = 2247) were males. Diabetes was found in over a third of those with HCC (n = 1032;39.5%). The most common etiology of HCC was NAFLD (n = 927;35.5%) followed by viral hepatitis B and C and harmful levels of alcohol. Among those with HCC, 27.9% (n = 744) had no cirrhosis. Higher proportion of cirrhotic HCC patients had alcohol as an etiological factor as compared to non-cirrhotic (17.5 vs. 4.7%, p ≤ 0.001). NAFLD was an etiological factor for a higher proportion of non-cirrhotic HCC patients as compared to cirrhotic HCC (48.2 vs. 30.6%, p ≤0.001). Diabetics more commonly had non-cirrhotic HCC (50.5 vs. 35.2%). The following factors were associated with an occurrence of cirrhotic HCC: male gender (OR 1.372 and 95% CI 1.070–1.759), age above 60 years (OR 1.409 and 95% CI 1.176–1.689), HBV (OR 1.164 and 95% CI 0.928–1.460), HCV (OR 1.228 and 95 CI 0.964–1.565) and harmful consumption of alcohol (OR 3.472 and 95% CI 2.388–5.047). The adjusted odds of non-cirrhotic patients having NAFLD was 1.553 (95% CI 1.290–1.869). This large multi-centric study demonstrates that NAFLD is the most important risk factor for development of both cirrhotic and non-cirrhotic HCC in India and has overtaken viral hepatitis. Awareness campaigns and large-scale screening are required to reduce the high burden of NAFLD-related HCC in India.


Introduction
The sequential progression of chronic liver disease to fibrosis and to cirrhosis often culminates in neoplasia. The preneoplastic setting of the cirrhotic background provides a conducive environment for cellular transformation. The etiological spectrum of cirrhosis has varied among different geographical populations worldwide. Countries where alcohol consumption is high have reported it as the leading cause of cirrhosis, whereas countries where alcohol consumption is low or prohibited have shown viral hepatitis as the most common cause of cirrhosis [1]. Primary liver cancer is the sixth most commonly diagnosed cancer and the third leading cause of cancer death worldwide in 2020, with approximately 906,000 new cases and 830,000 deaths. Primary liver cancer predominantly includes hepatocellular carcinoma (HCC), which comprises 75%-85% of the total case load in most countries [2].
Cirrhosis is a well-known cause for a person to develop HCC [3]. Reports from tertiary care centers in India on HCC indicate that 70-97% of patients with HCC at the time of diagnosis had underlying cirrhosis of liver [4]. HCC usually develops on a background of oxidative stress and inflammation, triggered by chronic infection with hepatitis B or C virus (HBV or HCV), aflatoxin exposure, excess alcohol consumption, obesity, diabetes, and smoking [5].
The major risk factors for development of HCC appear to be in transition, with the prevalence of HBV and HCV declining and excess body weight and diabetes increasing in many regions [6]. In India 84% children had received the primary three doses of Hepatitis B vaccine in 2020 which is close to the key service target to achieve HBV elimination by 2030. For HCV, even though currently there is no vaccine available, an 8-week to 12-week course of orally administered, direct-acting antiviral agents appear to cure HCV infection in most instances [7]. On the other hand, the incidence and prevalence of HCC due to the epidemic of non-alcoholic fatty liver disease (NAFLD) is poised to become the leading cause of liver cancer [8].
We undertook a multicentric national sentinel surveillance for liver cirrhosis and HCC to assess the attributable risk factors for development of cirrhosis and HCC under the guardianship of the WHO collaborative center at the Institute of Liver and Biliary Sciences (ILBS), New Delhi. Data from hospital-based records of all these eleven participating centers was included. Data were accessed retrospectively from health records for the period from January 2017 to December 2018 and thereafter, patients who visited the outpatient and inpatient departments of the hospitals were subsequently enrolled. Overall, data from January 2017 till August 2022 was collected for analysis.

ILBS, New
Cases of cirrhosis were defined as those confirmed by radiological (multiphase CT scan/MRI) and/or histopathological investigations. Diagnosis of HCC was made in accordance with AASLD 2018 case definition guidelines [9]. The history of significant alcohol intake was elicited by AUDIT C questionnaire [10]. The operational definition of NAFLD was those cases that had diabetes, excluding those with consumption of harmful levels of alcohol.
A dedicated online portal was created for uploading of data. A pre-tested, close-ended questionnaire in English was administered to candidates who visited the OPD and/ or IPD of the participating institutions. Data were collected in Microsoft Excel v. 2019 software and analyzed using IBM. SPSS statistics for Windows. Version 25.0. Armonk, NY: IBM; 2017. Continuous variables were summarized as median with interquartile range (IQR).
All quantitative variables were analyzed in terms of mean and standard deviation while qualitative variables were analyzed through proportions. Chi-square test was employed for testing differences in proportion for utilization pattern between different groups. p < 0.05 was considered the cutoff for statistical significance. Binary logistic regression was applied to determine the independent statistically significant relationship between etiological variables and occurrence of HCC with and without background cirrhosis. For the regression analysis the independent variables were recoded into binomial categories as follows: Age-less than 60 years and 60 years or above. Gender-male and female. Hepatitis B-present and not present.
Hepatitis C-present and not present. Alcohol-harmful consumption present and not present. NAFLD-present and not present.

Results
A total of 5798 participants with either cirrhosis or HCC or both were enrolled in the study. Of these 5054 patients had cirrhosis (86.7%) while 2664 patients had HCC (45.7%). A total of 1920 individuals (32.9%) had HCC with a background of cirrhosis. Majority of the participants were male (83.3% of total cirrhosis, 84.3% HCC, 86.1% noncirrhotic HCC and 83.7% cirrhotic HCC). Mean age was 58.2 ± 11.7 years for those with HCC (58.3 ± 11.6 for male and 57.9 ± 12.1 for female). It was 57.1 ± 11.9 years for those with cirrhotic HCC whereas it was 61.12 ± 10.9 years for those who had HCC without cirrhosis. Majority of the individuals belonged to upper socio-economic class according to updated BG Prasad's classification for 2021 (n = 4400; 76.7% and n = 2193; 82.3% for cirrhosis and HCC, respectively) ( Table 1).
Among those with HCC, 27.9% (n = 744) were non-cirrhotic HCC patients. NAFLD was significantly more commonly found in those with HCC without cirrhosis (n = 352; 48.2%) whereas alcohol consumption was found in a higher proportion of those with cirrhotic HCC as compared to the other (n = 336; 17.5%). Diabetics more commonly had noncirrhotic HCC (50.5 vs. 35.2%). Co-infection with hepatitis B and C was almost three times more commonly seen in cirrhotic HCC (1.4 vs. 0.5%) ( Table 3).
The clinical characteristics of cirrhotic and non-cirrhotic HCC patients are given in Table 4. Diagnosis through imaging techniques was the most common method for both cirrhotic (n = 1775; 92.4%) and non-cirrhotic HCC (n = 735; 98.8%). On comparing the blood parameters, non-cirrhotic HCC patients showed a higher median level of AST, albumin and prothrombin time as compared to cirrhotic HCC patients.
On applying logistic regression for HCC with and without background cirrhosis, it was found that the following factors were associated with occurrence of cirrhotic HCC: male gender (OR 1.372 and 95% CI 1.070-1.759), age above 60 years (OR 1.409 and 95% CI 1.176-1.689), HBV (OR 1.164 and 95% CI 0.928-1.460), HCV (OR 1.228 and 95 CI 0.964-1.565) and harmful consumption of alcohol (OR 3.472 and 95% CI 2.388-5.047). Even though chances of developing cirrhotic HCC were higher in those having Hepatitis B, the findings were not significant (Fig. 1).

Discussion
The study aimed to describe the etiological fractions of hepatocellular carcinoma both with and without the background of cirrhosis. In our study that enrolled a large number of patients with HCC, the most common risk factor for HCC was found to be NAFLD, viral hepatitis B and C, and consumption of harmful levels of alcohol. Diabetes was found in over a third of those with HCC. A small fraction of patients had inherited or autoimmune disorders. Over a fourth of those with HCC had no cirrhosis. NAFLD and diabetes were etiological factors for a higher proportion of non-cirrhotic HCC patients as compared to cirrhotic HCC. In our study, 64.6% of patients with HCC belonged to age group of 50-70 years and 20.5% were less than fifty years old. Male-to-female ratio was 5.4:1 The age-specific incidence varies in different areas of the world [11,12].The incidence of HCC is usually found to be higher in male and in those over 40 years old [13]. Non-cirrhotic HCC patients were older at diagnosis as compared to cirrhotic HCC but had a similar male to female ratio. A similar finding in age profile was observed by Tham et al. in a comparative study between cirrhotic and non-cirrhotic HCC patients in the UK [14] HCC in non-cirrhotic patients is often seen to be clinically silent in its early stages because of lack of symptoms and surveillance imaging; and higher hepatic reserve in this population. [15] Of all HCC patients in our study, non-cirrhotic HCC accounted for 27.9%, while the remainder had a background of cirrhosis. Chronic necroinflammation and hepatocellular regeneration due to cirrhosis is the background for genetic mutations to accumulate and cells to progress to malignancy [16]. In various other studies done worldwide, cirrhosis was reported in 60% to 99% of patients with HCC. [4,[16][17][18] HCC is an immensely complex condition and there are multiple factors involved in the etiology of HCC. The major risk factors for HCC include hepatitis B virus (HBV) and hepatitis C virus (HCV), alcoholic fatty liver disease (AFLD), non-alcoholic fatty liver disease (NAFLD) and some autoimmune and inherited disorders.
Viral infection with Hepatitis B was seen in 21.2% of all HCC patients. A quarter (25.2%) of HBV-related HCC occurred in non-cirrhotic patients. Similar findings were reported in other studies [19,20]. Hepatitis C infection was seen in less than a fifth (18%) of those with HCC. However,  HCC in chronic Hepatitis C patients primarily occurs in the background of cirrhosis and is related to a necroinflammatory state with tissue damage, regeneration, repair and fibrosis [21,22]. This was corroborated in our findings where majority (76.2%) of HCC patients with HCV had a background of cirrhosis. HCV was found in 23.7% of non-cirrhotic HCC patients which is higher than figures found in other studies [23]. Significant alcohol intake was found in 13.9% of all HCC patients and accounted for 4.7% of all non-cirrhotic HCC and 17.5% of cirrhotic HCC patients. Close to 9.5% of HCC patients who consumed harmful levels of alcohol had no background of cirrhosis. Several studies have shown heavy alcohol intake in patients with non-cirrhotic HCC; however, most of these were not statistically significant [16]. The lack of availability of biopsy from non-tumorous liver parenchyma for all patients was a barrier in identifying the true prevalence of NAFLD among our patients. Therefore, those with HCC who had an underlying history of diabetes mellitus and no alcohol intake were considered to be having NAFLD as an etiology. While more than a third (39.5%) of the total patients with HCC had diabetes, 35.5% had diabetes with no history of harmful intake of alcohol. Over a third of those (37.9%) had no cirrhosis as well. A strong association has been reported between fatty liver disease and HCC in non-cirrhotic patients [24,25]. With recent advances in viral hepatitis, especially hepatitis C virus (HCV) treatment, NAFLD-related HCC is being recognized as a growing burden [26,27]. Several studies show that diabetes mellitus and obesity that are intrinsic to NAFLD and Metabolic Syndrome cause the release of multiple pro-inflammatory cytokines like IL-6, TNF-alpha, leptin and resistin and decrease release of adiponectin and these mechanisms in turn accentuate the development of hepatic steatosis and inflammation within the liver and subsequently can lead to the development of HCC [27,28]. Genetic factors like the rs641738 C > T variant of the MBOAT7 gene has been shown to increase the risk of hepatic steatosis and progressive liver disease in NAFLD patients and it is also associated with an increased risk of HCC in non-cirrhotic patients with NAFLD [29,30]. A relatively new-identified link between NAFLD and cellular senescence has also shown to have a role in NAFLD-related HCC. Hepatic cellular senescence is strongly associated with DNA damage, activation of oncogenes and aggravation of hepatic fibrosis due to recruitment of fibroblasts by senescent cells. A complex of cytokines, metalloproteases, chemokines, growth factors and matrix remodeling factors constitute the senescence associated secretory phenotype (SASP) which further aggravates inflammation, senescence to neighboring cells and facilitates tumorigenesis [31].
The etiological attributions in the study findings are in contrast to the findings from other reported studies from India, which showed viral hepatitis B and C to be commonest etiology for HCC [4,11,[32][33][34][35]. Close to half of the total HCC patients in our study who had neither HBV or HCV had diabetes as a risk factor. Similar finding was observed in a study by David et al. [36] Overall, geographical differences which in turn result in variations in prevalence of various etiological risk-factors for HCC might have an impact on their attribution fractions [37].
The strength of our study is the fact that it enrolled a large number of study participants from multiple centers across the country. However, being a cross-sectional study design meant it was not possible to determine any causal relationship between HCC and its etiological risk factors.

Conclusion
The findings of the present study suggest that NAFLD may become one of the most common etiological factors for HCC, both with and without background cirrhosis. HCC in patients with viral hepatitis usually appears in background of cirrhosis. As etiology of non-cirrhotic HCC varies from its cirrhotic counterpart, it becomes imperative that we focus on surveillance strategies in the patient population at risk, to attenuate the disease burden at presentation and ameliorate the prognosis of these patients.