Primary liver cancer is a significant global health concern, representing the seventh most common type of cancer and the second leading cause of cancer-related deaths [1]. The countries with the highest incidence of liver cancer mainly typically show lower levels of economic development, and most cases of liver cancer occur in geographically diverse countries, including those in North and West Africa (Gambia, Egypt, and Guinea) and East and Southeast Asia (Cambodia, Mongolia, and Vietnam) [2,3]. The major aetiologies of primary liver cancer include chronic hepatitis B virus (HBV) and hepatitis C virus (HCV) infections, alcoholic liver disease, and non-alcoholic fatty liver disease (NAFLD). However, the aetiologies greatly vary from region to region. More specifically, the most important risk factors in China and East Africa are chronic HBV infection and aflatoxin contamination in food, respectively, and in other countries such as Egypt and Japan, chronic HCV infection is often the main cause. In Western countries, chronic HCV infection, alcohol consumption, and obesity/diabetes are the most common causes of liver cancer.

The Global Burden of Disease (GBD) database was constructed to provide a tool for quantifying health losses from hundreds of diseases, injuries, and risk factors. The GBD was initiated and maintained by a consortium of more than 3,600 researchers in nearly 200 countries. The database comprises a comprehensive catalogue of surveys, censuses, vital statistics, and other health-related data such as death and disability [4]. Therefore, the GBD database can provide helpful insights for understanding the nature of liver cancer, how the challenges are changing over time, and how to allocate health resources more effectively.

Because effective antivirals against chronic HBV and HCV infections have recently become available, the global distribution of the dominant aetiologies has changed. In particular, due to universal screening for HCV infection and direct-acting antivirals (DAAs) against chronic HCV infection, HCV is expected to be eliminated by 2030. However, trends in liver cancer based on specific aetiology and sociodemographic status have rarely been reported and compared.

In this study, we aimed to describe and analyse the updated temporal trends in aetiologies and sociodemographic status from 1990 to 2019 to provide insights into appropriate global intervention strategies.

In the present study, we showed that the crude numbers of DALYs, deaths, and incident cases of liver cancer significantly increased between 1990 and 2019. However, age-standardized DALYs, mortality, and incidence rates have declined. The discrepancy between the number and rates of DALYs, deaths, and cases might be explained by the aging population structure [9,10]. With continued aging and population growth, the number and rate of incidence, as well as years of life lost and mortality, are projected to increase in the near future [11-13]. Concordant with previous studies, HBV remains the leading cause of liver cancer mortality and incidence, followed by HCV, alcohol consumption, and NASH/NAFLD [14]. Age-standardized mortality rates of liver cancer due to HBV and HCV have decreased during the past decade, possibly due to the effect of HBV vaccination and nucleos(t)ide analogues (NUCs) [15] and novel DAAs against HCV [16,17]. However, age-standardized mortality rates of liver cancer due to alcohol consumption and NASH/NAFLD have increased.

Patterns of liver cancer burden also vary significantly between regions and countries. In 2019, the liver cancer burden due to HBV was high in HBV-endemic regions such as East Asia, high-income Asia-Pacific, Southeast Asia, and Southern and Western Sub-Saharan Africa [18]. However, most of these regions are experiencing or will have a significant decrease in the burden of liver cancer due to HBV through universal HBV immunization programs and HBV screenings [18-20]. In contrast, the overall age-standardized mortality and incidence rates were low in the high-income countries of North America and Australasia, but increased to >80% in 2019 compared with 1990. In addition, the GBD 2019 data showed an inverse trend in the age-standardized incidence rates of acute HBV infection, cirrhosis, and other chronic diseases due to HBV compared with HBV-attributable liver cancer in these regions (Supplementary Figs. 1–6). Therefore, despite the current use of potent NUCs, liver cancer progression rates are not well controlled [21-24]. In the era of pre-NUCs, patients with chronic HBV infection would die before de novo hepatocarcinogenesis due to hepatic decompensation. Because HCC risk cannot be eliminated using only potent NUCs, the development of novel treatments is required to effectively eradicate closed circular DNA in hepatocytes. Furthermore, a more effective HCC surveillance strategy should be established to detect patients at an early stage and implement curative treatment.

In contrast to HBV infection, liver cancer due to HCV infection has a different threat pattern. Countries with high SDI were more affected by liver cancer due to HCV infection, with the high-income Asia-Pacific region showing the highest age-standardized mortality rates. However, with the development of DAAs with sustained virological response rates near 95%, the incidence and mortality caused by liver cancer due to HCV are expected to significantly decrease in the next decades [25-30]. Notably, in high-income North America and Australasia, the age-standardized mortality rates have doubled in 2019 compared with 1990. This trend, however, is likely due to the 20-to 40-year lag from HCV infection to HCC progression [31]. The effect of the 1945–1965 birth cohort in the United States with significantly high HCV infection rates [32,33] and the use of injectable drugs in Australia [34] could be causing this increase. Thus, even with careful measures, the burden of liver cancer due to HCV infection is likely to increase before the effects of intervention occur. In contrast to the past decade, when mainly high and high-middle SDI quintiles showed a decrease in age-standardized DALY rates, lowand middle-income countries are also likely to benefit from DAAs because of the production of cost-effective generic DAAs in the near future [35,36]. Only 21% of the HCV-infected population knew their diagnosis and only 62% were treated with DAAs in 2019 according to a WHO report [37], with many low- and middle-income countries lacking current standard diagnostic methods [36]. Thus, systematic and universal screening and appropriate referrals for HCV in endemic regions should be recommended with development of novel cost-effective diagnostic tests [38-41]. However, to prevent infection, an effective HCV vaccine is needed [42,43].

Contrary to liver cancer due to HBV and HCV, the age-standardised incidence and mortality of liver cancer due to alcohol consumption are increasing. Assuming an 8- to-9- year latency period of alcohol-induced liver cancer [44,45], the increasing trend of liver cancer due to alcohol consumption is concordant with the WHO Global information system on alcohol and health data (alcohol, total per capita [15+] consumption). A sharp increase was observed from 5.3 litres (95% confidence interval [CI] 5–6.5) in 2005 to 6.1 litres (95% CI 5.8–6.5) in 2010 [46]. If the trend of liver cancer due to alcohol consumption follows the alcohol consumption data, the number of liver cancers due to alcohol consumption is likely to remain steady in the next decade as alcohol consumption remains steady or possibly decreases (2019 alcohol consumption, 5.8, 95% CI, 5.5–6.2). Among the GBD regions, Central Asia, high-income North America, and Eastern Europe have experienced significant increase in age-standardized liver cancer mortality due to alcohol consumption. This is in agreement with previous studies reporting higher liver cancer mortality rates due to alcohol consumption in Central Asia and Eastern Europe in the 2012 to 2017 period [47]. In addition, alcohol consumption was recently suggested as a main reason for the increase in age-standardized mortality rate of HCC in the United States [48].

Furthermore, the liver cancer burden due to NASH/NAFLD is markedly increasing. As the prevalence of NAFLD significantly increases [49], and more than 25% of the global population suffers from NAFLD [50-52], liver cancer due to NASH/NAFLD will also likely increase for a prolonged period. Modelling studies have predicted that NAFLD-related HCC incidence will increase in the near future, with the United States experiencing a 137% increase in incidence rate by 2030 [53]. Similarly, China, France, Germany, Italy, Japan, and the United Kingdom will experience noticeable increase by 2030 [54]. Thus, controlling obesity and diabetes in the population [55,56], promoting weight loss [57,58], promoting a balanced diet [59], and suggesting chemoprevention for patients with comorbidities such as diabetes, obesity, and cardiovascular diseases[57,60] should be implemented in regions with a high number of NASH/NAFLD patients. Particular attention should be given to Central Asia, Australasia, and high-income North America where age-standardized DALY rates increased more than two-fold compared with those in 1990.

Liver cancer burden in the Republic of Korea has shown a declining trend during the past decade, with gradually decreasing age-standardized DALYs and mortalities. This could be attributed to the widespread use of NUCs for HBV, DAAs against HCV, and better treatment options for liver cancer patients. However, in agreement with the recent report on chronic liver disease [61], liver cancer burden due to alcohol consumption has increased in recent years. Although alcohol consumption in the Republic of Korea has declined in recent years, alcohol consumption continuously increased from 1998 to 2015 [62]. Thus, further measures to control alcohol consumption are needed to reduce the burden of alcohol-related liver cancer.

Sex and age differences were observed in liver cancer burden. Males tended to have a higher burden of liver cancer due to all major aetiologies, except NASH/NAFLD. The HBV genome has been suggested to interact with androgens, leading to increased expression of oncoproteins [63,64]. Oestrogen receptor expression and epigenetic mechanisms are associated with HCV-HCC [64,65]. and gender differences exist in alcohol consumption [66]. Various mechanisms have been proposed, including the protective role of oestrogen, HCC-promoting role of androgens, and additional role of sex chromosomes [67]. Furthermore, the age group with the highest incidence in females was older and postmenopausal, which may be partially explained by the progression of liver fibrosis that can accelerate after menopause [68], and because postmenopausal hormone therapy has a protective effect against liver cancer [69].

The main strength of the present study includes the analysis of the most recent data on the global burden of liver cancer, focusing on DALYs of major liver cancer aetiologies. To the best of our knowledge, this study is the first to analyse indepth DALYs of liver cancer based on aetiology. However, this study had several limitations. First, because data from the GBD 2019 database were analysed, the same limitations of the original dataset were applicable to the current study. Major limitations include the availability of primary data and the use of statistical modelling to generate estimates. For regions that lacked sufficient quality data, GBD estimates relied on predictive covariates and modelling measures. In addition, even when primary data were available, delay in data reporting and data obtained without using the preferred case definition could limit accuracy of the estimates. Thus, results need to be carefully interpreted regarding uncertainty intervals. In addition, due to the lack of data, various types of cancer have been widely identified as liver cancer, which hampered accurate analysis of gender and age differences. Third, the GBD 2019 database lacks detailed information other than epidemiological data such as vaccination policy changes, antiviral therapy use, and public health measures. Thus, this detailed information could not be included in the analysis.

In conclusion, despite substantial advances in prevention, diagnosis, and treatment, liver cancer continues to be an ongoing global threat. In particular, the liver cancer burden due to alcohol consumption and NASH/NAFLD is increasing and will continue to significantly increase if an effective intervention is not implemented. Because the burden of liver cancer significantly varies based on geographic region, SCI, sex, and age group, a more tailored approach should be implemented.