Global, regional, and national burden of liver cancer due to hepatitis B, 1990–2021 and projections to 2051: a systematic analysis of the Global Burden of Disease Study 2021

Abstract

Purpose

Using the Global Burden of Disease 2021 data, this study reports the global, regional, and national disease burden of liver cancer due to hepatitis B (LCDHB) from 1990 to 2021, stratified by age, sex, and sociodemographic index (SDI), and projects future burden to 2051.

Methods

We examined incidence, deaths, and disability-adjusted life years (DALYs) of LCDHB. Age-standardized incidence rates (ASIR), age-standardized death rates (ASDR), and age-standardized DALYs rates were analyzed (1990–2021). Spearman correlation assessed age-standardized rates–SDI relationships. The Bayesian age-period-cohort (BAPC) model projected the burden to 2051.

Results

Compared to 1990, the number of LCDHB incidences, deaths, and DALYs increased by 46.9, 41.2, and 33.9% in 2021; however, from 1990 to 2021, the ASIR, ASDR, and age-standardized DALYs rate all exhibited a declining trend. In 2021, the highest ASIR occurred in East Asia, High-income Asia Pacific, and Western sub-Saharan Africa. At the national and regional levels, Mongolia, the Republic of Paraguay, and the Commonwealth of the Bahamas showed peak ASIR. ASIR was higher in males and increased with age, peaking at 85–89 age group for both males and females in 2021. A reverse U-shaped correlation existed between age-standardized DALYs and SDI during 1990–2021. BAPC projections indicate declining global ASIR, ASDR, and age-standardized DALYs rates (2021–2051).

Conclusion

Despite rising incidence, deaths, and DALYs, LCDHB treatment challenges persist, especially for males and elderly populations. Our findings on epidemiological trends and demographic variations provide crucial insights for policymakers addressing this global health burden.

Introduction

According to the latest GLOBOCAN estimates compiled by the International Agency for Research on Cancer, there were 865 000 new cases of liver cancer globally in 2022, with approximately 758 000 deaths because of liver cancer. It ranks sixth in incidence and third in mortality worldwide [1]. The most common type of primary liver cancer is hepatocellular carcinoma, followed by intrahepatic cholangiocarcinoma and other rare types (such as sarcomas and angiosarcomas) [2]. Liver cancer is often diagnosed at intermediate to advanced stages, making surgical options unfeasible, while systemic treatments have limited efficacy for advanced liver cancer [3]. The prognosis for liver cancer is poor, with an overall 5-year survival rate of only 21.7% [4].

Several modifiable risk factors have been identified in liver cancer, including infectious factors [hepatitis B virus (HBV) and hepatitis C virus], behavioral factors, and metabolic factors, which show significant regional variations [5–7]. Among these, HBV infection is the primary cause of liver cancer. Despite the effective implementation of global hepatitis B vaccination strategies, approximately 50% of hepatocellular carcinoma (HCC) cases worldwide are still attributable to HBV infection [8]. Notably, in East Asia and Oceania, the proportion of liver cancer due to hepatitis B (LCDHB) is significantly elevated compared with other causes [9]. It is estimated that chronic HBV infection affects approximately 296 million people globally, with 192 000 deaths attributed to HBV-related liver cancer in 2019, an increase from 156 000 in 2010 [10].

While previous studies have recorded the burden of LCDHB at national and regional levels [11,12], there is a lack of comprehensive information on the epidemiology and burden of LCDHB, including incidence, deaths, and disability-adjusted life years (DALYs), at global, regional, and national levels. In addition, the rapidly advancing field of LCDHB research requires nearly 2 years of data to remain relevant to public health policymakers, as prior studies only provided data up to 2019. Since the onset of the coronavirus disease 2019 (COVID-19) pandemic in 2019, global healthcare resources have tended to be allocated toward controlling and treating the pandemic. Data from the COVID-19 pandemic period (2019–2021) is particularly critical for analyzing the burden of LCDHB.

The 2021 Global Burden of Disease (GBD) study is a publicly accessible comprehensive database that records the burden of over 300 diseases and injuries across 204 countries and territories. Notably, GBD 2021 is the first publication to provide a comprehensive burden estimate for COVID-19 as an infectious disease, covering its impact on the burden of specific diseases [13], which was not included in previous GBD studies.

Therefore, to gain deeper insights into the burden of LCDHB and manage resource allocation for disease management and prevention, we analyzed the latest data on LCDHB incidence, deaths, and DALYs from the GBD 2021 study, covering the years 1990–2021 at global, regional, and national levels, stratified by sex, age, and sociodemographic index (SDI). Moreover, this study also forecasts the trends in global LCDHB burden over the next 30 years. We hope this research will provide valuable insights for policymakers in assessing the overall burden of LCDHB and promote the development of targeted prevention policies, thereby advancing the rational allocation of public health resources.

Methods

Data source and case definition

The GBD 2021 data (released on 16 May 2024) were obtained through the GBD Results Tool online platform of the Institute for Health Metrics and Evaluation (IHME) on 7 September 2024 (https://vizhub.healthdata.org/gbd-results/). This research is a secondary analysis of the publicly available aggregated anonymized data from GBD 2021. The 2021 GBD study conducted a comprehensive assessment of health risks associated with 371 diseases, injuries, and risk factors using the latest epidemiological data and enhanced standardization methods, covering 204 countries and territories, 21 regions, and seven super regions. The burden of disease and injury analysis in GBD 2021 estimated the incidence, prevalence, death, years lived with disability, years of life lost, DALYs, and health-adjusted life expectancy using 100 983 data sources. Data were sourced from vital registration systems, verbal autopsies, population censuses, household surveys, disease-specific registries, health service contact data, and other sources.

Using the GBD 2021 study, we extracted annual statistics on the incidence, deaths, and DALYs of LCDHB from 1990 to 2021, along with their corresponding age-standardized rates. In the GBD study, LCDHB is defined as malignant neoplasms of the liver specifically due to hepatitis B. The cancer diagnosis codes used in this study include C22.0–C22.1, C22.3–C22.8 from the International Classification of Diseases, 10^th Revision, and a proportion of C22.9 reflecting primary liver cancer. In this study, we extracted estimates of incidence, deaths, and DALYs from the GBD 2021 study, along with their corresponding 95% uncertainty intervals. Given that the GBD 2021 data is publicly available, the ethics committee of Nanjing University of Chinese Medicine approved an exemption for this study. Our research adhered to the Guidelines for Accurate and Transparent Health Estimates Reporting.

Sociodemographic index

The SDI is a composite measure that takes into account factors such as per capita lagged distribution income, average years of education for those aged 15 and above, and total fertility rates among women under 25, to determine the position of countries or other geographic areas within the development spectrum. The SDI is provided by the IHME, ranging from 0 to 1, where 0 represents the theoretical lowest level of development related to health outcomes, and 1 represents the theoretical highest level. In this study, the GBD database categorized regions into five SDI categories: low, low-middle, middle, middle-high, and high, to investigate the relationship between the burden of LCDHB and socioeconomic development.

Statistical analysis

To quantify the global burden trends of LCDHB, we used the percentage change values of ASIR, ASDR, and age-standardized DALYs rate. Positive or negative percentage change values were used to determine the increasing or decreasing trends of the LCDHB burden. The 95% uncertainty interval was identified as the 2.5^th and 97.5^th centiles of the ordered draws. The statistical analysis of the data in our study was performed using R version 4.2.3, and visualizations were created with the ‘ggplot2’ package. Spearman correlation was used to assess the association between SDI and age-standardized incidence rates of LCDHB. In this study, we utilized the BAPC model to predict the future trends of the burden of LCDHB. The reliability of this model has been demonstrated in previous studies [14–16]. The BAPC model employs integrated nested Laplace approximation (INLA), which directly avoids any Markov Chain Monte Carlo sampling techniques. Consequently, it also circumvents the issues of mixing and convergence and provides well-calibrated probabilistic predictions. The prediction intervals are not overly wide, making it superior to the generalized Lee-Carter model [17]. The calculation process was implemented through the ‘INLA’ package (version 0.0.36) and the ‘BAPC’ package (version 24.06.27) in the R program (version 4.2.3). A P value of less than 0.05 was considered statistically significant.

Results

Global Burden of liver cancer due to hepatitis B

In 2021, there were 206 365.7 new cases of LCDHB reported globally (95% uncertainty interval: 169 400.9–252 049.9), an increase of 46.9% from 109 478.6842 cases in 1990 (95% uncertainty interval: 127 365.5781–94 820.08889) (Table 1 and Fig. 1a). The global ASIR was 2.4 cases per 100 000 populations (95% uncertainty interval: 2–2.9), which represents a decline of 6.7% since 1990 (95% uncertainty interval: −23.9 to 14.7) (Table 1).

Table 1.

Incidence cases, deaths, and disability-adjusted life years for liver cancer due to hepatitis B in 2021, and percentage change in age-standardized rates per 100 000 population, by Global Burden of Disease region, from 1990 to 2021

Location	Incidence (95% UI)			Deaths (95% UI)			DALYs (95% UI)
	Number_2021 (95% UI)	ASRs per 100 000 (95% UI) in 2021	Percentage change in ASRs from 1990 to 2021	Number_2021 (95% UI)	ASRs per 100 000 (95% UI) in 2021	Percentage change in ASRs from 1990 to 2021	Number_2021 (95% UI)	ASRs per 100 000 (95% UI) in 2021	Percentage change in ASRs from 1990 to 2021
Global	206 365.7 (169 400.9–252 049.9)	2.4 (2–2.9)	−6.7 (−23.9 to 14.7)	181 194.3 (148 896.5–221 685.3)	2.1 (1.7–2.6)	−16.4 (−31.2 to 3)	5 668 199.3 (4 706 886.5–6 885 071.4)	65.4 (54.4–79.4)	−22.3 (−36 to −4.5)
High-income Asia Pacific	14 818.3 (11 974–18 162.6)	4 (3.2–4.8)	−26.9 (−41.8 to −7.6)	10 738.8 (8624.7–13 128.5)	2.7 (2.2–3.3)	−44.7 (−56.4 to −29.6)	260 756.9 (212 392.8–320 315)	77.5 (63.6–95.6)	−50.8 (−61.5 to −35.9)
High-income North America	3795.4 (3138.1–4501.9)	0.6 (0.5–0.8)	103.9 (94.7–112.4)	2839.4 (2330.4–3358.6)	0.5 (0.4–0.5)	80.9 (72.3–88.1)	75 464.3 (62 785.6–89 778.7)	13.6 (11.5–16)	72.3 (63.6–79.7)
Western Europe	5487.1 (4017.1–7391.6)	0.7 (0.5–0.9)	43.8 (34.5–54.2)	4637.8 (3414.1–6338.1)	0.6 (0.4–0.7)	22.6 (14.4–31.4)	114 155 (83 112.6–152 786.9)	15.8 (11.7–21.1)	18.1 (10.9–26.6)
Australasia	377.3 (264.1–528.7)	0.8 (0.6–1.1)	147 (115.2–183.4)	305.4 (212.1–430.7)	0.6 (0.4–0.9)	110.3 (81.6–142.3)	8370.9 (5831.4–11 517.9)	19.3 (13.6–25.8)	96.9 (70.4–125.9)
Andean Latin America	565.5 (400.2–777.4)	0.9 (0.7–1.3)	−15.1 (−35.5 to 10.7)	587.5 (411.5–817)	1 (0.7–1.4)	−15.7 (−35.9 to 9.4)	16 688.1 (11 894.7–22 491.8)	26.9 (19.1–36.3)	−20.6 (−39.8 to 2.6)
Tropical Latin America	915 (763.6–1081.2)	0.3 (0.3–0.4)	1.8 (−4.8 to 8.9)	924.9 (771.2–1095.2)	0.4 (0.3–0.4)	0.7 (−5.9 to 7.7)	28 606.5 (24 045.5–33 529.7)	10.9 (9.2–12.8)	−3.1 (−9.4 to 3.6)
Central Latin America	869.3 (648.2–1159.3)	0.3 (0.3–0.5)	−22 (−30 to −12.6)	885.3 (655.4–1190.1)	0.3 (0.3–0.5)	−22.9 (−31.2 to −13.7)	26 897.5 (20 334.5–35 081.5)	10.3 (7.8–13.4)	−24.1 (−32.7 to −14.6)
Southern Latin America	272.7 (188.9–389.6)	0.3 (0.2–0.5)	96.1 (61.4–132.4)	272.6 (187.7–392.1)	0.3 (0.2–0.5)	89.6 (56–124.2)	7489.8 (5327.8–10 344.6)	9.1 (6.5–12.5)	81.9 (48.4–116)
Caribbean	234.4 (164.8–317.4)	0.4 (0.3–0.6)	−14.5 (−28 to −0.6)	236 (166.2–319.3)	0.4 (0.3–0.6)	−16.6 (−29.6 to −3.5)	7121.7 (5166.2–9525.5)	13.5 (9.8–17.9)	−15.1 (−29 to −0.8)
Central Europe	1277.1 (900.8–1745.3)	0.6 (0.5–0.8)	−16.3 (−26.5 to −4.6)	1312 (924.7–1810.8)	0.6 (0.5–0.9)	−17.9 (−27.8 to −6.3)	34 676.4 (24 322.5–46 549.4)	18.4 (13.4-24.4)	−21.8 (−31.5 to −10.4)
Eastern Europe	1646.4 (1353.3–2004.6)	0.5 (0.4–0.6)	16.7 (6.3– 28.2)	1655.2 (1365.6–2007)	0.5 (0.4–0.6)	15 (5.3–26.5)	49 543.3 (40 975.8–60 027.9)	16.4 (13.7–19.6)	11.2 (1.6 –21.8)
Central Asia	1398.9 (971.2–1939.1)	1.5 (1.1–2.1)	−25.7 (−37.6 to −11.8)	1405.1 (966.1–1956.9)	1.6 (1.1–2.2)	−25.2 (−37.3 to −10.9)	46 762 (32 878.5–65 100.2)	49.2 (34.9–67.8)	−28.3 (−40.2 to −14.8)
North Africa and Middle East	6303.7 (4749–8154.2)	1.3 (0.9–1.6)	−1.4 (−27.6 to 21.8)	6219.2 (4680.6–8079)	1.3 (1–1.7)	−3.4 (−29.3 to 19.6)	198 545.3 (151 319.8–254 187.7)	37 (28–47.8)	−6.8 (−31.2 to 15.2)
South Asia	14 004.2 (11 808.4–16 799.7)	0.9 (0.7–1.1)	19.4 (1.7–39.1)	14 172.2 (11 904–16 996.1)	0.9 (0.8–1.1)	19.2 (1.7–38.8)	456 528.9 (386 459.9–542 189.9)	27.3 (23.2–32.6)	15.5 (−1.7 to 34.7)
Southeast Asia	17 700.8 (12 864.2–24 371.2)	2.5 (1.8–3.4)	−19.1 (−39.5 to 6.3)	17 131.5 (12 363.6–23 545.5)	2.4 (1.8–3.3)	−22 (−41.3 to 2.6)	564 561.2 (418 982.2–764 927.2)	75.5 (56–103.4)	−23 (−41.8 to 0.8)
East Asia	122 179.5 (95 828.7–157 144.4)	5.7 (4.5–7.3)	−12.8 (−35.1 to 17.4)	103 357.9 (80 924.5–132 256.4)	4.8 (3.8–6.1)	−25.8 (−44.4 to −0.7)	3 247 052 (2 546 030–4 193 951.3)	155.1 (122.1–199.6)	−29 (−47.4 to −3.5)
Oceania	149.4 (91.4–296.8)	1.7 (1–3.3)	−28.8 (−47 to 8.5)	147.4 (89.9–294)	1.7 (1–3.4)	−29.4 (−47.2 to 6.6)	5233.7 (3172.7–10 386.2)	52.6 (32.1–104.9)	−29.2 (−47 to 8)
Western sub-Saharan Africa	8892.1 (6801.9–11 252.1)	3.8 (2.9–4.7)	−35.9 (−54.4 to −0.4)	8912.3 (6855.6–11 345.6)	3.9 (3–4.9)	−35.7 (−54.5 to 0.8)	319 286.1 (246 128.2–406 001.6)	118.3 (90.8–149.9)	−36.5 (−54.6 to −0.3)
Eastern sub-Saharan Africa	3086.6 (2121.4–4672.5)	1.5 (1–2.3)	−19.3 (−44.1 to −37.5)	3089.5 (2114.6–4707.9)	1.6 (1.1–2.4)	−19.5 (−44.3 to 36.9)	112 354.5 (77 315.7–168 603.8)	47.8 (32.6–73)	−19.2 (−44.4 to 36)
Central sub-Saharan Africa	730.7 (325.8–1738.8)	1 (0.5–2.4)	−38.7 (−55.8 to −16)	724 (321.1–1761.3)	1.1 (0.5–2.6)	−38.6 (−56.2 to −14.9)	27 733.2 (12 495.1–68 107.8)	34.1 (15.1–82.8)	−38.9 (−57.3 to −14.1)
Southern sub-Saharan Africa	1661.3 (1330.6–2098.3)	2.5 (2–3.1)	30.3 (−25.3 to 147.3)	1640.6 (1309.4–2072.5)	2.5 (2–3.1)	30.4 (−25.5 to 146.4)	60 371.9 (48 434.7–76 245)	83.8 (67.4–105.6)	27.5 (−26.5 to 142.6)

ASR, age-standardized rate; DALY, disability-adjusted life years; UI, uncertainty interval.

Fig. 1.

Comparison of global LCDHB incidence cases, deaths, DALYs cases in 1990 and 2021. (a) Number of incidence cases. (b) Number of deaths. (c) Number of DALYs. DALY, disability-adjusted life years; LCDHB, liver cancer due to hepatitis B.

In 2021, LCDHB caused 181 194.3 deaths (95% uncertainty interval: 148 896.5–221 685.3), a 41.2% increase from 106 514.0762 deaths in 1990 (95% uncertainty interval: 124 291.2347–91 939.5928) (Table 1 and Fig. 1b). The global ASDR in 2021 was 2.1 (95% uncertainty interval: 1.7–2.6), which shows a decrease of 16.4% since 1990 (95% uncertainty interval: −31.2 to 3) (Table 1).

In 2021, the total number of DALYs for global LCDHB was 5 668 199.3 (95% uncertainty interval: 4 706 886.5–6 885 071.4), an increase of 33.9% from 3 748 179.417 in 1990 (95% uncertainty interval: 4 362 510.79–3 255 633.565) (Table 1 and Fig. 1c). The age-standardized DALYs rate in 2021 was 65.4 cases per 100 000 populations (95% uncertainty interval: 54.4–79.4), which indicates a 22.3% decline since 1990 (95% uncertainty interval: −36 to −4.5) (Table 1).

Regional burden of liver cancer due to hepatitis B

As shown in Table 1, in 2021, the regions with the highest number of new cases of LCDHB were East Asia [122 179.5 (95% uncertainty interval: 95 828.7–157 144.4)], Southeast Asia [17 700.8 (95% uncertainty interval: 12 864.2–24 371.2)], and high-income Asia Pacific [14 818.3 (95% uncertainty interval: 11 974–18 162.6)]. During the same period, the highest number of deaths caused by LCDHB occurred in East Asia [103 357.9 (95% uncertainty interval: 80 924.5–132 256.4)], Southeast Asia [17 131.5 (95% uncertainty interval: 12 363.6–23 545.5)], and South Asia [14 172.2 (95% uncertainty interval: 11 904–16 996.1)]. In terms of DALYs in 2021, East Asia reported the highest number [3 247 052 (95% uncertainty interval: 2 546 030–4 193 951.3)], followed by Southeast Asia [564 561.2 (95% uncertainty interval: 418 982.2–764 927.2)] and South Asia [456 528.9 (95% uncertainty interval: 386 459.9–542 189.9)].

In 2021, the ASIR of LCDHB was highest in East Asia [5.7 (95% uncertainty interval: 4.5–7.3)], high-income Asia Pacific [4 (95% uncertainty interval: 3.2–4.8)], and Western sub-Saharan Africa [3.8 (95% uncertainty interval: 2.9–4.7)], while it was lowest in Southern Latin America [0.3 (95% uncertainty interval: 0.2–0.5)], Tropical Latin America [0.3 (95% uncertainty interval: 0.3–0.4)], and Central Latin America [0.3 (95% uncertainty interval: 0.3–0.5)]. In the same year, the ASDR for LCDHB was highest in East Asia [4.8 (95% uncertainty interval: 3.8–6.1)], Western sub-Saharan Africa [3.9 (95% uncertainty interval: 3–4.9)], and high-income Asia Pacific [2.7 (95% uncertainty interval: 2.2–3.3)], while it was lowest in Southern Latin America [0.3 (95% uncertainty interval: 0.2–0.5)], Central Latin America [0.3 (95% uncertainty interval: 0.3–0.5)], and Tropical Latin America [0.4 (95% uncertainty interval: 0.3–0.4)]. In terms of age-standardized DALYs rate, East Asia [155.1 (95% uncertainty interval: 122.1–199.6)], Western Sub-Saharan Africa [118.3 (95% uncertainty interval: 90.8–149.9)], and Southern sub-Saharan Africa [83.8 (95% uncertainty interval: 67.4–105.6)] had the highest rates, while Southern Latin America [9.1 (95% uncertainty interval: 6.5–12.5)], Central Latin America [10.3 (95% uncertainty interval: 7.8–13.4)], and Tropical Latin America [10.9 (95% uncertainty interval: 9.2–12.8)] had the lowest.

From 1990 to 2021, the largest increase in ASIR of LCDHB was observed in Australasia [147 (95% uncertainty interval: 115.2–183.4)], high-income North America [103.9 (95% uncertainty interval: 94.7–112.4)], and Southern Latin America [96.1 (95% uncertainty interval: 61.4–132.4)], while the largest decreases were found in Central sub-Saharan Africa [−38.7 (95% uncertainty interval: −55.8 to −16)], Western sub-Saharan Africa [−35.9 (95% uncertainty interval: −54.4 to −0.4)], and Oceania [−28.8 (95% uncertainty interval: −47 to 8.5)]. During the same period, the greatest increases in ASDR were in Australasia [110.3 (95% uncertainty interval: 81.6–142.3)], Southern Latin America [89.6 (95% uncertainty interval: 56–124.2)], and high-income North America [80.9 (95% uncertainty interval: 72.3–88.1)], with the largest reductions occurring in high-income Asia Pacific [−44.7 (95% uncertainty interval: −56.4 to −29.6)], Central sub-Saharan Africa [−38.6 (95% uncertainty interval: −56.2 to −14.9)], and Western sub-Saharan Africa [−35.7 (95% uncertainty interval: −54.5 to 0.8)]. In 2021, the largest increases in age-standardized DALYs rates were seen in Australasia [96.9 (95% uncertainty interval: 70.4–125.9)], Southern Latin America [81.9 (95% uncertainty interval: 48.4–116)], and high-income North America [72.3 (95% uncertainty interval: 63.6–79.7)], while the regions with the largest decreases were high-income Asia Pacific [−50.8 (95% uncertainty interval: −61.5 to −35.9)], Central sub-Saharan Africa [−38.9 (95% uncertainty interval: −57.3 to −14.1)], and Western sub-Saharan Africa [−36.5 (95% uncertainty interval: −54.6 to −0.3)].

National burden of liver cancer due to hepatitis B

At the national level, the ASIR of LCDHB ranges from 0.2 to 17.2 cases per 100 000 population. The highest ASIRs are found in Mongolia [17.2 (95% uncertainty interval: 11–25.2)], the Republic of Paraguay [16.6 (95% uncertainty interval: 10.3–25)], and the Commonwealth of the Bahamas [13.3 (95% uncertainty interval: 8.9–18.6)], while the lowest ASIRs are recorded in the Federated States of Micronesia [0.2 (95% uncertainty interval: 0.2–0.3)], the Kingdom of Eswatini [0.2 (95% uncertainty interval: 0.2–0.3)], the Republic of El Salvador [0.2 (95% uncertainty interval: 0.2–0.4)], the Republic of Peru [0.2 (95% uncertainty interval: 0.1–0.3)], and the Republic of Rwanda [0.2 (95% uncertainty interval: 0.2–0.3)] (Fig. 2a and Table S1, Supplemental digital content 1, https://links.lww.com/EJGH/B203). In 2021, the ASDR for LCDHB ranged from 5.4 to 531.3 cases per 100 000 populations, with the highest ASDRs in the Republic of the Congo [531.3 (95% uncertainty interval: 327–792)], Mongolia [518.1 (95% uncertainty interval: 339.3–773.9)], and the Republic of Maldives [401.2 (95% uncertainty interval: 267.4–559)]. The lowest ASDRs were observed in the Kingdom of Morocco [5.4 (95% uncertainty interval: 3.4–7.9)], the Kingdom of Sweden [6.6 (95% uncertainty interval: 4.8–8.8)], and the Argentine Republic [6.7 (95% uncertainty interval: 4.8–9)] (Fig. 2b and Table S2, Supplemental digital content 1, https://links.lww.com/EJGH/B203). During the same period, the age-standardized DALYs rate for LCDHB ranged from 0.2 to 16.6 cases per 100 000 population. The highest rates were recorded in Mongolia [16.6 (95% uncertainty interval: 10.8–24.4)], the Republic of the Gambia [16.3 (95% uncertainty interval: 10.1–24.5)], and the Republic of Mali [12.8 (95% uncertainty interval: 8.6–17.7)], while the lowest rates were in the Argentine Republic [0.2 (95% uncertainty interval: 0.2–0.3)], the Kingdom of Morocco [0.2 (95% uncertainty interval: 0.1–0.3)], the Kingdom of Sweden [0.2 (95% uncertainty interval: 0.2–0.3)], the Republic of El Salvador [0.2 (95% uncertainty interval: 0.2–0.4)], and the United Mexican States [0.2 (95% uncertainty interval: 0.2–0.3)] (Fig. 2c and Table S3, Supplemental digital content 1, https://links.lww.com/EJGH/B203).

Fig. 2.

ASIR (a), ASDR (b), and age-standardized DALYs rate (c) of liver cancer due to hepatitis B per 100 000 population by country in 2021. ASDR, age-standardized death rates; ASIR, age-standardized incidence rates; DALY, disability-adjusted life years.

From 1990 to 2021, the country with the largest percentage increase in ASIR was the People’s Republic of China [170.6 (95% uncertainty interval: −26.1 to 1091.9)], followed by the Republic of Zambia [166.6 (95% uncertainty interval: 155.8–178.2)] and the Commonwealth of Dominica [153 (95% uncertainty interval: 119.1–191.7)]. Conversely, the countries with the largest percentage decrease were the Kingdom of Saudi Arabia [−85 (95% uncertainty interval: −86.5 to −83.5)], the Republic of Indonesia [−79.2 (95% uncertainty interval: −84.1 to −73.6)], and the Republic of Korea [−63.8 (95% uncertainty interval: −73.5 to −49.6)] (Table S1, Supplemental digital content 1, https://links.lww.com/EJGH/B203). During the same time frame, the Kingdom of Lesotho [193.4 (95% uncertainty interval: −21.7 to 1262.4)], the UK of Great Britain and Northern Ireland [157.2 (95% uncertainty interval: 148.4–166.9)], and the Republic of Poland [154.9 (95% uncertainty interval: 119.8–193.7)] saw the largest percentage increases in age-standardized death rates, while the Republic of Mauritius [−84.7 (95% uncertainty interval: −86.3 to −83.2)], the State of Kuwait [−81.8 (95% uncertainty interval: −86.4 to −76.4)], and the Republic of Korea [−67 (95% uncertainty interval: −76 to −53.3)] experienced the largest percentage decreases (Table S2, Supplemental digital content 1, https://links.lww.com/EJGH/B203). From 1990 to 2021, the UK of Great Britain and Northern Ireland [210 (95% uncertainty interval: 197.4–223.7)], the Kingdom of Lesotho [175.3 (95% uncertainty interval: −24.9 to 1124.7)], and Australia [161.6 (95% uncertainty interval: 121.5–207.3)] had the largest percentage increases in age-standardized DALYs rate, while the Republic of Mauritius [−84.3 (95% uncertainty interval: −85.9 to −82.7)], the State of Kuwait [−78.8 (95% uncertainty interval: −83.8 to −72.8)], and the Republic of Kazakhstan [−64 (95% uncertainty interval: −72.2 to −54.1)] had the largest percentage decreases in age-standardized DALYs rate (Table S3, Supplemental digital content 1, https://links.lww.com/EJGH/B203).

Burden of liver cancer due to hepatitis B by age and gender

In 2021, the ASIR of LCDHB was higher in males globally and increased with age. Both male and female ASIRs peaked in the 85–89 age group in 2021, followed by a decline. The number of new cases also increased with age, with males peaking in the 55–59 age group and females peaking in the 65–69 age group (Fig. 3a). In 2021, the ASDR for LCDHB was similarly higher in males and increased with age, with both male and female ASDRs peaking in the 85–89 age group, followed by a downward trend. Concurrently, the number of deaths because of LCDHB peaked in the male 55–59 age group and the female 65–69 age group (Fig. 3b). In 2021, the age-standardized DALYs rate for LCDHB reached its peak in the 65–69 age group for both males and females, while the number of DALYs peaked in males aged 50–54 and females aged 55–59 (Fig. 3c).

Fig. 3.

Number of incidence, deaths, and DALYs cases globally and ASIR, ASDR, and age-standardized DALYs rate of LCDHB per 100 000 population, by age and sex in 2021. Lines indicate prevalent cases with 95% uncertainty intervals for men and women. (a) Incidence; (b) deaths; (c) DALYs. ASDR, age-standardized death rates; ASIR, age-standardized incidence rates; DALY, disability-adjusted life years; LCDHB, liver cancer due to hepatitis B.

Liver cancer due to hepatitis B burden by sociodemographic index

At the regional level, we observed a reverse U-shaped correlation between the age-standardized DALYs rates of LCDHB and SDI from 1990 to 2021 (P = 0.004). The age-standardized DALYs rates exhibited exponential growth with increasing SDI, peaking at approximately 0.45 and then declining, with another increase in age-standardized DALYs when the SDI reached 0.75. From 1990 to 2021, East Asia, high-income Asia Pacific, Southeast Asia, and Central Asia exhibited rates above their expected SDI levels. In contrast, North Africa and the Middle East, Andean Latin America, Central Europe, the Caribbean, Central Latin America, Southern sub-Saharan Africa, Southern Latin America, Eastern Europe, Eastern sub-Saharan Africa, Western Europe, Australasia, and high-income North America were below their expected SDI levels in all years (Fig. 4a). There was no significant correlation between age-standardized incidence rates, mortality rates, and SDI (Figs. S1 and S2, Supplemental digital content 2, https://links.lww.com/EJGH/B204).

Fig. 4.

Age-standardized DALYs rates of LCDHB from 1990 to 2021, categorized by SDI across the 21 Global Burden of Disease regions and 204 countries and territories. (a) The 21 Global Burden of Disease regions. (b) A total of 204 countries and territories. The black line represents the expected value. (c) Temporal trajectories of key nations demonstrating SDI-DALYs dynamics. Arrows indicate progression from 1990 (tail) to 2021 (head), colored by country. DALY, disability-adjusted life years; LCDHB, liver cancer due to hepatitis B; SDI, sociodemographic index.

At the national level, a reverse U-shaped correlation between age-standardized DALY rates and SDI was observed from 1990 to 2021 (P ≤ 0.001). The age-standardized DALY rates increased with rising SDI, peaking at approximately 0.24 before showing an overall declining trend. From 1990 to 2021, countries such as Mongolia, Mali, Tonga, Guinea-Bissau, and Guinea were above their expected SDI levels. In contrast, Morocco, Sweden, Argentina, Mexico, and Mauritius remained below their expected SDI levels in all years (Fig. 4b). The correlations between ASIR and ASMR at the national level followed a similar pattern to that of age-standardized DALY rates (Figs. S3 and S4, Supplemental digital content 2, https://links.lww.com/EJGH/B204).

We specifically illustrated the SDI-DALY trajectories from 1990 to 2021 for four selected countries: China, Saudi Arabia, Japan, and the USA. Among these countries, China’s 2021 age-standardized DALYs rate was relatively high for its corresponding SDI value. The trajectory indicates that as SDI increased, the age-standardized DALYs rate initially rose before undergoing a continuous decline, reaching a relatively lower level by 2021. Saudi Arabia’s curve exhibits a more stable upward trend in SDI, with moderate fluctuations in its DALYs rate. Compared with China, Saudi Arabia maintained a lower age-standardized DALYs rate for a given SDI value. In contrast, Japan and the USA, as high-income developed countries, demonstrated relatively lower age-standardized DALYs rates, aligning with their advanced healthcare infrastructure and disease management strategies (Fig. 4c).

Future projections of the global burden of liver cancer due to hepatitis

It is expected that from 2021 to 2051, the global burden of LCDHB will change significantly, with varying trends across different indicators. The ASIR of LCDHB for both genders is anticipated to decline, dropping from approximately 3.8 cases per 100 000 populations in 2021 to about two cases per 100 000 populations by 2051, representing a decrease of approximately 47.3% over 30 years (Fig. 5a). The global ASDR of LCDHB is projected to fall from about 3.5 cases per 100 000 populations in 2021 to around 1.8 cases per 100 000 populations by 2051, amounting to a decrease of about 48.6% over the same period (Fig. 5b). Meanwhile, the age-standardized DALYs rate for global LCDHB is also expected to show a declining trend, decreasing from around 90 per 100 000 populations in 2021 to approximately 50 per 100 000 populations, which corresponds to a reduction of about 44.4% over 30 years (Fig. 5c). Notably, among the three indicators, although the gap between males and females seems to narrow over time, the ASIR, ASDR, and age-standardized DALYs rates for males remain consistently higher compared with females. It is important to note that the 95% uncertainty intervals for the incidence, mortality, and DALYs presented in this study reflect the uncertainty inherent in the data. These intervals were calculated by determining the 2.5 and 97.5% percentiles of ordered samples, rather than providing precise values. Because the data are sourced from multiple channels across different regions globally, with varying data quality and differences in disease diagnostic standards and data collection methods, the resulting 95% uncertainty intervals are relatively wide. This variability, to some extent, affects the accuracy and precision of the data. Nevertheless, these intervals help to provide an approximate range of the true data and offer a reference for assessing the reliability of the study’s findings.

Fig. 5.

Future projections of the global burden of LCDHB. (a) ASIR. (b) ASDR. (c) Age-standardized DALYs rate. ASDR, age-standardized death rates; ASIR, age-standardized incidence rates; DALY, disability-adjusted life years; LCDHB, liver cancer due to hepatitis B.

Discussion

This study provides the first comprehensive analysis of the global burden of LCDHB from 1990 to 2021, revealing significant differences in LCDHB burden among different regions, genders, and age groups. The findings indicate that in 2021, there were 206 365.7 new cases, 181 194.3 deaths, and 5 668 199.3 DALYs cases reported for LCDHB globally. Although the ASIR, ASDR, and age-standardized DALYs rate have declined over the past 31 years, the absolute numbers have increased. This trend is consistent with previous reports on the global burden of primary liver cancer [18]. Previous studies have suggested that the differences in the number and rate of DALYs, deaths, and cases may be explained by the aging population structure [3,18,19]. Currently, the global population’s age structure is undergoing significant changes. Between 1990 and 2019, the global population increased by 43% (from 5394.7 to 7713 million), while the number of individuals aged 65 and above surged by 114.6% (from 327.6 to 702.9 million). It is projected that by the end of 2030, one in six people globally will be aged 60 or older [20]. Population aging is also notably pronounced across regions. For instance, in East Asia, the proportion of individuals aged 65 and above was 6% in 1990, but this percentage increased to 12% by 2021 [21]. The growth and aging of the global population may contribute to the observed decoupling of trends in frequency, ASIR, ASDR, and age-standardized DALYs rate over time [22]. These high-burden regions exhibit a typical contradiction between the characteristics of HBV-endemic areas and the transitional phase of HBV prevention and control measures [23,24] – namely, the persistence of disease burden inertia in the early stages of universal HBV immunization and screening programs, followed by a gradual realization of long-term prevention benefits. Taking representative countries as examples, China established a nationwide neonatal HBV vaccination program, increasing coverage from 60% in 2000 to 99.38% in 2021 [25]. This has resulted in a synchronous decline in ASDR and age-standardized DALYs rates, although ASIR has shown an increasing trend. In contrast, Nigeria, located in Western sub-Saharan Africa, had a vaccination coverage rate of only 56% in 2021, but this still represented a substantial improvement from just 8% in 2004. This significant increase in coverage has contributed to a more pronounced reduction in LCDHB disease burden, with ASIR, ASDR, and age-standardized DALYs rates all exhibiting substantial declines. We hypothesize that in China, factors such as population aging and the rising prevalence of other liver-related risk factors due to lifestyle changes may offset the impact of HBV vaccination coverage on the overall LCDHB disease burden. Conversely, in Nigeria and other Western Sub-Saharan African countries, a relatively younger population structure and the more concentrated impact of vaccination on high-risk age groups may have contributed to a more substantial reduction in LCDHB burden.

We observed a declining trend in the ASDR in the high-income Asia Pacific region between 1990 and 2021; however, the ASDR remained high in 2021. This may be attributed to several factors: first, the rapid aging process in the region [26], with a large elderly population at high risk of infection, has extended the window for hepatocarcinogenesis [27] and has been compounded by a long-standing high burden of LCDHB [28]. In addition, liver cancer mortality is closely linked to the availability of healthcare resources for early disease detection and the opportunity for potentially curative treatments [29]. In recent years, advancements in liver cancer detection and diagnostic methods, particularly in countries such as Japan and South Korea [30], have become more widespread, which may, to some extent, contribute to the higher ASDR reported in the region.

From 1990 to 2021, the country with the largest percentage increase in age-standardized incidence rate was the People’s Republic of China, while the largest decrease was found in the Kingdom of Saudi Arabia. China is an area with a high prevalence of HBV, and the government has implemented various preventive measures such as vaccination programs, prevention of blood product transmissions, and interruption of mother-to-child transmission [31]; however, substantial work remains to reduce the burden of LCDHB. The Kingdom of Saudi Arabia has previously been described as a region with a high prevalence of HBV infection in the Middle East, but since 1989, the country has launched a vaccination program for all newborns and school-aged children. In addition, the development of healthcare facilities and improvements in socioeconomic status have significantly reduced the rate of HBV infections [32]. Over the past 31 years, many countries have seen a decline in LCDHB incidence, especially in previously heavily burdened Asian countries. Nevertheless, in some previously low-incidence areas, such as the USA, the incidence of LCDHB is increasing, indicating a clear need for strengthened prevention and control measures.

We also observed gender and age differences in the incidence of LCDHB. The incidence of LCDHB is generally higher in males compared with females. Studies have suggested that the synergistic interaction between HBV and androgens contributes to the development and progression of HCC [33]. The HBV X protein (HBx) mediates the c-Src kinase signaling pathway, leading to the phosphorylation of the androgen receptor, thereby enhancing androgen receptor activity. In addition, HBx regulates androgen receptor transcriptional activity through the glycogen synthase kinase 3β pathway, promoting HCC cell proliferation and migration [34]. Furthermore, the androgen receptor has been identified as a key factor in the male predominance of HBV-related HCC, as it facilitates HBV integration and point mutations in the promoter region of the telomerase reverse transcriptase (TERT) gene, leading to elevated TERT expression [35]. In addition, gender differences may also be attributed to other influencing factors, particularly smoking and alcohol consumption. According to WHO statistics, in 2021, the global smoking prevalence was 29% among men and only 5% among women [36]. Similarly, in 2019, approximately 6.7% of men were classified as heavy drinkers, compared with just 0.6% of women [37]. Globally, the prevalence of smoking and alcohol consumption is consistently higher in men than in women. Both smoking and alcohol consumption are major risk factors for liver cancer. Among individuals infected with HBV, the adverse effects of smoking and alcohol use are compounded, further exacerbating disease progression [38]. Because of their higher rates of smoking and alcohol consumption, men experience greater exposure to these risk factors, which significantly increases their risk of developing LCDHB compared with women.

We also found that the incidence of LCDHB in females is more pronounced in postmenopausal women. Previous reports have indicated that estrogen can protect hepatocytes from malignant transformation by downregulating interleukin 6 released by Kupffer cells, thus providing a protective effect for women [39]. Our research indicates that the ASIR for both males and females peaked in the 85–89 age group in 2021, followed by a downward trend. This phenomenon may be attributed to the lower vaccination rates among the elderly population and suboptimal immune responses because of age-related changes in the immune system [40]. As the global population ages, HBV infections among the elderly are expected to become more prevalent [41]. These results suggest that formulating precise prevention and control policies based on gender and age differences is extremely necessary to prevent the occurrence of LCDHB. Guidelines from the American Association for the Study of Liver Diseases, the European Association for the Study of the Liver, and the Asian Pacific Association for the Study of the Liver recommend semiannual abdominal ultrasound screening, with or without serum alpha-fetoprotein, for high-risk individuals (including chronic HBV subgroups and patients with liver cirrhosis of any etiology) to monitor for HCC [42–44]. Although current guidelines do not yet include specific recommendations for additional screening measures, based on the findings of this study, we propose that in LCDHB-endemic regions (such as East Asia), individuals aged 50 years and older, postmenopausal women, and those in the 85–89 age group should undergo annual HBV DNA testing [45] and transient elastography (FibroScan) screening [46]. These additional measures would help assess viral load and liver fibrosis progression, enabling the early detection of potential pathological changes.

Our study revealed a reverse U-shaped correlation between the age-standardized DALYs rates of LCDHB and SDI from 1990 to 2021. Although there is generally an association between LCDHB burden and SDI, it should not be considered in isolation. We found that the high burden of LCDHB is not limited to low SDI countries; we also observed relatively high LCDHB burdens in high SDI countries. Despite high-income countries possessing resources such as financial assets, healthcare infrastructure, and medical technologies, they are still not immune to the burden of HBV infections and LCDHB. One possible explanation for this phenomenon is the presence of competing risk factors in high SDI regions, with the increasing burden of nonalcoholic fatty liver disease (NAFLD) being particularly prominent. As the prevalence of obesity, diabetes, and metabolic syndrome continues to rise globally, NAFLD has increasingly overlapped with HBV-related liver diseases, further complicating the dynamics of disease burden. Recent studies have shown that HBV carriers with concurrent metabolic syndrome or NAFLD may experience an accelerated progression to liver fibrosis and HCC [47]. This etiological shift highlights the complexity of liver disease prevention and control in high SDI regions, where the reduction in LCDHB burden is partially offset by the growing impact of metabolic liver diseases. Therefore, when designing prevention and control strategies for LCDHB in high SDI regions, it is essential to integrate HBV prevention and treatment with metabolic liver disease management, ensuring the optimal allocation of healthcare resources to address the evolving epidemiological landscape of liver cancer.

Through BAPC model predictions, we observed that the global ASIR, ASDR, and age-standardized DALYs rates of LCDHB are expected to decline from 2021 to 2051. This result suggests that the primary prevention strategies against liver cancer through HBV vaccination are starting to show success. In 2016, the WHO set the goal of eliminating viral hepatitis as a public health problem by 2030, with specific targets for HBV aimed at reducing new infections by 95% and lowering the mortality rate by 65% from the baseline in 2015 [48]; however, if the current situation of underdiagnosis and undertreatment persists, the global HBV-related mortality rate is expected to increase by 39%. The challenges in treating LCDHB remain severe, and substantial work is still needed to address these issues.

The main strengths of this study include the analysis of the latest data from the GBD 2021 study, which provides the first comprehensive analysis of the global, regional, and national burden of LCDHB from 1990 to 2021, and examines differences between various age groups, genders, and SDIs while also forecasting the global LCDHB burden over the next 30 years; however, this study has several limitations. First, the accuracy of our study depends on the quality of the GBD 2021 data, which may be subject to underreporting and misdiagnosis of LCDHB, especially in low-income regions. Second, HBV infection may occur concurrently with other diseases (such as alcoholic liver disease and diabetes); this study focused solely on the burden of liver cancer caused by HBV without considering the multifactorial nature of the disease, which could lead to potential data distortion and should be analyzed in future research. Third, the data in this study are sourced from the GBD 2021 study and are therefore limited by the availability of original data and the statistical models used to generate estimates. Over-reliance on covariates and modeling measures may constrain the accuracy of the estimates. Fourth, the BAPC model has certain limitations, including high data requirements, sensitivity to covariance structures, complex assumptions, and significant computational demands. Its prediction mechanism primarily relies on the extrapolation of historical data trends. Under a stable environment without major interventions, this assumption remains reasonable; however, if unexpected interventions occur – such as the wider implementation of HBV vaccination programs, the emergence of new treatment methods, or major adjustments in public health policies – the model may fail to fully capture abrupt shifts in trends [49]. Consequently, the predicted outcomes may not precisely align with the actual future trajectory of LCDHB. Therefore, further accumulation of empirical evidence is necessary to comprehensively validate the accuracy of these projections.

Conclusion

With the implementation of primary prevention strategies for liver cancer, such as HBV vaccination, the ASIR, ASDR, and age-standardized DALYs rate of LCDHB have decreased from 1990 to 2021; however, the corresponding absolute numbers have still increased, indicating that LCDHB remains a persistent global threat. The burden of LCDHB varies by region, gender, and age. This study provides empirical evidence for understanding the global burden of LCDHB, which will aid in achieving causative prevention and in formulating targeted prevention and control policies for LCDHB at both global and regional levels.

Acknowledgements

We are indebted to individuals who have participated in this study or have helped with this article.

This study was supported by the National Natural Science Foundation of China (No. 82374539), the Major Project of Jiangsu Administration of Traditional Chinese Medicine (No. ZD202214), and the Jiangsu Province Chinese Medicine Leading Talents training object project (No. SLJ0327).

Conceptualization, data management, formal analysis, manuscript writing: P.Y. and W.H. Conceptualization, methodology, writing, review and editing: Y.X. and Q.L. Visualization and data management: X.S., J.Z., W.R., and Y.H. Data curation and investigation: Y.T. Conceptualization, supervision, and funding acquisition: P.S.

This research was approved by the Ethics Committee of Affiliated Hospital of Nanjing University of Chinese Medicine (Nanjing, China). The methods were carried out following the Declaration of Helsinki and its later amendments or comparable ethical standards. As data are available in a public, open-access repository, consent to participate is not applicable.

The data used are publicly available online on the website of the Institute for Health Metrics and Evaluation (https://vizhub.healthdata.org/gbd-results/).

Conflicts of interest

There are no conflicts of interest.