Global, regional, and national burden of primary liver cancer among middle-aged and elderly populations from 2000 to 2021 with a prediction to 2050: a cross-sectional analysis from Global Burden of Disease Study 2021

Abstract

Background:

Primary liver cancer (PLC) is a major cause of cancer mortality worldwide, particularly among middle-aged and elderly populations. Given increasing life expectancy, understanding PLC burden in these groups is essential.

Methods:

We obtained data about incidence, deaths, and disability-adjusted-life-years (DALYs) from 2000 to 2021 for PLC aged 55+ from Global Burden of Disease (GBD) 2021. Trends across sociodemographic index (SDI), etiology, age, and sex level were evaluated using age-standardized rates, estimated annual percentage change (EAPC), and percentage change. Bayesian-age-period-cohort (BAPC) model was applied to predict the global burden to 2050.

Results:

Globally, the cases of PLC burden among middle-aged and elderly populations increased by 74% in incidence, 67% in deaths, and 59% in DALYs. Despite rising case counts, EAPC for incidence, deaths, and DALYs declined with −0.13, −0.32, and −0.55. Trends varied by region and SDI level, with reductions in HBV- and HCV-related PLC but increases in non-alcoholic steatohepatitis (NASH) and alcohol use related PLC, particularly in high-SDI regions. Burden was the highest and increasing among those aged 75+, especially for NASH-related PLC. Globally, the male-to-female burden ratio was 2.19, though gender differences varied regionally. Finally, projection indicated that the incidence case will persistently rise from 2021 to 2050, while incidence rate will decrease first and then increase.

Conclusions:

Our study analyzed the global epidemiology of PLC among middle-aged and elderly populations from etiology, age and gender, and conducted a prediction outlook for the next 30 years. These findings contribute to the development of efficient and targeted health strategies to mitigate PLC burden.

Introduction

Primary liver cancer (PLC) is a prevalent malignant tumor, with nearly 865 269 new cases and 757 948 deaths reported globally in 2022^[1]. It ranks sixth among all cancers in incidence and is the third leading cause of cancer related mortality, following lung and colorectal cancers, thus significantly impacting global healthcare systems. The etiological factors for PLC primarily include infections with hepatitis B virus (HBV) and hepatitis C virus (HCV), alcohol consumption, nonalcoholic steatohepatitis (NASH), obesity and dietary exposure to aflatoxins^[2-4]. However, the principal risk factors vary by region. For instance, in China and Africa, HBV infection remains a major driver of liver cancer, whereas in Japan and Egypt, HCV infection is the predominant risk factor^[5-7]. Although there are a variety of treatments for PLC, including interventional, immunotherapy and targeted therapy, surgical resection and liver transplantation remain the main treatment^[8].

Globally, population aging is becoming increasingly severe. According to the United Nations’ World Population Prospects 2019, the number of elderly individuals is projected to double by 2050, reaching 1.5 billion^[9]. With this progression in aging, the incidence of various cancers is also rising. One study predicted that by 2035, individuals aged 65 and older will account for approximately 60% of all new cancer diagnoses^[10]. Similarly, a report on liver cancer indicates that the proportion of newly diagnosed liver cancer patients aged 70 and above increased from 39.6% (1998–2002) to 47.5% (2013–2016)^[11]. Like PLC risk factors, the pace and extent of population aging differ significantly across regions and over time. As a result, the PLC burden among the elderly is expected to manifest clear disparities between countries. However, most previous studies on PLC burden have focused on specific countries or regions, particular etiologies, or analyses in children and adolescents^[12–14]. Currently, research specifically examining the PLC burden in middle-aged and elderly populations is notably limited.

This study aims to utilize data from the Global Burden of Disease (GBD) 2021 to comprehensively analyze the global trends in PLC burden among middle-aged and elderly populations over the past two decades, and to project future disease burden. Our findings are intended to inform the allocation of healthcare resources, guide the formulation of targeted prevention strategies, and aid in prioritizing public health initiatives. Moreover, these insights may help reshape clinical management and surgical decision-making for PLC patients in aging populations.

Methods

Data source

The GBD 2021 provides the latest estimates on the burden of over 300 diseases and injuries across 21 GBD regions and 204 countries^[15]. All data are freely accessible and downloadable via the Global Health Data Exchange (https://ghdx.healthdata.org/). Details on statistical methods for data collection and modeling are available in previous studies^[16]. We extracted data including incidence, deaths, and DALYs of PLC among middle-aged and elderly populations (aged 55 and above). PLC was defined using ICD-10 code C22, and detailed methods for estimating incidence, deaths and DALYs rates have been described in multiple studies^[14,17]. In GBD framework, PLC etiologies are categorized into five types: HBV, HCV, alcohol consumption, NASH, and other causes (e.g., liver fluke infections, obesity, and aflatoxin exposure). The University of Washington Institutional Review Board reviewed and approved a waiver of informed consent for the GBD study. The work has been reported in line with the strengthening the reporting of cohort, cross-sectional, and case-control studies in surgery (STROCSS) guideline^[18].

SDI

The socio-demographic index (SDI) is a composite measure to assess a country’s development level, closely tied to population health outcomes. Specifically, SDI is the geometric mean of three factors: average educational attainment among individuals aged 15 years and older, total fertility rate among those under 25, and lag-distributed income per capita^[19]. An SDI score of 0 represents the lowest theoretical level of health-related development, while a score of 1 indicates the highest. In GBD 2021, countries are categorized into five SDI regions: low, low-middle, middle, high-middle, and high SDI^[15]. Supplemental Digital Content Table S1 (available at: http://links.lww.com/JS9/E469) provides SDI values for 204 countries worldwide in 2021, along with their respective GBD regions.

HIGHLIGHTS

• Global PLC incidence cases among middle-aged and elderly rose by 74% from 2000 to 2021, however declining incidence rates, with regional variations.

• NASH- and alcohol-related PLC increased, while HBV- and HCV-related PLC declined.

• Highest PLC burden was found in individuals aged 75+ and males.

• The predicted PLC incidence shows a U-shaped trend: decline until 2035, then rise through 2050.

DALYs

DALYs is the standard metric used to quantify disease burden, representing the total years of healthy life lost from disease onset to death. DALYs encompass years of life lost due to premature mortality and years lived with disability^[20].

Statistical analysis

We used percentage change to reflect the variation in the cases of incidence, deaths and DALYs from 2000 to 2021. The formula for this calculation is as follows:

$$Percentage=\frac{Case\in 2021-Case\in 2000}{Case\in 2000}$$

We used the age-standardized rate (ASR) and estimated annual percentage change (EAPC) to quantify temporal trends in incidence, deaths, and DALYs rates. ASR per 100 000 population was calculated using the following formula:

$$ASR=1000\ast \frac{\sum _{i=1}^A{a}_i{b}_i}{\sum _{i=1}^A{b}_i}$$

(α_i: the age-specific rate in i^th age group; b: the number of people in the corresponding i^th age group among the standard population; A: the number of age groups.)

EAPC is the common metric in epidemiological research, used to determine temporal changes in ASR^[21,22]. The EAPC calculation is based on a linear regression model fitted to the natural logarithm of the ASR, with time as the independent variable. By fitting a straight line to the natural log of each observed ASR over the study period, the slope of this line provides the EAPC^[22]. This method offers a straightforward way to quantify trends, with positive EAPC values indicating an increasing trend and negative values signaling a decrease in ASR over time. The formula for this calculation is as follows:

$$Y=\alpha +\beta X+\epsilon$$

$$EAPC=100\ast \left(\mathrm{e}\mathrm{x}\mathrm{p}\left(\beta \right)-1\right)$$

(X: year, Y: the natural logarithm of rates (such as incidence rate), α: the intercept, β: the slope, ε: the random error. The 95% confidence intervals of the EAPC are also derived from this fitted model.)

Additionally, we employed Pearson correlation analysis to assess the relationship between the SDI and both ASR and EAPC across the 204 countries and regions. We also used the Bayesian age-period-cohort (BAPC) model to predict future trends in the burden of PLC to 2050. This model utilized Bayesian inference, implemented via the integrated nested Laplace approximation (INLA) approach^[23]. In the GBD study, the 95% UIs are defined by averaging data from 1000 draws. The upper and lower bounds of the 95% UIs are determined by the 2.5th and 97.5th ranked values from these 1000 draws, respectively^[15]. Statistical analysis was conducted using R software. P value of less than 0.05 was considered statistically significant.

Results

Global level

In 2021, approximately 524 000 new cases of PLC were reported worldwide among adults aged 20 and over, with nearly 76% of these cases occurring in middle-aged and elderly individuals (Supplemental Digital Content Figure S1, available at: http://links.lww.com/JS9/E469), and around 80% of PLC-related adult deaths were concentrated in this age group. The absolute numbers of the PLC burden among middle-aged and elderly populations globally have risen significantly. From 2000 to 2021, new cases increased from 231 000 to 401 170 (a 74% growth), deaths rose from 229 000 to 383 000 (a 67% increase), and DALYs escalated from 5.183 to 8.26 million (a 59% rise) (Table 1; Supplemental Digital Content Tables S2 and S3, available at: http://links.lww.com/JS9/E469). Despite these absolute increases, the ASRs for incidence, deaths, and DALYs declined during this period (Fig. 1A). The EAPCs were −0.13 (−0.22 to −0.05), −0.32 (−0.41 to −0.23), and −0.55 (−0.64 to −0.46) for incidence, deaths and DALYs (Table 1; Supplemental Digital Content Tables S2 and S3, available at: http://links.lww.com/JS9/E469).

Table 1.

The incidence case, rate and EAPC of liver cancer among middle-aged and elderly people at global and region level between 2000 and 2021

Characteristics	Incidence case			Incidence rate
	2000 (103)	2021 (103)	Percentage	2000 (per 105)	2021 (per 105)	EAPC_IR
All	231.01 [217.75–248.32]	401.17 [364.56–443.76]	0.74	28.35 [26.73–30.48]	27 [24.53–29.86]	−0.13 [−0.22 to −0.05]
Etiology
HBV	78.23 [65.23–92.88]	127.41 [100.69–159.47]	0.63	9.6 [8.01–11.4]	8.57 [6.78–10.73]	−0.35 [−0.47 to −0.22]
HCV	84.51 [74.13–95.98]	138.92 [118.09–160.48]	0.64	10.37 [9.1–11.78]	9.35 [7.95–10.8]	−0.53 [−0.62 to −0.44]
Alcohol	43.04 [35.13–52.11]	83.76 [67.36–101.91]	0.95	5.28 [4.31–6.4]	5.64 [4.53–6.86]	0.45 [0.37 to 0.53]
NASH	16.45 [12.6–20.83]	35.03 [27.68–43.63]	1.13	2.02 [1.55–2.56]	2.36 [1.86–2.94]	0.96 [0.84 to 1.09]
Other	8.78 [6.85–11.03]	16.05 [12.29–20.05]	0.83	1.08 [0.84–1.35]	1.08 [0.83–1.35]	0.17 [0.07 to 0.26]
Age
55–59 years	34.34 [31.67–37.87]	59.77 [52.91–68.92]	0.74	16.75 [15.44–18.47]	15.1 [13.37–17.42]	−0.71 [−0.84 to −0.58]
60–64 years	43.16 [40.47–46.48]	66.69 [60.64–74.75]	0.55	23.16 [21.72–24.94]	20.84 [18.95–23.36]	−0.15 [−0.35 to 0.05]
65–69 years	48.52 [45.65–51.56]	76.6 [69.97–84.95]	0.58	31.8 [29.92–33.8]	27.77 [25.37–30.8]	−0.32 [−0.54 to −0.1]
70–74 years	44.81 [42.2–47.83]	66.64 [60.86–73.81]	0.49	37.62 [35.42–40.15]	32.37 [29.57–35.86]	−0.73 [−0.76 to −0.7]
75+ years	60.18 [54.74–65.17]	131.47 [113.71–143.58]	1.18	39.67 [36.08–42.96]	45.56 [39.41–49.76]	0.71 [0.5 to 0.93]
Sex
Male	153.94 [144.37–165.57]	265.13 [238.36–303.46]	0.72	40.44 [37.93–43.5]	37.9 [34.08–43.38]	−0.2 [−0.29 to −0.12]
Female	77.06 [70.34–85.4]	136.04 [120.67–150.51]	0.77	17.75 [16.2–19.67]	17.3 [15.34–19.14]	−0.04 [−0.13 to 0.06]
Region
High SDI	82.13 [76.77–85.18]	128.13 [115.25–136.41]	0.56	37.97 [35.5–39.38]	37.14 [33.41–39.54]	−0.24 [−0.36 to −0.11]
High-middle SDI	54.81 [50.19–59.8]	86.47 [75.22–99.32]	0.58	27.2 [24.9–29.67]	24.94 [21.7–28.65]	−0.18 [−0.32 to −0.03]
Middle SDI	60.35 [55.42–65.7]	125.19 [109.14–146.04]	1.07	27.14 [24.93–29.55]	26.64 [23.23–31.08]	0.25 [0.08 to 0.42]
Low-middle SDI	20.95 [18.56–25.16]	43.17 [39.31–47.89]	1.06	16.31 [14.45–19.6]	17.91 [16.31–19.86]	0.44 [0.42 to 0.46]
Low SDI	12.63 [9.88–16.92]	17.98 [14.91–22.29]	0.42	27.88 [21.81–37.33]	21.92 [18.17–27.16]	−1.12 [−1.19 to −1.05]
High-income Asia Pacific	49.75 [46.33–51.94]	53.75 [46.03–59.32]	0.08	106.96 [99.6–111.67]	76.24 [65.29–84.14]	−1.93 [−2.08 to −1.77]
High-income North America	10.79 [10.07–11.16]	31.05 [28.62–32.41]	1.88	16.57 [15.47–17.14]	27.59 [25.43–28.8]	2.67 [2.54 to 2.81]
Western Europe	26.28 [24.69–27.53]	43.35 [39.4–46.05]	0.65	24.34 [22.87–25.5]	29.07 [26.42–30.88]	1.15 [0.94 to 1.36]
Australasia	0.76 [0.69–0.82]	2.54 [2.25–2.84]	2.35	15.7 [14.26–17.1]	28.73 [25.46–32.2]	3.28 [3.09 to 3.48]
Andean Latin America	0.45 [0.38–0.52]	1.2 [0.92–1.52]	1.68	9.49 [8.06–11.13]	12.11 [9.33–15.32]	0.9 [0.72 to 1.09]
Tropical Latin America	1.33 [1.26–1.4]	3.76 [3.49–3.97]	1.82	6.43 [6.08–6.74]	8.5 [7.88–8.97]	1.37 [0.99 to 1.76]
Central Latin America	2.08 [1.99–2.15]	5.91 [5.36–6.48]	1.84	10.83 [10.36–11.19]	13.82 [12.54–15.16]	1.13 [1.09 to 1.17]
Southern Latin America	0.44 [0.4–0.47]	1.35 [1.21–1.48]	2.09	4.66 [4.33–4.99]	9.15 [8.19–10.07]	3.51 [3.16 to 3.86]
Caribbean	0.46 [0.42–0.52]	0.78 [0.68–0.89]	0.68	8.47 [7.74–9.48]	8.43 [7.37–9.57]	0.25 [−0.09 to 0.59]
Central Europe	3.85 [3.59–4.11]	5.65 [5.11–6.28]	0.47	13.65 [12.72–14.57]	15.27 [13.79–16.97]	0.5 [0.27 to 0.73]
Eastern Europe	4.51 [4.34–4.69]	6.7 [6.19–7.23]	0.48	8.97 [8.64–9.33]	10.79 [9.97–11.64]	0.78 [0.39 to 1.18]
Central Asia	2.98 [2.66–3.33]	4.4 [3.71–5.15]	0.48	35.36 [31.62–39.62]	30.22 [25.51–35.36]	−1.01 [−1.21 to −0.81]
North Africa and Middle East	7.66 [6.46–9.94]	17.6 [14.93–20.49]	1.3	21 [17.7–27.26]	23.09 [19.59–26.88]	0.62 [0.51 to 0.73]
South Asia	12.07 [10.62–13.76]	30.83 [27.75–34.26]	1.56	9.68 [8.52–11.04]	12.42 [11.18–13.8]	1.2 [1.12 to 1.29]
Southeast Asia	16.6 [14.58–18.98]	31.38 [24.56–40.22]	0.89	30.39 [26.71–34.76]	27.39 [21.44–35.11]	−0.66 [−0.75 to −0.57]
East Asia	74.34 [65.55–83.92]	138.5 [113.82–169.07]	0.86	40.17 [35.43–45.35]	35.32 [29.03–43.12]	−0.17 [−0.39 to 0.06]
Oceania	0.1 [0.07–0.18]	0.17 [0.11–0.3]	0.62	17.29 [11.91–30.66]	13.4 [9.08–24.1]	−1.49 [−1.62 to −1.35]
Western Sub-Saharan Africa	8.8 [6.49–12.18]	11.3 [9.41–13.21]	0.28	49.56 [36.54–68.62]	35.17 [29.26–41.1]	−1.61 [−1.69 to −1.53]
Eastern Sub-Saharan Africa	4.3 [3.57–5.3]	6.18 [4.82–8.1]	0.44	29.49 [24.49–36.3]	22.87 [17.81–29.95]	−1.36 [−1.43 to −1.29]
Central Sub-Saharan Africa	1.31 [0.64–2.73]	1.89 [0.93–3.93]	0.44	28.54 [13.94–59.54]	20.95 [10.28–43.54]	−1.52 [−1.72 to −1.32]
Southern Sub-Saharan Africa	2.17 [1.92–2.51]	2.88 [2.47–3.38]	0.32	38.16 [33.66–44.12]	29.55 [25.34–34.69]	−1.63 [−1.8 to −1.46]

EAPC: estimated annual percentage changes; GBD: Global Burden of Disease; SDI: socio-demographic index.

Figure 1.

Temporal trends of incidence, deaths and DALYs of primary liver cancer across middle-aged and elderly populations at global and region levels. (A) The incidence, deaths and DALYs rate from 1990 to 2021. (B) The EAPC of incidence, deaths and DALYs rate from 1990 to 2021. SDI: socio-demographic index; EAPC: estimated annual percentage change; DALYs: Disability-Adjusted Life Years.

SDI and GBD region level

At regional level, the high-SDI region bore the greatest burden in 2021, with incidence, deaths, and DALYs rates of 37.14, 31.61, and 598.67 per 100 000, respectively (Fig. 1A; Table 1). The low-middle-SDI region, despite having the lowest rates, exhibited notable upward trends, with EAPC of 0.44 (0.42 to 0.46), 0.39 (0.34 to 0.44), and 0.32 (0.26 to 0.37) for incidence, deaths and DALYs (Table 1; Supplemental Digital Content Tables S2 and S3, available at: http://links.lww.com/JS9/E469). Conversely, the low-SDI region experienced the sharpest declines, with EAPC of −1.12 (−1.19 to −1.05), −1.10 (−1.18 to −1.03), and −1.16 (−1.25 to −1.07) for incidence, deaths and DALYs. Most GBD regions reported increasing PLC burdens, with Australasia, High-income North America, and Southern Latin America experiencing the largest increases (Fig. 1B). Over the past two decades, the High-income Asia-Pacific consistently had the highest PLC burden (Table 1). Encouragingly, this region also recorded the steepest declines, with EAPC of −1.93 (−2.08 to −1.77), −2.14 (−2.26 to −2.02), and −3.28 (−3.43 to −3.14) for incidence, deaths and DALYs (Fig. 1B). Substantial declines were also observed in East Asia, Southeast Asia, and Sub-Saharan Africa (Fig. 1B).

National level

At national level, China recorded the highest number of PLC cases (133 258) in 2021, accounting for roughly one-third of the global total (Supplemental Digital Content Table S4, available at: http://links.lww.com/JS9/E469). Since 2000, PLC incidence has risen in nearly all countries, with the most significant increase observed in the Arab states (499%), and only 10 countries reported a decline, with Ukraine experiencing the steepest decrease (38%) (Supplemental Digital Content Table S4, available at: http://links.lww.com/JS9/E469). Mongolia had the highest incidence rate with 322 per 100 000, followed by Gambia and Mali, while Morocco recorded the lowest rate at 2.3 per 100 000 (Fig. 2A). Additionally, 79 countries showed rising incidence rates, with Poland, Guatemala, and UK demonstrating the fastest increases, with EAPC of 6.58, 5.49, and 4.78. Conversely, Zambia, Kazakhstan, and Ukraine showed the largest declines (Fig. 2B; Supplemental Digital Content Table S4, available at: http://links.lww.com/JS9/E469). Trends in deaths and DALYs largely paralleled incidence. China had the highest absolute number of deaths and DALYs, while Mongolia exhibited the highest rates of deaths and DALYs. Poland recorded the fastest increases in both deaths and DALYs rates, whereas Zambia showed the most significant declines (Supplemental Digital Content Figures S2 and S3, available at: http://links.lww.com/JS9/E469; Supplemental Digital Content Tables S5 and S6, available at: http://links.lww.com/JS9/E469).

Figure 2.

The global trends in incidence of primary liver cancer across middle-aged and elderly populations in 204 countries and territories. (A) The ASIR of primary liver cancer in 2021. (B) The EAPC in ASIR of primary liver cancer from 2000 to 2021. ASIR: age-standardized incidence rate; EAPC estimated annual percentage change.

Association between SDI and PLC burden

We identified a positive correlation between SDI and the EAPC of incidence rates across countries (R = 0.39, P < 0.001) (Fig. 3A). This suggests that higher SDI nations, such as in High-income North America and Western Europe, tend to experience faster incidence rate increases. In contrast, Western Sub-Saharan Africa countries showed the declining incidence rates. A nonlinear relationship between SDI and incidence rates was also evident: in countries with SDI below 0.5, lower SDI values were associated with higher incidence rates, whereas in countries with SDI above 0.75, higher SDI values corresponded to higher incidence rates (Fig. 3A). Similar trends were observed in the relationships between SDI and both death and DALYs rates, as well as their respective EAPC (Fig. 3B, 3C).

Figure 3.

The correlation between SDI and primary liver cancer burden at national level. (A) The correlation of incidence rate and EAPC of incidence rate with SDI in 2021. (B) The correlation of deaths rate and EAPC of deaths rate with SDI in 2021. (C) The correlation of DALYs rate and EAPC of DALYs rate with SDI in 2021. Each circle represents a country, color of circle represents the GBD region to which country belongs, and size of circle represents the number of cases. SDI: socio-demographic index; EAPC: estimated annual percentage change; DALYs: Disability-Adjusted Life Years; GBD: Global Burden of Disease.

Etiology pattern

From an etiological perspective, HCV and HBV infections remain the primary causes of PLC among middle-aged and elderly populations globally. From 2000 to 2021, HBV-related cases increased by 63%, HCV-related cases by 64%, alcohol-related cases by 95%, NASH-related cases by 113%, and cases due to other causes by 83% (Table 1). Despite annual declines in HBV- and HCV-related incidence rates, alcohol- and NASH-related rates have risen markedly (Fig. 4A). HBV-related PLC imposed a significant burden in middle and middle-high SDI regions, despite the downward trend. HCV-related PLC was most prevalent in high-SDI region. Alcohol- and NASH-related PLC incidence rates have risen significantly across all regions except low-SDI, with the sharpest increases observed in high-SDI region. Regionally, HCV-related PLC incidence rate was highest in High-income Asia Pacific, whereas HBV-related remained elevated in East Asia, and West Africa (Fig. 4B). Additionally, NASH- and alcohol-related incidence rates exhibited upward trends in nearly all regions, except High-income Asia Pacific and Sub-Saharan Africa (EAPC > 0) (Fig. 4C).

Figure 4.

The incidence burden of primary liver cancer across middle-aged and elderly populations by different etiologies at global and region levels. (A) The incidence rate trend from 2000 to 2021. (B) The incidence rate of global and GBD region in 2021. (C) The EAPC of incidence rate from 1990 to 2021. SDI: socio-demographic index; EAPC: estimated annual percentage change; NASH: nonalcoholic steatohepatitis.

The trend of deaths and DALYs by etiology closely mirrors the incidence. HCV-related PLC deaths now surpass those from HBV, despite global declines in both (Supplemental Digital Content Table S2, available at: http://links.lww.com/JS9/E469; Supplemental Digital Content Figure S4, available at: http://links.lww.com/JS9/E469). In contrast, alcohol- and NASH-related deaths and DALYs rate have risen significantly, particularly in high-SDI region (Supplemental Digital Content Figures S4 and S5, available at: http://links.lww.com/JS9/E469). Among these, NASH-related PLC death rate have shown the fastest growth in Australasia, while HBV-related PLC DALYs rate have declined most notably in High-income Asia Pacific.

At national level, the trends in HBV- and NASH-related PLC burdens were further analyzed. NASH-related PLC incidence rate have shown upward trends in countries such as China, Mongolia, Mexico, and Saudi Arabia, while HBV-related incidence rate have declined (Supplemental Digital Content Table S7, available at: http://links.lww.com/JS9/E469; Supplemental Digital Content Figure S6, available at: http://links.lww.com/JS9/E469). Similarly, in regions like South America and Middle East and North Africa, the growth in incidence of NASH-related is outpacing that of HBV. For deaths and DALYs, NASH-related burdens increasing more rapidly, compared to HBV, in most countries across North and South America, East Asia (Supplemental Digital Content Figures S7 and S8, available at: http://links.lww.com/JS9/E469; Supplemental Digital Content Tables S8 and S9, available at: http://links.lww.com/JS9/E469).

Age pattern

Globally, PLC incidence rates increase with age, peaking at 45.56 (39.41–49.76 per 100 000) in the 75+ age group (Table 1). From 2000 to 2021, incidence rates in the 75+ age group rose, while rates declined in all other age groups (Fig. 5A). This age-related trend was particularly pronounced in high-SDI region. In contrast, low-middle SDI region saw increases across all age groups, while low-SDI regions experienced declines (Fig. 5A). Etiology analysis in the 75+ age group indicates that NASH-related PLC demonstrated the fastest global growth (EAPC:1.3) (Fig. 5B). Nearly all regions experienced significant increases in NASH-related PLC incidence, except High-income Asia Pacific and Sub-Saharan Africa.

Figure 5.

The global incidence burden of primary liver cancer across middle-aged and elderly populations by age patterns. (A) The incidence rate by age group, global and 5 SDI regions. (B) The EAPC of incidence rate of liver cancer across 75+ years old populations from 1990 to 2021 at GBD region levels. (C) The EAPC of incidence rate of liver cancer across 75+ years old populations from 1990 to 2021 at national levels. SDI: socio-demographic index; ASIR: age-standardized incidence rate; EAPC estimated annual percentage change; NASH: nonalcoholic steatohepatitis.

At national level, PLC incidence in the 75+ age group surged most rapidly in North America, Western Europe, and Middle East, while Sub-Saharan Africa showed declining trends (Fig. 5C). Countries, like Poland, United Kingdom, exhibited the sharpest increases, whereas Mauritius, Bulgaria, and Ukraine showed the most significant declines (Fig. 5C; Supplemental Digital Content Table S10, available at: http://links.lww.com/JS9/E469). Age-specific trends for deaths and DALYs burdens closely mirrored those observed for incidence (Supplemental Digital Content Figures S9 and S10, available at: http://links.lww.com/JS9/E469; Supplemental Digital Content Tables S11 and S12, available at: http://links.lww.com/JS9/E469).

Sex pattern

Significant sex-based disparities existed in the PLC burden among middle-aged and elderly populations. Globally, in 2021, males accounted for approximately 265 000 new cases, nearly double that of females (136 000), with higher incidence rates in males (37.9 vs 17.3) (Table 1). Male incidence rates generally declined across most regions, with the steepest reductions in low-SDI areas (Supplemental Digital Content Figure S11, available at: http://links.lww.com/JS9/E469). The High-income Asia-Pacific region reported the highest incidence rates for both sexes and a male-to-female ratio of 2.45 (Fig. 6A). Except in Andean Latin America and Eastern Sub-Saharan Africa, male rates exceeded female rates across all regions. Australasia and Southern Latin America saw the most rapid male increases, while Africa and the High-income Asia-Pacific region showed the largest decreases (Fig. 6B). Notably, East Asia exhibited rising female and declining male incidence, whereas the Caribbean showed the reverse pattern. Trends in deaths and DALYs mirrored those of incidence (Supplemental Digital Content Figures S12 and S13, available at: http://links.lww.com/JS9/E469).

Figure 6.

The global incidence burden of primary liver cancer across middle-aged and elderly populations by sex patterns. (A) The incidence rate in male and female of global and GBD region in 2021. (B) The EAPC of incidence rate in male and female from 1990 to 2021. (C) The incidence rate ratio for male/female of liver cancer across middle-aged and elderly people in 2021 at national levels. SDI: socio-demographic index; EAPC: estimated annual percentage change.

The Maldives displayed the highest male-to-female incidence ratio at 6.05, while ratios below 1 were predominantly found in Sub-Saharan Africa (Fig. 6C). Poland showed the fastest increase in male incidence rates, surpassing the rate in females (EAPC: 8.32 vs 4.5) (Supplemental Digital Content Figure S14, available at: http://links.lww.com/JS9/E469; Supplemental Digital Content Table S13, available at: http://links.lww.com/JS9/E469). In contrast, countries such as the United States and Brazil exhibited faster growth in male incidence rates, whereas in China and Mongolia, female rates rose more rapidly, with male rates showing slight declines (Supplemental Digital Content Figure S14, available at: http://links.lww.com/JS9/E469; Supplemental Digital Content Table S13, available at: http://links.lww.com/JS9/E469). Similar sex-based differences were reflected in deaths and DALYs burdens (Supplemental Digital Content Table S13, available at: http://links.lww.com/JS9/E469).

Prediction of incidence burden

We used the BAPC model to project the PLC burden among middle-aged and elderly populations over the next 30 years. Global PLC cases are expected to rise continuously, with accelerated growth from 2035 to 2050, reaching an estimated 964 406 cases by 2050 (Fig. 7A; Supplemental Digital Content Table S14, available at: http://links.lww.com/JS9/E469). Incidence rates are projected to follow the “U” trend, declining to 26.5 by 2035 before peaking at 35.6 per 100 000 in 2050 (Fig. 7B). Male and female incidence rates will show similar trends, maintaining a 2:1 male-to-female ratio. The 75+ age group will consistently have the highest rates, with all age groups showing a nadir before subsequent increases (Fig. 7C, D; Supplemental Digital Content Table S15, available at: http://links.lww.com/JS9/E469). Etiology-specific projections indicate declines in HBV- and HCV-related PLC incidence until 2035, followed by increases, while alcohol- and NASH-related incidence will rise steadily (Fig. 7E, F, Supplemental Digital Content Table S16, available at: http://links.lww.com/JS9/E469).

Figure 7.

The prediction of incidence in liver cancer across middle-aged and elderly people from 2021 to 2050 globally. (A) Prediction of the number of incidence case by sex group. (B) Prediction of incidence rate by sex group. (C) Prediction of the number of incidence case by age group. (D) Prediction of incidence rate by age group. (E) Prediction of the number of incidence case by etiology group. (F) Prediction of incidence rate by etiology group. The solid line represents the real data, and the dashed line is the predicted data.

Discussion

With the progressive aging of the global population, the health issues and healthcare policy decisions facing older adults are gaining increasing importance. PLC, closely associated with advancing age^[2], has thus become a focal point of our study. Utilizing data from GBD 2021, we provided the first comprehensive assessment of PLC burden among middle-aged and elderly populations over the past 22 years. Globally, cases of PLC incidence, deaths, and DALYs among middle-aged and elderly people have all increased, a trend potentially linked to nearly 40% population growth over recent decades^[19,24]. Encouragingly, PLC incidence, deaths, and DALYs rate have shown a gradual decline over time.

However, this trend is uneven across regions: while High-income Asia Pacific, East Asia, Southeast Asia, and African regions experienced declining rates, North America, Western Europe, and parts of South America were seeing marked increases. Correlation analysis highlights a positive association between EAPC values and national SDI levels, indicating that countries with higher levels of socioeconomic development, such as Poland and the United Kingdom, tend to have more rapidly increasing burdens, whereas African countries with lower SDI levels show reductions. Mongolia had the highest burden globally in terms of incidence, deaths, and DALYs rate. And this may due to its high rates of HBV and HCV infections, including co-infections, coupled with substantial alcohol use and unbalanced dietary practices^[25].

As with previous studies^[13,26], HBV and HCV infections remained the predominant risk factors for PLC across middle-aged and elderly population, accounting for over 65%, though their burden has declined. Notably, NASH- and alcohol-related PLC incidence have surged since 2000, with the deaths doubling, highlighting their growing contribution to the PLC burden. Despite this shift, viral hepatitis eradication remains the cornerstone of liver cancer prevention. Key strategies for HBV control include vaccination, blood donor screening, and safe sexual practices^[27], while HCV prevention focuses on blood transfusion screening, infection control in healthcare, and reducing mother-to-child transmission^[28]. Although no HCV vaccine exists, curative treatments have significantly reduced HCV-related PLC deaths^[29]. The global decline in PLC incidence likely reflects decades of viral hepatitis prevention efforts, alongside advances in PLC diagnosis and treatment, contributing to reduced deaths and DALYs^[4,30].

The global burden and etiological patterns of PLC demonstrated significant geographical heterogeneity. In 2021, HBV-related PLC remained highly prevalent in East Asia, Southeast Asia, and sub-Saharan Africa, particularly in countries such as China, Mongolia, and Gambia, while being relatively low in the Americas and Europe. Notably, the past two decades have seen a resurgence of HBV-related PLC in countries like the United States, the United Kingdom, and Poland, which may be attributed to increased migration from HBV-endemic regions^[31–34], emphasizing the need to strengthen HBV prevention programs. In contrast, long-term vaccination programs and improved antiviral therapies have contributed to declining HBV-related PLC incidence in nations like China^[35]. However, due to the lag in vaccine efficacy, HBV-related PLC still remained a predominant type among middle-aged and elderly populations in China. HCV-related PLC is more prevalent in high-SDI regions, including Japan, South Korea, Western Europe, and North America. Although its global incidence is declining, HCV remains a key contributor to PLC due to its chronic disease course and strong association with cirrhosis^[28]. Furthermore, metabolic comorbidities such as obesity and diabetes have been identified as risk amplifiers for HCV-related hepatocarcinogenesis, which could explain the high HCV-related PLC burden in these countries^[36,37]. These underscore the necessity of integrating metabolic disease prevention with viral hepatitis control strategies to mitigate PLC burden in these regions.

In contrast to the gradual decline in virus-related PLC, NASH- and alcohol-related PLC is rising in most high- and middle-SDI regions, driven by the global obesity epidemic and sedentary lifestyles. NASH, a metabolic condition marked by hepatic steatosis and insulin resistance^[38], can lead to hepatocellular carcinoma even without cirrhosis. Estimates suggested that up to 13% of NASH patients may develop PLC within a decade^[39,40]. Dietary interventions, particularly adherence to the Mediterranean diet, which is rich in fruits, vegetables, whole grains, legumes, and healthy fats, have demonstrated protective effects against both hepatic steatosis and liver cancer development^[41,42]. Several cohort studies and meta-analyses have shown that higher adherence to the Mediterranean diet is associated with a lower risk of hepatocellular carcinoma^[43–45], likely due to its anti-inflammatory and insulin-sensitizing properties. In addition, regular physical activity not only contributes to weight control but also improves insulin sensitivity and reduces hepatic fat accumulation^[41]. Public health policies that encourage population-level shifts toward healthier dietary patterns and promote active lifestyles may be essential for reversing the upward trajectory of NASH-related PLC.

The age pattern analysis indicates that among the middle-aged and elderly population, those aged over 75 exhibited a more pronounced increase in PLC incidence, particularly in countries, like UK, Poland, the United States, and Nordic countries. Further analysis suggests that the increased burden was likely more attributable to the rise in NASH-related PLC, especially in regions like Western Europe and Australia. Previous studies have also noted this trend^[5,12]. Moreover, the evolution of NASH-related PLC is relatively rapid, and this difference in age patterns underscores the need to shift more focus to the prevention and treatment of PLC in individuals over 75 years old, especially NASH-related PLC. Promoting healthier diets and lifestyles may be crucial for PLC in this population. Globally, the male/female ratio for PLC incidence among middle-aged and elderly populations is 2.19, which aligns with traditional understanding. However, this ratio was significantly lower in low SDI region, particularly in African, likely due to cultural, economic, and conflict-related factors. In China, Mongolia, and Bangladesh, female incidence was rising while male incidence declines, whereas the opposite trend is observed in the U.S., Brazil, and North Africa. These findings suggests that the PLC burden of different gender groups should be given different levels of attention in specific countries.

Projections for PLC incidence rate among middle-aged and elderly populations reveal the U-shaped trend: a slight decline from 2021 to 2035, followed by an increase from 2035 to 2050. However, the patterns differ among etiological subgroups. While alcohol- and NASH-related PLC continue to increase, HBV- and HCV-related PLC incidence rates also exhibit a U-shaped trajectory. East Asia, High-income Asia-Pacific, and Western Europe, accounting for 60% of cases, were further analyzed and results show similar U-shaped trends (Supplemental Digital Content Figure S15, available at: http://links.lww.com/JS9/E469). Further analysis suggests that this may be driven by accelerating increases in case numbers alongside a slowdown, or even decline, in total population growth within these regions from 2035 to 2050. Overall, the post-2035 rise may be attributable to two factors: the resurgence of HBV- and HCV-related PLC and the deceleration of population growth in major PLC endemic regions. These findings underscore the long-term nature of PLC prevention efforts. Moving forward, it is crucial to continue promoting medical policies focused on dietary and lifestyle changes to address the ongoing rise of NASH- and alcohol-related PLC. At the same time, efforts to control viral hepatitis must not be relaxed in order to prevent a resurgence of HBV- and HCV-related PLC.

This study provides the first comprehensive assessment of PLC burden trends among middle-aged and elderly populations. These findings could assist government health agencies in formulating more efficient and targeted PLC prevention policies aimed at reducing this burden. However, several limitations exist. First, the reliability of GBD-based assessments depends on data quality, which may be compromised in developing countries due to misdiagnosis, underreporting, or overreporting. Second, trend analyses rely on historical data, potentially obscuring the effects of recent interventions. Lastly, data constraints prevented analysis of mixed etiology cases, comorbidities, complications, and histological subtypes, limiting a more nuanced understanding of disease patterns.

Conclusion

In summary, our study reveals a slight decline in PLC burden among middle-aged and elderly populations, with notable regional and national disparities. HBV- and HCV-related PLC are decreasing, while NASH- and alcohol-related cases are rising, especially in developed countries. The burden is highest and still increasing among those aged 75 and above, with a marked rise in NASH-related cases. Males generally experience a higher burden than females. Over the next 30 years, the incidence of PLC is likely to experience a U-shaped trend of first decline and then rise. These findings enhance awareness of the global burden of liver cancer across middle-aged and elderly and provide a basis for future policies on prevention, clinical management and other aspects of PLC.

Ethical approval

Not applicable.

Consent

Not applicable.

Sources of funding

This study received support from the Key Project of Research and Development Program of Shaanxi Province (No. 2020SF-072), the Innovation Project for Medical Integration in XJTU (YXJLRH2022062), and the Free Exploration Project of the Second Affiliated Hospital of Xi’an Jiaotong University (2020YJ(ZYTS)018). The funders played no role in the study design, data analysis, or manuscript preparation.

Author contributions

B.W.: conceptualization, writing – original draft, formal analysis, investigation, visualization; Y.X.: writing – review & editing, formal analysis; N.H.: writing – review & editing, investigation; J.L.: conceptualization, supervision, writing – review & editing; S.Z.: conceptualization, supervision, writing – review & editing, funding acquisition. All authors have reviewed and approved the final manuscript.

Conflicts of interest disclosure

The authors declare that they have no competing interests.

Research registration unique identifying number (UIN)

Name of the registry: Research Registry. Unique identifying number or registration ID: researchregistry11243. Hyperlink to your specific registration (must be publicly accessible and will be checked): Research Registry (knack.com).

Guarantor

Shu Zhang.

Provenance and peer review

Not commissioned, externally peer-reviewed.

Data availability statement

The datasets analyzed during the current study are available in the GHDx query tool (https://vizhub.healthdata.org/gbd-results/).

Declarations of artificial intelligence (AI)

The authors declare that AI was not used in the research and manuscript development.