Disease burden of biliary tract cancer in 204 countries and territories, 1990–2021: A comprehensive demographic analysis of the Global Burden of Disease Study 2021

Abstract

Background:

Biliary tract carcinomas (BTCs) are relatively rare but lethal primary malignant tumors derived from the biliary tract system. The burden of BTCs varies according to sex, age, region, and country, but limited attention has been paid to the burden of BTCs. We sought to explore the up-to-date data from the Global Burden of Disease Study (GBD) and expand findings by accessing the demographic features of BTC disease burden.

Methods:

Using the latest data from the GBD 2021, we evaluated and analyzed the distributions and patterns of BTC disease burden in various age groups, sexes, regions, and countries.

Results:

The number of incident cases, deaths, and disability-adjusted life-years (DALYs) tended to increase and peaked at 216,770 (95% uncertainty interval [UI]: 181,890–245,240), 171,960 (95% UI: 142,350–194,240), and 3,732,100 (95% UI: 3,102,900–4,317,000) person-years, respectively, in 2021. However, the average global age-standardized rates (ASRs) of incident cases, deaths, and DALYs shrunk by −11.46% (95% UI: −21.91 to 3.35%), −24.09% (95% UI: −33.19 to 16.88%), and −26.25% (95% UI: −35.53 to 18.36%), respectively, from 1990 to 2021. Meanwhile, the male/female ratio (male per 100 female) of incidence, deaths, and DALYs changed from 76.40, 75.41, and 74.72 to 86.89, 79.11, and 82.29, respectively. In 2021, the highest number of incident cases, deaths, and DALYs occurred in East Asia. The top three highest incidences, deaths, and DALYs were observed in China, India, and Japan, and the highest ASRs were observed in Chile in 2021. Analysis of the Human Development Index along with disease burden estimates of BTCs also suggests that the burden of the disease is related to the level of comprehensive development of the society.

Conclusion:

This study provided a comprehensive comparison of differences in the burden of disease across populations and over time, and further presented evidence concerning the formulation of prevention and control policies and etiologic studies for BTCs and proposed logical hypotheses to investigate.

Introduction

Primary malignant tumors derived from the biliary tract system are collectively referred to as biliary tract carcinomas (BTCs).^[1] BTCs are extremely lethal cancers with dismal prognoses despite the development of diagnostic and therapeutic methods.^[2,3] Because of the characteristics of high malignancy and refractoriness with insidious onset, only a small percentage of patients with BTC have a chance for radical resection, which is the only curative treatment for BTCs currently.^[4,5] The only risk factor supported by global statistical evidence is high body mass index, and other suspected risk factors, including benign biliary tract diseases like gallstone disease, alcohol, infections, and diabetes, are controversial and lack evidence from large samples.^[6–8]

It has been estimated that BTC is the sixth most prevalent global disease burden for digestive system cancers, with 0.2 million estimated new cases in 2019 according to the results from the Global Health Data Exchange (GHDx), and 122,462 new cases in 2022, according to the Global Cancer Statistics 2022.^[9,10] The geographical distribution of BTC and its burden vary globally with significant hotpot countries and territories including China, India, and Chile.^[10,11] The incidence, mortality, and prevalence of BTC also vary according to sex; the incidence is estimated to be higher in females than males globally.^[10,11] Considering the relatively lower overall incidence and regional variability of BTC, intergroup and geographic analyses based on up-to-date large-scale data are still lacking. Based on the latest worldwide data, comprehensive analyses are needed to inspire researchers and medics. To estimate disease burden systematically and comprehensively, the Global Burden of Disease (GBD) study consistently collected and structured the incidence, prevalence, mortality, and other burden measures and updated the 371 diseases and injuries in 204 countries and territories from 1990 to 2021.^[6,12] This study was to provide a robust and accurate analysis of the temporal status and trends of the disease burden of BTC by estimating data from the GBD Study 2021 (GBD 2021). In the current study, the disease burden of BTC was evaluated by analyzing the distributions and patterns of BTC in various age groups, sexes, regions, and countries by combining estimating measures including incidence, mortalities, and disability-adjusted life-years (DALYs).^[6] This information will be instructive and beneficial for researchers and policymakers to systematically identify the distribution and differentiation of disease burden in time and space while conducting relevant investigations and formulating public health policies.

Methods

Overview

The GBD 2021 was led by the Institute for Health Metrics and Evaluation (IHME) and provided consistent comprehensive disease burden estimates for major diseases, injuries, and related risk factors. The latest available data can be acquired from the GHDx (http://ghdx.healthdata.org/gbd-results-tool), including estimates of incidence, prevalence, death, DALYs, years of life lost (YLLs), and years lived with disability (YLDs) for 371 diseases and injuries in 204 countries and territories from 1990 to 2021, and the results were viewable online in GBD Compare (https://vizhub.healthdata.org/gbd-compare).^[12]

This study complied with the Guidelines for Accurate and Transparent Health Estimates Reporting statement.^[13] Informed consent, ethical approval, and patient and public involvement for this study were waived because GBD estimation is publicly available online and primary data at individual levels were excluded.

BTC definition

Anatomically, BTC is defined as primary malignant neoplasms of the biliary system, which consists of intrahepatic bile duct, extrahepatic bile duct, and gallbladder. The 10th revision of the International Classification of Diseases (ICD-10) criteria was adopted in the GBD 2021, in which gallbladder and biliary tract cancer (GBTC) is defined as a separate category for estimations.^[14,15] Therefore, in the current study, BTC includes extrahepatic cholangiocarcinoma and gallbladder cancer (coded as C23 through C24.9 in ICD-10), and intrahepatic cholangiocarcinoma (coded as C22.1) was excluded.

Data source

All measures were stratified by age groups (in 5-year intervals from 20 years to ≥95 years), sexes (male and female), and locations (countries and territories) according to the data structure provided by the GHDx website.^[12,16] The Human Development Index (HDI) is the geometric mean value published by the United Nations Development Program and calculated across combined indicators for three dimensions, including life expectancy, education, and income.^[17] HDI scores for each country were on a spectrum from 0 to 1 and covered almost all countries from 1990 to 2021.^[17]

Estimation and statistical analysis

The standardized methodological framework of GBD 2021 includes the estimation of both fatal and non-fatal measures of disease burden. In GBD 2021, BTC mortality data were based on the vital registration system and cancer registry system. Details of the data structure and methodology of estimations in GBD using previously mentioned data sources have been previously reported.^{[12,16,18,19]} The formula for age-standardized rate (ASR) calculation is as follows:

$$\text{ASR}=\frac{{{\displaystyle \sum }}_{i=1}^A{a}_i{w}_i}{{{\displaystyle \sum }}_{i=1}^A{w}_i}\times 100,000$$

ASR was calculated by summing up the products of age-specific rates in each age subgroup (a_i) and the standard number of people in the same age subgroup (w_i), dividing the sum by the total standard population weights (A and i denote the counts of age subgroups).^[20] The estimated annual percentage change (EAPC) was used to quantify disease burden trends and was calculated via the linear regression formulas fitted to the natural logarithm of the rates as follows (y refers to ln(ASR) and x denotes calendar year)^[16,21]:

$$\begin{array}{c}\text{y=α+βx+ε}\\ \text{EAPC}=\text{ exp}\left(\beta \right)-1\end{array}$$

Besides incidence, mortality, and DALY, to evaluate the burden of BTC across the world, age-standardized incidence rate (ASIR), age-standardized mortality rate (ASMR), and age-standardized DALY rate were used in this study. The uncertainty interval (UI) analysis was performed in the GBD study to reduce the impact of heterogeneity caused by sampling error and non-sampling variance. The latest burden of disease data from a global perspective was analyzed first, followed by the age and sex patterns and their trends on a global scale, to identify the underlying drivers. The distribution patterns of disease burden indicators were summarized at the regional and national levels, and a comparative analysis of representative regions and countries was conducted. In addition, the interplay between HDI and the burden of disease between 1990 and 2021 was explored using the linear regression model. All statistical analyses were performed with R software (Version 4.3.3, R Foundation, Vienna, Austria), with P <0.05 considered statistically significant.

Results

Global disease burden of primary BTC

Globally, all-age incidences and all-age deaths due to BTC consistently increased from 1990 to 2021. In 2021, 216,770 (95% UI: 181,890–245,240) new BTC cases were reported, which was the highest number so far. This number of BTC cases increased by 101.09% (95% UI: 77.14–120.21%) from 107,800 (95% UI: 96,890–117,510) cases in 1990 [Table 1]. The number of deaths and DALYs exhibited similar upward trends to the peak, increased 74.26% (95% UI: 54.20–91.45%) and 60.45% (95% UI: 40.37–78.10%) from 1990 to 2021, respectively [Table 1]. However, the ASR of disease burden measures demonstrated diametrically opposite changes with declines across the world. The average global ASIR of BTC accounted for 2.56 (95% UI: 2.16–2.89) per 100,000 population in 2021, descended by −11.46% (95% UI: –21.91% to –3.35) from 1990, while the ASMR descended by −24.09% (95% UI: –33.19% to –16.88) and the ASR of DALYs descended by −26.25% (95% UI: –35.53% to –18.36) [Table 1].

Table 1.

All-age incident cases, deaths, and DALYs for BTC and their percentage changes by sex from 1990 to 2021, globally.

Categories	Incident cases (95% UI)		Deaths (95% UI)		DALYs (95% UI)
	Numbers (thousands)	ASRs per 100,000	Numbers (thousands)	ASRs per 100,000	10,000 person-years	ASRs per 100,000
Total
1990	107.80 (96.89–117.51)	2.89 (2.59–3.15)	98.68 (88.94–109.19)	2.69 (2.39–2.97)	232.61 (205.45–258.30)	58.58 (52.21–64.88)
2021	216.77 (181.89–245.24)	2.56 (2.16–2.89)	171.96 (142.35–194.24)	2.04 (1.70–2.29)	373.21 (310.29–431.70)	43.20 (36.01–49.88)
Change (%)	101.09 (77.14–120.21)	−11.46 (–21.91 to –3.35)	74.26 (54.20–91.45)	−24.09 (–33.19 to –16.88)	60.45 (40.37–78.10)	−26.25 (–35.53 to –18.36)
Male
1990	40.92 (34.82–46.84)	2.46 (2.13–2.80)	36.71 (30.58–43.21)	2.27 (1.93–2.65)	91.11 (74.06–108.06)	49.55 (40.88–58.55)
2021	100.78 (77.37–116.40)	2.65 (2.06–3.04)	75.95 (57.12–89.00)	2.03 (1.54–2.36)	168.48 (122.95–199.66)	41.95 (30.95–49.56)
Change (%)	146.32 (110.84–182.12)	7.42 (−7.78−21.91)	106.88 (76.40–137.07)	−10.74 (−23.31−1.19)	84.91 (55.17–115.47)	−15.33 (–28.62 to –2.04)
Female
1990	66.88 (59.47–75.64)	3.22 (2.85–3.64)	61.97 (54.71–71.11)	3.01 (2.66–3.44)	141.50 (123.29–164.94)	66.31 (57.94–77.19)
2021	115.98 (94.59–132.85)	2.50 (2.03–2.86)	96.01 (77.02–111.37)	2.06 (1.65–2.39)	204.73 (162.44–240.07)	44.55 (35.35–52.31)
Change (%)	73.42 (51.23–94.53)	−22.55 (–32.18 to –13.31)	54.93 (34.78–73.62)	−31.42 (–40.16 to –23.23)	44.69 (25.27–63.18)	−32.81 (–42.02 to –24.38)

ASR: Age-standardized rate; BTC: Biliary tract carcinoma; DALY: Disability-adjusted life-year; UI: Uncertainty interval.

Sex and age patterns

The global disparity of BTC burden according to sex was similar from 1990 to 2021 and reflected higher incidence (115,980, 95% UI: 94,590–132,850), deaths (96,010, 95% UI: 77,020–111,370), DALYs (2,047,300, 95% UI: 1,624,400–2,400,700, person-years), and ASRs (ASIR: 2.50%, 95% UI: 2.03–2.86%, ASMR: 2.06%, 95% UI: 1.65–2.39%, ASR of DALYs: 44.55%, 95% UI: 35.35–52.31%) in women than incidence (110,780, 95% UI: 77,370–116,400), deaths (75,950, 95% UI: 57,120–89,000), DALYs (1,684,800, 95% UI: 1,229,500–1,996,600, person-years), and ASRs (ASIR: 2.65%, 95% UI: 2.06–3.04%, ASMR: 2.03%, 95% UI: 1.54– 2.36%, ASR of DALYs: 41.95%, 95% UI: 30.95–49.56%) in men [Table 1]. From 1990 to 2021, the sex ratio (male per 100 female) of incidence, deaths, and DALYs changed in parallel from 61.18, 59.24, and 64.39 to 86.89, 79.11, and 82.29, respectively [Figure 1A–C and Table 1]. Sex ratios of ASRs altered as well, with the enlargement of ASIR, ASMR, and ASR of DALYs from 76.40%, 75.42%, and 74.72% in 1990 to 106.00%, 98.54%, and 94.16%, respectively, in 2021 [Figure 1D–F and Table 1].

Figure 1.

Trends of incident cases, deaths, DALYs, ASIR, ASMR, and age-standardized DALY rate for BTC in both sexes (male and female) 1990–2021, globally. (A) Incident cases. (B) ASIR. (C) Deaths. (D) ASMR. (E) DALYs. (F) Age-standardized DALY rate. ASIR: Age-standardized incidence rate; ASMR: Age-standardized mortality rate; BTC: Biliary tract carcinoma; DALY: Disability-adjusted life-year.

The greatest upward shift of disease burden during the study period occurred in the elderly population ≥80 years old, in which the number of new BTC cases and deaths increased by eight and six times in the age group ≥95 years, respectively [Supplementary Table 1, http://links.lww.com/CM9/C235]. In the age groups from 65 years to 84 years, a relatively higher concentration of new BTC cases and deaths occurred in 2021. The proportion of cases in these age groups accounted for more than half of the total cases, including the subgroup of 70 years to 74 years, which accounts for the largest proportion of incidence (32,910, 95% UI: 27,650–37,281) and deaths (26,047, 95% UI: 21,448–30,073) [Figure 2A, B and Supplementary Table 1, http://links.lww.com/CM9/C235]. In comparison, higher DALYs clustered in younger age groups from 55 years to 74 years in 1990 and 2021, but the most significant variation at this stage occurred in groups ≥80 years old [Figure 2C and Supplementary Table 1, http://links.lww.com/CM9/C235].

Figure 2.

Distribution of incident cases, deaths, DALYs for BTC by sex and age groups in 2021, globally. (A) Incident cases. (B) Deaths. (C) DALYs. BTC: Biliary tract carcinoma; DALY: Disability-adjusted life-year; UI: Uncertainty interval.

In 2021, the number of incident cases peaked at ages 70–74 years in both men (15,918, 95% UI: 12,080–18,776) and women (16,992, 95% UI: 13,431–19,583) and decreased with age [Figure 2A and Supplementary Table 2, http://links.lww.com/CM9/C235]. The number of deaths climbed with age and peaked in the group of 70–74 years in men (12,007, 95% UI: 8,927–14,332) and 75–79 years in women (10,176, 95% UI: 9,148–11,321) before descending [Figure 2B and Supplementary Table 2, http://links.lww.com/CM9/C235]. DALYs peaked in the group of 65–69 years in women and men, and then started to decrease [Figure 2C and Supplementary Table 2, http://links.lww.com/CM9/C235]. The numbers of incidence, deaths, and DALYs were substantially higher in females than in males in most age groups, except for the age subgroups from 30 years to 44 years. In these groups, the numbers of new BTC cases in males exceeded the number in females [Figure 2A, B and Supplementary Table 2, http://links.lww.com/CM9/C235]. However, although the numbers of new cases, deaths, and DALYs in females were still generally higher than those in males in 2021, the EAPCs of these estimates in men significantly exceeded the EAPCs in women by almost twice from 1990 to 2021. Such overall differences were also similar in most of the age groups [Supplementary Tables 2 and 3, http://links.lww.com/CM9/C235].

Patterns at regional and national levels

In 2021, the highest incidence was estimated in East Asia (53,480, 95% UI: 37,560–68,790) followed by high-income Asia Pacific (36,720, 95% UI: 30,700–41,220), South Asia (32,820, 95% UI: 22,750–38,790), Western Europe (25,450, 95% UI: 22,480–27,380) and high-income North America (14,130, 95% UI: 12,810–14,900) [Table 2]. The numbers of deaths and DALYs in East Asia (Deaths: 39,220, 95% UI: 28,020–51,030; DALYs: 889,300, 95% UI: 634,700–1,160,100 person-years), South Asia (33,700, 95% UI: 23,310–39,890; 865,400, 95% UI: 591,800–1,024,300), and high-income Asia Pacific (27,840, 95% UI: 23,020–31,230; 422,700, 95% UI: 365,400–468,100) were significantly and exponentially higher than the deaths and DALYs in other regions in 2021 [Table 2].

Table 2.

All-age incident cases, deaths and DALYs for BTC along with their ASRs in 2021 and their EAPCs from 1990 to 2021 in 21 GBD regions.

GBD regions	Incident cases (95% UI)			Deaths (95% UI)			DALYs (95% UI)
	Numbers, in thousands	ASRs per 100,000	EAPCs of ASR (%)	Numbers, in thousands	ASRs per 100,000	EAPCs of ASR (%)	10,000 person-years	ASRs per 100,000	EAPCs of ASR (%)
Andean Latin America	2.33 (1.77–3.04)	3.99 (3.04–5.20)	–19.19 (–34.60 to –1.50)	2.33 (1.77–3.01)	4.03 (3.07–5.19)	–24.66 (–38.88 to –8.88)	5.39 (4.11–7.04)	90.54 (68.89–118.02)	–26.86 (–41.45 to –9.72)
Australasia	1.84 (1.61–1.98)	3.35 (2.98–3.60)	–0.29 (–8.19 to 6.75)	0.63 (0.55–0.68)	1.10 (0.97–1.17)	–37.69 (–42.47 to –33.93)	1.19 (1.08–1.27)	22.80 (20.96–24.13)	–39.16 (–43.22 to –35.60)
Caribbean	0.54 (0.47–0.62)	1.00 (0.87–1.15)	–38.72 (–46.27 to –30.01)	0.53 (0.46–0.61)	0.98 (0.85–1.13)	–42.07 (–49.20 to –33.46)	1.24 (1.06–1.47)	23.11 (19.73–27.29)	–40.68 (–48.42 to –30.74)
Central Asia	0.65 (0.58–0.74)	0.83 (0.74–0.94)	–22.21 (–34.43 to –8.79)	0.66 (0.58–0.75)	0.86 (0.76–0.97)	–22.60 (–34.86 to –9.48)	1.69 (1.48–1.91)	19.98 (17.63–22.62)	–26.56 (–38.05 to –13.81)
Central Europe	6.41 (5.83–6.99)	2.80 (2.55–3.05)	–36.78 (–41.43 to –31.40)	5.98 (5.46–6.51)	2.58 (2.36–2.81)	–44.06 (–48.06 to –39.79)	12.10 (11.13–13.17)	55.32 (50.98–60.11)	–44.61 (–48.63 to –40.33)
Central Latin America	5.36 (4.78–5.96)	2.16 (1.93–2.40)	–49.44 (–54.60 to –44.00)	5.32 (4.74–5.89)	2.16 (1.93–2.39)	–52.81 (–57.61 to –47.80)	12.74 (11.35–14.23)	50.01 (44.62–55.87)	–52.21 (–57.43 to –46.81)
Central Sub–Saharan Africa	0.17 (0.11–0.24)	0.33 (0.22–0.46)	7.24 (–19.31 to 46.16)	0.17 (0.11–0.24)	0.36 (0.24–0.51)	6.95 (–19.52 to 45.67)	0.47 (0.31–0.67)	8.17 (5.46–11.66)	4.31 (–22.86 to 44.39)
East Asia	53.48 (37.56–68.79)	2.49 (1.74–3.19)	13.31 (–14.10 to 44.47)	39.22 (28.02–51.03)	1.85 (1.31–2.40)	–20.02 (–38.53 to 1.62)	88.93 (63.47–116.01)	40.45 (28.79–52.44)	–23.17 (–41.85 to –0.78)
Eastern Europe	6.53 (6.05–7.03)	1.84 (1.70–1.98)	11.62 (1.22 to 21.52)	4.72 (4.35–5.10)	1.32 (1.21–1.43)	–12.48 (–20.95 to –4.25)	10.38 (9.55–11.25)	29.82 (27.37–32.28)	–16.25 (–24.48 to –7.63)
Eastern Sub–Saharan Africa	1.48 (1.04–1.97)	0.95 (0.67–1.26)	–8.49 (–27.89 to 20.24)	1.56 (1.09–2.07)	1.04 (0.73–1.39)	–7.75 (–26.98 to 20.25)	4.09 (2.84–5.51)	23.33 (16.37–31.01)	–13.95 (–33.52 to 15.72)
High–income Asia Pacific	36.72 (30.70–41.22)	6.57 (5.62–7.33)	–30.74 (–37.27 to –24.39)	27.84 (23.02–31.23)	4.80 (4.13–5.34)	–42.92 (–48.34 to –37.46)	42.27 (36.54–46.81)	86.56 (75.72–96.27)	–49.80 (–54.74 to –44.53)
High–income North America	14.13 (12.81–14.90)	2.14 (1.96–2.25)	–15.17 (–17.49 to 13.10)	5.98 (5.37–6.32)	0.88 (0.80–0.93)	–36.71 (–38.55 to –35.09)	12.23 (11.38–12.79)	19.26 (18.06–20.10)	–34.96 (–36.64 to –33.44)
North Africa and Middle East	5.94 (4.36–7.30)	1.40 (1.03–1.73)	–0.18 (–16.68 to 17.42)	5.59 (4.10–6.85)	1.37 (1.01–1.69)	–9.03 (–24.20 to 6.71)	13.96 (10.08–16.85)	29.82 (21.76–36.22)	–12.98 (–26.72 to 3.03)
Oceania	0.04 (0.02–0.05)	0.52 (0.35–0.66)	–17.90 (–33.32 to –0.87)	0.04 (0.03–0.05)	0.56 (0.38–0.70)	–18.64 (–33.50 to –1.55)	0.11 (0.07–0.14)	13.20 (8.77–17.19)	–19.45 (–35.58 to 0.41)
South Asia	32.82 (22.75–38.79)	2.28 (1.59–2.71)	36.14 (1.37 to 61.96)	33.70 (23.31–39.89)	2.40 (1.67–2.85)	33.24 (–0.71 to 58.23)	86.54 (59.18–102.43)	56.52 (38.80–66.92)	28.15 (–5.06 to 52.30)
Southeast Asia	12.04 (8.76–15.16)	1.93 (1.41–2.43)	14.83 (–13.66 to 46.45)	11.16 (8.22–14.18)	1.85 (1.37–2.35)	3.86 (–21.76 to 31.06)	27.23 (20.10–34.38)	40.73 (30.05–51.36)	–2.89 (–26.58 to 22.19)
Southern Latin America	4.44 (4.08–4.71)	5.06 (4.67–5.36)	–41.34 (–44.41 to –37.72)	3.99 (3.67–4.23)	4.51 (4.15–4.78)	–49.42 (–51.98 to –46.28)	8.79 (8.25–9.26)	102.44 (96.35–107.96)	–49.41 (–52.00 to –46.30)
Southern Sub–Saharan Africa	0.46 (0.33–0.55)	0.84 (0.59–0.98)	22.49 (–3.21 to 42.29)	0.48 (0.34–0.56)	0.89 (0.62–1.04)	20.60 (–4.65 to 40.28)	1.24 (0.87–1.47)	20.59 (14.51–24.26)	18.98 (–5.23 to 37.76)
Tropical Latin America	5.83 (5.38–6.12)	2.28 (2.10–2.40)	–27.42 (–30.45 to –24.32)	5.87 (5.38–6.17)	2.31 (2.11–2.43)	–31.56 (–34.58 to –28.61)	13.96 (13.13–14.54)	53.58 (50.27–55.87)	–30.23 (–32.93 to –27.27)
Western Europe	25.45 (22.48–27.38)	2.57 (2.31–2.73)	–29.70 (–32.82 to –27.11)	16.11 (14.09–17.44)	1.55 (1.39–1.66)	–49.55 (–51.92 to –47.57)	28.40 (25.72–30.28)	31.05 (28.61–32.81)	–50.99 (–52.98 to –49.21)
Western Sub–Saharan Africa	0.10 (0.06–0.12)	0.05 (0.04–0.07)	19.96 (–13.44 to 47.97)	0.10 (0.07–0.13)	0.06 (0.04–0.07)	21.26 (–13.58 to 49.80)	0.27 (0.17–0.33)	1.30 (0.84–1.59)	11.71 (–16.84 to 38.65)
Total	216.77 (181.89–245.24)	2.56 (2.16–2.89)	–0.11 (–0.22 to 0.03)	171.96 (142.35–194.24)	2.04 (1.70–2.29)	–0.24 (–0.33 to 0.17)	373.21 (310.29–431.70)	43.20 (36.01–49.88)	–0.26 (–0.36 to 0.18)

ASR: Age-standardized rate; BTC: Biliary tract carcinoma; DALY: Disability-adjusted life-year; EAPC: Estimated annual percentage change; GBD: Global Burden of Disease.

The ASRs of incidence, deaths, and DALYs for the top three GBD regions were estimated and varied significantly. High-income Asia Pacific had the highest ASIR (6.57, 95% UI: 5.62–7.33, per 100,000) and ASMR (4.80, 95% UI: 4.13–5.34, per 100,000), Southern Latin America had the highest ASR of DALYs (102.44, 95% UI: 96.35–107.96, per 100,000) [Table 2]. Among all GBD regions, South Asia, Central Sub-Saharan Africa, Southern Sub-Saharan Africa, and Western Sub-Saharan Africa exhibited increases in all ASRs of incidence, deaths, and DALYs from 1990 to 2021, while other GBD regions exhibited varying degrees of decline in ASIR, ASMR, or/and ASR of DALYs. Even in East Asia, which had the highest incidence and growing ASIR, the ASRs of death and DALYs showed a descending trend [Table 2].

The disease burden of BTC revealed divergent differences in estimates and trends among countries and territories around the world during the study period [Supplementary Table 4, http://links.lww.com/CM9/C235]. In 2021, the highest numbers of incident cases, deaths, and DALYs were observed in China, India, and Japan, while the highest ASRs were observed in Chile [Supplementary Table 4, http://links.lww.com/CM9/C235].

In 2021, China, India, and Japan had the highest numbers of incident cases, deaths, and DALYs. China had extremely high numbers but relatively low ASR rankings of 37th in ASIRs, 43rd in both ASMRs and ASRs of DALYs in 204 countries and territories. In contrast, Japan ranked high in all ASRs, ranking 4th in ASIRs, 5th in ASMRs, and 9th in ASRs of DALYs [Supplementary Table 4, http://links.lww.com/CM9/C235]. From 1990 to 2021, India demonstrated a strong growth trend in ASRs of incidence (47.50%, 95% UI: 4.73–77.25%), death (44.57%, 95% UI: 1.75–73.66%) and DALYs (37.08%, 95% UI: 65.55–3.23%). In contrast, ASRs in China and Japan declined, with the following changes in ASRs of death (China: −20.24%, 95% UI: −39.19 to 2.21%, Japan: −43.21%, 95% UI: −47.55 to –40.54%) and DALYs (−23.24%, 95% UI: –42.26 to –0.09%; −49.92%, 95% UI: −52.62 to –48.05%). While the ASR of incidence in China exhibited a growth trend (EAPC: 13.62%, 95% UI: –14.39 to 45.88%), EAPC of the ASIR in Japan still decreased (−33.17%, 95% UI: −29.15 to 37.99%) [Supplementary Table 4, http://links.lww.com/CM9/C235]. Unlike other countries ranked in the top 10 of ASRs in 2021, Chile, with the highest ASRs, still ranked high in the list of EAPCs from 1990 to 2021, ranking 21st in the EAPC of ASIR, 13th in the EAPC of ASMR, and 9th in the EAPC of the ASR of DALYs [Supplementary Table 4, http://links.lww.com/CM9/C235].

Disease burden and HDI

During the study period, the HDI influenced the disease burden of BTC in different countries and territories. An analysis of the relationship between disease burden measures and HDI of different countries demonstrated that the ASRs of incidence, deaths, and DALYs of BTC showed similar patterns across different levels of HDI, but the ASRs also showed different trends of interrelationship as the HDI values increased [Supplementary Figure 1, http://links.lww.com/CM9/C235]. Results of linear regression showed that the ASRs of incidence, deaths, and DALYs of BTC distributed across all levels of HDI show no continuous linear relationships with the HDI value of different countries and territories (ASIR: coefficient = 0.14, P = 0.000, R² = 0.099; ASMR: coefficient = 0.084, P = 0.002, R² = 0.046; ASR of DALYs: coefficient = 1.4, P = 0.015, R² = 0.031).

Discussion

The IHME released complete data of GBD 2021 on April 16, 2024. Based on the updated data, we evaluated the disease burden of BTC by analyzing the distribution and differentiation of disease burden in time and space. In this study, we reported the incidence, deaths, DALYs, their ASRs, and EAPCs of these indicators for BTC over the period from 1990 to 2021. On a global scale, the total burden of BTC has been increasing from 1990 to 2021, while the epidemiology of sexes, age groups, regions, and countries varied. Each time, with the update of a new version of the GBD study, the latest dataset is revised, resulting in localized differences between the most recent estimates and previously disclosed GBD estimates for the entire period. The present study provides a systematic and up-to-date assessment of disease burden due to BTC using the latest data from GBD 2021, not only including the most recent data from 2020 and 2021, but also choosing a different analytical perspective, i.e., based on a comparison of the data to identify age and sex groups, countries, and regions with significant changes in the burden of disease and to probe the potential clinical public health reasons behind them.^[22]

The distribution of BTC has been considered strong and the similarities and variations of BTC disease burden had been demonstrated in the latest and former studies.^[1,10,23,24] Although the global incidence, deaths, and DALYs of BTC remain relatively low compared to other types of digestive system cancers, the number still grew to a new peak during the study period.^[9,24,25] Previous studies reported ascending trends of BTC incidence in different countries and subnational regions and called for greater efforts to prevent and treat the disease.^[26–28] Currently, BTC remains a significantly growing component of the global cancer disease burden and may continue to aggravate the cancer burden from an absolute number perspective.

The increased total number of BTC cases and the decreased ASRs of BTC may be due to the continuously changing global demographics, the development of BTC treatments, and increased attention to BTCs by researchers, medics, and public health departments in the last decades.^[4,29,30] Apart from these factors, the status of treatments for other benign diseases at the same primary site of BTC may interfere with the statistics; the development and popularization of treatments for benign diseases originating in the gallbladder and bile ducts, such as cholecystectomy, might have a protective effect on BTC.^[31]

From 1990 to 2021, global ASRs for incidences, deaths, and DALYs for both sexes exhibited an overall downward trend, however, BTC burdens trends were significantly different in males vs. females; the magnitude of the decline in these estimates was much smaller among males than among females. A unique feature of BTC is that, over time, reports based on national and global data demonstrated significantly more morbidity, morbidity, and mortality in women than in men, with a similar distribution of disease burden.^{[3,9,23,24,26–28]} While a significant male/female ratio discrepancy in most BTC disease burden estimates, except for the ASIR, remains, the changes in sex ratios have markedly narrowed.

To understand the sex ratio fluctuations in incidence and disease BTC burdens, a population-based perspective can be approached. The overall morbidity and mortality of other digestive cancers have been higher in men than in women for several biopsychosocial reasons. This gap is gradually increasing, possibly due to underlying driving forces in common with the fact that the burden of disease in men with BTC is catching up with that in women.^[9,10,19] Benign biliary tract diseases, including gallbladder stones, are often considered risk factors for BTC, and previous reports showed that such diseases were more prevalent in women than in men. The widespread availability of techniques to treat benign biliary tract diseases, such as cholecystectomy, has increased and affects females more than males.^[31–33]

In the clinical practice of BTC, epidemiologic evidence for identifying at-risk populations has been an important reference for healthcare professionals to conduct directed screening and make a preliminary diagnosis.^[3,5,27] During the study period, the gender ratio of BTC disease burden showed more women than men, but the difference between genders showed a gradual decrease at the same time, which suggests that the medics should reconsider the significance and influence of the gender factor on the risk of BTC according to the latest statistics.^[33–35] In addition, based on the data of 2021, this study summarized the age group containing more than half of the new cases and deaths, which on the one hand can be regarded as the high-risk age group for the diagnosis of BTC, and on the other hand, most deaths and morbidities are concentrated in the same age group, which also confirms that the prognosis of BTC is still extremely poor and difficult to be detected at an early stage.

The distribution of disease burden among different age groups suggests gender differences in the older age groups between genders should guide the development of secondary prevention strategies. Within the study period, the peaks and concentrations of deaths were in younger age groups in males than in females. This phenomenon may derive from gender differences in the changing age structure, and it could also derive from that there were more new cases or cases with poorer prognoses in males than in females in the younger age groups, which in turn might provide circumstantial evidence for etiological differences of BTC in males vs. females.^[34,35]

Policymakers tailor their policies to the conditions and needs of the local area. Based on our analysis of disease burden statistics in 21 GBD regions and 204 countries and territories, Asia and Latin America are still the hotpot regions of BTC disease burden. Studies based on the geographical distribution of the disease burden are important references for the development of public health policies in different regions and countries.^[35–37]

Regional differences reflect the trends and status of disease prevention in each country and region, which is the synthesizing embodiment of demographic background, level of medical care, public health policies, and level of economic development of each country and region. For example, in countries such as India, where the absolute numbers are high, and the ASRs are increasing, and in countries such as Chile, where the levels of ASRs are high and have sustained growth, more healthcare resources are needed to reach the milestone of reversing the growth trend. Data from Japan and China suggest that countries with historically high numbers and ASRs of disease burdens can reduce or reverse the growth in disease burden by sustained public health efforts over decades.

Localized research on BTC may have played a positive role. Local researchers and medical professionals have continued to present new findings on etiology researches, and studies on risk factors, diagnosis, and treatment options of BTC, and have each developed localized diagnostic and treatment guidelines and procedures.^[36,37] For example, in Shanghai, China, a population-wide case-control study of biliary tract cancer in Shanghai was initiated as early as 1997 and delved into the possible etiology of biliary tract cancer.^[38,39] In Japan, the frequently updated biliary tract cancer treatment guidelines provide a detailed and standardized treatment process that can be replicated.^[40,41] The results of such localized studies are very likely to contribute to the improvement of BTC prevention, diagnosis, and treatment in the region. The premise of the above results is the increase in the level of integrated social development, which can be reflected to some extent in the development of HDI during the study period.^[17]

Analysis of the HDI introduced in this study also suggests that the burden of the disease is related to the level of comprehensive development of the society exists, however, the trend supporting this relationship is not continuous. It is worth noting that the linear relationship between these ASRs and HDI is opposite and most significant in countries with very high HDI as opposed to low and medium HDI.

This phenomenon of showing different correlation trends at different HDI index stages suggests, on the one hand, that there is a close relationship between the level of social development and the level of BTC disease burden, like other digestive cancers.^[22,23] On the other hand, there are more complex risk factors for BTC disease burden, which, from the point of view of the development of the HDI index, influence each other, promote each other, or neutralize each other.

Significant differences were observed in multiple fields, including trends and distributions of incidence, mortality, and sexes, between the results of this study and other global data.^[42] A possible explanation for the differences is the misclassification and under-reporting of cancer cases in undeveloped regions. Considering the anatomical definition of the biliary system, there is still room for improvement in the way GBD data are counted.^[14,15] We made logical assumptions and explanations for the variations of disease burden estimates of different sexes and age subgroups, however, these hypotheses need further verification to elucidate the significance of these variations in etiology and disease prevention and control.^[6]

Conclusion

Globally, the number of incident cases, deaths, and DALYs for BTC reached a new peak in 2021, but the ASRs for these metrics declined between 1990 and 2021. The changes in the sex ratio of BTC incidence and other disease burden indicators may correspond with gender differences in the incidence of benign diseases and the promotion of relative interventions. Geographic differences in the burden of disease for BTC, especially among different countries and territories, provide effective references for public health departments to develop locally adapted policies. Taken together, this study provides evidence concerning the formulation of prevention and control policies and etiologic studies for BTC and proposes logical hypotheses to investigate.

Acknowledgments

We especially sincerely thank the health experts who promote the collection of data, and especially appreciate the contributions of all GBD 2021 collaborators.

Funding

This study is supported by grants from The Key Program of Shanghai Science and Technology Commission (No. YDZX20193100004049), the National Natural Science Foundation of China (Nos. 82273016 and 82303937), the Open Project of State Key Laboratory of Oncogenes and Related Genes (No. KF2120), the National Key Project of Research and Development Program of China (Nos. 2021YFC2500404 and 2021YFC2500405), and the University-level Scientific Fund of Shanghai University of Medicine and Health Sciences (No. SSF-24-16-02).

Conflicts of interest

None.

Data Sharing

All data on disease burden estimates in this study were acquired from the GHDx (http://ghdx.healthdata.org/gbd-results-tool).