Abstract
Background
Biliary tract cancers (BTCs) in East Asia have traditionally been linked to inflammation-related etiologies. However, metabolic risk factors are increasingly contributing to BTC development. Temporal trends in BTC epidemiology remain underexplored in the Korean population. This study aimed to review 13-year trends in BTCs, focusing on metabolic and inflammatory comorbidities, in a southeastern Korean cancer center historically characterized by a high incidence of inflammation-related BTCs.
Methods
This retrospective observational study analyzed 2,001 patients diagnosed with BTC at a southeastern Korean cancer center between 2009 and 2021. Data on general demographics (age, sex, smoking, alcohol), metabolic factors (hypertension, body mass index (BMI) ≥ 25 kg/m2, diabetes mellitus, hyperlipidemia), and inflammatory conditions (viral hepatitis, liver cirrhosis, bile duct stone, liver fluke infection) were collected from electronic medical records. Annual trends in clinical and demographic variables were evaluated using the Cochran–Armitage trend test.
Results
The cohort comprised 680 patients with gallbladder cancers and 586 patients with intrahepatic, 367 patients with perihilar, and 368 patients with distal cholangiocarcinomas. Gallbladder cancer was more common in females, whereas intrahepatic and distal cholangiocarcinomas were more prevalent in males. Over 13 years, the prevalence of hypertension and BMI ≥ 25 kg/m2 significantly increased (both p < 0.001). The prevalence of diabetes mellitus remained stable at approximately 26%. Although the proportion of patients with inflammatory factors increased from 38.9% to 49.7%, this trend was not statistically significant (p = 0.646).
Conclusions
These findings indicate an ongoing epidemiological transition in BTC risk profiles in Korea, with metabolic factors, such as hypertension and obesity, playing an increasingly prominent role.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12876-026-04714-8.
Keywords: Biliary tract cancer, Epidemiology, Metabolic syndrome, Chronic inflammation
Background
Biliary tract cancers (BTCs) are rare but highly aggressive malignancies, encompassing gallbladder cancer (GBC), intrahepatic cholangiocarcinoma (iCCA), perihilar cholangiocarcinoma (pCCA), and distal cholangiocarcinoma (dCCA) [1–3]. While traditionally considered uncommon, the global incidence of BTCs has increased over the past decade [1, 4]. Because BTC is anatomically difficult to access, it is frequently diagnosed at an advanced stage, and the availability of effective therapeutic options remains limited; both delayed diagnosis and restricted treatment options independently contribute to its poor overall prognosis [5, 6]. While early detection is the most effective strategy for improving outcomes, reliable and efficient methods for early diagnosis remain lacking. Therefore, primary prevention remains a critical approach for reducing disease burden and improving long-term outcomes.
In East Asia, BTC has historically been associated with chronic inflammatory conditions, such as liver fluke infection and bile duct stones. Notably, pCCA strongly correlates with Clonorchis sinensis infection [2, 3]. Between 2011 and 2020, the Ministry of Health and Welfare had conducted a community-based clonorchiasis control program in southern Korea, screening 23,000–42,000 residents annually and reducing infection rates from 11.1% to 3.8%. Furthermore, in recent decades, the adoption of Westernized lifestyles has rapidly transformed the epidemiological landscape of East Asian countries. An aging population and increased consumption of high-calorie diets have led to the rise in metabolic risk factors such as obesity, hypertension, and diabetes, resembling the risk profiles commonly observed in Western populations [7, 8]. This epidemiologic transition, often referred to as the “global cancer transition,” reflects a rising incidence of cancers—such as breast, prostate, and colorectal cancer—historically prevalent in the West, now emerging in Asia alongside persistent burdens of infection-related malignancies [9].
Moreover, regional differences in BTC prevalence within East Asia are evident. For example, southeastern Korea has historically exhibited high BTC rates due to endemic liver fluke infections. However, in these regions, metabolic factors are becoming increasingly relevant, altering the traditional risk profile of BTCs [1, 2].
Despite this changing landscape, few studies have examined the evolution of BTC epidemiology in regions where inflammation-related cancers are still prevalent [10–12]. Understanding the dynamic interplay between declining infection risk factors and increasing metabolic status is essential for developing tailored prevention and treatment strategies in East Asia.
This study aimed to provide a comprehensive 13-year overview of BTC trends, with a particular focus on demographic patterns, including metabolic and inflammatory comorbidities, at a southeastern Korean cancer center historically characterized by a high incidence of inflammation-related BTCs.
Methods
Study design and population
This retrospective observational study was conducted at a single, representative cancer center in southeastern Korea, a region historically characterized by a high prevalence of inflammation-related BTCs. We analyzed 2,001 patients with pathologically or radiologically confirmed BTCs between January 2009 and December 2021. BTC subtypes were classified as GBC (n = 680), iCCA (n = 588), pCCA (n = 368), or dCCA (n = 369) based on anatomical and histological criteria. Among the total cohort, four patients were diagnosed with overlapping BTC subtypes (GBC + iCCA, n = 2; GBC + dCCA, n = 1; and iCCA + pCCA, n = 1). To maintain consistency in subtype-specific analyses, these patients were assigned to the clinically predominant lesion: the two GBC + iCCA cases and one with GBC + dCCA case were included in the GBC group, whereas the iCCA + pCCA case was included in the iCCA group. Consequently, the final mutually exclusive subtype counts were: GBC (n = 680), iCCA (n = 586), pCCA (n = 367), and dCCA (n = 368), which totaled the entire study cohort (n = 2,001), without duplication. This approach ensured that the final subtype-specific count matched the total cohort size (n = 2,001).
Data collection
Clinical and demographic data were extracted from electronic medical records. Variables were categorized into three domains: (1) general demographic factors, including age, sex, smoking status, and alcohol consumption; (2) metabolic factors, defined as the presence of hypertension, body mass index (BMI) ≥ 25 kg/m2, diabetes mellitus, or hyperlipidemia; and (3) inflammatory factors, defined as documented viral hepatitis, liver cirrhosis, bile duct stones, or liver fluke (Clonorchis sinensis) infection. Data were collected annually to assess temporal trends over the 13-year study period.
Time series analysis was performed on almost all variables, and variables with known epidemiological relevance in other cancer types were selected.
Statistical analysis
All variables were systematically screened for missing values. Records with missing data in key variables required for each specific analysis were excluded using a complete-case approach. The proportion of missing data was low and distributed randomly across study years, minimizing the risk of systematic bias. Descriptive statistics were used to summarize the demographic and clinical characteristics of BTC subtypes. Continuous variables are presented as mean ± standard deviation or median with interquartile range, while categorical variables are reported as frequencies and percentages. Chi-squared or Fisher’s exact tests were used to compare categorical variables across years. To assess temporal trends in the annual incidence of BTCs and their anatomical subtypes, linear regression analysis was performed using the year of diagnosis as the independent variable and the number of new cases as the dependent variable. Goodness-of-fit was evaluated using the coefficient of determination (R²), and statistical significance was determined using p-values derived from the regression model. Statistical significance was set at p < 0.05.
Variables included in temporal trend analyses were selected a priori based on established clinical relevance and previous literature on BTC risk factors. Only variables with sufficient data completeness and clinically interpretable temporal patterns were subjected to formal trend testing. Temporal trends of categorical variables were assessed using the Cochran–Armitage trend test only when the variable demonstrated visually assessed monotonic or near-monotonic patterns over time. Variables exhibiting irregular or non-monotonic temporal distributions were summarized descriptively, without formal trend testing, to avoid potential model misspecification and misinterpretation of statistical significance.
All statistical analyses were conducted using IBM SPSS Statistics for Windows, Version 26.0 (IBM Corp., Armonk, NY, USA) and R software, Version 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria) [13, 14].
Ethics statement
This study was conducted in accordance with the ethical principles of the Declaration of Helsinki and was approved by the Institutional Review Board of Pusan National University Hospital (IRB No. 2508-026-154). The requirement for informed consent was waived due to the retrospective nature of the study.
Results
Patient demographic characteristics according to tumor location
The mean age at diagnosis was 70.0 ± 10.3 years, with pCCA patients being the oldest (72.1 ± 9.7 years) and iCCA patients the youngest (68.4 ± 10.7 years). Females were predominant in the GBC group (55.3%), whereas males were predominant in the iCCA (68.1%) and dCCA (59.5%) groups. Among 1,834 patients with available data, 26.7% had a history of smoking, and 25.7% reported alcohol use, with smoking (34.2%) and alcohol consumption (31.8%) being the most frequent factors among patients with iCCA.
Overall, 43.7% of the patients exhibited one or more inflammatory factors (viral hepatitis, liver cirrhosis, bile duct stone, or liver fluke infection). The prevalence of viral hepatitis was the highest in iCCA (15.5%), whereas bile duct stones were common across all subtypes, particularly in dCCA (32.9%). Liver fluke infection was observed in 5.0% of patients, with slightly higher prevalence in pCCA (5.7%) and dCCA (6.3%).
Metabolic factors (defined as the presence of hypertension, diabetes mellitus, hyperlipidemia, or elevated BMI) were observed in 65.6% of the patients, with the highest frequency in dCCA (70.4%). Hypertension was the most common metabolic condition (46.7%), and diabetes mellitus was present in 26.3% of the patients. The mean BMI was 23.1 ± 3.4 kg/m², with most patients in the normal range (66.6%); overweight and obesity were more frequent in GBC (28.8%) and dCCA (27.7%) (Table 1).
Table 1.
Baseline demographic, clinical, and laboratory characteristics of patients with biliary tract cancer according to tumor location
| Variable | Total (n = 2001) | GBC (n = 680) | iCCA (n = 586) | pCCA (n = 367) | dCCA (n = 368) |
|---|---|---|---|---|---|
| General factors | |||||
| Sex | |||||
| male | 1,123 (56.1) | 304 (44.7) | 399 (68.1) | 201 (54.8) | 219 (59.5) |
| female | 878 (43.9) | 376 (55.3) | 187 (31.9) | 166 (45.2) | 149 (40.5) |
| Age (years) | |||||
| mean ± SD | 70.0 ± 10.3 | 69.9 ± 10.4 | 68.4 ± 10.7 | 72.1 ± 9.7 | 70.7 ± 9.5 |
| median (IQR) | 71.0 (63.0–77.0) | 72.0 (63.0–77.0) | 69.0 (62.0–76.0) | 74.0 (66.0–79.0) | 72.0 (64.0–77.0) |
| ≤ 49yrs | 71 (3.5) | 29 (4.3) | 27 (4.6) | 6 (1.6) | 9 (2.4) |
| 50-64yrs | 497 (24.8) | 165 (24.3) | 173 (29.5) | 74 (20.2) | 85 (23.1) |
| 65-79yrs | 1,094 (54.7) | 372 (54.7) | 308 (52.6) | 206 (56.1) | 208 (56.5) |
| ≥ 80yrs | 339 (16.9) | 114 (16.8) | 78 (13.3) | 81 (22.1) | 66 (17.9) |
| Smoking (n = 1,834) | |||||
| Yes | 490 (26.7) | 130 (21.5) | 188 (34.2) | 90 (26.3) | 82 (24.3) |
| No | 1,344 (73.3) | 476 (78.5) | 361 (65.8) | 252 (73.7) | 255 (75.7) |
| Alcohol (n = 1,834) | |||||
| Yes | 471 (25.7) | 134 (22.1) | 174 (31.8) | 74 (21.6) | 89 (26.4) |
| No | 1,363 (74.3) | 473 (77.9) | 373 (68.2) | 269 (78.4) | 248 (73.6) |
| Inflammatory factors | |||||
| Yes | 874 (43.7) | 279 (41.0) | 275 (46.9) | 158 (43.1) | 162 (44.0) |
| No | 1,127 (56.3) | 401 (59.0) | 311 (53.1) | 209 (56.9) | 206 (56.0) |
| Viral hepatitis (n = 1,916) | |||||
| Yes | 184 (9.6) | 46 (7.1) | 87 (15.5) | 28 (8.0) | 23 (6.5) |
| No | 1,732 (90.4) | 606 (92.9) | 475 (84.5) | 321 (92.0) | 330 (93.5) |
| Liver cirrhosis (n = 1,841) | |||||
| Yes | 112 (6.1) | 29 (4.6) | 48 (8.9) | 17 (5.0) | 18 (5.4) |
| No | 1,729 (93.9) | 598 (95.4) | 493 (91.1) | 320 (95.0) | 318 (94.6) |
| Bile duct stone | |||||
| Yes | 608 (30.4) | 212 (31.2) | 161 (27.5) | 114 (31.1) | 121 (32.9) |
| No | 1,393 (69.6) | 468 (68.8) | 425 (72.5) | 253 (68.9) | 247 (67.1) |
| Liver fluke infection | |||||
| Yes | 101 (5.0) | 27 (4.0) | 30 (5.1) | 21 (5.7) | 23 (6.3) |
| No | 1,900 (95.0) | 653 (96.0) | 556 (94.9) | 346 (94.3) | 345 (93.8) |
| Metabolic factors | |||||
| Yes | 1,312 (65.6) | 439 (64.6) | 364 (62.1) | 250 (68.1) | 259 (70.4) |
| No | 689 (34.4) | 241 (35.4) | 222 (37.9) | 117 (31.9) | 109 (29.6) |
| HT (n = 1,936) | |||||
| Yes | 905 (46.7) | 313 (47.7) | 247 (43.3) | 168 (47.5) | 177 (49.7) |
| No | 1,031 (53.3) | 343 (52.3) | 323 (56.7) | 186 (52.5) | 179 (50.3) |
| DM (n = 1,929) | |||||
| Yes | 508 (26.3) | 166 (25.3) | 124 (21.9) | 114 (32.4) | 104 (29.4) |
| No | 1,421 (73.7) | 490 (74.7) | 443 (78.1) | 238 (67.6) | 250 (70.6) |
| Hyperlipidemia (n = 1,731) | |||||
| Yes | 246 (14.2) | 77 (13.5) | 62 (12.0) | 48 (15.0) | 59 (18.2) |
| No | 1,485 (85.8) | 492 (86.5) | 455 (88.0) | 273 (85.0) | 265 (81.8) |
| BMI (kg/m²) (n = 1,654) | |||||
| mean ± SD | 23.1 ± 3.4 | 23.4 ± 3.5 | 22.9 ± 3.5 | 22.7 ± 3.2 | 23.3 ± 3.3 |
| median (IQR) | 22.9 (20.9–25.1) | 23.2 (20.9–25.4) | 22.7 (20.8–25.0) | 22.5 (20.8–24.7) | 23.2 (21.1–25.4) |
| < 18.5 (underweight) | 128 (7.7) | 34 (6.2) | 52 (10.8) | 25 (8.0) | 17 (5.5) |
| 18.5–24.9 (normal) | 1,101 (66.6) | 358 (65.0) | 311 (64.5) | 227 (72.3) | 205 (66.8) |
| 25-29.9 (overweight) | 376 (22.7) | 138 (25.0) | 103 (21.4) | 56 (17.8) | 79 (25.7) |
| ≥ 30 (obese) | 49 (3.0) | 21 (3.8) | 16 (3.3) | 6 (1.9) | 6 (2.0) |
Data are presented as number (%) for categorical variables and mean ± standard deviation or median (interquartile range) for continuous variables. BTC subtypes include GBC gallbladder cancer, iCCA intrahepatic cholangiocarcinoma, pCCA perihilar cholangiocarcinoma, and dCCA distal cholangiocarcinoma. Variables assessed include sex, age, smoking, alcohol use, inflammatory factors (viral hepatitis, liver cirrhosis, bile duct stone, liver fluke infection), and metabolic factors (HT hypertension, DM diabetes mellitus, hyperlipidemia, BMI body mass index )
Temporal trends in incidence and epidemiologic characteristics
The annual number of BTC diagnoses increased steadily from 36 in 2009 to 288 in 2020, before declining to 151 in 2021. GBC and iCCA demonstrated the most pronounced increases over time, whereas pCCA and dCCA exhibited modest growth (Table 2; Fig. 1). Linear regression analysis revealed a significant upward trend in the annual number of newly diagnosed BTC cases between 2009 and 2021 (R² = 0.390; p = 0.022). Subtype-specific analyses revealed notable increases in GBC (R² = 0.331, p = 0.040), pCCA (R² = 0.499, p = 0.007), and dCCA (R² = 0.354, p = 0.032). In contrast, iCCA showed an upward trend that did not reach statistical significance (R² = 0.227, p = 0.100) (Fig. 2).
Table 2.
Annual number of newly diagnosed biliary tract cancer (BTC) cases by tumor location from 2009 to 2021
| BTC Subtype | 2009 | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GBC | 11 | 20 | 50 | 83 | 59 | 50 | 32 | 47 | 60 | 56 | 55 | 97 | 60 |
| iCCA | 9 | 17 | 45 | 64 | 53 | 48 | 30 | 36 | 39 | 83 | 41 | 86 | 35 |
| pCCA | 9 | 12 | 32 | 31 | 15 | 27 | 19 | 24 | 34 | 53 | 35 | 43 | 33 |
| dCCA | 7 | 17 | 35 | 31 | 20 | 23 | 18 | 23 | 31 | 32 | 46 | 62 | 23 |
| Total | 36 | 66 | 162 | 209 | 147 | 148 | 99 | 130 | 164 | 224 | 177 | 288 | 151 |
BTC Biliary tract cancer, GBC Tumor subtypes include gallbladder cancer, iCCA intrahepatic cholangiocarcinoma, pCCA perihilar cholangiocarcinoma, and dCCA distal cholangiocarcinoma. Data are presented as absolute numbers for each calendar year
Fig. 1.
Annual number of newly diagnosed biliary tract cancer cases by tumor subtype from 2009 to 2021. Tumor subtypes include gallbladder cancer (GBC), intrahepatic cholangiocarcinoma (iCCA), perihilar cholangiocarcinoma (pCCA), and distal cholangiocarcinoma (dCCA)
Fig. 2.
Annual trends in newly diagnosed biliary tract cancers (BTCs) from 2009 to 2021. Line graphs show the number of new cases per year for overall BTCs and by anatomical subtype: gallbladder cancer (GBC), intrahepatic cholangiocarcinoma (iCCA), perihilar cholangiocarcinoma (pCCA), and distal cholangiocarcinoma (dCCA). R² values are shown for each regression model to indicate the strength of linear trend
The proportion of older patients (age ≥ 65 years) increased significantly, from 47.2% in 2009 to 79.5% in 2021 (p < 0.001), indicating a progressive aging of the BTC patient population. The male-to-female ratio remained relatively stable over time with no significant trend (p = 0.055). Smoking and alcohol consumption patterns remained largely stable throughout the study period. The proportion of patients with any inflammatory risk factor (e.g., viral hepatitis, liver cirrhosis, bile duct stone, and liver fluke infection) showed a modest increase from 38.9% in 2009 to 49.7% in 2021; however, this change was not statistically significant (p = 0.646) (Table 3; Fig. 3a and b).
Table 3.
Temporal trends in clinical and metabolic comorbidities among patients with biliary tract cancer from 2009 to 2021
| Variable/Year | 2009 (n = 36) |
2010 (n = 66) |
2011 (n = 162) |
2012 (n = 209) |
2013 (n = 147) |
2014 (n = 148) |
2015 (n = 99) |
2016 (n = 130) |
2017 (n = 164) |
2018 (n = 224) |
2019 (n = 177) |
2020 (n = 288) |
2021 (n = 151) |
p-value |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| General demographic factors | ||||||||||||||
| Sex (male) | 20 (55.6) | 39 (59.1) | 95 (58.6) | 121 (57.9) | 92 (62.6) | 91 (61.5) | 55 (55.6) | 64 (49.2) | 92 (56.1) | 127 (56.7) | 88 (49.7) | 153 (53.1) | 86 (57.0) | 0.055 |
| Age ≥ 65 (years) | 17 (47.2) | 41 (62.1) | 114 (70.4) | 133 (63.6) | 99 (67.3) | 107 (72.3) | 67 (67.7) | 94 (72.3) | 118 (72.0) | 165 (73.7) | 131 (74.0) | 227 (78.8) | 120 (79.5) | < 0.001 |
| Smoking (n = 1,834) | 6 (17.1) | 15 (23.8) | 49 (34.3) | 66 (35.9) | 31 (22.5) | 31 (23.1) | 27 (31.0) | 25 (21.7) | 43 (30.1) | 61 (29.6) | 34 (20.7) | 65 (23.7) | 37 (25.0) | 0.069 |
| Alcohol (n = 1,834) | 7 (20.0) | 19 (30.2) | 42 (29.4) | 43 (23.2) | 36 (26.1) | 40 (29.9) | 23 (26.4) | 22 (19.3) | 37 (25.9) | 61 (29.6) | 38 (23.2) | 65 (23.7) | 38 (25.7) | 0.523 |
| Inflammatory factors | 14 (38.9) | 25 (37.9) | 69 (42.6) | 84 (40.2) | 63 (42.9) | 66 (44.6) | 46 (46.5) | 68 (52.3) | 80 (48.8) | 99 (44.2) | 76 (42.9) | 104 (36.1) | 75 (49.7) | 0.646 |
| Viral hepatitis (n = 1,916) | 4 (11.4) | 7 (11.1) | 8 (5.3) | 21 (10.6) | 18 (12.9) | 22 (15.6) | 9 (9.5) | 9 (7.1) | 18 (11.8) | 26 (12.1) | 20 (11.6) | 12 (4.3) | 10 (6.7) | 0.110 |
| Liver cirrhosis (n = 1,841) | 3 (10.0) | 4 (7.5) | 15 (12.0) | 12 (6.3) | 8 (5.7) | 5 (3.6) | 8 (8.8) | 5 (4.1) | 7 (4.9) | 16 (7.7) | 11 (6.6) | 9 (3.2) | 9 (6.0) | 0.043 |
| Bile duct stone | 7 (19.4) | 16 (24.2) | 47 (29.0) | 50 (23.9) | 42 (28.6) | 45 (30.4) | 33 (33.3) | 54 (41.5) | 59 (36.0) | 72 (32.1) | 57 (32.2) | 75 (26.0) | 51 (33.8) | 0.085 |
|
Liver fluke infection |
2 (5.6) | 0 (0.0) | 7 (4.3) | 11 (5.3) | 7 (4.8) | 6 (4.1) | 4 (4.0) | 11 (8.5) | 9 (5.5) | 4 (1.8) | 4 (2.3) | 20 (6.9) | 16 (10.6) | 0.066 |
| Metabolic factors | 24 (66.7) | 40 (60.6) | 90 (55.6) | 144 (68.9) | 83 (56.5) | 85 (57.4) | 65 (65.7) | 73 (56.2) | 104 (63.4) | 147 (65.6) | 127 (71.8) | 217 (75.3) | 113 (74.8) | < 0.001 |
| HT (n = 1,936) | 16 (44.4) | 26 (40.6) | 51 (34.0) | 89 (44.7) | 62 (43.7) | 56 (39.4) | 40 (41.7) | 53 (41.7) | 79 (52.0) | 103 (46.8) | 94 (54.3) | 150 (52.6) | 86 (57.3) | < 0.001 |
| DM (n = 1,929) | 8 (22.2) | 20 (31.3) | 39 (26.4) | 60 (30.0) | 25 (17.6) | 36 (25.4) | 29 (30.2) | 38 (30.2) | 40 (26.5) | 59 (27.1) | 53 (30.8) | 71 (25.0) | 30 (20.0) | 0.577 |
| Hyperlipidemia (n = 1,731) | 8 (24.2) | 11 (18.6) | 26 (18.2) | 31 (16.6) | 13 (9.4) | 13 (9.6) | 5 (5.9) | 7 (6.3) | 12 (8.8) | 23 (11.8) | 24 (15.5) | 35 (14.1) | 38 (36.9) | 0.242 |
|
BMI ≥ 25 (kg/m²) (n = 1,654) |
6 (17.6) | 9 (15.0) | 25 (19.5) | 39 (22.3) | 27 (21.8) | 28 (24.8) | 19 (26.0) | 18 (18.2) | 37 (29.8) | 50 (26.7) | 42 (29.2) | 82 (33.1) | 43 (29.7) | < 0.001 |
Data are presented as number (%) of patients for each variable by year of diagnosis. Variables include sex (male), age ≥ 65, smoking, alcohol use, inflammatory factors (viral hepatitis, liver cirrhosis, bile duct stone, liver fluke infection), and metabolic factors (HT hypertension, DM diabetes mellitus, hyperlipidemia, BMI ≥ 25 body mass index ≥ 25). p values indicate statistical significance for trend over time
Fig. 3.
Temporal trends in the proportion of each biliary tract cancer (BTC) subtype from 2009 to 2021, expressed as a percentage of all BTC cases diagnosed each year. a Variables represent general demographic factors, including sex (male), age ≥ 65 years, smoking, and alcohol consumption. b Variables represent inflammatory factors, including viral hepatitis, liver cirrhosis, bile duct stones, and liver fluke infection. c Variables represent metabolic factors, including hypertension (HT), diabetes mellitus (DM), hyperlipidemia, and elevated body mass index (BMI ≥ 25 kg/m²). Cochran–Armitage trend test was applied only to HT and BMI ≥ 25 kg/m²
Over the study period, metabolic risk factors exhibited a clear upward trajectory. The prevalence of hypertension increased from 44.4% in 2009 to 57.3% in 2021 (p < 0.001). Similarly, the proportion of patients with BMI ≥ 25 kg/m2 rose from 17.6% in 2009 to 29.7% in 2021 (p < 0.001). After adjustment for age in multivariable logistic regression models, the temporal increase in the prevalence of hypertension and BMI ≥ 25 remained statistically significant, indicating that these trends cannot be explained solely by aging of the study population (Supplementary Table 1). In contrast, diabetes mellitus remained stable (p = 0.577), and hyperlipidemia did not exhibit significant changes (Table 3; Fig. 3c).
Discussion
This study presents a comprehensive 13-year analysis of BTCs in southeastern Korea, a region historically characterized by a high prevalence of inflammation-related cancers. Although inflammation remains an important oncogenic mechanism, our findings show that its overall contribution has remained relatively stable over time. In contrast, we observed a rising burden of metabolic characteristics, such as hypertension and elevated BMI, reflecting a gradual shift toward a Westernized risk profile. These temporal trends were statistically confirmed using the Cochran–Armitage trend test, which demonstrated significant linear increases in metabolic characteristics over the study period.
Traditionally, BTC in East Asia has been closely associated with chronic inflammatory etiologies, including liver fluke infections and bile duct stones. However, the marked increase in hypertension and overweight/obesity indicates that Western-style metabolic risk factors are becoming more prevalent in the BTC population [1, 9]. In this study, hypertension emerged as the most prevalent comorbidity among patients with BTC, affecting nearly half of the cohort and increasing significantly over the 12-year study period (from 44.4% in 2009 to 57.3% in 2021). Similarly, the proportion of patients with elevated body mass index (BMI ≥ 25 kg/m2) demonstrated a steady upward trend, increasing from 17.6% to 29.7%. These findings align with the broader epidemiological shifts observed in the general population, driven by sedentary lifestyles, dietary transitions toward energy-dense foods, and an aging population. This transition mirrors the broader trends observed in East Asian cancer epidemiology, where breast, prostate, and colorectal cancers have also increased in tandem with industrialization, sedentary lifestyles, and dietary westernization [9].
The increasing prevalence of hypertension and obesity in patients with BTC has important clinical and etiological implications. Both conditions induce chronic low-grade inflammation, promote insulin resistance, and disrupt bile acid metabolism [1, 15, 16]. These metabolic disturbances may create a procarcinogenic hepatic and biliary microenvironment, facilitating genetic and epigenetic changes that predispose to tumorigenesis [17–19].
The prevalence of diabetes mellitus remained relatively stable over time (average: 26.3%), consistent with existing literature linking diabetes mellitus to hepatic steatosis, non-alcoholic fatty liver disease (NAFLD), and an increased risk of hepatobiliary malignancies [20, 21]. Hyperlipidemia did not exhibit significant temporal variation but was more common in dCCA, potentially reflecting a subtype-specific involvement in lipid metabolism.
These epidemiological trends underscore the importance of incorporating metabolic health assessments into the care for patients with BTC [20, 22]. Given the interplay between metabolic status and cancer outcomes, personalized treatment approaches that consider these comorbidities may improve prognosis and quality of life [6, 23]. From 2009 to 2021, the inflammatory characteristics associated with chronic biliary inflammation in patients with BTC remained prevalent. Conditions such as viral hepatitis, liver cirrhosis, bile duct stones, and liver fluke infection did not show statistically significant upward trends.
Chronic inflammation plays a key role in BTC development. Persistent bile duct injury—caused by mechanical irritation from gallstones or bile duct strictures, chemical irritation from cholestasis, or parasitic infection with Clonorchis sinensis—leads to continuous activation of inflammatory pathways [24]. Over time, this inflammatory microenvironment promotes dysplasia and malignant transformation [25–27].
Between 2011 and 2020, the Ministry of Health and Welfare had implemented a community-based screening and chemotherapy program for clonorchiasis in southern Korea. The program screened 23,000–42,000 residents annually, reducing egg-positive rates from 11.1% to 3.8% and lowering prevalence to below 5% in most endemic areas [10]. Consequently, infection-related inflammation has markedly declined. However, preexisting conditions such as primary sclerosing cholangitis or recurrent choledocholithiasis continue to predispose bile ducts to chronic injury, maintaining a residual inflammatory risk for tumor development [1, 28].
In this study, the overall incidence of BTC increased between 2009 and 2021. Although inflammatory factors are known contributors to the development of dCCA and pCCA, the concurrent rise in metabolic risk factors appears to be associated with the sustained increase in these subtypes. However, given annual fluctuations in subtype case numbers, these observations should be interpreted with caution. This overall trend likely reflects increased diagnostic awareness and advances in imaging technologies. In addition, the temporary decline across all subtypes observed in 2021 may be attributable to healthcare disruptions caused by the COVID-19 pandemic.
This study had some limitations. First, its retrospective, single-center design may introduce selection bias and limit generalizability. Our institution functions as a major regional referral center for complex hepatobiliary diseases, including BTC; however, other tertiary hospitals exist in southeastern Korea. Consequently, the study population may not fully represent all BTC patients in the region, and caution is warranted when extrapolating these findings to broader populations. Second, although metabolic factors, such as hypertension and elevated BMI, were more common among patients with BTC, these trends parallel the rising prevalence of metabolic syndrome in the general population due to westernized lifestyles. Therefore, the observed temporal association does not confirm causality. Third, the Cochran–Armitage trend test was applied only to categorical variables that exhibited visually confirmed monotonic or near-monotonic temporal patterns. Variables with non-monotonic distributions over time were summarized descriptively, as formal trend testing could violate inappropriate model assumptions and artificially inflate statistical significance. Although this conservative approach enhances the validity of the trend analysis, it may reduce sensitivity to detect complex temporal patterns in some variables. Fourth, while metabolic characteristics such as hypertension, BMI, diabetes mellitus, or hyperlipidemia were assessed, data on medication use, lifestyle factors (e.g., diet and physical activity), and duration or severity of comorbidities were unavailable, potentially confounding the observed associations. Fifth, temporal changes in diagnostic practices may have influenced the observed increase in BTC incidence over the 13-year study period. Although imaging advancements may have improved incidental detection of BTCs, core diagnostic methods (ERCP-guided or percutaneous biopsies) and referral center availability remained largely unchanged, minimizing bias from diagnostic inconsistency. However, lack of year-specific diagnostic data limits our ability to fully assess detection-related bias. Sixth, inflammatory status was inferred from clinical documentation and laboratory findings rather than standardized inflammatory indices or histopathology, introducing potential misclassification bias. Therefore, prospective controlled studies incorporating mechanistic analyses are necessary to clarify the etiological roles of metabolic and inflammatory factors in the development of BTC. Seventh, a small number of cases with overlapping BTC subtypes were assigned to the predominant lesion rather than excluded, which may have introduced minor classification bias; however, their impact on overall results is expected to be negligible.
Conclusion
In this 13-year retrospective study, we observed a significant increase in metabolic risk factors—particularly hypertension and obesity (BMI ≥ 25 kg/m2)—among patients with BTCs. These findings reflect a broader epidemiological shift and highlight the growing relevance of metabolic syndrome in the contemporary risk landscape of BTCs. While causal inferences cannot be drawn from this observational data, our results support the importance of public health strategies that incorporate metabolic risk assessment and prevention in regions historically focused on inflammation-related BTCs.
Supplementary Information
Acknowledgements
Not applicable.
Abbreviations
- BTC
Biliary tract cancer
- GBC
Gallbladder cancer
- iCCA
Intrahepatic cholangiocarcinoma
- pCCA
Perihilar cholangiocarcinoma
- dCCA
Distal cholangiocarcinoma
- BMI
Body mass index
- HT
Hypertension
- DM
Diabetes mellitus
Authors’ contributions
AJ and JL contributed to the study design and data collection. AJ performed the data analysis. JL supervised the project. AJ drafted the manuscript. Both AJ and JL contributed to the interpretation of the results and critically revised the manuscript.
Funding
This study was supported by Biomedical Research Institute Grant (202200140001) Pusan National University Hospital.
Data availability
The datasets used and analyzed during this study are available from the corresponding author upon reasonable request.
Declarations
Ethical approval and consent to participate
This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Institutional Review Board of Pusan National University Hospital (IRB No. 2508-026-154). The requirement for informed consent was waived due to the retrospective nature of the study.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Banales JM, Marin JJG, Lamarca A, Rodrigues PM, Khan SA, Roberts LR, et al. Cholangiocarcinoma 2020: the next horizon in mechanisms and management. Nat Rev Gastroenterol Hepatol. 2020;17(9):557–88. 10.1038/s41575-020-0310-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Woo S, Kim Y, Hwang S, Chon HJ. Epidemiology and genomic features of biliary tract cancer and its unique features in Korea. J Liver Cancer. 2025;25(1):41–51. 10.17998/jlc.2025.02.27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Zhou J, Tan G, Zhang L, Xie G, Chen W, Zhang X, et al. Epidemiology of biliary tract cancer in China: a narrative review. Chin J Cancer Res. 2024;36(5):474–88. 10.21147/j.issn.1000-9604.2024.05.02. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Koshiol J, Yu B, Kabadi SM, Baria K, Shroff RT. Epidemiologic patterns of biliary tract cancer in the United States: 2001–2015. BMC Cancer. 2022;22(1):1178. 10.1186/s12885-022-10286-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Beaulieu C, Lui A, Yusuf D, Abdelaziz Z, Randolph B, Batuyong E, et al. A Population-based retrospective study of biliary tract cancers in Alberta, Canada. Curr Oncol. 2021;28(1):417–27. 10.3390/curroncol28010044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zhao DY, Lim KH. Current biologics for treatment of biliary tract cancers. J Gastrointest Oncol. 2017;8(3):430–40. 10.21037/jgo.2017.05.04. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Huang J, Ngai CH, Deng Y, Tin MS, Lok V, Zhang L, et al. Cancer incidence and mortality in Asian countries: a trend analysis. Cancer Control. 2022;29:10732748221095955. 10.1177/10732748221095955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Fu M, Peng Z, Wu M, Lv D, Li Y, Lyu S. Current and future burden of breast cancer in Asia : a GLOBOCAN data analysis for 2022 and 2050. Breast. 2025;79:103835. 10.1016/j.breast.2024.103835. [DOI] [PMC free article] [PubMed]
- 9.Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74(3):229–63. 10.3322/caac.21834. [DOI] [PubMed] [Google Scholar]
- 10.Lee MR, Shin HE, Back SO, Lee YJ, Lee HI, Ju JW. Status of helminthic infections in residents around river basins in the Republic of Korea for 10 years (2011–2020). Korean J Parasitol. 2022;60(3):187–93. 10.3347/kjp.2022.60.3.187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Namboonrueng K, Charoensuk L, Pinlaor S, Laothong U, Suttiprapa S, Suwannatrai AT, et al. Environmental contamination of Opisthorchis viverrini eggs and its impact on transmission control in rural northeast Thailand. Infect Dis (Lond). 2025;57(5):397–406. 10.1080/23744235.2024.2443010. [DOI] [PubMed] [Google Scholar]
- 12.Pupacdi B, Loffredo CA, Budhu A, Rabibhadana S, Bhudhisawasdi V, Pairojkul C, et al. The landscape of etiological patterns of hepatocellular carcinoma and intrahepatic cholangiocarcinoma in Thailand. Int J Cancer. 2024;155(8):1387–99. 10.1002/ijc.35034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.IBM Corp. IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY: IBM Corp; 2019. [Google Scholar]
- 14.R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. 2021. Available from: https://www.R-project.org/
- 15.Chen TI, Chen MH, Yin SC, Lin CJ, Lam TK, Huang CW, et al. Associations between metabolic syndrome and cholangiocarcinoma risk: a large-scale population-based cohort study. Hepatology. 2025. 10.1097/HEP.0000000000001312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Park JH, Hong JY, Park YS, Kang G, Han K, Park JO. Persistent status of metabolic syndrome and risk of cholangiocarcinoma: A Korean nationwide population-based cohort study. Eur J Cancer. 2021;155:97–105. 10.1016/j.ejca.2021.06.052. [DOI] [PubMed] [Google Scholar]
- 17.Bojan A, Pricop C, Ciocoiu M, Vladeanu MC, Bojan IB, Badulescu OV, et al. Environmental and metabolic risk factors linked to gallbladder dysplasia. metabolites. 2024;14(5). 10.3390/metabo14050273. [DOI] [PMC free article] [PubMed]
- 18.Borena W, Edlinger M, Bjørge T, Häggström C, Lindkvist B, Nagel G, et al. A prospective study on metabolic risk factors and gallbladder cancer in the metabolic syndrome and cancer (Me-Can) collaborative study. PLoS ONE. 2014;9(2):e89368. 10.1371/journal.pone.0089368. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Fu K, Yang X, Wu H, Gong J, Li X, Diabetes. PKM2 affect prognosis in patients with intrahepatic cholangiocarcinoma. Oncol Lett. 2020;20(5):265. 10.3892/ol.2020.12128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Saengboonmee C, Seubwai W, Lert-Itthiporn W, Sanlung T, Wongkham S. Association of diabetes mellitus and cholangiocarcinoma: update of evidence and the effects of antidiabetic medication. Can J Diabetes. 2021;45(3):282–90. 10.1016/j.jcjd.2020.09.008. [DOI] [PubMed] [Google Scholar]
- 21.Li J, Han T, Xu L, Luan X. Diabetes mellitus and the risk of cholangiocarcinoma: an updated meta-analysis. Prz Gastroenterol. 2015;10(2):108–17. 10.5114/pg.2015.49004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Chaiteerakij R, Yang JD, Harmsen WS, Slettedahl SW, Mettler TA, Fredericksen ZS, et al. Risk factors for intrahepatic cholangiocarcinoma: association between metformin use and reduced cancer risk. Hepatology. 2013;57(2):648–55. 10.1002/hep.26092. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Gutmans J, Mechahougui H. Biliary tract cancers: advances in diagnostic and management. Explor Target Antitumor Ther. 2025;6:1002328. 10.37349/etat.2025.1002328. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Wise C, Pilanthananond M, Perry BF, Alpini G, McNeal M, Glaser SS. Mechanisms of biliary carcinogenesis and growth. World J Gastroenterol. 2008;14(19):2986–9. 10.3748/wjg.14.2986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Roy S, Glaser S, Chakraborty S. Inflammation and progression of cholangiocarcinoma: role of angiogenic and lymphangiogenic mechanisms. Front Med (Lausanne). 2019;6:293. 10.3389/fmed.2019.00293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Kongpetch S, Jusakul A, Ong CK, Lim WK, Rozen SG, Tan P, et al. Pathogenesis of cholangiocarcinoma: from genetics to signalling pathways. Best Pract Res Clin Gastroenterol. 2015;29(2):233–44. 10.1016/j.bpg.2015.02.002. [DOI] [PubMed] [Google Scholar]
- 27.Braconi C, Patel T Cholangiocarcinoma: new insights into disease pathogenesis and biology. Infect Dis Clin North Am. 2010;24(4):871–84, vii. 10.1016/j.idc.2010.07.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Chen J, Zhang S. The Role of Inflammation in Cholestatic Liver Injury. J Inflamm Res. 2023;16:4527–40. 10.2147/JIR.S430730. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets used and analyzed during this study are available from the corresponding author upon reasonable request.