Abstract
Background
Combined hepatocellular‐cholangiocarcinoma (HCC‐CC) is an aggressive primary liver cancer. However, the clinical features are not clearly understood because of limited literature and the complex nature of both hepatocellular carcinoma (HCC) and cholangiocarcinoma (CC).
Methods
The records of 100,754 patients with newly diagnosed liver cancer between 2004 and 2013 were obtained from the Taiwan Cancer Registry. The primary outcome measures were overall survival and local recurrence‐free survival. The median follow‐up time was 60 months (29–120 months).
Results
HCC‐CC tended to share some characteristics with HCC, including increased frequency of stage I cases, high individual tumor rates, and similar patterns of viral hepatitis B and hepatitis C infections. In contrast, HCC‐CC showed malignant behavior similar to that of CC, as high‐grade tumor cell differentiation and presentation of jaundice were predominant in HCC‐CC and CC compared with HCC. Overall survival and local recurrence‐free survival rates of HCC‐CC were between HCC and CC rates. The mortality rate of HCC‐CC was 79.2% (HCC, 77.5%; CC, 93.5%) and the local recurrence rate of HCC‐CC was 65.3% (HCC, 74.6%; CC, 88.4%). Surgical treatment was an independent factor for the long‐term prognosis of HCC‐CC, whereas transarterial chemoembolization (TAcE) promoted survival in both surgical and nonsurgical groups.
Conclusion
Our data confirmed that, although it reflects the malignant behavior of CC, HCC‐CC should mainly be characterized as a subtype of HCC. With careful selection of patients, curative resection and TAcE might benefit the survival of patients with HCC‐CC.
Implications for Practice
Combined hepatocellular‐cholangiocarcinoma (HCC‐CC) is a rare cancer that shares demographic characteristics, as well as survival probabilities, with both hepatocellular carcinoma and cholangiocarcinoma. It occurs frequently in patients with hepatitis B virus infection, cirrhotic liver background, and early‐stage disease. Compared with 20% of initial resection rates of its counterparts, HCC‐CC has higher initial resection rate (55%). Although short‐term overall survival is inferior to HCC, its long‐term overall survival is similar with HCC.
Keywords: Liver cancer, Hepatocellular carcinoma, Cholangiocarcinoma, Mixed hepatocellular‐cholangiocarcinoma
Short abstract
This article analyzes population‐based data of a study that aimed to determine the clinical prognosis of and to address the lack of comprehensive information on combined hepatocellular‐cholangiocarcinoma.
Introduction
Combined hepatocellular‐cholangiocarcinoma (HCC‐CC) is a primary liver malignancy that is composed of two major liver cancers: hepatocellular carcinoma (HCC) from hepatocytes and cholangiocarcinoma (CC) from bile duct epithelial cells [1, 2]. This uncommon type of liver cancer has gradually drawn attention as it has become more frequently reported, with the incidence increasing to 14.2% in specific groups [3, 4, 5]. Documented HCC‐CC is considered to be aggressively malignant and is associated with poor oncological outcomes. However, because of the low incidence of HCC‐CC, a detailed understanding of the demographic characteristics and clinical features has thus far been limited [6, 7, 8].
HCC‐CC has been shown to possess distinctive clinicopathologic features and prognoses compared with either HCC or CC alone [9]. Many studies have attempted to describe an association between HCC‐CC and other primary liver malignancies; however, most of these were confined to a single‐center study of patients who underwent curative surgical treatments. Thus the prognosis of HCC‐CC, compared with that of HCC and CC, remains unclear [10, 11, 12]. It has been suggested that HCC‐CC is a variant of classic HCC, based on epithelial characteristics that suggest that HCC‐CC shares similar etiological features with HCC [13, 14]. However, the varying incidence of primary liver malignancies in diverse geographical areas reflects different levels of risk factors in different populations [15, 16, 17]. Nevertheless, this hypothesis remains to be confirmed in a large patient population.
This study sought to analyze population‐based data to determine the clinical prognosis and to address the lack of comprehensive information on HCC‐CC.
Material and Methods
Study Population
The patients in this study were identified from two national databases, the Taiwan Cancer Database (TCDB) and the Taiwan Cancer Registry. Eligible patients were diagnosed with International Classification of Diseases for Oncology (ICD‐O‐3) and C22 (liver cancer) between January 2004 and December 2013. Data were retrieved from the registries to analyze HCC, CC, and HCC‐CC. The Taiwan Cancer Registry and the TCDB are prospective databases organized by the government to facilitate academic research. Both databases (from the Health Promotion Administration) collect 70% of national newly diagnosed patients with liver cancer and provided cancer staging, treatment profiles, and follow‐up status of individual patients [18, 19]. The institutional review board at Taipei City Hospital approved this study and waived informed consent (No: TCHIRB‐10802007‐W).
Cohort Analysis
Data were collected from all patients who were aged less than 85 years at the time of diagnosis. Patients with incomplete or indeterminate pathology reports with respect to HCC‐CC, HCC, or CC (ICD‐O‐3 histology code 817/816/818) were excluded from analysis.
Independent Variables
Relevant demographic, clinical, pathological, and therapeutic data of patients with HCC were retrieved from the TCDB. Possible prognostic variables included patient characteristics (age, gender, comorbidity), disease characteristics (tumor size, tumor number, histological grade of cell differentiation, vascular invasion, cirrhosis), and therapeutic factors. The use of nucleoside/nucleotide analogs indicated patients took lamivudine, entecavir, or telbivudine [20]. Comorbidities included myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, dementia, chronic pulmonary disease, connective tissue disease, peptic ulcer disease, mild liver disease, diabetes without complications, diabetes with chronic complications, hemiplegia (or paraplegia), moderate or severe renal disease, cancer without metastasis, moderate or severe liver disease, metastatic solid tumors, and AIDS [21]. Information on comorbidities (Charlson Comorbidity Index, Deyo version) and nucleoside/nucleotide analog usage was obtained by searching a third database, the National Health Insurance Research Database (NHIRD). Because the insurance system in Taiwan is a single‐payer system, the NHIRD covers health care billing services for all patients; therefore, data related to comorbidity and medication can be accurately retrieved.
Dependent Variables
The primary endpoint was overall survival, defined as the percentage of patients who were alive for a certain period of time after diagnosis of HCC. The secondary endpoint was local recurrence‐free survival, defined as the percentage of patients who remained free of local recurrence for a certain period of time after diagnosis of HCC. Survival status was validated by linkage to a fourth database, the Cause of Death Data (from the Ministry of Health and Welfare). Date of death and cause of death were also available if death occurred.
Statistical Analysis
Demographic, clinical, pathological, and therapeutic variables were reported as percentage or mean ± SD. Categorical variables were compared by χ2 test and continuous variables were compared by one‐way ANOVA. Univariate analysis was conducted for all possible variables by Cox proportional hazards model. Potential prognostic factors identified by univariate analysis (p < .10) were selected for multivariate analysis. A multivariate Cox model for overall survival was conducted to assess the independent prognostic value of nucleoside analogs. Multivariate subgroup analyses including cirrhosis or surgery were also conducted. Hazard ratios (HRs) with 95% confidence intervals (CIs) were reported. Finally, stage‐specific Kaplan‐Meier plots and log‐rank test were performed to calculate survival estimates and significance. We used SAS 9.2 software (SAS Institute Inc., Cary, NC) for the initial database merge process and SPSS Statistics 21 software (IBM Corp., Armonk, NY) for data management and statistical analysis. All p values were two‐sided and the significance level was set at p ≤ .05.
Results
A total of 100,754 consecutive patients with newly diagnosed liver cancer between January 2004 and December 2013 were retrieved from the national database (Figure 1). Following exclusion of those with uncertain pathology and those older than 85 years, a total of 95,149 patients were included in this study. The final cohort consisted of 530 patients with HCC‐CC (0.6%), 88,062 patients with HCC (92.5%), and 6,613 patients with CC (7%). The demographic, clinical, and interventional characteristics are summarized in Table 1. The median age of the HCC‐CC group was 60.0 years (22–85 years), as HCC and CC groups were 64.0 years (3–85 years) and 66.5 years (17–85 years), respectively. The diagnosis of HCC‐CC was less frequent in those >70 years of age (22.5%) compared with HCC (32.9%) and CC (38.9%). HCC‐CC tended to share more characteristics with HCC than with CC. Male patients were predominant in HCC‐CC (73.6%) and HCC (72.7%) but not in CC (54.2%). The most frequent stage in both HCC‐CC and HCC was stage I (41.7% and 36.3%), in contrast to CC, in which stage IV was the most prevalent stage (50.7%). HCC‐CC presented as a single tumor in 51.3% of patients, similar to 52.7% of patients with HCC. In contrast, CC presented as a single tumor in only 31.5% of patients. Additionally, vascular invasion of the tumor was not present in 51.3% of patients with HCC‐CC and 52.7% of patients with HCC, whereas this figure was much lower in patients with CC (31.6%). Cirrhosis was observed in 60.8% of patients with HCC‐CC, 76.5% of patients with HCC, and 24.7% of patients with CC. Jaundice within the 2 months prior to diagnosis was higher in patients with CC (6.5%), followed by patients with HCC‐CC (1.9%) and patients with HCC (0.3%). Patients with HCC‐CC and patients with HCC had similar rates of infection with hepatitis B virus (HBV), hepatitis C virus (HCV), or both. However, most patients with CC (up to 68.2%) had neither HBV nor HCV infection. Nevertheless, HCC‐CC tended to share malignant behavior with CC, as high‐grade tumor cell differentiation was predominant in cases of HCC‐CC (59.1%) and CC (40.7%) compared with HCC (27.1%).
Figure 1.
Flow chart of patient enrollment in this study. After exclusion of patients with incomplete or indeterminate pathology reports and those more than 85 years of age, a total of 95,149 patients were analyzed. Abbreviations: CC, cholangiocarcinoma; HCC, hepatocellular carcinoma; HCC‐CC, mixed hepatocholangiocarcinoma.
Table 1.
Demographic, clinical, and therapeutic characteristics of patients with HCC‐CC, HCC, and CC (2004–2013, n = 95,149)
| Characteristics | All patients, n = 95,149 | HCC‐CC, n = 530 | HCC, n = 88,006 | CC, n = 6,613 | p value |
|---|---|---|---|---|---|
| Age group, yr | <.001 | ||||
| <40.0 | 3,671 (3.9) | 20 (3.8) | 3,521 (4.0) | 130 (2.0) | |
| 40.0–44.9 | 3,881 (4.1) | 34 (6.4) | 3,644 (4.1) | 203 (3.1) | |
| 45.0–54.9 | 16,644 (17.5) | 112 (21.1) | 15,561 (17.7) | 971 (14.7) | |
| 55.0–69.9 | 39,269 (41.3) | 245 (46.2) | 36,285 (41.3) | 2,739 (41.4) | |
| ≥70.0 | 31,622 (33.3) | 119 (22.5) | 28,934 (32.9) | 2,569 (38.9) | |
| Gender | <.001 | ||||
| Male | 67,952 (71.4) | 390 (73.6) | 63,976 (72.7) | 3,586 (54.2) | |
| Female | 27,197 (28.6) | 140 (26.4) | 24,030 (27.3) | 3,027 (45.8) | |
| Comorbidity | <.001 | ||||
| ≤1 | 29,202 (30.7) | 174 (32.8) | 27,406 (31.1) | 1,622 (24.5) | |
| 2, 3, 4 | 23,870 (25.1) | 142 (26.8) | 22,210 (25.2) | 1,518 (23.0) | |
| >4 | 42,021 (44.2) | 214 (40.4) | 38,339 (43.6) | 3,468 (52.4) | |
| Unknown | 56 (0.1) | 0 (0) | 51 (0.1) | 5 (0.1) | |
| Body mass index | <.001 | ||||
| <30.0 kg/m2 | 13,782 (14.5) | 89 (16.8) | 12,514 (14.2) | 1,179 (17.8) | |
| ≥30.0 kg/m2 | 9,631 (10.1) | 104 (19.6) | 8,940 (10.2) | 587 (8.9) | |
| Unknown | 71,736 (75.4) | 337 (63.6) | 66,552 (75.6) | 4,847 (73.3) | |
| Smoking status | <.001 | ||||
| Smoker | 15,013 (15.8) | 99 (18.7) | 13,606 (15.5) | 1,308 (19.8) | |
| Nonsmoker | 10,258 (10.8) | 98 (18.5) | 9,565 (10.9) | 595 (9.0) | |
| Unknown | 69,878 (73.4) | 333 (62.8) | 64,835 (73.7) | 4,710 (71.2) | |
| Drinking habit | <.001 | ||||
| Never alcohol intake | 16,180 (64.0) | 127 (64.8) | 14,636 (63.2) | 1,417 (74.3) | |
| Habitual drink | 3,287 (13.0) | 21 (10.7) | 3,089 (13.3) | 177 (9.3) | |
| Quit or rare | 5,796 (22.9) | 48 (24.5) | 5,436 (23.5) | 312 (16.4) | |
| Unknown | 7,354 | 35 | 6,715 | 604 | |
| Primary tumor size | <.001 | ||||
| <5.0 cm | 35,985 (37.8) | 247 (46.6) | 34,337 (39.0) | 1,401 (21.2) | |
| 5.1–10.0 cm | 16,138 (17.0) | 124 (23.4) | 14,163 (16.1) | 1,851 (28.0) | |
| ≥10.0 cm | 10,745 (11.3) | 71 (13.4) | 9,950 (11.3) | 724 (10.9) | |
| Unknown | 32,281 (33.9) | 88 (16.6) | 29,556 (33.6) | 2,637 (39.9) | |
| Tumor number | <.001 | ||||
| Single | 48,777 (51.3) | 272 (51.3) | 46,420 (52.7) | 2,085 (31.5) | |
| Multiple | 29,346 (30.8) | 102 (19.2) | 27,165 (30.9) | 2,079 (31.4) | |
| Unknown | 17,026 (17.9) | 156 (29.4) | 14,421 (16.4) | 2,449 (37.0) | |
| Surgery as initial treatment | <.001 | ||||
| No | 76,564 (80.5) | 236 (44.5) | 71,083 (80.8) | 5,245 (79.3) | |
| Yes | 18,585 (19.5) | 294 (55.5) | 16,923 (19.2) | 1,368 (20.7) | |
| Resection margin | <.001 | ||||
| Involved | 1,178 (1.2) | 30 (5.7) | 868 (1.0) | 280 (4.2) | |
| Free | 16,375 (17.2) | 263 (49.6) | 15,123 (17.2) | 989 (15.0) | |
| Unknown | 77,596 (81.6) | 237 (44.7) | 72,015 (81.8) | 5,344 (80.8) | |
| Tumor cell grade | <.001 | ||||
| Well/moderately differentiated | 21,298 (22.4) | 143 (27.0) | 19,862 (22.6) | 1,293 (19.6) | |
| Poorly differentiated or undifferentiated | 8486 (8.9) | 207 (39.1) | 7,390 (8.4) | 889 (13.4) | |
| Unknown | 65,365 (68.7) | 180 (34.0) | 60,754 (69.0) | 4,431 (67.0) | |
| Vascular involvement | <.001 | ||||
| No | 48,838 (51.3) | 272 (51.3) | 46,476 (52.8) | 2,090 (31.6) | |
| Yes | 31,802 (33.4) | 124 (23.4) | 28,892 (32.8) | 2,786 (42.1) | |
| Unknown | 14,509 (15.2) | 134 (25.3) | 12,638 (14.4) | 1,737 (26.3) | |
| Stage | <.001 | ||||
| 0 | 7 (0) | 0 (0) | 7 (0) | 0 (0) | |
| I | 28,284 (35.0) | 166 (41.7) | 27,247 (36.3) | 871 (16.5) | |
| II | 16,827 (20.8) | 84 (21.1) | 16,177 (21.5) | 566 (10.7) | |
| III | 24,429 (30.2) | 73 (18.3) | 23,188 (30.9) | 1,168 (22.1) | |
| IV | 11,275 (14.0) | 75 (18.8) | 8,516 (11.3) | 2,684 (50.7) | |
| Perioperative α‐fetoprotein level | <.001 | ||||
| ≤400 ng/ml | 16,251 (72.3) | 139 (76.0) | 14,655 (70.5) | 1,457 (97.0) | |
| 401–6000 ng/ml | 3,532 (15.7) | 32 (17.5) | 3,465 (16.7) | 35 (2.3) | |
| >6000 ng/ml | 2,697 (12.0) | 12 (6.6) | 2,675 (12.9) | 10 (0.7) | |
| Unknown | 72,669 | 347 | 67,211 | 5,111 | |
| Cirrhosis of the liver | <.001 | ||||
| No | 25,875 (27.2) | 208 (39.2) | 20,687 (23.5) | 4,980 (75.3) | |
| Yes | 69,274 (72.8) | 322 (60.8) | 67,319 (76.5) | 1,633 (24.7) | |
| Jaundice | <.001 | ||||
| Nil | 94,479 (99.3) | 520 (98.1) | 87,773 (99.7) | 6,186 (93.5) | |
| Jaundice at (before) initial diagnosis | 670 (0.7) | 10 (1.9) | 233 (0.3) | 427 (6.5) | |
| Hepatitis virus infection | <.001 | ||||
| History of HBV infection | 37,337 (39.2) | 250 (47.2) | 35,755 (40.6) | 1,332 (20.1) | |
| History of HCV infection | 23,918 (25.1) | 78 (14.7) | 23,295 (26.5) | 545 (8.2) | |
| History of HBV and HCV infection | 12,703 (13.4) | 60 (11.3) | 12,416 (14.1) | 227 (3.4) | |
| No history of HBV/HCV infection | 21,191 (22.3) | 142 (26.8) | 16,540 (18.8) | 4,509 (68.2) | |
| Nucleos(t)ide analogue user | <.001 | ||||
| No | 79,606 (83.7) | 414 (78.1) | 73,022 (83.0) | 6,170 (93.3) | |
| Yes | 15,543 (16.3) | 116 (21.9) | 14,984 (17.0) | 443 (6.7) | |
| History of fulminant hepatitis/liver failure | .860 | ||||
| Nil | 93,908 (98.7) | 522 (98.5) | 86,856 (98.7) | 6,530 (98.7) | |
| Yes | 1,241 (1.3) | 8 (1.5) | 1,150 (1.3) | 83 (1.3) | |
| TAcE | <.001 | ||||
| No | 46,263 (48.6) | 300 (56.6) | 41,362 (47.0) | 4,601 (69.6) | |
| Yes | 48,886 (51.4) | 230 (43.4) | 46,644 (53.0) | 2,012 (30.4) | |
| Sorafenib | <.001 | ||||
| Yes | 91,843 (96.5) | 507 (95.7) | 84,728 (96.3) | 6,608 (99.9) | |
| No | 3,306 (3.5) | 23 (4.3) | 3,278 (3.7) | 5 (0.1) | |
| Overall survival | <.001 | ||||
| Dead | 74,836 (78.7) | 420 (79.2) | 68,235 (77.5) | 6,181 (93.5) | |
| Alive | 20,313 (21.3) | 110 (20.8) | 19,771 (22.5) | 432 (6.5) | |
| Cancer‐specific survival | <.001 | ||||
| No | 53,006 (55.7) | 255 (48.1) | 52,225 (59.3) | 526 (8.0) | |
| Yes | 42,143 (44.3) | 275 (51.9) | 35,781 (40.7) | 6,087 (92.0) | |
| Disease‐free survival | <.001 | ||||
| Yes | 17,011 (17.9) | 132 (24.9) | 16,147 (18.3) | 732 (11.1) | |
| No | 78,138 (82.1) | 398 (75.1) | 71,859 (81.7) | 5,881 (88.9) | |
| Local relapse‐free survival | <.001 | ||||
| Yes | 23,269 (24.5) | 184 (34.7) | 22,321 (25.4) | 764 (11.6) | |
| No | 71,880 (75.5) | 346 (65.3) | 65,685 (74.6) | 5,849 (88.4) | |
| Recurrence status | <.001 | ||||
| No | 23,269 (24.5) | 184 (34.7) | 22,321 (25.4) | 764 (11.6) | |
| Local recurrence | 55,960 (58.8) | 307 (57.9) | 50,912 (57.9) | 4,741 (71.7) | |
| Distant metastasis | 698 (0.7) | 6 (1.1) | 609 (0.7) | 83 (1.3) | |
| Persistence | 15,222 (16.0) | 33 (6.2) | 14,164 (16.1) | 1,025 (15.5) |
Values are presented as number (ratio), and characters in the parentheses indicate the reference group. The comorbidity index was codified according to the Charlson Comorbidity Index (Deyo version) I and the stage according to the 6th edition American Joint Committee on Cancer classification on liver cancer. Because of rounding, percentages may not add up to 100%.
Abbreviations: CC, cholangiocarcinoma; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCC‐CC, hepatocellular‐cholangiocarcinoma; HCV, hepatitis C virus; TAcE, transarterial chemoembolization.
Surgery is the first choice in the treatment of HCC‐CC, as 55.5% of patients with HCC‐CC underwent surgical resection (compared with19.2% for HCC and 20.7% for CC). The median duration between diagnosis and initial surgery for HCC‐CC, HCC, and CC was 0.8, 0.9, and 0.7 months, respectively. The prognosis of HCC‐CC tended to fall between that of HCC and CC, with a mortality rate for HCC‐CC at the end of study of 79.2% (HCC, 77.5%; CC, 93.5%). The local recurrence rate of HCC‐CC at the end of study was 65.3% (HCC, 74.6%; CC, 88.4%). The median follow‐up time was 60 months (range, 29–120).
The survival analysis is illustrated in Figure 2 with Kaplan‐Meier survival plots for HCC‐CC, HCC, and CC. The clinical prognosis of HCC‐CC shared similar trends with HCC. The overall survival curve for HCC‐CC fell between those of HCC and CC but tilted toward that of HCC. The local recurrence‐free curve for HCC‐CC fell below the line for HCC after 6 years of follow‐up. Both HCC‐CC curves were consistent with the HCC curves, rather than the CC curves. The 1‐, 3‐, and 5‐year overall survival rates in patients with HCC‐CC were 53.5%, 34.3%, and 25.7%, respectively. The 1‐, 3‐, and 5‐year local recurrence‐free rates in patients with HCC‐CC were 48.2%, 37.7%, and 26.9%, respectively. Moreover, HCC‐CC shared similar risk factors that were associated with clinical prognosis with HCC. Univariate Cox hazard models of overall survival in patients with HCC‐CC indicated that age, tumor cell grade, surgery, TNM stage, tumor size, tumor number, tumor vascular invasion, comorbidity, oral antiviral therapy, and perioperative α‐fetoprotein levels were all significantly associated with survival (Table 2).
Figure 2.
Kaplan‐Meier curves for overall survival and local recurrence‐free survival comparing the three pathological types in patients with hepatocellular carcinoma (HCC), cholangiocarcinoma (CC), and mixed hepatocholangiocarcinoma (HCC‐CC). The prognosis of HCC‐CC is more comparable to that of HCC than that of CC. At the end of study, HCC‐CC demonstrated a 79.2% mortality rate (HCC, 77.5%; CC, 93.5%) and a 65.3% recurrence rate (HCC, 74.6%; CC, 88.4%)
Table 2.
Univariate analysis of prognostic factors for overall survival in patients with primary liver cancer (2004–2013)
| Variables | HCC‐CC, n = 530 | HCC, n = 88,006 | CC, n = 6,613 | |||
|---|---|---|---|---|---|---|
| HR (95% CI) | p value | HR (95% CI) | p value | HR (95% CI) | p value | |
| Age (<45.0 yr) | .005 | <.001 | <.001 | |||
| 45.0–54.9 yr | 0.45 (0.25–0.83) | .010 | 1.00 (0.93–1.07) | .909 | 1.11 (0.87–1.40) | .413 |
| 55.0–64.9 yr | 0.86 (0.52–1.45) | .579 | 0.86 (0.81–0.92) | <.001 | 1.20 (0.96–1.50) | .117 |
| 65.0–74.9 yr | 0.89 (0.47–1.39) | .433 | 0.98 (0.92–1.05) | .585 | 1.34 (1.07–1.68) | .010 |
| ≥75.0 yr | 1.37 (0.74–2.55) | .323 | 1.37 (1.28–1.46) | <.001 | 1.90 (1.51–2.38) | <.001 |
| Gender (male) | .196 | <.001 | .225 | |||
| Female | 0.79 (0.56–1.13) | 0.87 (0.84–0.90) | 0.95 (0.87–1.03) | |||
| Comorbidity (≤1) | <.001 | <.001 | <.001 | |||
| 2, 3, 4 | 1.40 (0.91–2.15) | .127 | 1.06 (1.02–1.11) | .003 | 1.11 (0.98–1.26) | .090 |
| >4 | 2.47 (1.67–3.64) | <.001 | 1.41 (1.36–1.46) | <.001 | 1.43 (1.29–1.59) | <.001 |
| Unspecified | 1.38 (0.66–2.89) | .400 | 0.22 (0.03–1.54) | .127 | ||
| Diabetes mellitus (No) | .931 | .009 | .811 | |||
| Yes | 0.99 (0.72–1.35) | 0.96 (0.94–0.99) | 1.01 (0.93–1.10) | |||
| Hypertension (No) | .308 | .176 | .131 | |||
| Yes | 1.17 (0.86–1.59) | 1.02 (0.99–1.05) | 1.07 (0.98–1.16) | |||
| Body mass index (<30.0 kg/m2) | <.001 | <.001 | <.001 | |||
| ≥30.0 kg/m2 | 1.08 (0.76–1.53) | .662 | 0.78 (0.76–0.81) | <.001 | 0.80 (0.72–0.89) | <.001 |
| Unknown | 2.68 (1.76–4.08) | <.001 | 1.55 (1.49–1.60) | <.001 | 1.16 (1.05–1.28) | .003 |
| Smoking status (nonsmoker) | <.001 | <.001 | .009 | |||
| Smoker | 1.13 (0.81–1.60) | .469 | 1.31 (1.27–1.35) | <.001 | 1.12 (1.01–1.24) | .028 |
| Unknown | 2.68 (1.74–4.12) | <.001 | 1.78 (1.72–1.85) | <.001 | 1.15 (1.04–1.28) | .006 |
| Drinking habit (never alcohol intake) | <.001 | <.001 | .101 | |||
| Habitual drink | 1.02 (0.58–1.80) | .937 | 1.35 (1.29–1.42) | <.001 | 0.97 (0.82–1.14) | .670 |
| Quit or rare | 1.08 (0.73–1.62) | .694 | 1.20 (1.16–1.25) | <.001 | 1.06 (0.93–1.20) | .413 |
| Unknown | 2.65 (1.77–3.98) | <.001 | 1.73 (1.66–1.79) | <.001 | 1.13 (1.02–1.24) | .021 |
| Differentiation (well or moderate) | .001 | <.001 | <.001 | |||
| Poor or none | 1.72 (1.12–2.65) | .014 | 1.42 (1.33–1.52) | <.001 | 1.54 (1.34–1.78) | <.001 |
| Unknown | 2.30 (1.49–3.55) | <.001 | 3.13 (3.00–3.26) | <.001 | 2.39 (2.15–2.66) | <.001 |
| Initial liver resection (No) | <.001 | <.001 | <.001 | |||
| Liver resection | 0.30 (0.22–0.42) | <.001 | 0.24 (0.23–0.26) | <.001 | 0.27 (0.24–0.31) | <.001 |
| Liver transplantation | 0.48 (0.15–1.53) | .215 | 0.20 (0.16–0.26) | <.001 | 0.50 (0.13–2.01) | .330 |
| Initial surgery (no) | <.001 | <.001 | <.001 | |||
| Yes | 0.31 (0.22–0.42) | 0.24 (0.23–0.25) | 0.28 (0.24–0.31) | |||
| Resection margin (positive) | <.001 | <.001 | <.001 | |||
| Negative | 1.05 (0.50–2.20) | .889 | 0.48 (0.41–0.57) | <.001 | 0.53 (0.42–0.67) | <.001 |
| Unspecified | 3.59 (1.73–7.45) | .001 | 2.05 (1.75–2.40) | <.001 | 2.14 (1.74–2.63) | <.001 |
| TNM stage, 6th edition (I) | <.001 | <.001 | <.001 | |||
| II | 2.13 (1.24–3.66) | .006 | 1.72 (1.64–1.81) | <.001 | 2.23 (1.84–2.71) | <.001 |
| III | 5.17 (2.72–9.84) | <.001 | 5.40 (5.18–5.63) | <.001 | 3.52 (2.72–4.55) | <.001 |
| IV | 6.80 (4.06–11.39) | <.001 | 10.14 (9.64–10.65) | <.001 | 4.93 (4.20–5.79) | <.001 |
| TNM stage, 7th edition (I) | <.001 | <.001 | <.001 | |||
| II | 2.07 (1.21–3.55) | .008 | 1.72 (1.64–1.80) | <.001 | 2.23 (1.83–2.70) | <.001 |
| III | 4.91 (2.60–9.28) | <.001 | 5.41 (5.19–5.65) | <.001 | 3.51 (2.72–4.54) | <.001 |
| IV | 6.32 (3.82–10.46) | <.001 | 8.09 (7.73–8.47) | <.001 | 4.92 (4.20–5.77) | <.001 |
| Unspecified | 2.12 (1.38–3.25) | .001 | 3.33 (3.18–3.49) | <.001 | 2.59 (2.19–3.07) | <.001 |
| Tumor size, cm (<3.0) | <.001 | <.001 | <.001 | |||
| 3.0–4.9 | 1.62 (0.96–2.72) | .069 | 1.53 (1.46–1.60) | <.001 | 1.54 (1.26–1.87) | <.001 |
| 5.0–9.9 | 3.41 (2.05–5.67) | <.001 | 2.99 (2.86–3.12) | <.001 | 2.51 (2.09–3.00) | <.001 |
| ≥10.0 | 4.61 (2.62–8.13) | <.001 | 5.62 (5.37–5.88) | <.001 | 2.69 (2.20–3.28) | <.001 |
| Unspecified | 5.25 (2.96–9.31) | <.001 | 3.82 (3.64–4.00) | <.001 | 2.80 (2.34–3.36) | <.001 |
| Tumor size (<5.0 cm) | <.001 | <.001 | <.001 | |||
| 5.0–9.9 cm | 2.56 (1.75–3.73) | <.001 | 2.51 (2.42–2.61) | <.001 | 1.88 (1.68–2.11) | <.001 |
| ≥10.0 cm | 3.46 (2.20–5.44) | <.001 | 4.71 (4.53–4.91) | <.001 | 2.02 (1.75–2.33) | <.001 |
| Unknown | 3.93 (2.48–6.25) | <.001 | 3.21 (3.08–3.34) | <.001 | 2.10 (1.87–2.37) | <.001 |
| Tumor size, cm (<5.0) | <.001 | <.001 | <.001 | |||
| ≥10.0 | 2.82 (2.01–3.96) | <.001 | 3.18 (3.08–3.29) | <.001 | 1.92 (1.72–2.14) | <.001 |
| Unknown | 3.92 (2.47–6.23) | <.001 | 3.18 (3.05–3.32) | <.001 | 2.10 (1.87–2.37) | <.001 |
| Tumor number (single) | <.001 | <.001 | <.001 | |||
| ≥2 | 4.64 (3.02–7.12) | <.001 | 4.83 (4.67–4.99) | <.001 | 2.91 (2.59–3.27) | <.001 |
| Unknown | 1.74 (1.23–2.47) | .002 | 2.87 (2.76–2.99) | <.001 | 2.05 (1.84–2.28) | <.001 |
| Vascular invasion (no) | <.001 | <.001 | <.001 | |||
| Yes | 4.74 (3.14–7.15) | <.001 | 5.07 (4.91–5.24) | <.001 | 2.91 (2.62–3.25) | <.001 |
| Unknown | 1.64 (1.14–2.34) | .007 | 2.57 (2.47–2.68) | <.001 | 1.82 (1.62–2.05) | <.001 |
| TAcE (no) | .107 | .176 | .378 | |||
| Yes | 1.29 (0.95–1.75) | 1.02 (0.99–1.05) | 0.96 (0.88–1.05) | |||
| Cirrhosis (no) | .357 | <.001 | .744 | |||
| Yes | 0.87 (0.64–1.18) | 1.38 (1.34–1.43) | 0.98 (0.89–1.09) | |||
| Nucleos(t)ide analog user (yes) | .010 | <.001 | <.001 | |||
| No | 1.59 (1.12–2.26) | 1.32 (1.27–1.36) | 1.41 (1.24–1.61) | |||
| Nucleos(t)ide analog usage (nil) | .042 | <.001 | <.001 | |||
| NS to NS | 0.16 (0.02–1.12) | .064 | 0.60 (0.52–0.71) | <.001 | 0.50 (0.24–1.04) | .063 |
| NS to NT or NT to NS | 1.07 (0.39–2.90) | .896 | 0.52 (0.46–0.59) | <.001 | 0.14 (0.05–0.45) | .001 |
| NT to NT | 0.66 (0.45–0.95) | .025 | 0.80 (0.77–0.83) | <.001 | 0.75 (0.66–0.86) | <.001 |
| Nucleos(t)ide analog start time (nonuser) | .011 | <.001 | <.001 | |||
| >3 mo before surgery | 0.52 (0.24–1.12) | .094 | 0.37 (0.34–0.41) | <.001 | 0.26 (0.15–0.48) | <.001 |
| Between preoperative 3.1–12 mo after surgery | 0.46 (0.27–0.78) | .004 | 0.47 (0.43–0.50) | <.001 | 0.46 (0.36–0.59) | <.001 |
| Since 12 mo after surgery | 1.06 (0.65–1.74) | .823 | 1.30 (1.25–1.36) | <.001 | 1.00 (0.86–1.16) | .982 |
| Sorafenib (yes) | .304 | <.001 | .838 | |||
| No | 0.76 (0.44–1.29) | 0.68 (0.65–0.71) | 0.87 (0.22–3.46) | |||
| Virus‐related status (never HBV and HCV infection) | <.001 | <.001 | <.001 | |||
| HBV | 0.38 (0.31–0.48) | <.001 | 0.59 (0.58–0.61) | <.001 | 0.69 (0.65–0.74) | <.001 |
| HCV | 0.72 (0.54–0.96) | .027 | 0.62 (0.6–0.63) | <.001 | 0.78 (0.71–0.85) | <.001 |
| HBV and HCV | 0.39 (0.28–0.55) | <.001 | 0.53 (0.51–0.54) | <.001 | 0.55 (0.48–0.64) | <.001 |
| Chemoradiotherapy (nil) | .005 | <.001 | <.001 | |||
| Chemotherapy | 1.65 (0.93–2.94) | .087 | 1.77 (1.69–1.86) | <.001 | 1.08 (0.93–1.24) | .318 |
| Radiotherapy | 1.85 (1.27–2.69) | .001 | 2.76 (2.66–2.86) | <.001 | 1.25 (1.17–1.34) | <.001 |
| Chemoradiotherapy | 0.96 (0.13–6.81) | .964 | 2.23 (1.96–2.54) | <.001 | 0.85 (0.66–1.09) | .201 |
| Perioperative α‐fetoprotein level (≤400 ng/ml) | <.001 | <.001 | <.001 | |||
| 401–6,000 ng/ml | 1.18 (0.74–1.90) | .485 | 2.10 (2.01–2.20) | <.001 | 1.42 (1.00–2.02) | .049 |
| >6,000 ng/ml | 2.10 (1.09–4.05) | .027 | 4.46 (4.26–4.68) | <.001 | 2.17 (1.17–4.05) | .014 |
| Unknown | 2.16 (1.50–3.11) | <.001 | 2.08 (2.01–2.15) | <.001 | 1.19 (1.09–1.29) | <.001 |
Characters in the parentheses indicate the reference group. The comorbidity index was codified according to the Charlson Comorbidity Index (Deyo version) I and the stage according to the 6th edition American Joint Committee on Cancer classification on liver cancer.
Abbreviations: CC, cholangiocarcinoma; CI, confidence interval; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCC‐CC, hepatocellular‐cholangiocarcinoma; HCV, hepatitis C virus; HR, hazard ratio; TAcE, transarterial chemoembolization.
Multivariate Cox hazard models of overall survival are shown in Table 3. The models suggested that age, gender, tumor size, tumor vascular invasion, comorbidity, surgery, and hepatitis virus infection were all significant factors. Significant predictors of local recurrence in HCC‐CC included tumor cell grade, tumor size, surgery, and hepatitis virus infection. In patients with HCC‐CC, women had better overall survival (HR, 0.63; 95% CI, 0.50–0.80; p < .001) but not local recurrence‐free survival (HR, 0.80; 95% CI, 0.62–1.03; p = .089). Similarly, compared with younger patients (<45 years), middle‐aged patients (45–54 years) had better overall survival (HR, 0.62; 95% CI, 0.42–0.92; p = .017) but not local recurrence‐free survival (HR, 0.68; 95% CI, 0.45–1.05; p = .080). Compared with patients without hepatitis infection, patients with HCC‐CC with HBV infection had worse overall survival (HR, 1.66; 95% CI, 1.20–2.30; p = .002) and local recurrence‐free survival (HR, 1.55; 95% CI, 1.07–2.27; p = .022).
Table 3.
Multivariate analysis of prognostic factors for overall survival and local recurrence in hepatocellular‐cholangiocarcinoma (2004–2013)
| Variables | All patients | Surgery patients, overall survival | Nonsurgery patients, overall survival | |||||
|---|---|---|---|---|---|---|---|---|
| Overall survival | Local recurrence | |||||||
| HR 95% CI | p value | HR 95% CI | p value | HR 95% CI | p value | HR 95% CI | p value | |
| Age, yr (<45.0) | .012 | .356 | .009 | .200 | ||||
| 45.0–54.9 | 0.62 (0.42–0.92) | .017 | 0.68 (0.45–1.05) | .080 | 0.72 (0.38–1.36) | .306 | 0.51 (0.30–0.89) | .017 |
| 55.0–64.9 | 1.03 (0.72–1.49) | .858 | 0.93 (0.62–1.39) | .709 | 1.51 (0.82–2.77) | .184 | 0.72 (0.43–1.20) | .207 |
| 65.0–74.9 | 0.84 (0.56–1.26) | .405 | 0.90 (0.58–1.40) | .627 | 1.23 (0.64–2.36) | .531 | 0.64 (0.37–1.12) | .118 |
| ≥75.0 | 0.84 (0.54–1.31) | .439 | 0.86 (0.54–1.39) | .543 | 0.95 (0.43–2.13) | .908 | 0.66 (0.37–1.16) | .144 |
| Gender (male) | <.001 | .089 | .181 | .001 | ||||
| Female | 0.63 (0.50–0.80) | 0.80 (0.62–1.03) | 0.77 (0.53–1.13) | 0.57 (0.41–0.79) | ||||
| Differentiation (well or moderate) | .218 | .234 | .394 | .671 | ||||
| Poor or none | 1.18 (0.91–1.53) | 1.20 (0.89–1.63) | 1.15 (0.84–1.57) | 1.12 (0.66–1.90) | ||||
| Tumor size (<5.0 cm) | .032 | .015 | .012 | .068 | ||||
| 5.0–9.9 cm | 1.30 (0.98–1.72) | .067 | 1.38 (1.00–1.90) | .052 | 1.19 (0.80–1.77) | .392 | 1.77 (1.09–2.86) | .020 |
| ≥10.0 cm | 1.66 (1.17–2.36) | .004 | 1.87 (1.28–2.75) | .001 | 2.11 (1.30–3.44) | .003 | 2.01 (1.14–3.55) | .017 |
| Comorbidity index (≤1) | .042 | .924 | .118 | .326 | ||||
| 2, 3, 4 | 0.99 (0.75–1.29) | .912 | 1.02 (0.76–1.38) | .889 | 0.99 (0.66–1.47) | .944 | 0.96 (0.64–1.44) | .840 |
| >4 | 1.29 (1.01–1.66) | .041 | 1.06 (0.80–1.41) | .700 | 1.40 (0.97–2.03) | .076 | 1.22 (0.85–1.75) | .292 |
| Cirrhosis (no) | .086 | .476 | .035 | .593 | ||||
| Yes | 1.22 (0.97–1.52) | 1.09 (0.86–1.40) | 1.44 (1.03–2.02) | 0.92 (0.67–1.26) | ||||
| Initial surgery (no) | <.001 | <.001 | ||||||
| Yes | 0.37 (0.28–0.48) | 0.29 (0.22–0.40) | ||||||
| Tumor number (single) | .292 | .113 | .929 | .056 | ||||
| ≥2 | 1.35 (0.77–2.36) | 1.66 (0.89–3.10) | 1.05 (0.35–3.17) | 2.06 (0.98–4.31) | ||||
| Vascular invasion (no) | .048 | .250 | .569 | .237 | ||||
| Yes | 1.66 (1.00–2.75) | 1.39 (0.79–2.42) | 1.36 (0.47–3.96) | 1.44 (0.79–2.63) | ||||
| Virus‐related status (HBV infection) | <.001 | .011 | <.001 | <.001 | ||||
| HCV | 1.66 (1.20–2.30) | .002 | 1.55 (1.07–2.27) | .022 | 2.41 (1.54–3.77) | <.001 | 1.17 (0.72–1.89) | .526 |
| HBV and HCV | 1.04 (0.72–1.48) | .852 | 0.91 (0.59–1.40) | .662 | 1.00 (0.61–1.62) | .987 | 1.16 (0.65–2.07) | .620 |
| Never HBV and HCV infection | 2.15 (1.61–2.88) | <.001 | 1.55 (1.12–2.15) | .009 | 2.00 (1.24–3.22) | .005 | 2.28 (1.55–3.34) | <.001 |
| TAcE (no) | .917 | .508 | .002 | <.001 | ||||
| Yes | 1.01 (0.82–1.25) | 1.09 (0.85–1.39) | 1.64 (1.20–2.25) | 0.56 (0.41–0.77) | ||||
| Nucleoside/nucleotide antiviral therapy (yes) | .210 | .279 | .979 | .279 | ||||
| No | 0.83 (0.61–1.11) | 0.83 (0.59–1.16) | 1.01 (0.67–1.51) | 0.77 (0.48–1.24) | ||||
| Chemoradiotherapy (nil) | .508 | .797 | .485 | .564 | ||||
| Chemotherapy | 1.41 (0.76–2.62) | .272 | 1.36 (0.65–2.87) | .419 | 1.25 (0.49–3.22) | .642 | 1.32 (0.58–3.02) | .512 |
| Radiotherapy | 1.21 (0.80–1.83) | .378 | 1.07 (0.69–1.66) | .759 | 1.59 (0.83–3.05) | .158 | 0.78 (0.43–1.43) | .424 |
| Chemoradiotherapy | 2.25 (0.29–17.28) | .437 | 1.86 (0.24–14.50) | .553 | 0.67 (0.03–14.14) | .794 | ||
Comorbidity index codified according to Charlson Comorbidity Index (Deyo version) l.
Abbreviations: CI, confidence interval; HBV, hepatitis B virus; HCV, hepatitis C virus; HR, hazard ratio; TAcE, transarterial chemoembolization.
Surgery benefited the overall survival and local recurrence‐free survival of patients with HCC‐CC (Table 4). Compared with nonsurgical patients, the surgical outcomes were associated with tumor size <5 cm and cirrhosis of the liver. Female nonsurgical patients had better overall survival (HR, 0.57; 95% CI, 0.41–0.79; p = .001) and local recurrence‐free survival (HR, 0.60; 95% CI, 0.43–0.83; p = .002). Moreover, transarterial chemoembolization (TAcE) also improved both overall survival (HR, 0.56; 95% CI, 0.41–0.77; p < .001) and local recurrence‐free survival (HR, 0.62; 95% CI, 0.44–0.86; p = .004) in nonsurgical patients.
Table 4.
Multivariate analysis of prognostic factors for overall and local recurrence‐free survival in patients with hepatocellular‐cholangiocarcinoma (2004–2013, n = 530)
| Variables | Surgery as initial treatment | |||||||
|---|---|---|---|---|---|---|---|---|
| Yes (n = 294) | No (n = 236) | |||||||
| Overall survival | Local recurrence | Overall survival | Local recurrence | |||||
| HR 95% CI | p value | HR 95% CI | p value | HR 95% CI | p value | HR 95% CI | p value | |
| Age (<45.0 yr) | .009 | .389 | .200 | .390 | ||||
| 45.0–54.9 yr | .72 (.38–1.36) | .306 | .72 (.34–1.53) | .391 | .51 (.30–.89) | .017 | .56 (.32–.99) | .045 |
| 55.0–64.9 yr | 1.51 (.82–2.77) | .184 | 1.03 (.50–2.13) | .927 | .72 (.43–1.20) | .207 | .77 (.45–1.32) | .341 |
| 65.0–74.9 yr | 1.23 (.64–2.36) | .531 | 1.20 (.55–2.63) | .642 | .64 (.37–1.12) | .118 | .75 (.42–1.33) | .33 |
| ≥75.0 yr | .95 (.43–2.13) | .908 | .77 (.28–2.09) | .608 | .66 (.37–1.16) | .144 | .75 (.42–1.33) | .32 |
| Gender (male) | .181 | .499 | .001 | .002 | ||||
| Female | .77 (.53–1.13) | 1.17 (.74–1.86) | .57 (.41–.79) | .60 (.43–.83) | ||||
| Differentiation (well or moderate) | .394 | .627 | .671 | .606 | ||||
| Poor or none | 1.15 (.84–1.57) | 1.10 (.74–1.64) | 1.12 (.66–1.90) | 1.15 (.67–1.99) | ||||
| Tumor size (<5.0 cm) | .012 | .001 | .068 | .079 | ||||
| 5.0–9.9 cm | 1.19 (.80–1.77) | .392 | 1.37 (.83–2.25) | .215 | 1.77 (1.09–2.86) | .02 | 1.75 (1.06–2.89) | .030 |
| ≥10.0 cm | 2.11 (1.30–3.44) | .003 | 2.77 (1.57–4.89) | <.001 | 2.01 (1.14–3.55) | .017 | 2.06 (1.15–3.68) | .015 |
| Comorbidity (≤1) | .118 | .529 | .326 | .632 | ||||
| 2, 3, 4 | .99 (.66–1.47) | .944 | 1.30 (.80–2.10) | .286 | .96 (.64–1.44) | .84 | .84 (.56–1.27) | .409 |
| >4 | 1.40 (.97–2.03) | .076 | 1.24 (.76–2.01) | .396 | 1.22 (.85–1.75) | .292 | .98 (.67–1.43) | .923 |
| Cirrhosis (no) | .035 | .193 | .593 | .376 | ||||
| Yes | 1.44 (1.03–2.02) | 1.33 (.87–2.04) | .92 (.67–1.26) | .86 (.62–1.20) | ||||
| Tumor number (single) | .929 | .903 | .056 | .109 | ||||
| ≥2 | 1.05 (.35–3.17) | .93 (.26–3.28) | 2.06 (.98–4.31) | 1.94 (.86–4.37) | ||||
| Vascular invasion (no) | .569 | .551 | .237 | .282 | ||||
| Yes | 1.36 (.47–3.96) | 1.45 (.43–4.94) | 1.44 (.79–2.63) | 1.44 (.74–2.78) | ||||
| Virus‐related status (never HBV and HCV infection) | <0.001 | .004 | <.001 | .064 | ||||
| HBV | 2.41 (1.54–3.77) | <.001 | 2.24 (1.28–3.95) | .005 | 1.17 (.72–1.89) | .526 | 1.13 (.67–1.91) | .645 |
| HCV | 1.00 (.61–1.62) | .987 | .70 (.36–1.36) | .286 | 1.16 (.65–2.07) | .62 | 1.07 (.59–1.95) | .824 |
| HBV and HCV | 2.00 (1.24–3.22) | .005 | 1.53 (.83–2.83) | .171 | 2.28 (1.55–3.34) | <.001 | 1.69 (1.12–2.55) | .013 |
| TAE (no) | .002 | <.001 | <.001 | .004 | ||||
| Yes | 1.64 (1.20–2.25) | 2.12 (1.42–3.15) | .56 (.41–.77) | .62 (.44–.86) | ||||
| NA (yes) | .979 | .945 | .279 | .407 | ||||
| No | 1.01 (.67–1.51) | 1.02 (.61–1.69) | .77 (.48–1.24) | .81 (.50–1.33) | ||||
| Chemoradiotherapy (nil) | .485 | .482 | .564 | .594 | ||||
| Chemotherapy | 1.25 (.49–3.22) | .642 | .72 (.21–2.48) | .607 | 1.32 (.58–3.02) | .512 | 1.44 (.54–3.84) | .466 |
| Radiotherapy | 1.59 (.83–3.05) | .158 | 1.13 (.53–2.40) | .748 | .78 (.43–1.43) | .424 | .82 (.46–1.45) | .49 |
| Chemoradiotherapy | .67 (.03–14.14) | .794 | .11 (.00–2.73) | .178 | ||||
Characters in the parentheses indicate reference group.
Abbreviations: CC, cholangiocarcinoma; CI, confidence interval; HBV, hepatitis B virus; HCC‐CC, hepatocellular‐cholangiocarcinoma; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HR, hazard ratio; NA, nucleos(t)ide analog user; TAE, transarterial chemoembolization.
Discussion
The clinical prognosis of HCC‐CC remained unclear in regard to the rare prevalence of the primary liver malignancy and the complicated histological features of malignant transformation in both liver and bile duct cells [22, 23]. Our study is a representative of the Asian country with the high prevalence of viral hepatitis B, along with liver cirrhosis and liver malignancy. In this analysis, HCC‐CC continued to present as a rare type of primary liver cancer compared with HCC and CC, with a ratio of approximately 1:92:7. The natural history of HCC‐CC tended to be more similar to HCC than to CC, as were important clinical and demographic factors. With a median follow‐up of 60 months, the long‐term prognosis of HCC‐CC corresponded to a mortality rate of 79.2% (HCC, 77.5%; CC, 93.5%) and a recurrence rate of 65.3% (HCC, 74.6%; CC, 88.4%) at the end of the study period.
Previously published data comparing HCC‐CC, HCC, and CC have conflicting results. A recent analysis from the American Cancer Data Base suggested that the long‐term prognosis of patients with HCC‐CC was more similar to patients with CC than to those with HCC [24]. However, our findings confirm the results of other studies showing that patients with HCC‐CC had a natural history and prognosis that was more comparable to HCC than CC [3, 25, 26]. In contrast to patients with HCC‐CC and patients with HCC, the patients with CC had a much lower rate of liver cirrhosis and hepatitis virus infection compared with those with HCC and HCC‐CC. Our data suggest that the different pathophysiology should be clarified between the two groups while considering the higher jaundice rate of HCC‐CC to HCC. The patients with HCC‐CC shared higher rates of liver cirrhosis and hepatitis virus infection with HCC instead of CC. Furthermore, the high resectability of HCC‐CC suggested that the majority of tumors were located away from hepatic hilum. Apart from obstructive jaundice that frequently presented in hilar CC, it is suggested that the liver decompensation‐associated jaundice should be outlined in HCC‐CC, because cirrhotic status and hepatitis virus infection were the significant prognostic factors in overall survival and local recurrence‐free survival. The demographic and histological findings are compatible with our analysis on overall survival and local recurrence‐free survival, which suggest that HCC‐CC should be considered to be a malignant subtype of HCC.
Chronic liver disease and subsequent liver injury represent strong oncogenic factors in the development of HCC‐CC [27]. In this study, conducted in a representative Asian country with a high prevalence of viral hepatitis B, liver cirrhosis, and liver malignancy, virus‐related hepatitis B and the presence of liver cirrhosis were poor prognostic factors for those who underwent surgery. HCC‐CC tended to share an early‐stage presentation and single tumor presentation with HCC in contrast to the frequent presentation of late‐stage CC. Considering the similar cirrhotic status and hepatitis virus infection with HCC, our analysis reflected the fact that HCC‐CC might share the intensive screen protocol with HCC HBV and HCV infection; this may be the result of the high prevalence of viral hepatitis in our population, which contributed to early detection of both HCC‐CC and HCC. The early detection of HCC‐CC might explain the high resection rate and better survival, even though HCC‐CC also shares some characteristics of CC. In contrast, the higher rate of late‐stage CC reflects the urgent need for appropriate tools and protocols for early detection of CC.
The optimal treatment for HCC‐CC is still debated, given that it represents a complex pathophysiological combination of HCC and CC [28, 29, 30]. In our analysis, neither adjuvant chemotherapy nor radiotherapy was associated with improved prognosis. In contrast, our data confirmed that curative resection should remain the first choice in the treatment of HCC‐CC [31]. With a higher rate of hepatitis virus infection, it is interesting that nucleoside analog usage was not associated with a lower risk of recurrence for HCC‐CC, in contrast to the benefit that it provides for HCC [26]. However, the duration of nucleoside analog usage was not determined in our study, which might have interfered with the interpretation of the potential benefits of nucleoside analog usage in patients with HCC‐CC. Moreover, our findings were consistent with previous reports that transarterial chemoembolization was a significant factor in improving overall survival and recurrence‐free survival in HCC‐CC [32]. Our data suggest that HCC‐CC tumors are more similar to HCC tumors than to CC tumors in terms of eligibility for TAcE, which might imply that the treatment of HCC‐CC should be similar to the treatment protocol for HCC.
We acknowledge that our study has several limitations, including difficulties in recruiting a sufficient number of total cases of HCC‐CC. For instance, patients aged more than 85 years were not recruited because of the 80‐year life expectancy of our population. However, the exclusion criteria might have prevented the over estimation of the disease progress when focusing on cancer‐related overall survival. Although the study cases were exclusively analyzed by pathology reports, the diagnosis of HCC‐CC is still difficult and historical controversies remain, despite frequent updates of the key pathological findings [33]. Moreover, our study involved a long‐term observation period of 10 years. Thus, it is possible that missed cases introduced inaccuracies in our analysis. Importantly, the prognosis of CC differs greatly with regard to the lesion site. Although extrahepatic disease was carefully excluded from our analysis, the addition of mixed extrahepatic CC and intrahepatic CC cases might have decreased survival and interfered with the interpretation when comparing HCC‐CC, HCC, and CC. Additionally, our data did not provide good explanations for all prognostic factors, such as gender correlation. The interpretation of these factors is difficult, considering the limited reports regarding this rare primary liver malignancy. Nevertheless, considering that this population‐based data set includes at least 70% of patients with newly diagnosed liver malignancies in Taiwan (an Asian country with particular viral hepatitis profiles and liver disease patterns), the results might be reflective of other populations that share similar demographic characteristics. Although the results of this population‐based study represent a real‐world setting, more specific comparisons on demographic and pathophysiologic characteristics should be considered in the future.
Conclusion
From a molecular genetics perspective, HCC‐CC and CC are different types of primary liver cancer, which highlights the need for accurate treatments specific for each type of liver cancer to optimize the prognosis [34, 35, 36]. Our data confirmed that, although it reflects the malignant behavior of CC, HCC‐CC should mainly be characterized as a subtype of HCC. With careful selection of patients, curative resection and TAcE might benefit the survival of patients with HCC‐CC.
Author Contributions
Conception/design: Yao‐Jen Chang
Provision of study material or patients: Yun‐Jau Chang
Collection and/or assembly of data: Po‐Da Chen, Li‐Ju Chen, Yun‐Jau Chang
Data analysis and interpretation: Li‐Ju Chen, Yun‐Jau Chang
Manuscript writing: Po‐Da Chen, Li‐Ju Chen, Yun‐Jau Chang
Final approval of manuscript: Po‐Da Chen, Li‐Ju Chen, Yao‐Jen Chang, Yun‐Jau Chang
Disclosures
The authors indicated no financial relationships.
Acknowledgments
We gratefully thank the Health and Welfare Data Science Center, Ministry of Health and Welfare Bureau, Taiwan, for providing data for analysis and the Liver Disease Prevention & Treatment Research Foundation for financial support.
Disclosures of potential conflicts of interest may be found at the end of this article.
No part of this article may be reproduced, stored, or transmitted in any form or for any means without the prior permission in writing from the copyright holder. For information on purchasing reprints contact commercialreprints@wiley.com. For permission information contact permissions@wiley.com.
Reference
- 1.Sobin LH. The international histological classification of tumours. Bull World Health Organ 1981;59: 813–819. [PMC free article] [PubMed] [Google Scholar]
- 2.Goodman ZD, Ishak KG, Langloss JM et al. Combined hepatocellular‐cholangiocarcinoma. A histologic and immunohistochemical study. Cancer 1985;55: 124–135. [DOI] [PubMed] [Google Scholar]
- 3.Jarnagin WR, Weber S, Tickoo SK et al. Combined hepatocellular and cholangiocarcinoma: demographic, clinical, and prognostic factors. Cancer 2002;94:2040–2046. [DOI] [PubMed] [Google Scholar]
- 4.Lee CC, Wu CY, Chen JT et al. Comparing combined hepatocellular‐cholangiocarcinoma and cholangiocarcinoma: A clinicopathological study. Hepatogastroenterology. 2002;49:1487–1490. [PubMed] [Google Scholar]
- 5.Koh KC, Lee H, Choi MS et al. Clinicopathologic features and prognosis of combined hepatocellular cholangiocarcinoma. Am J Surg 2005;189:120–125. [DOI] [PubMed] [Google Scholar]
- 6.Lee JH, Chung GE, Yu SJ et al. Long‐term prognosis of combined hepatocellular and cholangiocarcinoma after curative resection comparison with hepatocellular carcinoma and cholangiocarcinoma. J Clin Gastroenterol 2011;45:69–75. [DOI] [PubMed] [Google Scholar]
- 7.Yin X, Zhang BH, Qiu SJ et al. Combined hepatocellular carcinoma and cholangiocarcinoma: Clinical features, treatment modalities, and prognosis. Ann Surg Oncol 2012;19:2869–2876. [DOI] [PubMed] [Google Scholar]
- 8.Gentile D, Donadon M, Lleo A et al. Surgical treatment of hepatocholangiocarcinoma: A systematic review. Liver Cancer 2020;9:15–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Stavraka C, Rush H, Ross P. Combined hepatocellular cholangiocarcinoma (cHCC‐CC): An update of genetics, molecular biology, and therapeutic interventions. J Hepatocell Carcinoma 2019;6:11–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kassahun WT, Hauss J. Management of combined hepatocellular and cholangiocarcinoma. Int J Clin Pract 2008;62:1271–1278. [DOI] [PubMed] [Google Scholar]
- 11.Lee WS, Lee KW, Heo JS et al. Comparison of combined hepatocellular and cholangiocarcinoma with hepatocellular carcinoma and intrahepatic cholangiocarcinoma. Surg Today 2006;36:892–897. [DOI] [PubMed] [Google Scholar]
- 12.Chi M, Mikhitarian K, Shi C et al. Management of combined hepatocellular‐cholangiocarcinoma: A case report and literature review. Gastrointest Cancer Res 2012;5:199–202. [PMC free article] [PubMed] [Google Scholar]
- 13.Llovet JM, Burroughs A, Bruix J. Hepatocellular carcinoma. Lancet 2003;362:1907–1917. [DOI] [PubMed] [Google Scholar]
- 14.Zhou YM, Sui CJ, Zhang XF et al. Influence of cirrhosis on long‐term prognosis after surgery in patients with combined hepatocellular‐cholangiocarcinoma. BMC Gastroenterol 2017;17:25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Brenner H.Long‐term survival rates of cancer patients achieved by the end of the 20th century: A period analysis. Lancet 2002;360:1131–1135. [DOI] [PubMed] [Google Scholar]
- 16.Harlan LC, Hankey BF. The surveillance, epidemiology, and end‐results program database as a resource for conducting descriptive epidemiologic and clinical studies. J Clin Oncol 2003;21:2232–2233. [DOI] [PubMed] [Google Scholar]
- 17.Wang AQ, Zheng YC, Du J et al. Combined hepatocellular cholangiocarcinoma: Controversies to be addressed. World J Gastroenterol 2016;22:4459–4465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Chang YJ, Chung KP, Chang YJ et al. Long‐term survival of patients undergoing liver resection for very large hepatocellular carcinomas. Br J Surg 2016;103:1513–1520. [DOI] [PubMed] [Google Scholar]
- 19.Chang YJ, Chen LJ, Chang YJ et al. Population‐based matching comparison between radiofrequency ablation and percutaneous ethanol or acetic acid injection for hepatocellular carcinoma. Eur J Surg Oncol 2020;46:1703–1710. [DOI] [PubMed] [Google Scholar]
- 20.Wu CY, Chen YJ, Ho HJ et al. Association between nucleoside analogues and risk of hepatitis B virus‐related hepatocellular carcinoma recurrence following liver resection. JAMA 2012;308:1906–1914. [DOI] [PubMed] [Google Scholar]
- 21.Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD‐9‐CM administrative databases. J Clin Epidemiol 1992;45:613–619. [DOI] [PubMed] [Google Scholar]
- 22.Bergquist JR, Groeschl RT, Ivanics T et al. Mixed hepatocellular and cholangiocarcinoma: A rare tumor with a mix of parent phenotypic characteristics. HPB (Oxford) 2016;18:886–892. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Cabuk F, Bassullu N, Turkmen I et al. The prognostic relationship between histopathological and immunohistochemical features of hepatocellular carcinoma, intrahepatic cholangiocarcinoma and mixed type. Pol J Pathol 2020;71:79–86. [DOI] [PubMed] [Google Scholar]
- 24.Spolverato G, Bagante F, Tsilimigras D et al. Management and outcomes among patients with mixed hepatocholangiocellular carcinoma: A population‐based analysis. J Surg Oncol 2019;119:278–287. [DOI] [PubMed] [Google Scholar]
- 25.Chantajitr S, Wilasrusmee C, Lertsitichai P et al. Combined hepatocellular and cholangiocarcinoma: Clinical features and prognostic study in a Thai population. J Hepatobiliary Pancreat Surg 2006;13:537–542. [DOI] [PubMed] [Google Scholar]
- 26.Garancini M, Goffredo P, Pagni F et al. Combined hepatocellular‐cholangiocarcinoma: A population‐level analysis of an uncommon primary liver tumor. Liver Transpl 2014;20:952–959. [DOI] [PubMed] [Google Scholar]
- 27.Chu KJ, Lu CD, Dong H et al. Hepatitis B virus‐related combined hepatocellular‐cholangiocarcinoma: Clinicopathological and prognostic analysis of 390 cases. Eur J Gastroenterol Hepatol 2014;26:192–199. [DOI] [PubMed] [Google Scholar]
- 28.Salimon M, Prieux‐Klotz C, Tougeron D et al. Gemcitabine plus platinum‐based chemotherapy for first‐line treatment of hepatocholangiocarcinoma: An AGEO French multicentre retrospective study. Br J Cancer 2018;118:325–330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Hong TS, Wo JY, Yeap BY et al. Multi‐institutional phase II study of high‐dose hypofractionated proton beam therapy in patients with localized, unresectable hepatocellular carcinoma and intrahepatic cholangiocarcinoma. J Clin Oncol 2016;34:460–468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ejaz A, Cloyd JM, Pawlik TM. Advances in the diagnosis and treatment of patients with intrahepatic cholangiocarcinoma. Ann Surg Oncol. 2020;27:552–560. [DOI] [PubMed] [Google Scholar]
- 31.Wang J, Wang F, Kessinger A. Outcome of combined hepatocellular and cholangiocarcinoma of the liver. J Oncol 2010;2010:917356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Kim JH, Yoon HK, Ko GY et al. Nonresectable combined hepatocellular carcinoma and cholangiocarcinoma: Analysis of the response and prognostic factors after transcatheter arterial chemoembolization. Radiology 2010;255: 270–277. [DOI] [PubMed] [Google Scholar]
- 33.Sciarra A, Park YN, Sempoux C. Updates in the diagnosis of combined hepatocellular‐cholangiocarcinoma. Hum Pathol 2020;96:48–55. [DOI] [PubMed] [Google Scholar]
- 34.Itoyama M, Hata M, Yamanegi K et al. Expression of both hepatocellular carcinoma and cholangiocarcinoma phenotypes in hepatocellular carcinoma and cholangiocarcinoma components in combined hepatocellular and cholangiocarcinoma. Med Mol Morphol 2012;45:7–13. [DOI] [PubMed] [Google Scholar]
- 35.O'Connor K, Walsh JC, Schaeffer DF. Combined hepatocellular‐cholangiocarcinoma (cHCC‐CC): A distinct entity. Ann Hepatol 2014;13:317–322. [PubMed] [Google Scholar]
- 36.Liu ZH, Lian BF, Dong QZ et al. Whole‐exome mutational and transcriptional landscapes of combined hepatocellular cholangiocarcinoma and intrahepatic cholangiocarcinoma reveal molecular diversity. Biochim Biophys Acta Mol Basis Dis 2018; 1864:2360–2368. [DOI] [PubMed] [Google Scholar]