Background
Combined hepatocellular-cholangiocarcinoma (cHCC-CC) is a distinct type of primary hepatic cancer distributed across all races in the world. cHCC-CC was first reported by Allen in 1949 [
1] and comprises ingredients of hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC) [
2]. The histological features of cHCC-CC include hepatocytes that produce bile and bile duct epithelial cells that produce mucin [
3].
At present, the gold standard of cHCC-CC diagnosis is histopathological examination, and the histopathological findings in biopsy specimen are that two progenitor cells of monoclonal origin are differentiated into hepatocytes and bile duct cells, which are closely integrated in the same tumor cell [
4‐
6]. However, preoperative diagnosis of cHCC-CC is hard because of its particular imaging characteristics with similar features of HCC and ICC [
7]. Although there may be a few features implying cHCC-CC on enhanced computer tomography and enhanced magnetic resonance imaging, previous studies have indicated that cHCC-CC are often misdiagnosed as either HCC or ICC [
8,
9]. In addition, Portolani N showed that accuracy of cHCC-CC diagnosis in preoperative percutaneous liver biopsy is 11.1%, and the remaining cases are misdiagnosed as metastatic carcinoma, HCC, or ICC [
10]. Currently, there is no unified standard for the diagnosis and treatment of cHCC-CC. Treatment of cHCC-CC mainly includes hepatic resection, liver transplantation (LT), transarterial chemoembolization (TACE), radiofrequency ablation (RAF), and percutaneous ethanol injection.
Unlike those of HCC or ICC, there is limited understanding of the clinical and prognostic features of cHCC-CC. Previous publications have been largely based on case series and retrospective studies which involve small research object populations from single institutions with finite statistical power [
11‐
20]; thus, there is a lack of effective clinical evidence to help standardize the diagnosis and treatment and improve the prognosis of patients with cHCC-CC. In addition, most of these studies are histopathological in nature, and the demographic and prognostic features of the disease remain unclear. Despite improvements in and utilization of imaging technologies, operation method, combined adjuvant therapy, the incidence, and survival rates of cHCC-CC remain unknown. Therefore, the Surveillance, Epidemiology, and End Results (SEER)-18 database (2000–2014) was used to explore trends in the incidence and incidence-based (IB) mortality and survival for cHCC-CC. We also examined the independent predictors of mortality.
Discussion
cHCC-CC is a rare malignancy from the liver with distinctive clinicopathological and prognostic features from those of HCC and ICC. The reported frequency of cHCC-CC among cases with primary liver carcinoma varies widely from 0.4 to 14.2% [
1,
2,
11,
25‐
28]. Our study aimed to explore the trends in cHCC-CC incidence, IB mortality, and survival. We also examined the independent predictors of mortality.
Based on our analysis, cHCC-CC incidence continued to increase in the US population during our study period, and the rate of increase in incidence was steady on the whole, at approximately 4.59% per year. The steady increase in the incidence of cHCC-CC might manifest that main prevention strategies for cHCC-CC have not improved much in recent years. The current analysis on the APC of cHCC-CC incidence provides an important support for projecting and assessing cancer control programs. In addition, IB mortality for cHCC-CC displays similar changes. The steady increase in IB mortality may imply that the treatment of cHCC-CC has not improved or progressed much in recent years. More attention should be thus paid to this malignant tumor type, and further studies are warranted to explore therapies to benefit these patients.
In our analysis, males (67.4%) were more commonly affected with cHCC-CC than females, and the patients were apt to be older than 60 years when they were diagnosed (mean age at diagnosis = 62.9 years). In previous studies, cHCC-CC was indicated to be analogous to HCC with respect to demographic and clinical characteristics, with an apparent male dominance and an average age at presentation between 50 and 60 years [
17,
29‐
31]. cHCC-CC had a medium prognosis with respect to HCC and ICC, accord with the results of previous researches [
31‐
34]. In our analysis, the 1-, 3-, and 5-year survival rates were 43.4%, 21.5%, and 17.1%, respectively. Based on the literature, the survival rates for HCC are 49%, 19%, and < 10% at 1, 3, and 5 years, respectively [
35], and those for ICC are 38.9%,18.5%, and 14.6% at 1, 3, and 5 years, respectively [
36,
37].
Several studies on cHCC-CC have been centered on Asian populations because of the higher incidence of primary liver cancer in Asian [
18,
29,
30]. In our study, African-American race was a negative prognostic factor relative to other races (American-Indian/Alaska Native, Asian/Pacific Islander). The results could be explained by differences in risk factors or prevention awareness between these racial groups. Therefore, when screening for liver malignancy in African-American populations, we should be cautious of cHCC-CC rather than being confined to HCC and ICC. These measures may also lead to a shift in health care resources associated with cHCC-CC. The prevention and control measures of HCC may also benefit to cHCC-CC patients. To the best of our knowledge, Asian populations were at a high risk of HCC; thus, national public health department might implement more rigorous and elaborate prevention strategies in these populations. Hence, the prevention and control measures of cHCC-CC may be extrapolated from those of HCC to determine strategies to prevent and control cHCC-CC in general.
In our cohort, surgical treatment (LT and resection) at an early stage generally resulted in a better prognosis than interventions at an advanced stage. The best mean survival was observed among patients undergoing LT for localized disease (86.7 months; 95% CI 66.6–106.9). In addition, patients undergoing resection for localized cHCC-CC had favorable mean survival (68.1 months; 95% CI 49.9–86.3). However, as many patients were already at an advanced stage at diagnosis, they were not candidates for LT or resection. Further investigations and efforts should thus be focused on early diagnosis and treatment. In addition, comprehensive treatment of advanced disease is worth exploring.
Based on our results, tumor size > 5 cm, and regionalized and distant SEER stages were independently associated with an increased risk of death, consistent with the results of previous studies [
31‐
34]. Therefore, these factors indicative of unfavorable prognosis might provide support for the decision of clinical treatment program and risk assessment.
Using data from the SEER database between 1988 and 2009, Mattia Garancini found that in comparison with HCC and ICC, cHCC-CC displays medium demographic, clinical, and survival characteristics, and factors affecting the prognosis of cHCC-CC patients are usually limited to sources of apparent survival benefit [
32]. We carried out a more comprehensive assessment of cHCC-CC, such as the analysis of incidence and IB mortality trends by APC and taking advantage of sex and stage to examine trends. However, there are many limitations to this study. Many of the limitations of population-based datasets, such as selection bias and missing information, have been previously described [
38]. Furthermore, the SEER database does not provide information on surgical volume and patient comorbidities.
Conclusions
The diagnosis and management of cHCC-CC are changing in the US population, with adverse mortality and survival outcomes. We identified a steady increase in the overall incidence of cHCC-CC, which indicates that main prevention strategies for cHCC-CC have not improved much in recent years and that cHCC-CC needs to be taken seriously. Our data may thus be helpful for designing and evaluating cancer control programs. The IB mortality data suggest that the treatment of cHCC-CC has not progressed much in recent years, and further studies are warranted to explore therapies to benefit these patients. Additionally, surgical treatment (LT and resection) at an early stage generally results in good prognosis, suggesting that further investigations and efforts should be focused on early diagnosis and treatment and that comprehensive treatment of advanced diseases is worth exploring.