Disease-free survival of 12 patients with combined hepatocellular carcinoma (HCC) and cholangiocarcinoma (CC), 476 patients with HCC, and 25 patients with intrahepatic CC after hepatic resection, from January 1, 1991, to December 31, 2000. The difference among the 3 groups was not significant (P = .44).
Cumulative overall survival of 12 patients with combined hepatocellular carcinoma (HCC) and cholangiocarcinoma (CC), 476 patients with HCC, and 25 patients with intrahepatic CC after hepatic resection, from January 1, 1991, to December 31, 2000. P = .02 for the patients with HCC-CC vs the patients with HCC and P = .38 for the patients with HCC-CC vs the patients with CC.
Liu C, Fan S, Lo C, Ng IO, Lam C, Poon RT, Wong J. Hepatic Resection for Combined Hepatocellular and Cholangiocarcinoma. Arch Surg. 2003;138(1):86-90. doi:10.1001/archsurg.138.1.86
Copyright 2003 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2003
Combined hepatocellular carcinoma (HCC) and cholangiocarcinoma (CC) is an uncommon form of primary liver cancer with features of hepatocellular and biliary epithelial differentiation. The clinical significance of this disease entity is poorly understood. The aim of the present study was to determine the operative and survival outcomes of patients with HCC-CC who underwent hepatic resection.
Tertiary referral center.
The study comprised 12 patients (aged 28-72 years) with HCC-CC (HCC-CC group) who underwent hepatic resection between January 1, 1991, and December 31, 2000. The diagnosis was based on a combination of histological, immunohistochemical, and, if necessary, electron microscopic examinations of the resected specimen.
Main Outcome Measures
Clinicopathological data and operative and survival outcomes of the HCC-CC group were compared with those of 476 patients with HCC (HCC group) and 25 patients with intrahepatic CC (CC group) who underwent hepatic resection during the study period.
Ten patients (83%) in the HCC-CC group underwent major hepatic resection. The operative morbidity and mortality were 17% (2 of 12 patients) and 0%, respectively. The clinicopathological variables of the HCC-CC group resembled those of the HCC group more closely than those of the CC group. The incidence of positive hepatitis B serologic test results in the HCC-CC group (7 [58%] of 12 patients) was intermediate between that of the HCC group (392 [82%] of 476 patients) and that of the CC group (5 [20%] of 25 patients) (P<.001). Underlying chronic liver disease was significantly less common in the CC group (5 [20%] of 25 patients) than in either the HCC-CC group (9 [75%] of 12 patients) or the HCC group (421 [88%] of 476 patients) (P<.001). After hepatic resection, the median disease-free survival of the HCC-CC, HCC, and CC groups was 10, 18, and 24 months, respectively (P = .44). The overall median survival was shortest in the HCC-CC group (17 months) and was not significantly different from that of the CC group (26 months) (P = .38), but was significantly worse than that of the HCC group (52 months) (P = .02).
Although the clinicopathological variables of patients with HCC-CC were most similar to those of patients with HCC, patients with HCC-CC had a significantly worse survival outcome after hepatic resection when compared with patients with HCC. Further studies on postoperative adjuvant therapy and multimodality treatment for recurrent disease are required to prolong the survival of these patients.
COMBINED hepatocellular and cholangiocarcinoma (HCC-CC) is a rare tumor in which dual differentiation toward hepatocytes and bile duct epithelia coexists in the same tumor or in the same liver.1 Hepatocellular carcinoma–CC was first described by Allen and Lisa2 in 1949, and was classified into 3 subtypes. Since then, there have been a few reported series regarding the clinicopathological features of HCC-CC. However, little has been reported on the outcome of these patients after curative hepatic resection, and the reported results of resection have been inconsistent. Nakamura et al3 reported a median survival of 48 months after hepatic resection for 6 patients with HCC-CC, whereas a median survival of less than 12 months was reported by others.4 It was previously suggested that the outcomes of hepatic resection of patients with HCC-CC were not different from those of patients with HCC.4,5 Patients with HCC-CC were also described as not associated with chronic liver disease, resulting in a higher resectability rate compared with patients with HCC.5 In the present study, we describe a series of 12 patients with HCC-CC (HCC-CC group) who underwent hepatic resection, and compare their clinicopathological variables and operative outcomes with those of 476 patients with HCC (HCC group) and those of 25 patients with intrahepatic CC (CC group).
Between January 1, 1991, and December 31, 2000, 520 patients with primary liver cancer underwent hepatic resection at the Department of Surgery, Queen Mary Hospital. Among them, 12 patients (2%) were diagnosed as having hepatocellular and biliary epithelium differentiation based on a combination of histological, immunohistochemical,6,7 and, when necessary, electron microscopic examinations of the resected specimen.4 Hepatocellular differentiation was recognized by the presence of a trabecular pattern, polygonal cells with abundant eosinophilic cytoplasm, bile production and intracytoplasmic hyaline globules, and staining positive for α-fetoprotein. Features of biliary epithelial differentiation included a glandular architecture that was composed of small cuboidal to low columnar cells with round vesicular nuclei with no prominent nucleoli, mucin production, and an accompanying fibrous stroma within the tumor.
All patients diagnosed as having primary liver cancer during the study period underwent a preoperative examination for hepatic resection using a standard protocol. Preoperative investigations included blood biochemistry tests, a determination of the serum α-fetoprotein level, the attainment of a chest x-ray film, percutaneous ultrasonography, a computed tomographic scan of the thorax and abdomen, and an indocyanine green clearance test. Hepatic resection followed the standard technique described previously,8 and ultrasonic dissection was used for parenchymal transection.9 In selected patients with a large tumor situated at the right lobe of the liver, the technique of anterior approach hepatic resection was used.10 All patients received the same perioperative care by the same team of surgeons, and were nursed in the intensive care unit during the early postoperative course. All intraoperative complications and postoperative morbidities were recorded prospectively. Hospital mortality was defined as death during the same period of hospitalization as for hepatic resection. After hepatic resection, patients were followed up with a serum α-fetoprotein assay, ultrasonography, and the attainment of a chest x-ray film every 3 months for the first 2 years and then once a year; and a computed tomographic scan of the abdomen every 6 months. Disease-free survival time was calculated from the date of hepatic resection to the date when recurrence was diagnosed.
The clinical data of all patients were recorded prospectively in a computerized database by a single research assistant. The clinicopathological data and the operative and survival outcomes of the HCC-CC group were compared with those of the HCC and the CC groups, who underwent hepatic resection during the study period. Tumor staging was performed in accordance with the Union Internationale Contre le Cancer TNM staging system.11 Statistical analysis was performed by the χ2 test or the Fisher exact test to compare discrete variables. The Mann-Whitney test was used to compare continuous variables. Survival analysis was estimated by the Kaplan-Meier survival method and compared by the log-rank test. Statistical analysis was performed with Statistical Product and Service Solutions computer software for Windows (SPSS Inc, Chicago, Ill). P<.05 was considered statistically significant.
During the 10-year study period, 12 patients with HCC-CC underwent hepatic resection. The median age of these patients was 60 years, and there were 8 men (67%) and 4 women (33%). All the patients were Chinese. The clinical data of these 12 patients and of the 476 patients with HCC and the 25 patients with CC who underwent hepatic resection during the same study period are presented in Table 1. The incidence of positive hepatitis B serologic test results in the HCC-CC group (58%) was intermediate between that of the HCC group (82%) and that of the CC group (20%) (P<.001). Other clinical variables of the HCC-CC group resembled those of the HCC group more closely than those of the CC group (Table 1).
The extent of hepatic resection is listed in Table 2.12 Ten patients (83%) in the HCC-CC group underwent major hepatic resection, with excision of 3 or more Couinaud segments.13 Hepatic resections performed in patients with HCC-CC were similar in magnitude to those performed in patients with HCC or CC, in which major hepatic resection was performed in 309 (65%) and 16 (64%) of the patients, respectively (P = .42). The median operative blood loss in the HCC-CC group was 1.8 L, and half of the patients did not receive any intraoperative transfusion (Table 3). There was no hospital mortality in the HCC-CC group, while the hospital mortality in the HCC and CC groups was 5% and 4%, respectively (P = .73). Histological examination of the resected specimens showed that the TNM tumor staging of all 3 groups was comparable (Table 4). Nine patients (75%) in the HCC-CC group had evidence of chronic liver disease, including chronic active hepatitis (n = 6) and cirrhosis (n = 3). The incidence was comparable to that of the HCC group (421 [88%] of 476 patients). By contrast, chronic liver disease was significantly less common in the CC group (5 [20%] of 25 patients) (P<.001). Multifocal disease was significantly (P = .01) more frequently seen in the HCC-CC group than in the HCC and the CC groups, while the incidence of lymphovascular permeation was comparable in all 3 groups.
The median disease-free survival following hepatic resection of the HCC-CC, HCC, and CC groups was 10, 18, and 24 months, respectively (P = .44) (Figure 1). Eleven patients (92%) in the HCC-CC group had recurrent disease on follow-up. All of them had recurrences in the liver remnant, and none had documented lymph node metastasis. Two of them also had an extrahepatic recurrence in the peritoneal cavity, while another patient had lung and brain metastases. Five patients underwent transarterial chemoembolization (TACE) for the treatment of recurrent disease. The median number of treatment sessions was 1 (range, 1-3). None of the patients underwent a second resection of the recurrent tumor. Recurrence also occurred in 289 patients (61%) in the HCC group. Of these, intrahepatic recurrences were found in 234 patients. Multimodal treatment for recurrent disease was provided for these patients,14 and 163 (70%) of them underwent TACE.15 The median number of treatment sessions was 4 (range, 1-20). The overall median survival of the HCC-CC group was 17 months, which was significantly worse than that of the HCC group (52 months) (P = .02), but was not statistically different from that of the CC group (26 months) (P = .38) (Figure 2).
Primary liver cancer that comprises elements of HCC and CC is only encountered infrequently. Therefore, the clinical experience of many patients from a single center is limited. Few reports5,16 have described, in detail, the surgical treatment and the outcomes after curative hepatic resection for HCC-CC. The incidence of HCC-CC in reported series of primary liver cancer varies considerably—from 2.4% reported by Goodman et al17 to 14.2% from the series of Allen and Lisa.2 In the estimation of the incidence of HCC-CC, it was important to include only patients who had liver cancer and underwent hepatic resection, because HCC-CC could be missed or misdiagnosed as HCC in patients who underwent a biopsy only. In the present study, 520 patients who underwent hepatic resection for primary liver cancer were examined, and 12 patients with HCC-CC were identified, resulting in an incidence of 2%.
The diagnosis of HCC-CC depends on the demonstration of dual differentiation of hepatocellular and biliary epithelial features. A pseudoglandular pattern that can be seen in patients with HCC should not be taken as a component of CC. Detection of mucin production is an important factor in diagnosing the biliary component in patients with HCC-CC. In patients in whom mucin production cannot be demonstrated, but a strong suggestion of a CC component is present, electron microscopic studies could be useful.4
The cell origin of the tumor cells of HCC-CC remains unclear. As for the histogenesis, 3 possibilities have been proposed: (1) double cancer; (2) the cancer initially developed from either the hepatocytes or the bile duct epithelium, with subsequent differentiation to other components; and (3) the cancer developed in the intermediate cells that subsequently differentiate into HCC and CC.16 Goodman et al17 classified HCC-CC into 3 groups, namely, collision tumors (type 1), transition tumors (type 2), and fibrolamellar tumors (type 3). It was also suggested that type 1 and 2 tumors were clinically more like HCC than CC and should, therefore, be considered as variants of HCC rather than true intermediate types between the usual HCC and CC. In the present study, the clinical features of patients with HCC-CC closely resembled those of patients with HCC. However, the survival outcomes after hepatic resection were significantly different between the 2 groups of patients, suggesting that the histogenesis of HCC-CC cannot be explained by only one hypothesis.
It was previously suggested that coexisting chronic liver diseases were rarely seen in patients with HCC-CC. Jarnagin et al5 attributed the high resectability rate (78%) of their patients with HCC-CC to the fact that none of them had underlying liver cirrhosis. In the present study, 9 (75%) of the 12 patients with HCC-CC had underlying chronic liver disease and 3 (25%) had liver cirrhosis. The reasons for the difference in the observation are unclear, but might reflect the difference in the cause of liver cancer between Western and Eastern societies and the difference in the selection criteria of patients for hepatic resection. More aggressive approaches for hepatic resection in patients with primary liver cancer who have underlying chronic liver disease may result in an apparently higher incidence of coexisting liver cirrhosis in patients with HCC-CC.
The present study shows that resection of HCC-CC results in a significantly worse survival outcome compared with hepatic resection of HCC. The difference in the disease-free survival between the 2 groups of patients did not reach statistical significance, and is most likely related to the small number of patients in the HCC-CC group. The overall survival was significantly worse in the HCC-CC group. The better overall survival in patients with HCC might be related to the treatment rendered after disease recurrence. Intrahepatic recurrence was the most common site of recurrent disease in most patients with HCC, and all of these patients underwent additional treatment with TACE. Transarterial chemoembolization was an effective treatment modality for HCC in patients with unresectable disease and postresection recurrence.15,18 Although patients with HCC-CC who had recurrence in the liver also underwent TACE, the tumors were usually much less vascular than HCC on angiography, resulting in poor uptake of chemotherapeutic agents. Embolization was also less effective in the treatment of the disease than of HCC, because the tumors are much less vascular. In the future, local ablative therapy, such as radiofrequency ablation, should be considered in patients with an intrahepatic recurrence of HCC-CC to prolong survival of this group.19- 21
In conclusion, patients with HCC-CC have a significantly worse survival outcome after hepatic resection when compared with patients with HCC. Further studies on postoperative adjuvant therapy and multimodality treatment for recurrent disease are required to prolong the survival of these patients.
Corresponding author and reprints: Chi-Leung Liu, MS, FRCS(Edin), Department of Surgery, University of Hong Kong Medical Centre, Queen Mary Hospital, 102 Pokfulam Rd, Hong Kong (e-mail: email@example.com).
Accepted for publication August 24, 2002.