The current study was performed to clarify whether the presence of residual carcinoma in situ at ductal resection margins differs prognostically from residual invasive ductal lesions in patients undergoing surgical resection for extrahepatic cholangiocarcinoma.
A retrospective analysis of 84 patients with extrahepatic cholangiocarcinoma who underwent surgical resection was conducted. The ductal resection margin status was classified as negative (n = 64 patients), positive with carcinoma in situ (n = 11 patients), or positive with invasive carcinoma (n = 9 patients). The median follow-up period was 105 months.
Ductal margin status was found to be a strong independent prognostic factor by both univariate (P = 0.0002) and multivariate (P = 0.0039) analyses. The outcome after surgical resection was comparable between patients with negative ductal margins (median survival time of 45 months; cumulative 10-year survival rate of 40%) and those with positive ductal margins with carcinoma in situ (median survival time of 99 months; cumulative 10-year survival rate of 23%; P = 0.4742). In patients with positive ductal margins, the outcome was found to be significantly better in patients with residual carcinoma in situ than in those with residual invasive carcinoma (median survival time of 21 months; cumulative 5-year survival rate of 0%; P = 0.0003). Of 11 patients with residual carcinoma in situ, 4 died of tumor recurrence and the initial site of the disease recurrence was local. All 9 patients with residual invasive carcinoma died of disease recurrence (local recurrence with or without distant metastases) within 40 months after surgical resection.
Surgical resection is the treatment of choice for patients with resectable extrahepatic cholangiocarcinoma.1–4 Surgical resection margin status is a critical prognostic factor,5, 6 and the prognosis of patients with positive ductal margins has been generally considered to be poor.4, 7–17 Several authors, however, have suggested that a small proportion of patients with positive surgical resection margins may survive > 5 years.7–12 The reported incidence of a positive surgical resection margin in patients who have undergone a resection with curative intent can range from 9% to as high as 74%,7–18 suggesting the need for a closer investigation of this association with poor prognosis. In our experience, there have been some long-term survivors with residual carcinoma in situ present at bile duct stumps, although patients with a residual invasive ductal lesion always faced a poor outcome. Therefore, we hypothesized that carcinoma in situ at ductal resection margins is prognostically different from residual invasive tumor.
In the current study, the group of patients with positive ductal margins was subdivided into those with carcinoma in situ and those with invasive carcinoma. This study was intended to clarify whether residual carcinoma in situ at ductal resection margins differs prognostically from residual invasive ductal disease in patients with extrahepatic cholangiocarcinoma.
MATERIALS AND METHODS
A total of 87 patients with carcinoma arising from the extrahepatic bile ducts (extrahepatic cholangiocarcinoma) underwent surgical resection with curative intent in the study department during the 15-year period between January 1988 and December 2002. Patients with carcinoma arising in the cystic duct were included according to the TNM staging system.5 Three patients who also had gallbladder carcinoma were excluded. The remaining 84 patients formed the basis of the current retrospective study, which included 33 women and 51 men with a median age of 66 years (range, 35–86 years). All patients were Japanese. Jaundice was noted at the time of presentation in 69 patients, 65 of whom received percutaneous (n = 62) or endoscopic (n = 3) biliary drainage preoperatively, 2 of whom underwent surgical cholecystostomy before definitive surgery, and 2 of whom underwent no preoperative biliary decompression procedure.
Primary Tumor Location
The location of the primary tumor was classified as hilar or nonhilar according to the classification of Bismuth et al.,13, 19 in which hilar cholangiocarcinoma was defined as a tumor involving the primary ductal confluence with or without extension into more proximal bile ducts. In the current series, there were 41 patients with hilar cholangiocarcinoma and 43 with nonhilar tumors.
Surgical Resection Procedures
Surgical resection procedures depended on the location of the primary tumor. Forty-eight patients underwent a hepatectomy with bile duct resection, 19 underwent a Whipple procedure or a pylorus-preserving pancreaticoduodenectomy, 9 underwent a bile duct resection, and a combined hepatectomy and pancreaticoduodenectomy was performed in 8 patients. Nine patients also underwent a combined resection of contiguous tissues, which included the portal vein (seven patients), the hepatic artery (one patient), the inferior vena cava (one patient), the stomach (two patients), the duodenum (one patient), and the transverse colon (two patients).
All 84 resected patients also underwent dissection of the regional lymph nodes, which included the pericholedocal, retroportal, cystic duct, right celiac, posterosuperior pancreaticoduodenal, and hepatic artery node groups.5, 20–22 In patients undergoing a pancreaticoduodenectomy, the right portion of the superior mesenteric lymph nodes22 also was dissected. Fifty-six patients with suspected regional lymph node disease underwent dissection of the paraaortic lymph nodes21, 22 in addition to the regional lymphadenectomy, and 4 had involvement of the paraaortic lymph nodes, classified as pathologic distant metastasis (pM1).5
Complications occurred after resection in 58 patients (69%). Biliary fistula was the most common (n = 20), followed by pancreatic fistula (n = 7), intraabdominal abscess (n = 7), intraabdominal bleeding (n = 7), cholangitis (n = 5), hepatic insufficiency (n = 3), pneumonia (n = 3), chyloascites (n = 2), liver abscess (n = 1), methicillin-resistant Staphylococcus aureus enterocolitis (n = 1), pseudomembranous enterocolitis (n = 1), and ileus (n = 1). Three patients died within 30 days after surgical resection, giving a surgical mortality rate of 3.6%.
Adjuvant treatment after surgical resection was administered at the discretion of the individual surgeon. Twenty-one patients received oral administration of 5-fluorouracil or its derivatives within approximately 1 year. Four patients with positive ductal resection margins received adjuvant local radiotherapy with (two patients) or without (two patients) systemic chemotherapy.
Patient Follow-Up after Resection
Patients undergoing a surgical resection were followed regularly in outpatient clinics every 1–6 months, with a median follow-up period of 105 months (range, 3–185 months). At the time of disease status assessment, 38 patients had died of tumor recurrence and 3 patients had died of suppurative cholangitis. Nine patients had died of other causes with no evidence of tumor recurrence. Three patients were alive with recurrent disease and the remaining 31 patients were alive without disease.
Resected specimens were submitted to the Department of Surgical Pathology in our hospital for histologic evaluation, in which experienced pathologists independently examined all the specimens without knowledge of the clinical details. Histologic findings were described according to the TNM staging system.5 The extent of the primary tumor was determined through the examination of multiple sections (median of 10 sections; range, 2–49 sections) of the entire lesion in each resected specimen. Histopathologic type was classified into papillary or nonpapillary.5, 23, 24 Histologic grade was determined based on areas of the tumor having the highest grade.5 Adenocarcinoma was identified as the primary tumor in 81 patients (papillary in 10 patients and nonpapillary in 71 patients) and adenosquamous carcinoma was identified in 3 patients (adenosquamous carcinoma was treated as a nonpapillary tumor in the current study). A total of 2047 lymph nodes taken from the 84 patients who underwent a surgical resection (median of 22 lymph nodes per patient) were examined histologically for metastases. A representative section measuring 3-μm thick was cut from each lymph node and stained with hematoxylin and eosin.
Histologic Assessment of the Ductal Resection Margin
Ductal resection margin status was assessed histologically by either frozen-section examination of ductal stumps (n = 71) or pathologic examination of resected specimens (n = 13). Margin status was classified into negative, positive with carcinoma in situ, or positive with invasive carcinoma (Fig. 1). A positive ductal margin with both carcinoma in situ and invasive carcinoma was classified as positive with invasive carcinoma.
In the current study, severe dysplasia was included in the category of carcinoma in situ, because it is usually difficult or even impossible to distinguish between these two epithelial lesions, as described by Albores-Saavedra et al.23, 24 Indeed, a distinction between severe dysplasia and carcinoma in situ was impossible in one of our patients. Mild to moderate dysplasia that was inflammatory, regenerative, or hyperplastic in nature was excluded. Carcinoma in situ or severe dysplastic epithelium may extend into intramural glands such as the sacculi of Beale or metaplastic pyloric-type glands. Discrimination between such intramural epithelial lesions (pseudoinvasion) and invasive carcinoma was made on the basis of the histologic criteria proposed by Albores-Saavedra et al.23
To elucidate factors influencing long-term survival after surgical resection, 15 conventional variables (Table 1)4, 5, 25, 26 together with ductal resection margin status were entered into univariate and multivariate analyses.
Table 1. Clinicopathologic Characteristics of 84 Patients Undergoing Surgical Resection for Extrahepatic Cholangiocarcinoma According to Ductal Margin Status
Medical records and survival data were obtained for all 84 patients. The causes of death were determined from the medical records. The follow-up period was defined as the interval between the date of surgical resection and that of the last follow-up. Deaths from cancer and/or suppurative cholangitis were treated as failure cases and those from other causes as censored cases. Survival curves were constructed using the Kaplan–Meier method and differences in survival were evaluated using the log-rank test. The Cox proportional hazards model was used to identify factors that were independently associated with postresectional survival. In this model, a stepwise selection is used for variable selection with entry and removal limits of P < 0.1 and P > 0.15, respectively. The stability of this model was confirmed using a step-backward and step-forward fitting procedure; the variables identified as having an independent effect on survival were identical with both procedures. Clinical features and pathologic tumor-related factors were compared among the patient groups according to ductal margin status using the Pearson chi-square test or Fisher exact test. All statistical evaluations were performed using the SPSS 11.5J software package (SPSS Japan Inc., Tokyo, Japan). All tests were two-sided and a P value < 0.05 was considered statistically significant.
Overall survival rates after resection were 42% at 5 years and 29% at 10 years. The ductal resection margin was negative in 64 patients, positive with carcinoma in situ in 11 patients, and positive with invasive carcinoma in 9 patients. Clinicopathologic characteristics were comparable between the three groups with the exception that perineural invasion was more frequent in patients with a positive margin with invasive carcinoma (P = 0.027) (Table 1).
Factors Influencing Long-Term Survival after Surgical Resection
Univariate analysis revealed that ductal margin status, pathologic lymph node metastasis (pN) classification, pathologic distant metastasis (pM) classification, size of the primary tumor, and histologic grade were significantly associated with long-term survival (Table 2). These five variables were entered into multivariate analysis, and pM classification, ductal margin status, size of the primary tumor, and histologic grade remained independently associated with long-term survival (Table 2).
Table 2. Factors Influencing Long-Term Survival after Surgical Resection
Impact of Ductal Margin Status on Long-Term Survival after Surgical Resection
The outcome after surgical resection was found to be better in patients with a negative ductal margin than in those with a positive ductal margin, although the difference was not statistically significant (P = 0.1516) (Fig. 2). When the positive ductal margin group was subdivided into those with carcinoma in situ and those with invasive carcinoma, there were further significant differences observed. The outcome after resection was comparable between patients with a negative margin and those with a positive margin with carcinoma in situ (P = 0.4742) (Fig. 3A). However, the outcome was significantly better in patients with a positive margin with carcinoma in situ than in those with a positive margin with invasive carcinoma (P = 0.0003) (Fig. 3B). All 9 patients with positive ductal margins with invasive carcinoma died of local recurrence (with or without distant metastases) within 40 months of undergoing surgical resection.
Patterns of Failure in 11 Patients with Positive Ductal Margins with Carcinoma in Situ
In 1 of the 11 patients with a positive ductal margin with carcinoma in situ (Case no. 2 in Table 3), it was impossible to distinguish between carcinoma in situ and severe dysplasia histologically. In the remaining patients, the epithelial lesions at the ductal resection margins were found to be carcinoma in situ.
Table 3. Eleven Patients with Positive Ductal Margins with Carcinoma in Situ
Primary tumor site
Type of surgical resection
Ductal margin status
Sites of recurrence
pT: pathologic primary tumor classification; pN: pathologic lymph node metastasis classification; pM: pathologic distant metastasis classification; F: female; G2: moderately differentiated; BD: resection of the extrahepatic bile ducts; N: regional lymph node dissection; CIS: carcinoma in situ; NED: (alive with) no evidence of disease; M: male; G3: poorly differentiated; PD: Whipple pancreaticoduodenectomy; HEP: hepatic; LYM: lymph nodes; DOD: died of disease; WR: wedge resection of the gallbladder bed; G1: well differentiated; ERH: extended right hepatectomy; PPPD: pylorus-preserving pancreaticoduodenectomy; PER: peritoneum; DOO: died of other causes.
Among the patients who had positive ductal margins with carcinoma in situ, 4 died of tumor recurrence, 2 died of other causes, and 5 survived without disease (Table 3); the median disease-free survival time was 89 months and the median survival time was 99 months. In all four of the patients who died of the disease, the initial site of disease recurrence was local, suggesting that carcinoma in situ at ductal resection margins may cause late local recurrences.
Although ductal margin status is an established prognostic factor in patients with extrahepatic cholangiocarcinoma,4–17 a small proportion of patients with positive resection margins may still survive in the long term.7–12 We also observed four 5-year survivors with positive ductal margins (Fig. 2), a fact that inspired the current study. To our knowledge, the current study is the first to demonstrate that residual carcinoma in situ differs prognostically from residual invasive ductal disease in patients with extrahepatic cholangiocarcinoma.
In the current study, which examined extrahepatic cholangiocarcinoma, 4 patients with residual carcinoma in situ at the ductal stumps died of local disease recurrence 26 months, 54 months, 99 months, and 112 months after surgical resection, respectively (Table 3). For gallbladder carcinoma, we previously reported that 2 patients with residual carcinoma in situ at the cystic ductal stumps died of local recurrence 66 months and 76 months, respectively, after undergoing cholecystectomy.27 Taken together, these findings suggest that it is very likely that in patients with biliary malignancies, the progression of carcinoma in situ to invasive carcinoma at ductal resection margins requires several years. We believe this finding explains why patients with residual carcinoma in situ may survive in the long term.
In the current study, ductal margin status appeared to independently affect postresectional survival. Because the extent of ductal spread of cholangiocarcinoma is difficult to determine preoperatively on imaging studies,28–30 intraoperative frozen-section examination of ductal resection margins is indispensable in achieving negative margins. When ductal margin status is found to be positive on frozen-section examination, discrimination between carcinoma in situ and invasive carcinoma is important. Residual carcinoma in situ may cause late local disease recurrences, whereas residual invasive ductal lesions cause early local recurrences.
The current study has limitations. First, it was a retrospective analysis of a small number of patients, and second, the follow-up period was < 5 years in 21 patients. However, we believe that these limitations did not significantly influence the outcome of the study due to the fact that the differences between groups were too marked to have resulted from these biases.
Among patients undergoing surgical resection for extrahepatic cholangiocarcinoma, invasive carcinoma at ductal resection margins appears to have a strong adverse effect on survival, whereas residual carcinoma in situ does not. It is therefore clear that when ductal resection margin status is positive on frozen-section examination, discrimination between carcinoma in situ and invasive carcinoma is clinically important.