AIM: To analyze the prognostic factors influencing the overall survival of peripheral cholangiocarcinoma (PCC) patients undergoing surgical treatment during 25 years at a single institution.

METHODS: This study retrospectively reviewed prospectively collecting data about 373 patients with histologically proven PCC who underwent surgical treatment between 1977 and 2001.

RESULTS: Three hundred and seventy-three PCC patients (159 men and 214 women) underwent surgical treatment from 1977 to 2001. Among them, 187 PCC patients underwent hepatectomy and 135 had curative resection (curative resectability rate: 36.2%). The follow-up duration ranged from 1.05 to 167.6 mo (mean/median = 14.1/7.2 mo). Overall cumulative survival rates at 1, 3, and 5 years were 32.5%, 9.2%, and 4.1%, respectively. Univariate log-rank analysis identified the following as adverse influences on overall survival: presence of symptoms, absence of mucobilia, elevated CEA and CA 19-9 levels, non-papillary tumor type, receiving non-hepatectomy, advanced tumor staging, lack of post-operative chemotherapy, and radiotherapy. Meanwhile, multivariate Cox’s proportional hazard analysis demonstrated that absence of mucobilia, non-papillary tumor type, advanced tumor staging, non-hepatectomy, and lack of post-operative chemotherapy were the five independent prognostic factors that adversely affected overall survival.

CONCLUSION: Favorable overall survival of PCC patients undergoing surgical treatment depends on early tumor stage, presence of mucobilia, papillary tumor type, hepatic resection, and post-operative chemotherapy.

Peripheral cholangiocarcinoma (PCC) is a malignant tumor occurring in the liver or arising from the second or more distal branches of the intrahepatic bile ducts[1]. The incidence of PCC exhibits wide geographical variation, and generally accounts for between 5% and 30% of primary liver cancers[1]. According to the Japan Liver Cancer Society, histologically proven PCC represented 3.58% of all primary liver cancers[2].

Although no specific etiological factor has been identified for PCC, various well-recognized predisposing conditions involving chronic inflammation of the bile ducts exist. These conditions include infestation with the liver flukes Opisthorchis viverrini in Thailand[3] and Clonorchis sinensis in other Asian countries[4,5]; hepatolithiasis, prevalent in countries such as Taiwan and Japan[6-8]; primary sclerosing cholangitis[9,10]; choledochal cysts[11,12]; and Caroli’s disease[4,5].

Clinically, PCC remains challenging because it is difficult to diagnose in its early stages, and patients typically do not present because of vague symptoms until the disease is quite advanced. Curative resection with clear margins and without vascular or lymphatic invasion is infrequent and recurrence is common. Three- to five-year survival rates even with resection remains dismal[13-18]. Moreover, radiation therapy or current chemotherapy does not significantly improve long-term survival rates[16,19]. Tumor biological behavior and early intrahepatic and extrahepatic spread limit the efficacy of surgery, and yet complete surgical removal currently is the curative option for PCC. Although liver transplantation has provided an alternative in the surgical management of PCC, high rates of PCC recurrence following transplantation limit liver transplantation for PCC patients[20]. Despite considerable progress in diagnostic procedures, surgical techniques, and adjuvant therapies, the prognosis of PCC remains extremely poor, because most tumors are advanced by the time of diagnosis. The literature contains few reports of successful surgical treatment of PCC and evaluation of its prognostic factors, and most such reports deal with small numbers of cases[21-25]. This study examines the influence of clinicopathologic characteristics of PCC undergoing surgical treatment on PCC patient’s overall survival to improve the prognosis of PCC patients.

From January 1977 to December 2001, 608 consecutive patients with histologically proven bile duct adenocarcinoma underwent surgical treatment at the Department of Surgery, Chang-Gung Memorial Hospital, Taipei, Taiwan. Based on tumor location, the 608 patients were classified into common bile duct cancer (n = 102; 16.8%), bile duct adenocarcinoma with hilar invasion (n = 133; 21.9%), and PCC (n = 373; 61.3%). PCC was defined as carcinoma arising from second order or more distal branches of the intrahepatic ducts. IP-PCC is histologically defined as intraductal papillary neoplasia of the liver (IPN-L) type 3 and 4, as described previously[8]. Type 3 showed an IPN-L lined by in situ and microinvasive adenocarcinoma, and type 4 showed with types 2 and 3 biliary lesions with variable invasion of adenocarcinoma[8]. Meanwhile, curative resection was defined as negative resection margin observed during histopathological examination. The 373 PCC patients comprised 159 men and 214 women with a mean age of 57.8 years (range 28-93 years). Table 1 lists the clinical manifestations and histological classifications. Right upper abdominal pain with tenderness was the most common symptom and sign on admission. Table 2 lists the distribution of the operative procedures, and indicates that 187 patients had hepatectomy, with a 36.2% curative resectability rate (curative resection cases: 135/373). Meanwhile, Table 3 lists the causes of surgical mortality, and reveals a surgical mortality rate of 6.7% (25/373). Surgical mortality is defined as the death occuring within one month after surgery. PCC patients undergoing non-hepatectomy procedure had a significantly higher surgical mortality (2.7 vs 10.8%; P = 0.002). Sepsis due to biliary tract infection is the main cause of surgical mortality. Laboratory tests were conducted on the day before surgery. Serum carbohydrate antigen 19-9 (CA 19-9) and carcinoembryonic antigen (CEA) were measured by radioimmunoassay. The tumors were preoperatively evaluated by abdominal ultrasonography (US), endoscopic retrograde cholangiopancreatography, percutaneous transhepatic cholangiography, computed tomography (CT), magnetic resonance image with cholangiopancreatography (MRCP), and hepatic arteriography, as appropriate. Tumor stage was defined according to the pathological tumor node metastasis (pTNM system) classification proposed by the UICC. Stages I and II were conditioned as early-stage, and stages III and IV as advanced-stage PCC. Adjuvant chemotherapy was systemically performed with a 5-flurouracil-based regimen due to either a positive section margin or local recurrence. Meanwhile, adjuvant radiotherapy was performed with intra-operative radiotherapy, external beam radiotherapy and/or brachytherapy due to either a positive section margin or local recurrence.

This study used univariate and multivariate analysis to calculate overall survival rates of 312 PCC patients after surgical treatment in terms of 17 clinicopathologic factors.

In univariate analysis, patients’ symptoms, preoperative tumor marker (CEA and CA19-9), presence of mucobilia, tumor spreading pattern, treatment option, tumor staging, and post-operative adjuvant therapy (chemotherapy and radiotherapy) significantly influenced overall survival of PCC patients after surgical treatment. Applying the multivariate Cox’s proportional hazard model to data from this investigation revealed that long-term overall survival depends on the presence of mucobilia, hepatectomy, early tumor stage, post-operative adjuvant chemotherapy, and radiotherapy.

Patients had PCC diagnosed asymptomatically and incidentally had a favorable survival, demonstrated by univariate log-rank survival analysis. This study revealed the fact that PCC remains clinically challenging because it is difficult to diagnose in its early stages, and patients typically do not present because of vague symptoms until the disease is quite advanced. So we should seek a certain novel tumor marker for the early detection of PCC in high-risk patient groups.

Diagnostic adjuncts for PCC, such as a serum marker, are useful for the clinical management of this disease. Several investigators have reported that CA 19-9 and CEA determinations are useful for diagnosing PCC in primary sclerosing cholangitis[26]. These investigators also demonstrated that serum CA 19-9 values are related to tumor burden, and suggested that values were elevated in patients with PCC related to unresectable disease[26]. A previous Japanese study has also demonstrated that elevated serum tumor marker, including CEA and CA19-9, was a predictor of a dismal prognosis[21]. Although the reason is unclear, this study also demonstrated that elevated CEA and CA 19-9 were predictors of dismal prognosis in univariate log-rank analysis.

The association between intrahepatic stone and PCC is well recognized.

Mucobilia occurs in various conditions, including biliary papillomatosis, cholangiocarcinoma, and biliary cystadenoma and cystadenocarcinoma of the liver[27]. The presence of mucobilia is important for the early diagnosis of PCC, even preoperatively[28]. Our previous study demonstrated that mucobilia was a prognostic variable predicting better survival in PCC patients undergoing surgical treatment[29]. This study confirmed the same observation and reemphasizes the importance of recognition in this specific clinical presentation as a hope of favorable overall survival of PCC patients.

Papillary type PCC, namely intraductal papillary neoplasia of the liver, is a specific type of biliary epithelial neoplasia with frequent gastrointestinal metaplasia associated with overproduction of mucin and mucobilia[8]. The special nature of this condition explains why more papillary adenocarcinoma and mucobilia were noted in this study. Tumor spreading type was also a favorable prognostic variable for PCC survival[29-33]. Patients with intraductal papillary type PCC had displayed significantly better survival rates than periductal infiltrating or mass forming type PCC. Meanwhile, this study confirmed that the absence of papillary growing in PCC patients undergoing surgical treatment increased the chances of reducing long-term survival by 1.946 times according to multivariate Cox’s proportional hazard analysis.

Although a previous Japanese report has proposed a new staging system mainly based on mass forming type PCC to challenge the TNM staging system[25], this study revealed that TNM staging system remains the most reliable prognostic factor for predicting the outcome of PCC patients undergoing surgical treatment for various types of PCC. Patients with more advanced tumors displayed a significantly worse prognosis, revealed not only by univariate log-rank analysis but also multivariate Cox’s proportional hazard analysis.

Hepatic resection is the preferred treatment for PCC[1]. This study demonstrated that PCC patients undergoing hepatic resection had significantly better survival than those without hepatic resection[29,34]. However, prognosis of PCC in our series is poor when compared with the other series[35]. The fact that only about 50.1% of 373 PCC patients receive hepatic resection and more advanced tumor staging patients exist may explain the poor prognosis in this study. This poor prognosis may be caused by the large number of PCC patients presenting with advanced tumors and unresected status (76.9% and 49.9%, respectively). Hepatic resection significantly prolongs the survival of PCC patients, as demonstrated by multivariate Cox’s proportional hazard analysis.

The utilization of adjuvant chemotherapy and radiation in PCC remains controversial[1]. However, this study showed that postoperative adjuvant chemotherapy was independent significant predictors for PCC patients with good long-term survival undergoing surgical treatment. Several authors have reported varying degrees of success with chemotherapy, either by systemic route or by intra-arterial hepatic route in PCC treatment[35-37]. The efficacy of chemotherapy has not been evaluated prospectively in patients with PCC[35]. No good chemotherapy regimen is currently available, but 5-flurocil (5-FU) is the standard base regimen. Several combinations with 5-FU have been reported to produce transient partial remission in a small portion of patients[35-37].

In conclusion, long-term survival of PCC patients undergoing surgical treatment depends on early tumor stage, presence of mucobilia and papillary tumor type, hepatic resection, and post-operative chemotherapy.