Dual-organ invasion is associated with a lower survival rate than single-organ invasion in distal bile duct cancer: A multicenter study

The revised criteria of the 8th American Joint Committee on Cancer (AJCC) cancer staging system consider depth of invasion as one of the factors that determine stage in distal bile duct (DBD) cancer, but exclude adjacent organ invasion. The aims were to evaluate the association between adjacent organ invasion and relapse-free survival (RFS) and overall survival (OS) after curative surgical resection of DBD cancer and to propose optimal criteria for predicting clinical outcomes. In this retrospective cohort study, 378 patients with DBD cancer treated in multi-institutions between 1996 and 2013 were investigated. This study evaluated the relationship between clinicopathologic parameters and adjacent organ invasion and used organ invasion to compare the survival times of each group. Among 204 patients with adjacent organ invasion, 152 were in the single-organ invasion group and 52 were in the dual-organ invasion group based on a review of microscopic slides. In univariate and multivariate analyses, patients with dual-organ invasion had a shorter RFS and OS time than those with single-organ invasion. Organ invasion should be included as one of the factors that determine the AJCC stage; this might ultimately help to predict better the survival rate of patients with DBD cancer.

The AJCC staging system according to adjacent organ invasion is still practically used in various cancers. Even in PBD cancer, criteria based on adjacent structures is still being used to determine the 8 th AJCC staging. The adjacent organ invasion might still provide important information for determining advanced-stage DBD cancer with successful tumour removal, although it was excluded from the T criteria, which only consider DOI. In the previous 7 th T criteria of the DBD cancer, the T1 and T2 stages were defined as "Tumour confined to the bile duct" and "tumour invading beyond the wall of the bile duct", respectively, whereas the T3 stage included organ invasion, such as the gallbladder, pancreas, duodenum, or other adjacent organs. Nevertheless, an inadequate number of validation studies was the main reason that organ invasion was excluded from the 8 th AJCC staging system. Previously published data were reported based on the results of analyses of both DBD and PBD cancer, because both cancers were categorized as "extrahepatic bile duct cancer" until the 6 th edition of the AJCC staging manual 1,[4][5][6][7][8] . Past studies on the relationship between patient survival and organ invasion had inadequate numbers of study participants, which is why they have little prognostic significance 9,10 .
We validated the prognostic impact of the 8 th and 7 th AJCC staging system in 374 patients with DBD cancer. We investigated the survival rate according to the number of infiltrating organ to further enhance prognostic accuracy. The aims of this study were to evaluate the association between adjacent organ invasion and relapse-free survival (RFS) and overall survival (OS) after curative surgical resection of DBD cancer and to suggest supplementary criteria for predicting clinical outcomes.

Comparisons of Clinicopathological Parameters in Patients with and without Organ Involvement.
Of all 378 patients, 204 patients had adjacent organ invasion. Patients with organ invasion had a significantly higher incidence of infiltrative gross type, poorly histological grade, lymphovascular invasion, perineural invasion and margin involvement, compared to those without organ invasion (all P < 0.05) ( Table 2).

Comparisons of Clinicopathological Parameters Between Single-and Dual-organ Invasion.
Of 204 patients with adjacent organ invasion, 152 patients had adjacent single-organ invasion as follows: pancreas, 146; duodenum, 4; gallbladder, 2. Fifty-two patients had dual-organ invasion as follows: pancreas and duodenum, 51; duodenum and gallbladder, 1. Patients with dual-organ invasion had a significantly higher incidence of infiltrative gross type and advanced N stage, compared to those with single-organ invasion (all P < 0.05) ( Table 3).

Discussion
In the new 8 th AJCC staging manual, cholangiocarcinoma is classified based on its anatomic location into three subtypes: intrahepatic, perihilar, and distal portion. Importantly, DBD cancers comprise 20-30% of all cholangiocarcinoma and are clinically silent, with symptoms only developing at an advanced stage 11 . Beginning in the 7 th edition, and continuing in the 8 th edition, the T and N criteria are different for PBD and DBD and are based  on their distinct biological behavior, natural course, and therapeutic plan 11,12 . In particular, the factors that were specifically associated with patient survival were DOI, nodal metastasis, lymphatic/perineural invasion as well as pancreatic invasion, and resection margin involvement, to name a few 1,4-6 . Interestingly, among the above prognostic indicators, DOI and number of metastatic regional lymph nodes were applied in the new 8 th T and N criteria for DBD cancer, because the previous 7 th AJCC staging was described as having vague T criteria resulting in wide inter-observer variation. Therefore, the 8 th AJCC staging system suggested a cut-off value of DOI measured in millimeters to reduce inter-observer variation 1,13,14 . To determine the stage of tumours in various organs (lip and oral cavity, cervix uteri, vulva, and melanoma of the skin), DOI was adopted in the 8 th AJCC staging system. These organs have simple anatomical structures and are relatively distant from tumour-adjacent organs. In other words, a long distance from the tumour origin inhibits direct tumour infiltration into other organs. However, the DBD is located near various organs and has a relatively complex anatomical structure. In organs that are close to tumour origins, such as the nasal cavity, paranasal sinus, and larynx, the T criteria are classified based on direct tumour infiltration into adjacent organs. A study of DBD cancer showed that the presence or absence of adjacent organ invasion was associated with a significant difference in survival 9 . Nevertheless, for DBD cancer, the 7 th AJCC staging manual categorizes T criteria based on adjacent organ invasion, but organ invasion is no longer described in the 8 th AJCC T criteria, especially T3 stage 3 . A study by Ebata et al. showed that presence or absence of adjacent organ invasion created a significant difference in survival 9 . In our results, patients with organ invasion show lower RFS and OS than those without organ invasion. Notably, there were significant differences of RFS or OS between single-and dual-organ invasion. However, there was little survival difference when the 8 th AJCC T criteria were adopted for DBD cancer. An explanation for this is that the interval of invasion depth among T1, T2, and T3 groups was widened. In our study, the categories of DOI were as follows: no invasion, 4.5 mm; single-organ invasion, 8.2 mm; dual-organ invasion, 10.7 mm.
The recommended 8 th AJCC T criteria for DOI are 5-12 mm and >12 mm in the T2 and T3 groups, respectively. Hong et al. performed a study of 222 patients who underwent surgery at a single center and whose tumours included both perihilar and/or distal tumours (perihilar, 111 cases; distal 101 cases; perihilar and distal, 10 cases) 3 . In survival models to determine the cut-off value of DOI, only 101 patients had DBD cancer, whereas 110 patients had PBD cancer. The cut-off values for the measured DOI were calculated in both perihilar and DBD cancers without considering their distinct biological behaviors 11,12 . In a validation study, there was no survival     difference between groups (T2 versus T3) 15 . Recently, a multicenter study of 179 patients with only DBD cancer was designed using a smaller range of DOI and revealed a significant survival difference among groups (<3 mm, 4-10 mm, >11 mm) 13 . Thus, controversy exists regarding the relationship between clinical outcome and the criteria for DOI. In summary, patients with dual-organ invasion showed lower survival rate than patients with single-organ invasion in DBD cancer, although there is no survival difference between the T2 and T3 groups based on DOI defined by the 8 th AJCC staging system. In the prognosis prediction with advanced T groups, adjacent organ invasion could enhance prognostic accuracy. Consequently, the significant difference in survival between single-and dual-organ invasion could be considered to supplement the T criteria using DOI to guide therapy and standardize the 8 th AJCC staging system.

Materials and Methods
Case Selection. Tumour with their center located between the confluence of the cystic duct and common hepatic duct and the Ampulla of Vater (excluding ampullary cancer) are considered DBD cancer in reference with 8 th AJCC stage. A total of 404 cases of patients diagnosed with DBD cancer at multi-institutions (Eulji Hospital, Kangbuk Samsung, Hanyang Guri, Hallym University Sacred Heart Hospital, Gangneung Asan) in Korea between January 1, 1996 and December 31, 2013 were collected for this study. This study included data obtained from previously conducted research study 13 . As for twenty-six patients who died within 90 days after surgery or who had few representative slide for microscopic review were excluded from this study.
The following clinicopathological parameters were recorded: age, gross type, histological grade, size, 8 th AJCC stage, lymph node metastasis, adjacent organ invasion (pancreas, duodenum, gallbladder), lymphovascular invasion, perineural invasion, margin involvement, relapse, and survival. Grossly, the tumours were classified as papillary, nodular, and infiltrative, and the tumour size was measured along its greatest dimension. Hematoxylin and eosin-stained slides with representative tumour section were reviewed by at three pathologists (KWM, DHK, EKK). The DOI from the basal lamina of the adjacent normal epithelium to the most deeply advanced tumour cells was measured in reference with previous study 3 .
The mean and median age of the remaining 378 patients was 63 and 64 years, respectively. The male to female ratio was 253:125. The surgical treatment included the Whipple procedure in 153 (40.5%), pylorus-preserving pancreaticoduodenectomy in 125 (33.1%), and extended bile duct resection in 101 (26.4%) patients. In multi-institutions, the indication of surgical procedure for DBD cancer was as follows: pylorus-preserving pancreaticoduodenectomy was performed on patients with (a) no evidence of tumour extension to the pylorus, (b) chances to preserve pylorus artery and (c) no ulcer in pylorus. Whipple's operation was conducted when a patient did not belong to the above PPPD indication. Extended bile duct resection was conducted when (a) tumour was positioned at mid-portion and did not invade adjacent organs and (b) safety margin is confirmed by frozen section of tumour.

Statistical Analysis. Correlations between clinicopathological parameters and adjacent organ invasion
were analysed using the Chi-square test and the linear-by-linear association. Relapse-free survival is defined as the time elapsed from the date of treatment to the date of progression such as a local recurrence, new lymph node metastasis or distant organ metastasis. Overall survival was defined as the time from the date of treatment