Adjuvant chemoradiotherapy for positive hepatic ductal margin on cholangiocarcinoma

Abstract Aim This study evaluated the effects of postoperative adjuvant chemoradiotherapy (A‐CRT) for positive hepatic ductal margin (HM+) in extrahepatic cholangiocarcinoma (EHCC). Methods Patients with EHCC who underwent surgical resection between 2002 and 2014 were included in this retrospective study. For patients with HM+, A‐CRT was conducted. The clinical effect of A‐CRT for HM+ on the survival and recurrence and prognostic factors of EHCC was reviewed. Results Among 340 patients, the hepatic ductal margin was negative in 296 and positive in 44. Of the 44 patients with HM+, 22 received postoperative A‐CRT, and 22 did not. Hepatic stump recurrence occurred in 19 patients. The incidence was significantly higher in patients with HM+ (20%, 9/44) than in those with negative hepatic ductal margin (HM−) (3%, 10/296) (P < .001). Among the patients with HM+, the incidence was almost identical between the patients with and without A‐CRT: 23% (5/22) in HM+/CRT− and 18% (4/22) in HM+/CRT+ patients (P = .999). The median survival time was 49 months in HM−, 43 months in HM+/CRT−, and 49 months in HM+/CRT+ patients. The differences were not significant among the groups. A multivariate analysis revealed CA 19‐9 ≥ 300 U/mL, combined vascular resection, histologic grade G2/G3, and lymph node metastasis to be significant prognostic factors. However, the performance of postoperative A‐CRT did not contribute to prolonging survival. Conclusion A‐CRT for HM+ in patients with EHCC did not affect the survival or stump recurrence.

In cases with microscopically positive resection margins after aggressive surgery, adjuvant local treatment is required for the elimination of residual tumor to improve the survival. Several studies have reported that adjuvant chemoradiotherapy (A-CRT) may improve the survival in patients who had R1 resection. 6,7 However, these studies included patients receiving adjuvant radiotherapy alone, brachytherapy, and intraoperative radiotherapy, and the survival benefits of adjuvant concurrent CRT after R1 resection have not fully been investigated.
In the authors' institution, postoperative A-CRT targeting the biliary stump at the hepatic hilum in patients with HM+ has been conducted aiming to reduce stump recurrence. The aim of this study was to review the effects of postoperative CRT for HM+ in EHCC.

| ME THODS
Data from consecutive patients with EHCC treated at the authors' institution between 2002 and 2014 were obtained from a prospectively collected database and reviewed retrospectively. Patients with intrahepatic cholangiocarcinoma involving the hepatic hilum, distant metastasis, and in-hospital mortality were excluded from the analysis. The incidence of stump recurrence, postoperative survival, and prognostic factors was evaluated by referencing the HM status or CRT practice. TNM classifications were determined according to the UICC system, 8th edition. The study was approved by the institutional review board.

| CRT for HM+
Our standard of treatment for EHCC has been surgery alone, regardless of the tumor stage. CRT was targeted at patients with HM+.
During this study period, HM+ was defined as a positive hepatic ductal margin with both invasive carcinoma and carcinoma in situ (CIS). After providing some information -as (a) positive hepatic ductal margin is correlated with stump recurrence, and (b) A-CRT is intended to prevent stump recurrence at the hepatic ductal stump; (c) however, there has been no prospective study of A-CRT for a positive ductal margin, and (d) several retrospective studies have revealed controversial results -the decision on whether or not to receive A-CRT was left to the patient. However, A-CRT was not intended for cases with a positive distal ductal margin. If recurrence at the distal margin stump alone occurs in the future, then additional pancreatoduodenectomy is considered to be a treatment option. In addition, the radiation field did not cover the regional lymph nodal basin because lymph node dissection was systematically performed in order to not leave remnant lymph nodes.
Three-dimensional conformal radiotherapy was planned to deliver a total of 50.4 Gy at 1.

| Statistical analyses
All statistical analyses were performed using the SPSS software pro-     Next, these patients were divided according to the degree of HM+.

| RE SULTS
In 37 patients with HM+ with CIS, the incidence was almost iden-  Table 2). The incidence of each recurrence site was identical.
Of the 19 patients with stump recurrence, isolated stump recurrence was detected in nine, including five with HM−, 2 with HM+/CRT−, and two with HM+/CRT+. The other 10 patients had recurrence at multiple sites, including the liver in four, peritoneum in four, lung in four, lymph node in two, and local in two (overlapped in patients).
After the detection of stump recurrence, five patients were treated with gemcitabine, three patients were treated with gemcitabine and cisplatin, two patients were treated with S-1, and nine patients received the best supportive care. Aside from these treatments, nine patients underwent percutaneous transhepatic biliary drainage.    Figure 5A).
The OS was also nearly the same among the groups, with an MST of 36 months in HM− patients, 36 months in HM+/CRT− patients, and 37 months in HM+/CRT + patients ( Figure 5B).

| D ISCUSS I ON
Complete resection is the mainstay treatment for patients with EHCC. 9 However, surgeons occasionally face the issue of a positive resection margin, especially at the hepatic ductal stump. 3,10 Even if complete resection is achieved, one of the most common patterns of failure for EHCC is locoregional recurrence. 11,12 To control stump recurrence and prolong the prognosis, A-CRT was administered in these patients.
However, we failed to reveal any marked benefit of A-CRT. A-CRT for HM+ was not effective for improving the survival or stump recurrence.
Stratified by the degree of HM+, the stump recurrence rate in patients with HM+ with CIS was nearly the same. Although few patients with HM+ had invasive carcinoma, two of the three who received A-CRT developed stump recurrence. In that sense, A-CRT does not seem to be effective for HM+ with CIS or invasive carcinoma.
Several previous studies evaluated the effect of A-CRT on reducing the locoregional recurrence for cholangiocarcinoma. There have been conflicting results regarding the effect of adjuvant radiotherapy with and without chemotherapy after curative surgical resection. 6,[13][14][15] Some studies have shown that postoperative adjuvant radiotherapy has no influence on survival 13,14 . In contrast, others have reported survival advantages of CRT. 6,[15][16][17] Two recent meta-analyses found that adjuvant therapy including radiotherapy for cholangiocarcinoma decreased the risk of death compared to surgery alone, especially in cases with lymph node metastases or a positive surgical margin. 18,19 However, another meta-analysis by Zhu 20 revealed that CRT was not effective for margin-positive disease. In all of these studies, radiation was delivered to the tumor bed and regional lymph nodes. In our series, radiation was strictly targeted at the hepatic ductal stump. This is therefore the first report to evaluate the effect of CRT on controlling stump recurrence.
In the present study, a CA19-9 value ≥300 U/mL, combined vascular resection, histological grade G2/G3, and lymph node metastases were found to be significant prognostic factors. In particular, a higher CA19-9 value, lymph node metastasis, and poor differentiation influenced the development of distant metastases. 12,21,22 In the present study, 10 of 19 patients with stump recurrence also had multiple-site recurrence, especially distant metastasis. All of these patients had at least one prognostic factor. In contrast, only two of nine patients with isolated stump recurrence had these prognostic factors. In patients at a high risk of distant metastases, the efficacy of CRT for local control seems to be low, and more powerful systemic chemotherapy regimens are needed.
Chemotherapy for advanced cholangiocarcinoma has been gradually established since 2005. 23 In the current series, the CRT regimen was relatively outdated. not sufficient to draw broad interpretations. In particular, the number of stump recurrence events was low. A slight increase in the events can affect the statistical analyses. A longer follow-up period is necessary, as some cases of stump recurrence develop more than 5 years later.

| CON CLUS ION
CRT for HM+ was not effective for improving survival or stump recurrence in patients with EHCC.