辅助放(化)疗在肝外胆管细胞癌中的作用:荟萃分析

Role of adjuvant (chemo)radiotherapy for resected extrahepatic cholangiocarcinoma: a meta-analysis # # 辅助放(化)疗在肝外胆管细胞癌中的作用:荟萃分析 first-author corresp ‡Corresponding Author ‡Corresponding Author †E-mail:liailin0312@163.com †E-mail:liailin0312@163.com liailin0312@163.com Department of Radiation Oncology, the First Hospital of China Medical University, Shenyang 110001, China Department of Radiation Oncology, the First Hospital of China Medical University, Shenyang 110001, China Background: 2 Adjuvant (chemo)radiotherapy (A(C)RT) may be an important supplement to surgery for extrahepatic cholangiocarcinoma (EHCC). However, whether all patients would achieve benefits from A(C)RT and which adjuvant regimen, adjuvant radiotherapy (ART) or adjuvant chemoradiotherapy (ACRT), would be preferred, are still undetermined. The low incidence of EHCC makes it difficult to carry out randomized controlled trials (RCTs); therefore, almost all clinical studies on radiotherapy are retrospective. We have conducted a meta-analysis of these retrospective studies. Methods: 2 We conducted a meta-analysis of current retrospective studies using PubMed, Embase, and ClinicalTrials databases. All studies published in English that were related to A(C)RT and which analyzed overall survival (OS), disease-free survival (DFS), or locoregional recurrence-free survival (LRFS) were included. Estimated hazard ratios (HRs) were calculated for OS, DFS, and LRFS. Results: 2 Data from eight studies including 685 patients were included. Our analysis showed that A(C)RT significantly improved OS (HR 0.69, 95% confidence interval (CI) 0.48–0.97, P =0.03), DFS (HR 0.60, 95% CI 0.47–0.76, P < 0.0001), and LRFS (HR 0.27, 95% CI 0.17–0.41, P < 0.00001) of EHCC overall. In subgroups, patients with microscopically positive resection margin (R1) could achieve a benefit from A(C)RT (HR 0.44, 95% CI 0.27–0.72, P =0.001). No statistically OS difference was observed in negative resection margin (R0) subgroup (HR 0.98, 95% CI 0.30–3.19, P =0.98).Significant OS benefit was found in patients who received concurrent ACRT (HR 0.40, 95% CI 0.26–0.62, P < 0.0001), while the result of ART without chemotherapy showed no significant benefit (HR 1.14, 95% CI 0.29–4.50, P =0.85). In the distal cholangiocarcinoma subgroup, no significant difference was seen when ACRT and ART were included (HR 0.61, 95% CI 0.14–2.72, P =0.52), but a significant difference was seen when analyzing the concurrent ACRT only (HR 0.29, 95% CI 0.13–0.64, P =0.002). Conclusions: 2 A(C)RT may improve OS, DFS, and LRFS in EHCC patients, especially in those with R1 resection margins. ACRT may be superior to ART especially in distal patients. 目的 2 分析术后辅助放(化)疗在肝外胆管细胞癌中的作用,找到具体哪些临床亚组能从辅助放(化)疗中获益. 创新点 2 本研究报告了辅助放(化)疗可以改善肝外胆管细胞癌患者尤其是显微镜下切缘阳性患者的总生存期、无病生存期和无转移生存期,为这一无定论的临床问题提供循证证据. 方法 2 我们检索了截至2019年10月,收录在PubMed、Embase和ClinicalTrials三个数据库中关于术后辅助放(化)疗在肝外胆管细胞癌中的作用的文献.经过筛选,最终有8篇文献符合纳入标准,并采用RevMan软件进行数据分析. 结论 2 辅助放(化)疗可以改善肝外胆管细胞癌患者的总生存期、无病生存期和无转移生存期,尤其是在显微镜下阳性切缘的患者中. 辅助同步放化疗可能优于单纯辅助放疗,尤其是在远端胆管癌患者中. Adjuvant (chemo)radiotherapy Extrahepatic cholangiocarcinoma Meta-analysis Disease-free survival Overall survival 辅助放(化)疗 肝外胆管细胞癌 荟萃分析 无病生存期 总生存期 Introduction 1 Cholangiocarcinoma is a rare malignancy with poor prognosis. The incidence of cholangiocarcinoma is between 0.35 to 2.00 per 100 000 annually in the Western world, but in Asia the incidence could be up to 40 times the rate observed in Western countries (Bridgewater et al., 2016 2016 ). Extrahepatic cholangiocarcinoma (EHCC) comprises approximately 75% of all cholangiocarcinomas (Doherty et al., 2017 2017 ). The five-year overall survival (OS) rate of EHCC ranges from 2% to 30%, depending on stage (Chaiteerakij et al., 2014 2014 ). EHCC is further subdivided into perihilar cholangiocarcinoma and distal cholangiocarcinoma based on anatomic location. Curative surgery provides the only chance for cure. However, it was reported that fewer than one-third of patients were deemed to have resectable disease at the time of diagnosis (Khan et al., 2012 2012 ), and even after resection, the postoperative five-year OS rates for perihilar cholangiocarcinoma and distal cholangiocarcinoma were 24.2% and 39.8%, respectively (Ishihara et al., 2016 2016 ). Adjuvant therapy has been used to improve the outcome of EHCC and has received increasing attention in the past ten years. In the latest National Comprehensive Cancer Network (NCCN) guidelines, adjuvant treatment regimens for EHCC based on adjuvant chemoradiotherapy (ACRT) are recommended. However, whether all EHCC patients or certain subgroups could benefit from ACRT or adjuvant radiotherapy (ART) is still undetermined because of the limited clinical data (NCCN, 2019 2019 ). To date, almost all clinical data available are from retrospective studies, because the low incidence of EHCC makes it difficult to recruit enough patients for randomized controlled trials (RCTs). Moreover, the inclusion criteria of patients and adjuvant (chemo)radiotherapy (A(C)RT) regimens were not uniform in those retrospective studies. Several studies also included gallbladder and other bile duct tumors. Meta-analysis is a reasonable alternative when large randomized trials are not feasible. A previous meta-analysis (Bonet Beltrán et al., 2012 2012 ) discussed the value of A(C)RT after curative resection of EHCC; however, the study included intraluminal brachytherapy mixed with external beam radiotherapy, as well as gallbladder cancer and other extrahepatic biliary tract tumors with EHCC. Consequently, the results of the study were of little value in present clinical practice. Therefore, we performed a meta-analysis to explore the benefits of A(C)RT vs. surgery alone in EHCC. Methods 1 Search strategy 2 We searched for relevant studies in PubMed, Embase, and ClinicalTrials (https://clinicaltrials.gov) databases with no publication type or time restrictions (up to Oct. 10, 2019). The main search terms were as follows: extrahepatic/perihilar cholangiocarcinoma, biliary tract cancers, adjuvant (chemo)radiotherapy, postoperative (chemo)radiotherapy, radiotherapy/radiation therapy after surgery/resection and clinical trial. Meeting abstracts and presentations at the American Society of Clinical Oncology Annual Meetings and the European Society of Medical Oncology Congresses from 2009 to 2019 were searched manually. The reference lists of some key articles were also searched manually. This meta-analysis was registered at the International Prospective Register of Systematic Reviews (No. CRD42020149802). Inclusion and exclusion criteria 2 Inclusion criteria were: (1) studies involving patients with clearly diagnosed EHCC; (2) studies comparing treatments between ACRT or ART and surgery only, which meant that studies had to include patients who underwent surgery alone as a comparator group; (3) studies reporting hazard ratios (HRs) of OS and/or disease-free survival (DFS), or studies in which these data could be calculated; and (4) studies reported in English. Exclusion criteria were: (1) relevant data that were not reported and could not be calculated; (2) studies that enrolled intraluminal brachytherapy in the A(C)RT group; (3) studies that enrolled adjuvant chemotherapy (ACT) without ART in the experimental group; and (4) studies that enrolled patients with gallbladder cancer or intrahepatic cholangiocarcinoma together with EHCC but specific data for EHCC were not provided. The outcomes assessed were OS, DFS, and locoregional recurrence-free survival (LRFS) of patients in the two groups. OS was defined as time to death or to the end of follow-up. DFS was defined as time to any recurrence or death, whichever occurred first. LRFS was defined as the time from the date of surgery to the date of treatment failure in the postsurgical tumor bed (Kim YJ et al., 2017 2017 ). Two authors reviewed study titles and abstracts, and studies that satisfied the inclusion criteria were retrieved for full-text evaluation. Data of the trials selected for detailed analysis were extracted independently by two authors, and disagreements were resolved by discussion. We extracted the following data from the selected studies: study period, center, total number of patients, age, sex, pathology, primary site of tumor, tumor-node-metastasis (TNM) stage, lymphovascular invasion, perineural invasion, resection margin status, lymph node (LN) status, total and fraction dose of ART, with concurrent chemotherapy or not, chemotherapy regimen, follow-up time, and HRs of OS, DFS, and LRFS of the two groups (Tables 1 1 and S1 S1 ). Also, we paid attention to radiation-induced toxicity. Assessment of study quality 2 Because the included studies were all retrospective, we used the Newcastle-Ottawa Scale (NOS) (Wells et al., 2019 2019 ) for assessing study quality. Studies that received 7–9 stars were considered to be high-quality, those that received 4–6 stars were defined as moderate-quality, and those that received 0–3 stars were regarded as low-quality (Table S2 S2 ). Statistical analysis 2 The primary outcome evaluated was OS, and the secondary outcomes were DFS and LRFS. Estimated HRs were calculated for OS, DFS, and LRFS. If the HR and its 95% confidence interval (CI) were not provided in the original article, they were calculated from available reported data as described previously (Tierney et al., 2007 2007 ). We calculated I 2 and Q to evaluate the heterogeneity of the included studies. If P was < 0.1 and I 2 was ≥50%, the heterogeneity was considered significant (Lau et al., 1997 1997 ; Higgins and Thompson, 2002 2002 ). When significant heterogeneity between studies was observed, the random-effects model was used; otherwise, the fixed-effects model was used (DerSimonian and Laird, 1986 1986 ). Publication bias was assessed using the Egger’s test by the funnel plot method (Egger et al., 1997 1997 ). Publication bias was considered to be present if the 2-tailed P value with Egger’s test was < 0.05. We also sub-grouped patients by resection margin status, LN metastasis, adjuvant therapy regimen (ACRT or ART), and the site of the tumor (perihilar or distal). Statistical analyses were done using Review Manager Version 5.3 software (Cochrane collaboration, Oxford, UK) and publication bars were derived with the STATA 11.0 package (Stata Corp., College Station, TX, USA). Results 1 Baseline characteristics of included studies 2 Our initial search identified 433 studies. We excluded 406 studies by title and abstract. We read the full-text of the remaining 27 articles and excluded 19; among which, seven were based on the National Cancer Database (NCDB)/Surveillance, Epidemiology, and End Results (SEER) database; one included patients undergoing intraoperative radiotherapy (IORT); two included patients undergoing intraluminal brachytherapy; three included patients who received ACT without ART in the adjuvant therapy group; five studies included intrahepatic cholangiocarcinoma, gallbladder cancer, or ampullary cancer, while the data for EHCC were not provided; one study received only three stars in the NOS scale and was regarded as low-quality. Ultimately, we identified eight articles (Itoh et al., 2005 2005 ; Hughes et al., 2007 2007 ; Gwak et al., 2010 2010 ; Matsuda et al., 2013 2013 ; Im et al., 2016 2016 ; Kim MY et al., 2016 2016 ; Kim YS et al., 2016 2016 ; Kim YJ et al., 2017 2017 ) that met our inclusion criteria. The inclusion process is described in Fig. 1 1 . In these eight articles, patients of the A(C)RT group received external beam radiotherapy, the mean dose ranging from 50.4 to 52.3 Gy. The concurrent chemotherapy agents used in the A(C)RT group included 5-fluorouracil (5-FU), capecitabine, and gemcitabine (Table 1 1 ). All eight studies were retrospectively designed; seven received 4–6 stars and the other one received 7 stars according to the NOS, and were evaluated as moderate-and high-quality studies, respectively (Table S2 S2 ). NCDB: National Cancer Database; SEER: Surveillance, Epidemiology, and End Results; IHCC: intrahepatic cholangiocarcinoma; NOS: Newcastle-Ottawa Scale A(C)RT: adjuvant (chemo)radiotherapy; TNM: tumor-node-metastasis; 5-FU: 5-fluorouracil; CRT: concurrent chemoradiotherapy; R0: negative resection margin; R1: microscopic positive resection margin; R2: macroscopic positive resection margin; LN: lymph node. Two articles contained three groups, comparing ACRT vs. surgery alone and ART vs. surgery alone; thus, they were calculated respectively (Im et al., 2016 2016 ; Kim YS et al., 2016 2016 ). Data are expressed as median (range), except that from Matsuda et al. ( 2013 2013 ) Study 2 Study period Center Patient number (A(C)RT vs. control) Primary site TNM stage Gwak et al., 2010 2010 2 1997 to 2005 Korea 31 vs. 47 Perihilar/distal I−III Hughes et al., 2007 2007 2 1994 to 2003 1970 to 1992 USA 34 vs. 30 Distal II−III Im et al., 2016 2016 # # 2 Jan. 2001 to Dec. 2010 Korea 49 vs. 168 Perihilar/distal I−III Im et al., 2016 2016 # # 2 Jan. 2001 to Dec. 2010 Korea 29 vs. 168 Perihilar/distal I−III Itoh et al., 2005 2005 2 Apr. 1994 to Mar. 2004 Japan 11 vs. 8 Perihilar/distal I−III Kim MY et al., 2016 2016 2 Jan. 2000 to Dec. 2013 Korea 19 vs. 33 Perihilar/distal I−III Kim YJ et al., 2017 2017 2 1997 to 2015 Korea 23 vs. 36 Perihilar/distal Kim YS et al., 2016 2016 # # 2 Jan. 2001 to Dec. 2013 Korea 20 vs. 102 Distal I−III Kim YS et al., 2016 2016 # # 2 Jan. 2001 to Dec. 2013 Korea 9 vs. 102 Distal I−III Matsuda et al., 2013 2013 2 Jan. 2000 to Mar. 2010 Japan 11 vs. 25 Perihilar/distal I−IVB 7 7 Study Radiation therapy dose (Gy) * * 2 Concurrent chemotherapy used in A(C)RT group CRT patients number in A(C)RT group Resection margin status LN status 7 Gwak et al., 2010 2010 50.4 (45.0−54.0) 2 5-FU based 16/31 R0/R1 N(+)/N(−) Hughes et al., 2007 2007 50.4 (40.0−54.0) 2 5-FU based 34/34 R0/R1/R2 N(+)/N(−) Im et al., 2016 2016 # # 50.4 (41.4−54.0) 2 5-FU/gemcitabine based 49/49 R0/R1/R2 N(+)/N(−) Im et al., 2016 2016 # # 50.4 (41.4−54.0) 2 No 0/29 R0/R1/R2 N(+)/N(−) Itoh et al., 2005 2005 52.3 (37.8−79.8) 2 No 0/11 R0/R1/R2 N(+)/N(−) Kim MY et al., 2016 2016 50.4 (45.0−54.0) 2 5-FU/gemcitabine based 12/19 R0/R1 N(+)/N(−) Kim YJ et al., 2017 2017 50.4 (45.0−61.0) 2 5-FU/gemcitabine based 15/23 R0/R1/R2 N(+)/N(−) Kim YS et al., 2016 2016 # # 2 5-FU/capecitabine/gemcitabine based 20/20 R0 N(+)/N(−) Kim YS et al., 2016 2016 # # 2 No 0/9 R0 N(+)/N(−) Matsuda et al., 2013 2013 46.0−60.0 2 Unknown 5/11 R0/R1 N(+)/N(−) Two articles contained three groups, comparing ACRT vs. surgery alone and ART vs. surgery alone; thus, they were calculated respectively (Im et al., 2016 2016 ; Kim YS et al., 2016 2016 ). OS data could be obtained from all included studies. As for DFS, Gwak et al. ( 2010 2010 ) provided data of A(C)RT vs. surgery alone in the microscopic positive resection margin (R1) and negative resection margin (R0) subgroups, which were calculated respectively. Im et al. ( 2016 2016 ) explored progression-free survival (PFS, defined as time from date of resection to first reported recurrence or death) instead of DFS. We included the PFS in DFS analysis since all the patients had undergone surgery. There were three articles reporting LRFS (Im et al., 2016 2016 ; Kim MY et al., 2016 2016 ; Kim YJ et al., 2017 2017 ). Efficacy of A(C)RT on OS 2 The meta-analysis of all relevant studies showed that the A(C)RT group had significantly better OS than the surgery alone group, although these trials showed significant heterogeneity (HR 0.69, 95% CI 0.48–0.97, P =0.03, I 2 =53%, random-effects model; Fig. 2 2 ). To explore further which patients benefited from A(C)RT, we performed subgroup analysis firstly according to the resection margin status and LN metastasis status because these are important factors affecting prognosis of EHCC. Patients with R1 resection margin could achieve a benefit from A(C)RT with low heterogeneity (HR 0.44, 95% CI 0.27–0.72, P =0.001, I 2 =0%, fixed-effects model; Fig. 3a 3a ). There was no statistical OS difference observed in the R0 subgroup comparing A(C)RT with surgery only (HR 0.98, 95% CI 0.30–3.19, P =0.98, I 2 =78%, random-effects model; Fig. 3b 3b ). Only one article provided data for patients who had positive LN metastasis (LN(+)), thus an LN(+) subgroup could not be performed. Significant OS benefit was found in patients who received concurrent ACRT (HR 0.40, 95% CI 0.26–0.62, P < 0.0001, I 2 =0%, fixed-effects model; Fig. 3c 3c ) while the result of ART without chemotherapy showed no significant benefit (HR 1.14, 95% CI 0.29–4.50, P =0.85, I 2 =86%, random-effects model; Fig. 3d 3d ). We also did subgroup analysis according to the site of tumor. Because of limited data, we could not analyze a perihilar subgroup. In the distal cholangiocarcinoma subgroup, we did not get a significant difference when adjuvant chemoradiation and adjuvant radiation were analyzed (HR 0.61, 95% CI 0.14–2.72, P =0.52, I 2 =85%, random-effects model; Fig. 3e 3e ). However, it is worth noting that in the three studies included, two studies were concurrent chemoradiotherapy vs. surgery while one study was ART alone vs. surgery. So we performed an ACRT for distal cholangiocarcinoma subgroup, and as we expected, significant OS benefit was observed (HR 0.29, 95% CI 0.13–0.64, P =0.002, I 2 =0%, fixed-effects model; Fig. 3f 3f ). HR: hazard ratio; OS: overall survival; A(C)RT: adjuvant (chemo)radiotherapy; SE: standard error; CI: confidence interval (a) R1 subgroup; (b) R0 subgroup; (c) ACRT subgroup; (d) ART subgroup; (e) ACRT and ART in distal EHCC subgroup; (f) ACRT only in distal EHCC subgroup. HR: hazard ratio; OS: overall survival; A(C)RT: adjuvant (chemo)radiotherapy; R1: positive resection margin; R0: negative resection margin; ACRT: adjuvant chemoradiotherapy; ART: adjuvant radiotherapy; EHCC: extrahepatic cholangiocarcinoma; SE: standard error; CI: confidence interval Efficacy of A(C)RT on DFS and LRFS 2 Significant differences in DFS were observed between patients who did and did not receive A(C)RT (HR 0.60, 95% CI 0.47–0.76, P < 0.0001, I 2 =26%, fixed-effects model; Fig. 4 4 ). In subgroup analysis, patients who underwent R1 resection achieved a significant benefit in DFS from A(C)RT (HR 0.38, 95% CI 0.18–0.78, P =0.008, I 2 =63%, random-effects model; Fig. 5 5 ). There were inadequate data to perform further subgroup analyses. HR: hazard ratio; DFS: disease-free survival; A(C)RT: adjuvant (chemo)radiotherapy; SE: standard error; CI: confidence interval HR: hazard ratio; DFS: disease-free survival; A(C)RT: adjuvant (chemo)radiotherapy; SE: standard error; CI: confidence interval Four studies provided information about LRFS. LRFS of EHCC also benefited from A(C)RT overall (HR 0.27, 95% CI 0.17–0.41, P < 0.00001, I 2 =0%, fixed-effects model; Fig. 6 6 ). HR: hazard ratio; LRFS: locoregional recurrence-free survival; A(C)RT: adjuvant (chemo)radiotherapy; SE: standard error; CI: confidence interval Treatment toxicity 2 Radiation-induced toxicities, as reported in the selected studies, are shown in Table S3 S3 . The grade of toxicity was scored according to the Common Terminology Criteria for Adverse Events v.3.0. Kim MY et al. ( 2016 2016 ) reported that two patients experienced grade 3 duodenal ulcer at 14 and 36 months after ART. Kim YJ et al. ( 2017 2017 ) showed two patients suffered grade 3 toxicity among 23 patients, one patient experiencing severe nausea and vomiting and the other patient having decreased platelet and white blood cell count. No other grade 3 or greater toxicity was observed. Publication bias 2 Funnel plots of publication bias are shown in Fig. 7 7 . There was no publication bias for OS (Begg’s P =0.929, Egger’s P =0.933), DFS (Begg’s P =0.452, Egger’s=0.257), or LRFS (Begg’s P =0.308, Egger’s=0.301). (a) OS; (b) DFS; (c) LRFS. OS: overall survival; DFS: disease-free survival; LRFS: locoregional recurrence-free survival; HR: hazard ratio; s.e.: standard error Discussion 1 Adjuvant therapy including ACT, ART, and ACRT as an important supplement to surgery has attracted the attention of many researchers in recent years, but the benefits remain unclear. A randomized, controlled, multi-center, phase III study compared capecitabine with observation following resection of biliary tract cancer; the OS primary endpoint in the intention-to-treat population did not reach statistical significance while the per-protocol OS and recurrence-free survival analyses showed benefit (Primrose et al., 2019 2019 ). A meta-analysis indicated that ACT could improve OS in perihilar cholangiocarcinoma patients, but distal cholangiocarcinoma patients gained no benefit from ACT (Wang et al., 2019 2019 ). A phase II trial Southwestern Oncology Group (SWOG) S0809 analyzed the effect of gemcitabine and capecitabine followed by concurrent capecitabine and radiotherapy in resected EHCC (68%) and gallbladder carcinoma (32%). The results of the trial implied that ACRT could be an effective and promising treatment method (Ben-Josef et al., 2015 2015 ). Therefore we sought to elucidate the role and effective subgroups of A(C)RT. To date, there were no data from RCTs. We asked the following questions: how much benefit can A(C)RT bring to patients? Which subgroup of patients could benefit from A(C)RT? Which regimen is the better option? Therefore, we analyzed the existing retrospective data and performed a meta-analysis, hoping to answer these questions. According to our analyses for all patients, A(C)RT significantly improved OS, DFS, and LRFS in EHCC. A study, based on NCDB, which included 8741 patients between 1998 and 2006, also argued that A(C)RT was associated with survival improvement in EHCC (Hoehn et al., 2015 2015 ). This is consistent with our meta-analysis. There were some previous studies not supporting our results. Two studies based on SEER database showed that A(C)RT was not associated with improvement in long-term OS (Shinohara et al., 2009 2009 ; Vern-Gross et al., 2011 2011 ). However, one of them admitted that because of the lack of some key data, including margin status and use of combined chemotherapy, in the SEER database, they failed to determine whether A(C)RT was beneficial for the local control in some subgroups (Vern-Gross et al., 2011 2011 ). The other argued that, in carefully selected patients, such as those with positive margins, A(C)RT would be of benefit (Shinohara et al., 2009 2009 ). R0 resection is related with good prognosis but could be achieved in fewer than one-third of EHCC patients, and local clearance (R0 or R1 status) was independently associated with survival (Khan et al., 2012 2012 ). Therefore, we did subgroup analysis, based on resection margins, to explore the value of A(C)RT in R1 and R0 subgroups and we found that A(C)RT was beneficial in the R1 but not the R0 subgroup. One study on distal cholangiocarcinoma also indicated that long-term survival may be achieved in patients with R0 resection (Chua et al., 2017 2017 ). A study by Lee et al. ( 2018 2018 ) compared patients who underwent R0 resection without any adjuvant treatment with those who underwent R1 resection but received A(C)RT; as a result, no significant survival difference was observed. This suggested that A(C)RT ameliorated the negative effect of microscopic positive resection margins, which is consistent with our meta-analysis. Compared with ART, ACRT can increase radiation sensitivity of tumor cells, control microscopic residual tumor growth, and reduce distant recurrence from hematogenous spread, thus reducing the recurrence risk (Jarnagin et al., 2003 2003 ; Im et al., 2016 2016 ; Sahai and Kumar, 2017 2017 ). At the same time, ACRT may be associated with more toxicity. ACRT rather than ART has become an indispensable treatment in other digestive tract tumors, such as rectal cancer (Yoon et al., 2019 2019 ) and gastric cancer (Yu et al., 2019 2019 ). In our analysis, the ACRT subgroup showed significant OS improvement in EHCC rather than ART. Similarly, the study of Kim YS et al. ( 2017 2017 ) showed that ACRT significantly improved recurrence-free survival than in the “surgery only” group and suggested that ACRT appeared to be an appropriate treatment in perihilar cholangiocarcinoma after complete resection. Prospective RCTs are needed to confirm the superiority of ACRT over ART. Prognosis of EHCC is influenced by location (Nakeeb et al., 1996 1996 ). So we also explore the impact of A(C)RT on EHCC in different sites. In the distal cholangiocarcinoma subgroup, three studies were included. Among them, two studies compared ACRT with observation after resection, and one study compared ART (no chemotherapy performed) with observation after resection (Hughes et al., 2007 2007 ; Kim YS et al., 2016 2016 ). No significant benefit was seen in distal EHCC when including studies with ACRT and ART, but significant outcomes were seen when analyzing ACRT studies only, which was consistent with the proposition that ACRT rather than ART would benefit, as discussed above. Kim YS et al. ( 2016 2016 ) pointed out that the small size of the ART group might influence the evaluation of the efficacy of ART, and the adverse effect of radiotherapy might be obvious in the small group of people. Despite this, he argued that the survival benefit in the ACRT group was statistically significant. We need more studies to assess the effect of ART alone, but current studies show that distal cholangiocarcinoma patients could benefit from ACRT (Kim YS et al., 2016 2016 ). So we argue that for patients with resected distal cholangiocarcinoma, ACRT should be recommended. We also tried to perform the perihilar subgroup, but we failed because of the limited data. Only a few patients in these study experienced treatment-induced toxicity, which could not be analyzed statistically. No patient-reported outcomes could be analyzed. We have described the toxicity in results, hoping to provide a reference for further research. The following limitations should be considered when interpreting our study. First, because of the lack of RCTs, all included studies were retrospectively designed, and thus randomized control could not be achieved and clinicopathologically detailed covariates were not adequately adjusted. The unequal radiation dose and chemotherapy drugs might also lead to heterogeneity. Second, because of the limited number of articles, we could not make more detailed subgroups such as perihilar or macroscopic residual (R2) resection. Third, no direct comparison to ACT could be made, so we could not answer the question of ACT vs. ACRT. Fourth, the analysis was done at the level of data available from the full text of the paper, so there might be biases and confounding factors that were unaccounted for. Also, the studies included were overwhelmingly from East Asian populations, so it was unclear how well these findings might be applied to other non-East Asian patient populations. Funnel plots of OS, DFS, and LRFS were basically symmetrical, suggesting no publication bias. We also did Egger’s regression test (Egger et al., 1997 1997 ) and Begg’s rank correlation test (Begg and Mazumdar, 1994 1994 ) to evaluate publication bias; the P values were all greater than 0.1, which confirmed the outcome of funnel plots. We look forward to RCTs being performed in the future. We also suggest that more retrospective data be summarized, from more different medical centers and in more detailed subgroups. For those who would undergo ACRT, we need to conduct clinical trials to find out the best regimen of concurrent chemotherapy and the best program of radiotherapy. For those who could not tolerate chemotherapy, we suggest ART be an option since current studies cannot rule out the potential benefit of ART. Conclusions 1 A(C)RT may improve OS, DFS, and LFRS in EHCC patients, especially in those with R1 resection margins. ACRT may be superior to ART especially in distal patients. References CB Begg , M Mazumdar . Operating characteristics of a rank correlation test for publication bias . Biometrics , 1994 . 50 ( 4 ): 1088 - 1101 . 7786990 DOI: 10.2307/2533446 http://doi.org/10.2307/2533446 . Operating characteristics of a rank correlation test for publication bias E Ben-Josef , KA Guthrie , AB El-Khoueiry , 等 . SWOG S0809: a phase II intergroup trial of adjuvant capecitabine and gemcitabine followed by radiotherapy and concurrent capecitabine in extrahepatic cholangiocarcinoma and gallbladder carcinoma . J Clin Oncol , 2015 . 33 ( 24 ): 2617 - 2622 . 25964250 DOI: 10.1200/jco.2014.60.2219 http://doi.org/10.1200/jco.2014.60.2219 . SWOG S0809: a phase II intergroup trial of adjuvant capecitabine and gemcitabine followed by radiotherapy and concurrent capecitabine in extrahepatic cholangiocarcinoma and gallbladder carcinoma M Bonet Beltrán , AS Allal , I Gich , 等 . Is adjuvant radiotherapy needed after curative resection of extrahepatic biliary tract cancers? A systematic review with a meta-analysis of observational studies . Cancer Treat Rev , 2012 . 38 ( 2 ): 111 - 119 . 21652148 DOI: 10.1016/j.ctrv.2011.05.003 http://doi.org/10.1016/j.ctrv.2011.05.003 . Is adjuvant radiotherapy needed after curative resection of extrahepatic biliary tract cancers? A systematic review with a meta-analysis of observational studies JA Bridgewater , KA Goodman , A Kalyan , 等 . Biliary tract cancer: epidemiology, radiotherapy, and molecular profiling . Am Soc Clin Oncol Educ Book , 2016 . 35 e194 - e203 . 27249723 DOI: 10.1200/edbk_160831 http://doi.org/10.1200/edbk_160831 . Biliary tract cancer: epidemiology, radiotherapy, and molecular profiling R Chaiteerakij , WS Harmsen , CR Marrero , 等 . A new clinically based staging system for perihilar cholangiocarcinoma . Am J Gastroenterol , 2014 . 109 ( 12 ): 1881 - 1890 . 25384902 DOI: 10.1038/ajg.2014.327 http://doi.org/10.1038/ajg.2014.327 . A new clinically based staging system for perihilar cholangiocarcinoma TC Chua , A Mittal , J Arena , 等 . Resection margin influences survival after pancreatoduodenectomy for distal cholangiocarcinoma . Am J Surg , 2017 . 213 ( 6 ): 1072 - 1076 . 27751529 DOI: 10.1016/j.amjsurg.2016.09.049 http://doi.org/10.1016/j.amjsurg.2016.09.049 . Resection margin influences survival after pancreatoduodenectomy for distal cholangiocarcinoma R DerSimonian , N Laird . Meta-analysis in clinical trials . Control Clin Trials , 1986 . 7 ( 3 ): 177 - 188 . 3802833 DOI: 10.1016/0197-2456(86)90046-2 http://doi.org/10.1016/0197-2456(86)90046-2 . Meta-analysis in clinical trials B Doherty , VE Nambudiri , WC Palmer . Update on the diagnosis and treatment of cholangiocarcinoma . Curr Gastroenterol Rep, 19:2 , 2017 . DOI: 10.1007/s11894-017-0542-4 http://doi.org/10.1007/s11894-017-0542-4 . Update on the diagnosis and treatment of cholangiocarcinoma M Egger , GD Smith , M Schneider , 等 . Bias in meta-analysis detected by a simple, graphical test . BMJ , 1997 . 315 ( 7109 ): 629 - 634 . 9310563 DOI: 10.1136/bmj.315.7109.629 http://doi.org/10.1136/bmj.315.7109.629 . Bias in meta-analysis detected by a simple, graphical test HK Gwak , WC Kim , HJ Kim , 等 . Extrahepatic bile duct cancers: surgery alone versus surgery plus postoperative radiation therapy . Int J Radiat Oncol Biol Phys , 2010 . 78 ( 1 ): 194 - 198 . 19910130 DOI: 10.1016/j.ijrobp.2009.07.003 http://doi.org/10.1016/j.ijrobp.2009.07.003 . Extrahepatic bile duct cancers: surgery alone versus surgery plus postoperative radiation therapy JPT Higgins , SG Thompson . Quantifying heterogeneity in a meta-analysis . Stat Med , 2002 . 21 ( 11 ): 1539 - 1558 . 12111919 DOI: 10.1002/sim.1186 http://doi.org/10.1002/sim.1186 . Quantifying heterogeneity in a meta-analysis RS Hoehn , K Wima , AE Ertel , 等 . Adjuvant chemotherapy and radiation therapy is associated with improved survival for patients with extrahepatic cholangiocarcinoma . Ann Surg Oncol , 2015 . 22 ( S3 ): 1133 - 1139 . 25316488 DOI: 10.1245/s10434-015-4599-8 http://doi.org/10.1245/s10434-015-4599-8 . Adjuvant chemotherapy and radiation therapy is associated with improved survival for patients with extrahepatic cholangiocarcinoma MA Hughes , DA Frassica , CJ Yeo , 等 . Adjuvant concurrent chemoradiation for adenocarcinoma of the distal common bile duct . Int J Radiat Oncol Biol Phys , 2007 . 68 ( 1 ): 178 - 182 . 17276614 DOI: 10.1016/j.ijrobp.2006.11.048 http://doi.org/10.1016/j.ijrobp.2006.11.048 . Adjuvant concurrent chemoradiation for adenocarcinoma of the distal common bile duct JH Im , J Seong , IJ Lee , 等 . Surgery alone versus surgery followed by chemotherapy and radiotherapy in resected extrahepatic bile duct cancer: treatment outcome analysis of 336 patients . Cancer Res Treat , 2016 . 48 ( 2 ): 583 - 595 . 26323644 DOI: 10.4143/crt.2015.091 http://doi.org/10.4143/crt.2015.091 . Surgery alone versus surgery followed by chemotherapy and radiotherapy in resected extrahepatic bile duct cancer: treatment outcome analysis of 336 patients S Ishihara , A Horiguchi , S Miyakawa , 等 . Biliary tract cancer registry in Japan from 2008 to 2013 . J Hepato-Biliary-Pan Sci , 2016 . 23 ( 3 ): 149 - 157 . DOI: 10.1002/jhbp.314 http://doi.org/10.1002/jhbp.314 . Biliary tract cancer registry in Japan from 2008 to 2013 H Itoh , K Nishijima , Y Kurosaka , 等 . Magnitude of combination therapy of radical resection and external beam radiotherapy for patients with carcinomas of the extrahepatic bile duct and gallbladder . Dig Dis Sci , 2005 . 50 ( 12 ): 2231 - 2242 . 16416167 DOI: 10.1007/s10620-005-3040-8 http://doi.org/10.1007/s10620-005-3040-8 . Magnitude of combination therapy of radical resection and external beam radiotherapy for patients with carcinomas of the extrahepatic bile duct and gallbladder WR Jarnagin , L Ruo , SA Little , 等 . Patterns of initial disease recurrence after resection of gallbladder carcinoma and hilar cholangiocarcinoma: implications for adjuvant therapeutic strategies . Cancer , 2003 . 98 ( 8 ): 1689 - 1700 . 14534886 DOI: 10.1002/cncr.11699 http://doi.org/10.1002/cncr.11699 . Patterns of initial disease recurrence after resection of gallbladder carcinoma and hilar cholangiocarcinoma: implications for adjuvant therapeutic strategies SA Khan , BR Davidson , RD Goldin , 等 . Guidelines for the diagnosis and treatment of cholangiocarcinoma: an update . Gut , 2012 . 61 ( 12 ): 1657 - 1669 . 22895392 DOI: 10.1136/gutjnl-2011-301748 http://doi.org/10.1136/gutjnl-2011-301748 . Guidelines for the diagnosis and treatment of cholangiocarcinoma: an update MY Kim , JH Kim , Y Kim , 等 . Postoperative radiotherapy appeared to improve the disease free survival rate of patients with extrahepatic bile duct cancer at high risk of loco-regional recurrence . Radiat Oncol J , 2016 . 34 ( 4 ): 297 - 304 . 27951624 DOI: 10.3857/roj.2016.01879 http://doi.org/10.3857/roj.2016.01879 . Postoperative radiotherapy appeared to improve the disease free survival rate of patients with extrahepatic bile duct cancer at high risk of loco-regional recurrence YJ Kim , K Kim , SK Min , 等 . Role of adjuvant radiotherapy for localized extrahepatic bile duct cancer . Br J Radiol , 2017 . 90 ( 1071 ): 20160807 28118028 DOI: 10.1259/bjr.20160807 http://doi.org/10.1259/bjr.20160807 . Role of adjuvant radiotherapy for localized extrahepatic bile duct cancer YS Kim , IG Hwang , SE Park , 等 . Role of adjuvant therapy after R0 resection for patients with distal cholangiocarcinoma . Cancer Chemother Pharmacol , 2016 . 77 ( 5 ): 979 - 985 . 27017615 DOI: 10.1007/s00280-016-3014-x http://doi.org/10.1007/s00280-016-3014-x . Role of adjuvant therapy after R0 resection for patients with distal cholangiocarcinoma YS Kim , SY Oh , SI Go , 等 . The role of adjuvant therapy after R0 resection for patients with intrahepatic and perihilar cholangiocarcinomas . Cancer Chemother Pharmacol , 2017 . 79 ( 1 ): 99 - 106 . 27928598 DOI: 10.1007/s00280-016-3206-4 http://doi.org/10.1007/s00280-016-3206-4 . The role of adjuvant therapy after R0 resection for patients with intrahepatic and perihilar cholangiocarcinomas J Lau , JPA Ioannidis , CH Schmid . Quantitative synthesis in systematic reviews . Ann Int Med , 1997 . 127 ( 9 ): 820 - 826 . 9382404 DOI: 10.7326/0003-4819-127-9-199711010-00008 http://doi.org/10.7326/0003-4819-127-9-199711010-00008 . Quantitative synthesis in systematic reviews J Lee , SH Kang , OK Noh , 等 . Adjuvant concurrent chemoradiation therapy in patients with microscopic residual tumor after curative resection for extrahepatic cholangiocarcinoma . Clin Transl Oncol , 2018 . 20 ( 8 ): 1011 - 1017 . 29256155 DOI: 10.1007/s12094-017-1815-y http://doi.org/10.1007/s12094-017-1815-y . Adjuvant concurrent chemoradiation therapy in patients with microscopic residual tumor after curative resection for extrahepatic cholangiocarcinoma T Matsuda , H Fujita , N Harada , 等 . Impact of adjuvant radiation therapy for microscopic residual tumor after resection of extrahepatic bile duct cancer . Am J Clin Oncol , 2013 . 36 ( 5 ): 461 - 465 . 22706178 DOI: 10.1097/COC.0b013e31825494ab http://doi.org/10.1097/COC.0b013e31825494ab . Impact of adjuvant radiation therapy for microscopic residual tumor after resection of extrahepatic bile duct cancer A Nakeeb , HA Pitt , TA Sohn , 等 . Cholangiocarcinoma: a spectrum of intrahepatic, perihilar, and distal tumors . Ann Surg , 1996 . 224 ( 4 ): 463 - 475 . DOI: 10.1097/00000658-199610000-00005 http://doi.org/10.1097/00000658-199610000-00005 . Cholangiocarcinoma: a spectrum of intrahepatic, perihilar, and distal tumors NCCN . NCCN Clinical Practice Guidelines in Oncology-Hepatobiliary Cancers (2019 Version 1) . 2019 . https://www. nccn.org/professionals/physician_gls/pdf/hepatobiliary.pdf [Accessed on Nov. 8, 2019] . . NCCN Clinical Practice Guidelines in Oncology-Hepatobiliary Cancers (2019 Version 1) JN Primrose , RP Fox , DH Palmer , 等 . Capecitabine compared with observation in resected biliary tract cancer (BILCAP): a randomised, controlled, multicentre, phase 3 study . Lancet Oncol , 2019 . 20 ( 5 ): 663 - 673 . DOI: 10.1016/s1470-2045(18)30915-x http://doi.org/10.1016/s1470-2045(18)30915-x . Capecitabine compared with observation in resected biliary tract cancer (BILCAP): a randomised, controlled, multicentre, phase 3 study P Sahai , S Kumar . External radiotherapy and brachytherapy in the management of extrahepatic and intrahepatic cholangiocarcinoma: available evidence . Br J Radiol , 2017 . 90 ( 1076 ): 20170061 DOI: 10.1259/bjr.20170061 http://doi.org/10.1259/bjr.20170061 . External radiotherapy and brachytherapy in the management of extrahepatic and intrahepatic cholangiocarcinoma: available evidence ET Shinohara , N Mitra , MY Guo , 等 . Radiotherapy is associated with improved survival in adjuvant and palliative treatment of extrahepatic cholangiocarcinomas . Int J Radiat Oncol Biol Phys , 2009 . 74 ( 4 ): 1191 - 1198 . DOI: 10.1016/j.ijrobp.2008.09.017 http://doi.org/10.1016/j.ijrobp.2008.09.017 . Radiotherapy is associated with improved survival in adjuvant and palliative treatment of extrahepatic cholangiocarcinomas JF Tierney , LA Stewart , D Ghersi , 等 . Practical methods for incorporating summary time-to-event data into meta-analysis . 2007 . Trials, 8:16 . DOI: 10.1186/1745-6215-8-16 http://doi.org/10.1186/1745-6215-8-16 . Practical methods for incorporating summary time-to-event data into meta-analysis TZ Vern-Gross , AT Shivnani , K Chen , 等 . Survival outcomes in resected extrahepatic cholangiocarcinoma: effect of adjuvant radiotherapy in a surveillance, epidemiology, and end results analysis . Int J Radiat Oncol Biol Phys , 2011 . 81 ( 1 ): 189 - 198 . DOI: 10.1016/j.ijrobp.2010.05.001 http://doi.org/10.1016/j.ijrobp.2010.05.001 . Survival outcomes in resected extrahepatic cholangiocarcinoma: effect of adjuvant radiotherapy in a surveillance, epidemiology, and end results analysis ML Wang , ZY Ke , S Yin , 等 . The effect of adjuvant chemotherapy in resectable cholangiocarcinoma: a meta-analysis and systematic review . Hepatobiliary Pancreat Dis Int , 2019 . 18 ( 2 ): 110 - 116 . DOI: 10.1016/j.hbpd.2018.11.001 http://doi.org/10.1016/j.hbpd.2018.11.001 . The effect of adjuvant chemotherapy in resectable cholangiocarcinoma: a meta-analysis and systematic review GA Wells , B Shea , D O'Connell , 等 . The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analyses . 2019 . http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp [Accessed on Oct. 1, 2019] . . The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analyses JE Yoon , SY Lee , HD Kwak , 等 . Oncologic outcomes of postoperative chemoradiotherapy versus chemotherapy alone in stage II and III upper rectal cancer . Ann Coloproctol , 2019 . 35 ( 3 ): 137 - 143 . DOI: 10.3393/ac.2018.09.28 http://doi.org/10.3393/ac.2018.09.28 . Oncologic outcomes of postoperative chemoradiotherapy versus chemotherapy alone in stage II and III upper rectal cancer JI Yu , DH Lim , J Lee , 等 . Clinical outcomes and the role of adjuvant concurrent chemoradiation therapy in D2-resected LN-positive young patients (≤45 years) with gastric cancer . Anticancer Res , 2019 . 39 ( 10 ): 5811 - 5820 . . Clinical outcomes and the role of adjuvant concurrent chemoradiation therapy in D2-resected LN-positive young patients (≤45 years) with gastric cancer Compliance with ethics guidelines Xin-qi SHI, Jing-yu ZHANG, Hua TIAN, Ling-na TANG, and Ai-lin LI declare that they have no conflict of interest. This article does not contain any studies with human or animal subjects performed by any of the authors. 10.1631/jzus.B1900691 Project supported by the National Natural Science Foundation of China (No. 81402520) 2019-12-08 2020-03-20 2022-01-05 Copyright © Zhejiang University and Springer-Verlag GmbH Germany, part of Springer Nature 2020 2020 Journal of Zhejiang University-SCIENCE B (Biomedicine and Biotechnology) 07 21