Introduction
Cholangiocarcinomas arise from either the intrahepatic or extrahepatic bile ducts and are typically diagnosed at an advanced stage, presenting significant treatment challenges [1]. Radical surgery stands as the primary and sole curative treatment for patients with resectable tumors, followed by adjuvant chemotherapy (CHT) and radiotherapy (RT) due to the high rates of postoperative recurrence [2]. Findings from a comprehensive national cancer database study demonstrate that adjuvant chemoradiotherapy enhances survival among patients with resected biliary adenocarcinoma [3]. Nevertheless, the prognosis remains poor, with data from the RARECAREnet study indicating that even among patients undergoing surgical intervention, the documented 5-year survival rates remain below 20% [4].
While surgery remains the primary treatment, RT with or without systemic therapy is crucial for patients who are medically unfit for surgery or have unresectable tumors [5,6]. Both stereotactic body RT (SBRT) and conventionally fractionated RT (CFRT) are viable treatments for unresectable extrahepatic cholangiocarcinoma. SBRT is typically used for smaller target volumes due to its higher fraction doses, often excluding regional lymphatics from the target volumes. In contrast, CFRT typically encompasses regional lymphatics along with the primary tumor. However, there is a lack of randomized studies directly comparing SBRT and CFRT for unresectable extrahepatic cholangiocarcinoma. This study aims to compare the efficacy of SBRT and CFRT in treating unresectable extrahepatic cholangiocarcinoma.
Materials and Methods
1. Patients
The medical records of patients diagnosed with either intrahepatic or extrahepatic cholangiocarcinoma who received RT as part of multidisciplinary management at our department between 2000 and 2023 were retrospectively evaluated from individual patient files, archive records, and the electronic system of our hospital. Patients with resectable tumors, had intrahepatic cholangiocarcinoma, had lymphatic and/or distant metastasis at initial diagnosis, and those without follow-up information were excluded from the analysis. Following these criteria, 18 patients with unresectable localized extrahepatic cholangiocarcinoma definitively treated with either CFRT (n = 10) or SBRT (n = 8) were included.
2. Radiotherapy
All treatment decisions were made based on individual patient and tumor characteristics during the weekly multidisciplinary gastrointestinal oncology tumor board meetings at our hospital. All patients were deemed to have unresectable tumors, radiologically. During RT planning, four-dimensional simulation computed tomography (CT) scans were conducted for all patients to accommodate potential respiratory motions affecting the tumor. Target volumes for RT were delineated by integrating magnetic resonance imaging (MRI) and/or positron emission tomography/CT images with the simulation CT scans of the patients. The internal target volume (ITV) was contoured to encompass the entire macroscopic tumor extent for either CFRT or SBRT. For SBRT, a planning target volume (PTV) was generated as ITV + 3–5 mm. No elective nodal irradiation (ENI) was administered to SBRT patients. For patients undergoing CFRT, the clinical target volume (CTV) included ITV + 5 mm along with hepatoduodenal, common hepatic, periportal, pancreatoduodenal, and para-aortic lymphatics, and the PTV was created as CTV + 3–5 mm. Fig. 1 shows illustrative plan examples for CFRT and SBRT.
3. Toxicity and follow-up
Common Terminology Criteria for Adverse Events version 5.0 was used for evaluation of the treatment related toxicities. Patients underwent monitoring every 3 months for the initial 2 years, followed by biannual checks for the subsequent 3 years, and annual assessments thereafter. Thorax and abdomen CT and/or MRI was performed at follow-ups, in addition to the routine blood checks.
4. Statistical analysis
All statistical analyses were conducted using Statistical Package for the Social Sciences version 23.0 (IBM Corp., Armonk, NY, USA). In the analysis of variables between groups, the choice of statistical test depends on the distribution of the data. The independent samples t-test is used for normally distributed data, while the Mann-Whitney U test or Wilcoxon rank-sum test are employed for data that do not follow a normal distribution. Treatment response was evaluated by Response Evaluation Criteria in Solid Tumor 1.1 criteria [7]. Local control (LC) was defined as the absence of progression in the primary tumor and regional control (RC) was defined as the absence of regional lymphatic recurrences during follow-up. Local failure referred to progression at the irradiated tumor site, whereas regional failure indicated lymph node recurrence during the follow-up period. Time-related events were defined from the completion of RT to the last follow-up, death, or recurrence, whichever occurred first. Survival analysis was performed using Kaplan-Meier estimates and log-rank for comparison. A p-value of less than 0.05 was considered statistically significant. Multivariate analysis was not performed due to the low number of patients.
Results
1. Patient, tumor, and treatment characteristics
Patient, tumor, and treatment characteristics were summarized in Table 1. The median age of the patients was 55 years (range, 31 to 85). Eleven patients were female, while seven patients were male. All patients presented with cT4N0M0 disease as per the eighth American Joint Committee on Cancer staging system, with the majority diagnosed with Klatskin tumors (n = 16, 88.8%). In all patients, percutaneous stent placement was utilized for biliary drainage purposes. No patient had been treated solely with RT, and all had received neoadjuvant, concurrent, and/or adjuvant CHT.
Characteristics were well-balanced between CFRT and SBRT groups, except for administration of concurrent CHT. The rate of concurrent CHT administration in patients treated with CFRT was significantly higher compared to those treated with SBRT (100% vs. 25.0%, p = 0.021). The median total doses of CFRT and SBRT were 50.4 Gy (range, 45 to 59.4) in 25–33 fractions and 37.5 Gy (range, 27.5 to 50) in 3–5 fractions, respectively. The median biologically effective dose value calculated by assuming an α/β ratio of 10 (BED10Gy) of CFRT and SBRT groups were 59.5 Gy (range, 53.1 to 70.1) and 65.6 Gy (range, 42.6 to 112.5), respectively. The median tumor size was similar between the CFRT and SBRT groups. The concurrent CHT regimen is capecitabine (850 mg/m2 twice daily) for all patients. The most commonly used CHT regimen for both neoadjuvant (n = 7, 77.7%), and adjuvant (n = 11, 78.5%) treatment was gemcitabine (1,000 mg/m2) and cisplatin (25 mg/m2) with a median of 6 cycles administered.
2. Disease outcomes
In the first response assessment after RT, five patients (62.5%) in the SBRT arm exhibited a partial response (PR), and three patients (37.5%) exhibited stable disease (SD). In the CFRT arm, four patients (40.0%) exhibited a PR, and six patients (60.0%) exhibited SD. The median follow-up was 22 months (range, 7 to 138). Throughout the follow-up period, local failure occurred in one patient (12.5%) in the SBRT group and six patients (60.0%) in the CFRT group. Conversely, regional failure developed in five patients (62.5%) in the SBRT group and in two patients (20.0%) in the CFRT group. All regional failures in both the CFRT and SBRT groups occurred in the para-aortic region, with all failures occurring outside the PTV for SBRT and within the PTV for CFRT. In patients receiving SBRT, the LC rate was higher compared to those receiving CFRT (p = 0.041), whereas the RC rate was lower (p = 0.047). None of the patients underwent orthotopic liver transplantation or tumor resection during their follow-up. Fig. 2 depicts the LC and RC rates according to the RT technique.
During follow-up, all patients developed distant metastases (DM). The most common site of DM was the peritoneum (n = 10), followed by the lung (n = 3), pleura (n = 2), liver (n = 2), and bone (n = 1). The median time between RT and the development of DM is 9 months (range, 4 to 49). Two-year overall survival (OS), local-regional failure-free survival (LRFFS), and DM-free survival (DMFS) rates were 35.4%, 30.1%, and 24.7%, respectively. No patients succumbed to primary tumor-related complications; all observed fatalities were due to DMs.
3. Prognostic factors
Results of univariate analysis were summarized in Table 2. Age was the only significant prognostic factor for all survival rates. Patients ≤55 years old had significantly increased 2-year rates of OS (63.4% vs 11.5%, p = 0.031), LRFFS (50.8% vs. 0.0%, p = 0.012), and DMFS (50.1% vs.0.0%, p = 0.007) as compared with patients >55 years old. None of the other parameters were found to be prognostic for any of the survival rates.
4. Toxicities
None of the patients experienced either acute or late RT-related severe (≥grade 3) toxicity. The most common RT-related acute toxicities were fatigue (n = 13, 72.2%), nausea (n = 11, 61.1%), abdominal pain (n = 7, 38.8%), and diarrhea (n = 7, 38.8%), all ≤ grade 2. The toxicity rates were similar between patients treated with SBRT and those treated with CFRT + ENI. None of the patients experienced gastroduodenal late toxicity, such as stricture or bleeding.
Discussion and Conclusion
In this single-center study, we observed that SBRT without ENI provides better LC but worse RC than CFRT for patients with unresectable extrahepatic cholangiocarcinoma. However, the most common failure pattern in our cohort was DM and either RT technique or administration of ENI did not have an impact on survival.
When treating patients with extrahepatic unresectable cholangiocarcinomas, relying solely on systemic therapies might not be the optimal approach due to the common occurrence of symptoms such as pain and jaundice, which stem from the tumor’s location, as well as potential life-threatening complications like biliary obstruction, cholangitis, and liver dysfunction [8]. Hence, it is essential to prioritize a multimodal treatment approach that combines systemic therapies targeted at eliminating micrometastases for systemic disease control, along with local treatments aimed at controlling the primary tumor. A large analysis of a Surveillance, Epidemiology, and End Results database, involving 4,758 extrahepatic cholangiocarcinoma patients, revealed that administering RT, even for palliative purposes, conferred a significant OS benefit compared to patients who did not undergo surgery or RT (median OS, 9 vs. 4 months) [9]. While combining RT with systemic therapy can enhance prognosis and offer relief from tumor-related local symptoms, determining the most effective RT technique, field, and dose for this patient group remains unclear [10]. Both CFRT and SBRT are suitable RT techniques for treating unresectable extrahepatic cholangiocarcinoma, each with its own distinct characteristics. While CFRT applies treatments with low fraction doses over several weeks, SBRT administers treatments with high fraction doses, usually over a few days, making SBRT ideal for treating small and well-defined tumors.
There's a lack of randomized prospective studies directly comparing SBRT and CFRT for unresectable extrahepatic cholangiocarcinoma. Studies in the literature have reported median OS durations ranging from 9 to 16 months with CFRT doses ranging between 45 and 54 Gy [11-14]. Moreover, studies have demonstrated that dose escalation with intraluminal brachytherapy, following CFRT, improves LC [15,16]. In modern era, dose escalation can be accomplished through a non-invasive and advanced RT technique like SBRT, allowing for increased dose within the tumor while minimizing toxicity through rapid dose fall-of outside the target volume. However, despite some retrospective studies investigating the role of SBRT in patients with extrahepatic unresectable cholangiocarcinoma, there remains a notable absence of high-quality evidence in the literature [17-19]. Although Kopek et al. [17] achieved approximately 10 months of OS with a 45 Gy SBRT delivered in 3 fractions, they also reported high rates of duodenal toxicity with their regimen. On the other hand, Momm et al. [18] and Polistina et al. [19] reported favorable treatment responses and lower toxicity rates with SBRT regimens characterized by lower fraction doses and higher number of fractions. In a relatively more recent series, Kozak et al. [20] from Stanford Cancer Institute reported oncological outcomes and toxicity rates of 40 patients treated with a median 40 Gy dose of SBRT in 1–5 fractions, with 15 extrahepatic tumors. They noted that the median survival of patients was 23 months, and toxicities were deemed acceptable. Moreover, they highlighted that the majority of severe late toxicities were secondary to the tumor rather than SBRT itself. In our series, the median OS was 21 months, and none of the patients experienced either acute or late ≥ grade 3 toxicity, which is consistent with the Stanford’s experience.
Another crucial finding of our study is the achievement of higher LC with SBRT compared to CFRT. Considering studies in the literature demonstrating improved oncological outcomes with increasing BED values, it is worth noting that the median BED10Gy value in patients treated with SBRT in our cohort was higher than that in patients treated with CFRT, which may explain the increased LC rates [21]. However, despite achieving higher LC with SBRT in our study, the likely absence of ENI may have resulted in decreased RC compared to CFRT. In the previously mentioned study by Kozak et al. [20], it was also reported that perihilar tumors had a 1-year regional failure rate of 24%, and there was an associated increase in DM rates among patients experiencing regional failure. The authors also suggested that ENI could potentially be beneficial in this patient group.
According to our institutional policy, we only administer ENI to patients receiving CFRT, due to the concern of toxicity. However, as we know from pancreatic cancer data, the administration of upper abdominal 25 Gy in 5 fractions ENI is feasible and has been shown to reduce regional failure rates, with a minimally increased rate of acute toxicity [22]. In our cohort as well, we believe that the increased rate of regional failure observed in patients treated with SBRT could potentially be improved with the implementation of ENI. Therefore, the SBRT + ENI approach appears to be a promising and worthy avenue for investigation to improve the oncological outcomes of patients diagnosed with unresectable extrahepatic cholangiocarcinoma. Although DMs are the primary determinant of survival in our study, we propose that modern systemic treatments, such as immune checkpoint inhibitors and targeted therapies, will highlight the importance of locoregional control in this infrequently encountered patient group.
While our study highlights the role of RT technique and ENI for patients with unresectable extrahepatic cholangiocarcinoma, it has several limitations. The small number of patients and the short follow-up duration are constraints that limit the generalizability of the findings. Due to the retrospective design of the study, the impact of patients' baseline performance scores and carbohydrate antigen 19-9 values on oncological outcomes could not be examined. Additionally, it should be noted that patients in the CFRT arm received more concurrent CHT compared to SBRT, which should be taken into account when evaluating the rates of lower rate of regional failure. However, cholangiocarcinoma is a relatively rare and rapidly fatal condition. Therefore, we believe that evaluating various therapeutic approaches contributes significantly to the existing body of literature, particularly regarding unresectable tumors.
In conclusion, the prognosis of patients with unresectable extrahepatic cholangiocarcinoma remains bleak, highlighting the need for innovative treatment strategies. Although SBRT may achieve higher LC compared to CFRT, the absence of ENI negatively affects RC rates. Hence, there is a critical need for studies exploring hypofractionated RT schemes, focusing on delivering high radiation doses to the primary tumor through SBRT while administering lower doses of radiation to the regional lymphatics, electively.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Print 
