Stereotactic radiotherapy in intrahepatic cholangiocarcinoma: A systematic review.

Among biliary tract cancers, intrahepatic cholangiocarcinoma (ICC) has different characteristics compared with those in other sites. Current guidelines suggest several treatment options for ICC, including stereotactic body radiation therapy (SBRT). However, the role of SBRT in locally advanced ICC is unclear. The aim of the present study was to present a systematic review on the efficacy and safety of SBRT in ICC. A systematic review based on the PRISMA methodology was performed. Only papers reporting outcomes in terms of overall survival (OS) after SBRT in inoperable patients with ICC were included. Secondary aims were local control (LC), progression-free survival (PFS) and treatment-related toxicity. Six papers (145 patients) were included in the present analysis. SBRT was frequently used as a salvage treatment, since 28.6-66.7% of patients received previous systemic or local treatments. The median SBRT dose was 45 Gy delivered in 3-5 fractions. The median follow-up was 16 months, and median OS time was 14 months (range, 10-48 months). In one of the included studies, SBRT was significantly superior in terms of OS compared with both chemoradiation and trans-arterial-radio-embolization. The 1-year LC rate was 85% in one study, and 1-year PFS rates were 50 and 68% in two studies, respectively. Toxicity was generally not reported in detail or was reported including other sites of biliary cancers. Overall, limited evidence was available on the efficacy of SBRT in ICC, which should be further investigated in prospective studies with a larger number of patients. However, based on the available data, SBRT seems to produce similar results compared with other ICC treatments, with the advantage of being a very short and non-invasive therapy. Therefore, SBRT should be considered in selected patients with ICC. Copyright: © Bisello et al.

Introduction

Intrahepatic cholangiocarcinoma (ICC) is an aggressive disease representing the second most common liver malignancy (1). The incidence of ICC is increasing worlwide (2). Surgery is considered a potential option with an overall 5-year survival of about 25-30%. (3,4). Gemcitabine-based chemotherapy (CT) represents the standard therapy in unresectable or metastatic biliary cancers with reported median overall survival (OS) of approximately 12 months (5,6) Particularly, in case of unresectable ICC, both CT or concurrent chemoradiation (CRT) (7,8) are treatment options.

Currently, different local therapies are available for inoperable ICC. The role of trans-arterial chemo-embolization (TACE) has been investigated (9). A meta-analysis on TACE in ICC treatment has been recently published, demonstrating a survival benefit compared to standard systemic therapies, with a good toxicity profile (10) Similarly, radiofrequency ablation resulted an effective treatment option in unresectable ICC. In fact, it was investigated in a systematic review and meta-analysis, reported 82 and 47% 1-year and 3-year survival rates, respectively (11).

Stereotactic body radiation therapy (SBRT) is a modern radiotherapy technique delivering ablative radiation doses while sparing the surrounding normal tissues. Despite surrounding organs at risk (OAR) and particularly liver radiosensitivity, SBRT has been considered as a possible therapeutic option in inoperable ICC. Two systematic reviews explored the role of SBRT in cholangiocarcinoma, but systematic reviews specifically addressed to ICC are lacking (12,13). However, the latter have different characteristics compared to tumors in other sites of the biliary tract, both in terms of routes of spread and clinical management. Therefore, even considering the scarce availability of clear guidelines in this setting, we considered it important to analyze the role of this technique (effective in other gastrointestinal neoplasms, such as HCC and pancreatic carcinomas) in the specific field of intrahepatic biliary carcinomas. The aim of this analysis is to present a systematic review of the available literature in the setting of SBRT in ICC.

Materials and methods

Inclusion criteria

We included in this analysis prospective and retrospective studies reporting OS after SBRT in inoperable ICC. Only English language written papers were considered. Only studies enrolling at least 10 patients with inoperable, non-metastatic, primary, or recurrent ICC were considered. Papers including hepatocellular carcinoma (HCC) patients were considered only if results on subjects with ICC were reported separately. Studies regarding SBRT planned and performed with any technique and device were included in the analysis even if preceded or followed by systemic treatments or other local therapies.

Studies on adjuvant SBRT after radical surgery were excluded. Analyses on preclinical models, planning studies, study protocols, case reports, letters and commentaries, editorials, systematic reviews, and meta-analyses were excluded. Primary endpoint of the analysis was OS. Secondary endpoints were local control (LC), progression-free survival (PFS), and treatment related toxicity.

Literature search strategy

A systematic search was performed on PubMed database on June 1st 2020 including papers published from the earliest date until 31st May 2020. The following search strategy was used: (‘cholangiocarcinoma’ OR ‘cholangiocarcinomas’ OR (‘intrahepatic’ AND ‘cholangiocarcinoma’) OR ‘intrahepatic cholangiocarcinoma’) AND (radiosurgery OR ‘stereotactic body radiotherapy’ OR ‘stereotactic body radiation therapy’ OR ‘stereotactic body radiosurgery’ OR ‘stereotactic radiotherapy’ OR ‘stereotactic radiation therapy’ OR ‘stereotactic radiosurgery’ OR stereotaxis OR stereotaxic OR sbrt OR sbrs OR sabr OR ‘stereotactic ablative’). The analysis was based on the PRISMA methodology (14).

From the first search on PubMed, title, abstracts, and keywords were independently analyzed by two authors (ACC, FB). Any disagreement was solved by the senior author (RA). A second search based on previous reviews and meta-analyses was performed by two other authors (SB, AA). Full-text evaluation based on the above-mentioned inclusion criteria was independently carried out by the latter two authors. Data on tumor, treatments characteristic, and outcomes were separately collected by two authors (SB and FB), with the supervision of AA. Any disagreement was resolved by a consensus-based discussion with the 2 senior authors (SiC and MB). Draft correction and critical revision were subsequently performed by all the author, as expressed below.

Results

Search result

Ninety-three papers were identified from PubMed. Other 74 papers were identified by checking the references list. After duplicates removal, based on a first screening of titles and abstracts, 47 records were excluded. Forty-six papers were subjected to full-text evaluation and 40 were excluded for the reasons described in Fig. 1. Only 6 papers met our inclusion criteria. The flowchart of papers selection is detailed in Fig. 1.

Figure 1

Flowchart of paper selection. After an initial evaluation and removal of duplicates, 93 studies were first selected. Only English-written papers reporting the outcomes of ≥10 patients with unresectable ICC were considered. Outcomes must be referred to SBRT for ICC only. If other diseases or techniques were involved, results must be expressed separately. ICC, intrahepatic cholangiocarcinoma; SBRT, stereotactic body radiation therapy.

Literature review

Shen et al (15) published the results of a prospective analysis on 28 patients with locally advanced, non-metastatic ICC. All patients underwent SBRT (median prescribed dose: 45 Gy), delivered in three to five fractions using a robotic linear accelerator. Median OS and PFS were 15 and 11 months, respectively. A total of 15 patients (53.7%) manifested grade ≥3 acute toxicity. A higher overall response rate was recorded in patients receiving a biological effective dose (BED) >100 Gy (52.4 vs. 0.0%, P=0.023).

Weiner et al (16) published the results of a phase I/II trial on SBRT in patients with unresectable HCC or ICC. Twelve patients had ICC and two had a bi-phenotypic disease. Median dose was 55 Gy delivered in 5 fractions. With a median follow-up of 8.8 months, median and 1-year OS were 13.2 months and 51.0%, respectively. Seventeen patients (65.4%) showed grade ≥3 acute toxicity in the whole population (HCC and ICC patients).

Klein et al (17) reported a prospective assessment of quality of life in patients with liver tumors treated with SBRT (24-60 Gy in 6 fractions). Physical and functional items were recorded during and after treatment. Median OS was 12.1 months in the ICC sub-group (20 patients).

Jung et al (18) retrospectively analyzed 58 patients with primary or recurrent intra- or extra-hepatic cholangiocarcinoma treated with SBRT. Thirty-three patients had ICC. Considering the whole population included in the analysis, 91% (53/58) of patient received SBRT alone (median dose: 45 Gy in 3-5 fractions) while five patients (9%) received external beam radiotherapy plus SBRT boost (15 to 18 Gy in a single fraction). With a median follow-up of 10 months, 1-year OS and PFS were 45 and 26%, respectively. Two patients from the whole population experienced grade 3 acute toxicity. During follow-up, four patients showed grade 4 late toxicity and two patients with cholangitis and bile duct stenosis required hospitalization. Considering only patients with ICC, median OS was 10 months and actuarial 1- and 2-year OS were 39.0 and 18.0%, respectively.

Sebastian et al (19) retrospectively analyzed 141 patients with inoperable non-metastatic ICC treated with SBRT (27 patients), or TARE (60 patients), or CRT (54 patients). With a median follow up of 17 months, median OS was 48 months in the SBRT group, while median OS in TARE and CRT groups were 20 and 14 months, respectively. At multivariate analysis, after propensity weighting adjustment, SBRT was correlated with higher OS compared to CRT (P<0.0001) and TARE (P<0.019).

Kozak et al (20) retrospectively collected data on 40 patients treated with SBRT for unresectable intrahepatic (25 patients) or hilar (15 patients) cholangiocarcinoma. For the entire group of patients, 25% were previously treated with CT, 20.0% with surgery, 7.5% with radiofrequency ablation, and 12.5% with TACE. Median dose was 40 Gy (range: 26-50) in five fractions. In ICC patients, with a median follow-up of 18 months, median OS was 23 months while in-field local failure and regional failure rates were 40 and 8%, respectively. Considering the whole group, both acute and late grade ≥3 hepatobiliary toxicity were recorded in 17 patients.

Analysis of selected studies

Six studies were included in this analysis reporting data on 145 patients. Four studies were retrospective (15,18-20) one was prospective (17), and one was a phase I/II trial (16). Patients were treated between 2003 and 2017. Studies' characteristics are shown in Table I.

Table I

Characteristics of the studies included in the systematic review.

Author, year	Study design	Accrual period	No. of patients	Median follow-up (range), months	ICC features, %	Treatment before SBRT, %	(Refs.)
Shen et al, 2017	Retrospective	2009-2012	28	16.0 (3.0-42.0)	Peripheral lesions: 78.6, Central lesions: 21.4	TACE: 28.6	(15)
					Solitary lesions: 71.4
					Multiple lesions: 28.6
					AJCC stage: II 21.4, III
					67.9, IVA 10.7
Weiner et al, 2016	Phase I-II	2012-2014	12	8.8 (0.3-33.0)	NRS	CHT: 66.7	(16)
Klein et al, 2015	Prospective	2003-2011	20	NR	NR	NR	(17)
Jung et al, 2014	Retrospective	2005-2013	33	10.0 (1.0-97.0)	NRS	NRS	(18)
Sebastian et al, 2019	Retrospective	2004-2014	27	17.0 (NR)	AJCC stage: T1 59.3; T2 25.9; T3 11.1; T4 3.7	CHT: 40.7	(19)
Kozak et al, 2020	Retrospective	2003-2017	25	18.0 (NR)	Positive nodes: 15.0	CHT: 24.0	(20)
						Surgery: 20.0
						RFA: 7.5
						TACE: 12.5

AJCC, American Joint Committee on Cancer (31); CHT, chemotherapy; ICC, intrahepatic cholangiocarcinoma; NR, not reported; NRS, not reported separately; RFA, radiofrequency ablation; SBRT, stereotactic body radiation therapy; TACE, trans-arterial chemoembolization.

Median follow-up was reported in five studies (15,16,18-20) and ranged between 8.8 and 18.0 months (median: 16 months). In two studies (15,19) ICC characteristics were described in detail. In one report, treated lesions were classified as solitary or multiple and as central or peripheric (15). The same paper reported the disease stage as follows: Stage II: 21.4%, stage III: 67.9%, and stage IVA: 10.7% (15). Another paper reported the tumor stages as follows: T1: 59.3%, T2: 25.9%, T3: 11.1%, and T4: 3.7% (19). In one study a 15.0% rate of patients with nodal involvement was reported (20).

Most patients included in this analysis were previously treated with local or systemic therapies. In one trial 28.6% of patients were previously treated with TACE (15). Two studies enrolled patients previously treated with CT in 66.7% (16) and 40.7% (19) of cases, respectively. Another paper reported on patients who received several previous therapies: CT (24.0%), surgery (20.0%), radiofrequency ablation (7.5%), and TACE (12.5%) (20).

Table II synthesizes SBRT planning and delivery characteristics. In three studies, the planning target volume (PTV) was defined as the gross tumor volume (GTV) plus 5 mm margin (15-17), while in two studies, the PTV was defined as the internal target volume (ITV) plus 2-4 mm margin (18,20). Three studies reported details on dose prescription (15,16,18). The 95% isodose line had to cover 95% of the PTV in 2 studies (15,16), while in one study the 92-99% isodose line had to cover 95% of PTV (18).

Table II

Treatment planning and dose delivery.

Author, year	Target definition	Median GTV (range), ml	Dose prescription	Median dose, Gy/fractions	Dose range, Gy	Median BED10Gy (range)	(Refs.)
Shen et al, 2017	PTV: GTV + 5 mm	NR	100% isodose line to cover	45/3-5	36-54	85.5 (72.0-124.8)	(15)
			GTV 95% isodose line to cover PTV
Weiner et al, 2016	PTV: GTV + 5 mm	107.0 (16.9-625.9)[a]	95% isodose line to cover 95% PTV	55/5	40-55	99.0 (72.0-115.5)	(16)
Klein et al, 2015	PTV: GTV + 5 mm	NR	NR	NR/6	24-60[a]	NRS	(17)
Jung et al, 2014	PTV: ITV + 2-4 mm	40.0 (5.0-1287.0)	70-80% isodose line (Cyber Knife) 92-99% isodose line to cover 95% PTV (Rapid Arc)	45/3-5[b] EBRT + SBRT boost, 16/1	30-60[b] EBRT + SBRT boost, 15-18	86.0 (48.0-150.0)	(18)
Sebastian et al, 2019	NR	NR	NR	45/5	40-50	NR	(19)
Kozak et al, 2020	PTV: ITV + 2-3 mm[a]	NR	NR	40/5	26-50[a]	NR	(20)

aFor the whole population included in the analysis;

b91% SBRT only, 9% EBRT + SBRT boost. BED, biologically effective dose; EBRT, external beam radiation therapy; GTV, gross tumour volume; ITV, internal target volume; NR, not reported; NRS, not reported separately; PTV, planning target volume; SBRT, stereotactic body radiation therapy.

Median prescribed dose was reported in five studies (15,16,18-20), ranging between 45 and 55 Gy (median 45 Gy), delivered in 3-5 fractions. The biological equivalent dose (BED) was reported in 3 studies and ranged between 85.5 and 99.0 Gy (median: 86.0 Gy) (15,16,18).

Only one study reported actuarial LC data, with 85 and 71.0% 1- and 2-year LC rates, respectively (18). Moreover, one study reported 40 and 8% local and regional failure crude rates. Respectively (20). Two papers reported 1-year and median PFS. One-year PFS was 50% (15) and 68% (16) while median PFS was 11(15) and 24.7 months (16).

Actuarial OS was reported in three studies with 1-year rates ranging between 39 and 57% (median value 51%) (15,16,18). Furthermore, two-year OS rates were 32% (15) and 18% (18) in two studies. Median OS was reported in all papers and ranged between 10 and 48 months (median: 14 months) (15-20). Toxicity was generally not described in detail. One paper reported a broad range of grade ≥3 gastrointestinal and hematological toxicities while three papers reported toxicity related to the entire cohort (16,18,20). Two of these papers (16,18) reported grade ≥3 acute toxicity in 3.4% (18) and 65.4% (16) of patients and grade ≥3 late toxicity in 6.9% (18) and 69.2% (16) of patients, respectively. Another paper reported grade ≥3 acute and late hepatobiliary toxicity, both recorded in 42.5% of cases (20).

Only two papers reported tumor response (15,16). The first one, according to the RECIST criteria, reported 10.7, 35.7, 42.9 and 10.7% rates of complete response, partial response, stable disease, and progressive disease, respectively (15). In the other paper, according to European Association for the Study of the Liver criteria, 33.3% of patients had a radiographic response and 8.3% had a complete radiographic response (16). Table III summarizes the treatment outcomes.

Table III

Survival outcomes.

Author, year	LC rate,%	PSF rate, %	Median PFS time, months	OS rate, %	Median OS time, months	Toxicity criteria	Acute toxicity, grade ≥3, % (n)	Late toxicity, grade ≥3, % (n)	Tumour response, %	(Refs.)
Shen et al, 2017	NR	1-year: 50.0 2-year: 21.4	11.0	1-year: 57.1 2-year: 32.1	15.0	CTCAE v.4.0	Nausea: 3.6(1) Anorexia: 7.1(2) Vomiting: 3.6(1) Gastric ulcer: 3.6(1) Altered hepatic values: 25.0(7) Haematological toxicity: 10.8(3)	NR	Complete response: 10.7 Partial response: 35.7 Stable disease: 42.9 Progressive disease: 10.7 (RECIST criteria)	(15)
Weiner et al, 2016	NRS	1-year: 68.0	24.7	1-year: 51.0	13.2	CTCAE v.4.0	65.4(17)[a]	69.2(18)[a]	Radiographic response: 33.3 Complete radiographic response: 8.3 (EASL criteria)	(16)
Klein et al, 2015	NR	NR	NR	NR	12.1	NR	NR	NR	NR	(17)
Jung et al, 2014	1-year: 85.0 2-year: 71.0	NR	NR	1-year: 39.0 2-year: 18.0	10.0	CTCAE v.4.0	3.4(2)[a]	6.9(4)[a]	NR	(18)
Sebastian et al, 2019	NR	NR	NR	NR	48.0	NR	NR	NR	NR	(19)
Kozak et al, 2020	Local failures: 40.0 Regional failures: 8.0	NR	NR	NR	23.0	NR	Hepatobiliary: 42.5(17)[a] Non-hepatobiliary: 2.5(1)[a]	Hepatobiliary: 42.5(17)[a] Non-hepatobiliary: 2.5(1)[a]	NR	(20)

aData including other sites of biliary cancers. CTCAE, Common Terminology Criteria for Adverse Events (32); EASL, European Association for the Study of the Liver (33); LC, local control; NR, not reported; NRS, not reported separately; OS, overall survival; PFS, progression-free survival; RECIST, Response Evaluation Criteria In Solid Tumours (34).

Discussion

To the best of our knowledge, this is the first systematic review on SBRT in the treatment of inoperable, non-metastatic, ICC.

ICC is considered as a rare malignancy in the western world and the management of locally advanced disease is still debated. CT is the standard therapy for advanced ICC (5,6) but the prognosis is still poor. SBRT allows the delivery of ablative doses to the tumor with minimal irradiation of the surrounding tissues. For this reason, some researchers have proposed SBRT as a possible therapeutic option in biliary tract cancer (12,13,21-23). Furthermore, the intrahepatic site of these tumours, therefore within an organ with parallel functional organization, generally far from serial organs unlike extrahepatic biliary tumours, makes ICCs particularly suitable to be treated with SBRT.

Our study has obvious limitations. Firstly, the selection of papers led to the evaluation of only 6 studies. More evidence is available on SBRT in HCC. However, we selected only papers on ICC due to the higher liver radiation sensitivity of HCC patients. In fact, the latter often present an impaired liver function or a cirrhotic status that can lead to higher complication rate. In those cases, a more conservative approach in volume definition or dose delivery may be needed.

The majority of these papers had a retrospective design (15,18-20). Moreover, the number of patients enrolled in the various studies was limited (range of patients enrolled, 12-33 patients). Furthermore, several data were lacking thus limiting the possibility to compare the results. Information on the disease stage (15,19) and treatments prior to SBRT (15,16,19,20) was present only in a few of the papers. Finally, the case series were very heterogeneous in terms of patient characteristics, treatment planning and delivery methods, total dose, fractionation, and results evaluation. The latter issue severely limits the possibility of an inherent comparison between the analyzed studies and among SBRT and other treatments results.

However, there are some worth noting considerations from our review. The best median OS (48 months) was reported in the study of Sebastian et al (19). This result can be justified by the favorable tumor stage (cT1-2 in 85.2% of patients). Therefore, this data suggests that OS can be particularly favorable in patients with less extensive ICC. In fact, 5/37 patients undergoing SBRT had >5 years OS.

From the comparison between the case series with the highest median SBRT dose (55 Gy) (16) and those with the lowest median dose (40 Gy) (20), it can be observed that median OS was 13.2 months in the former and 23 months in the latter. This comparison seems to suggest the lack of a positive correlation between delivered dose and patients outcome. However, this result contrasts with a study on 79 patients with inoperable ICC where a significant relationship between delivered dose and LC and OS was recorded (24). This difference can be explained by the use of 15-30 fractions in most patients in the latter study.

In terms of toxicity, an analytical description of acute toxicity has been reported only in the paper by Shen et al (15). Considering gastrointestinal toxicity, two out of 28 enrolled patients had anorexia, one nausea, one vomiting, and one a gastric ulcer (15). This finding seems to suggest a reasonable tolerability of SBRT in this setting, particularly in terms of low gastrointestinal toxicity rates.

Three papers reported the overall percentage of patients with grade ≥3 acute toxicity, with a wide range of incidence (3.4-65.4%) (16,18,20). It can be observed that the highest acute toxicity rate was observed in the series with the highest median delivered dose (16). Furthermore, in the series with the worst acute toxicity, the median GTV volume was 107.0 ml (16) while in the series with the lower acute toxicity rate, the median PTV volume was 40.0 ml (18). Comparing late toxicity, the rates recorded in the two series were 69.2% (16) and 6.9% (18) respectively. Even within the limits of a comparison between different series, these data would suggest a negative impact on toxicity of higher doses and larger irradiated volumes.

Further considerations can be made by comparing the results of our review with those of other treatments of locally advanced ICC. In particular, our analysis showed that ICC patients undergoing SBRT had a median OS of 14.1 months (range, 10-48 months). However, it should be noted that SBRT was frequently used as a salvage treatment, since 28.6-66.7% of patients received previous systemic or local treatments (15,16,19,20). This may have affected the results recorded in our analysis.

CT is a standard treatment of advanced ICC. In a series of 26 patients with unresectable ICC, median OS was 18.4 months, slightly higher compared to our analysis (25). However, it should be considered that, in that study, CT was used as an up-front treatment while SBRT was often used as a salvage therapy in our analysis. That same publication reported 30.8 months median OS (P<0.001) in another group of patients undergoing CT plus intrahepatic arterial infusion (25). This data suggests that combinations of systemic and local therapies may improve outcome in patients with advanced ICC.

Concurrent CRT is another therapeutic option for advanced non-metastatic cholangiocarcinoma, and proper target volume definition could achieve good results (26). Median OS was 13.5 months in a retrospective multicentric study on 76 patients with unresectable biliary cancer (8). Moreover, the results on 27 patients with advanced cholangiocarcinoma who underwent CRT +/- brachytherapy boost were reported in another study (27). Median OS was similar (14 months) and the toxicity grade ≥3 rate was 37%. Although separate data for ICC patients were not reported in these studies, the results seem to suggest that outcomes after SBRT are similar to those of CRT, with a clear benefit in terms of shorter treatment duration for SBRT.

Trans-catheter arterial chemoembolization is another treatment for biliary tract cancer. A meta-analysis on unresectable cholangiocarcinoma reported 13.4 months median OS from the time of first treatment. Severe toxicity was recorded in 18.9% of patients with 0.7% rate of 30-days mortality rate (10). Again, there is no clear difference with the results of our analysis on SBRT.

Similarly, a meta-analysis of 12 studies on radio-embolization with 99-Ittrium microsphere reported 15.5 months median weighted OS. Also this result is similar to the one of our analysis on SBRT. Toxicity was mainly represented by fatigue (33%), pain (28%) and nausea (25%) (28).

Finally, some recent studies evaluated the potential role of proton beam radiation therapy in cholangiocarcinoma. In a retrospective study on unresectable or locally recurrent ICC, 34 patients were treated with photon-based radiotherapy while 32 patients were treated with protons. Two-year OS was 49 and 65%, respectively. At multivariate analysis, compared to photon-based radiotherapy, there was a trend for improved OS with protons (HR: 0.50; p: 0.05) (29). In another retrospective analysis (30) on 30 patients with unresectable cholangiocarcinoma treated with proton-therapy (median dose: 72.6 cobalt Gy equivalent) median OS was 19.3 months. However, 10% of patients developed grade ≥3 gastric/duodenal ulcers. This irradiation modality deserves further study in the ICC setting, based on the theoretical advantages of proton therapy in terms of dose distribution. However, the currently available results do not allow a comparison with SBRT.

In conclusion, although other options are available in this setting, preliminary literature data on SBRT in ICC seems to suggest that this treatment could be considered as a possible therapeutic option.

Further studies are needed in this field. These analyzes could be aimed to the: i) assessment of efficacy and tolerability of SBRT in ICC by large and prospective trials; ii) definition of optimal dose, fractionation and GTV to PTV margins in this setting; iii) comparison between SBRT and other ICC local treatments; considering the palliative aim of advanced ICC therapies, these studies should include an accurate assessment of treatments impact on quality of life; iv) feasibility and efficacy of combined modality treatments including SBRT (for example: SBRT combined with systemic therapies; SBRT as a boost after concurrent chemoradiation).