Liver transplantation for locally advanced intrahepatic cholangiocarcinoma treated with neoadjuvant therapy: a prospective case-series

Articles

Liver transplantation for locally advanced intrahepatic cholangiocarcinoma treated with neoadjuvant therapy: a prospective case-series Keri E Lunsford, Milind Javle, Kirk Heyne, Rachna T Shroff, Reham Abdel-Wahab, Nakul Gupta, Constance M Mobley, Ashish Saharia, David W Victor, Duc T Nguyen, Edward A Graviss, Ahmed O Kaseb, Robert S McFadden, Thomas A Aloia, Claudius Conrad, Xian C Li, Howard P Monsour, A Osama Gaber, Jean-Nicolas Vauthey, R Mark Ghobrial, on behalf of the Methodist–MD Anderson Joint Cholangiocarcinoma Collaborative Committee (MMAJCCC)

Summary

Background At present, intrahepatic cholangiocarcinoma is a contraindication for liver transplantation. However, previous studies in this field did not preselect patients on the basis of chemosensitivity or disease trajectory after neoadjuvant therapy. Experience with hilar cholangiocarcinoma has indicated that neoadjuvant therapy followed by liver transplantation in patients without disease progression results in a long-term survival benefit. We aimed to establish the potential efficacy of liver transplantation in patients with biologically responsive intrahepatic cholangiocarcinoma who have had sustained tumour stability or regression with neoadjuvant therapy. Methods In this prospective case-series, patients with locally advanced, unresectable intrahepatic cholangiocarcinoma, without extrahepatic disease or vascular involvement, were treated at a single liver transplant centre according to a non-randomised, centre-approved clinical management protocol with neoadjuvant chemotherapy followed by liver transplantation. Neoadjuvant therapy consisted of gemcitabine-based chemotherapy, such as gemcitabine–cisplatin or gemcitabine–capecitabine, with second-line or third-line therapies given per institutional standards. Patients with a minimum of 6 months of radiographic response or stability were listed for liver transplantation. The primary endpoints were overall survival and recurrence-free survival after liver transplantation, assessed with Kaplan-Meier analysis. This report includes interim data from the initial case-series treated under this ongoing clinical management protocol, censored on Dec 1, 2017. Findings Between Jan 1, 2010, and Dec 1, 2017, 21 patients were referred for evaluation and 12 patients were accepted, of whom six patients have undergone liver transplantation for intrahepatic cholangiocarcinoma. Three patients received livers from extended criteria deceased donors that would otherwise have been discarded, two from domino living donors, and one from a standard criteria liver donor. Median duration from diagnosis to transplantation was 26 months (IQR 17–33) and median follow-up from transplantation was 36 months (29–51). All patients received neoadjuvant chemotherapy while awaiting liver transplantation. Overall survival was 100% (95% CI 100–100) at 1 year, 83·3% (27·3–97·5) at 3 years, and 83·3% (27·3–97·5) at 5 years. Three patients developed recurrent disease at a median of 7·6 months (IQR 5·8–8·6) after transplantation, with 50% (95% CI 11·1–80·4) recurrence-free survival at 1, 3, and 5 years. Adverse events after liver transplantation included one patient with postoperative ileus (grade 3) and one patient with acute kidney injury requiring temporary dialysis (grade 4). Interpretation Selected patients with locally advanced intrahepatic cholangiocarcinoma who show pre-transplant disease stability on neoadjuvant therapy might benefit from liver transplantation. Funding None.

Introduction Cholangiocarcinoma, which arises from biliary epithelium, can be subdivided into distal, hilar, and intrahepatic subgroups.1 Although rare, intrahepatic cholangiocarcinoma constitutes the second most common primary liver cancer after hepatocellular carcinoma, and the worldwide incidence of intrahepatic cholangio­carcinoma is increasing.1 At present, complete surgical resection is the only widely accepted curative therapy for intrahepatic cholangiocarcinoma. However, outcomes are disappointing: despite resection, 5-year survival is only 25%.2 Mortality largely results from

tumour recurrence, with 53–79% of patients experiencing tumour recurrence.3–7 About 83% of recurrences are local, occurring within the first 2 years after resection,8 suggesting inadequate local tumour control with resection in this highly infiltrative cancer. Non-surgical approaches, including systemic chemo­ therapy, local ablative therapies, and radiation, are also suboptimal, with minimal long-term survival. Most patients are unresectable because of tumour location, size, or multifocality. Liver transplantation has been previously investigated for unresectable intrahepatic cholangiocarcinoma.9–11 However, by contrast with

www.thelancet.com/gastrohep Published online March 13, 2018 http://dx.doi.org/10.1016/S2468-1253(18)30045-1

Lancet Gastroenterol Hepatol 2018 Published Online March 13, 2018 http://dx.doi.org/10.1016/ S2468-1253(18)30045-1 J C Walter Jr Transplant Center, Sherrie and Alan Conover Center for Liver Disease and Transplantation, Weill Cornell Medical College (K E Lunsford MD, C M Mobley MD, A Saharia MD, D W Victor MD, R S McFadden MD, Prof X C Li PhD, H P Monsour MD, Prof A O Gaber MD, Prof R M Ghobrial MD), and Department of Pathology and Genomic Medicine (D T Nguyen PhD, E A Graviss PhD), Houston Methodist Hospital and Research Institute, Houston, TX, USA; Department of Gastrointestinal Medical Oncology, Division of Cancer Medicine (Prof M Javle MD, R T Shroff MD, R Abdel-Wahab MD, A O Kaseb MD), and Hepato-Pancreato-Biliary Section, Department of Surgical Oncology (T A Aloia MD, C Conrad MD, Prof J-N Vauthey MD), University of Texas MD Anderson Cancer Center, Houston, TX, USA; Department of Oncology (K Heyne MD) and Department of Radiology (N Gupta MD), Houston Methodist Hospital, Houston, TX, USA; and Department of Clinical Oncology, Assiut University, Assiut, Egypt (R Abdel-Wahab) Correspondence to: Prof R Mark Ghobrial, J C Walter Jr Transplant Center, Sherrie and Alan Conover Center for Liver Disease and Transplantation, Weill Cornell Medical College, Houston Methodist Hospital, Houston, TX 77030, USA rmghobrial@ houstonmethodist.org

1

Articles

Research in context Evidence before this study We searched PubMed up to Nov 1, 2017, for English-language articles using the search terms “cholangiocarcinoma”, “intrahepatic cholangiocarcinoma”, “hilar cholangiocarcinoma”, “biliary tract cancer”, “bile duct cancer”, and “liver transplantation”. We identified 23 studies of liver transplantation for cholangiocaricinoma, few of which included patients with intrahepatic cholangiocarcinoma. Almost all of the patients with intrahepatic cholangiocarcinoma had incidental tumours identified at liver transplant and did not receive neoadjuvant therapy. Generally, data from these trials did not support liver transplantation for the treatment of intrahepatic cholangiocarcinoma. Experience with hilar cholangiocarcinoma suggests that neoadjuvant therapy before transplantation results in high relapse-free and overall survival. Added value of this study To our knowledge, this study is the first prospective series of patients undergoing liver transplantation for intrahepatic

favourable results for hepatocellular carcinoma, outcomes for intrahepatic cholangiocarcinoma have been poor, with 18–25% overall and recurrence-free survival after 5 years.12–14 Thus, intrahepatic cholangio­ carcinoma is considered by most centres to be a formal contraindication to liver transplantation. Similar to intrahepatic cholangiocarcinoma, liver transplantation outcomes for hilar cholanigocarcinoma were initially poor; however, an analysis of the United Network of Organ Sharing (UNOS) database found a significant survival benefit for patients with known hilar cholangiocarcinoma receiving pre-transplant anticancer therapy compared with patients transplanted for incident disease.14 Studies15–20 subsequently reported improved survival for hilar cholangiocarcinoma treated with neoadjuvant chemoradiation followed by liver transplantation. A multicentre study16 then reported 65% actuarial survival at 5 years after liver trans­ plantation. Thus, liver transplantation has become the preferred treatment for patients with unresectable, early-stage hilar cholangiocarcinoma. Response to neoadjuvant therapy probably offers a means to select patients with hilar cholangiocarcinoma who might benefit from transplantation. Liver transplantation for intrahepatic cholangio­carci­ noma is mostly limited to incidental or misdiagnosed tumours identified on explant pathology,21,22 with most patients not receiving neoadjuvant therapy. One retrospective analysis20 did include a small cohort of patients receiving neoadjuvant therapy. Pre-transplant therapy appeared to decrease disease recurrence;20,23 however, reports did not distinguish between hilar cholangiocarcinoma and intrahepatic cho­ langio­ carcinoma. In 2016, a multicentre, international, 2

cholangiocarcinoma after neoadjuvant therapy. In this initial series, six patients with locally advanced, unresectable intrahepatic cholangiocarcinoma with at least 6 months’ response or stability after neoadjuvant chemotherapy were offered liver transplantation under a centre-approved clinical management protocol. For these patients, 5-year overall survival was 83·3% and recurrence-free survival was 50%. Neither tumour volume nor multifocality affected the incidence of disease recurrence after liver transplantation. Implications of all the available evidence These data suggest that sustained response to neoadjuvant therapy might identify patients with intrahepatic cholangiocarcinoma who would benefit from liver transplantation. A larger prospective trial is needed to better define prognostic factors and investigate the efficacy of liver transplantation for intrahepatic cholangiocarcinoma.

retrospective study10 investigated outcomes of liver trans­plantation in 48 patients with intrahepatic chol­ angio­ carcinoma who had not received neoadjuvant chemotherapy or had received locoregional therapy. In that study, 5-year overall survival was 65% for early (≤2 cm) intrahepatic cholangiocarcinoma and 45% for advanced (>2 cm) intrahepatic cholangiocarcinoma. This finding suggests that liver transplantation might be a viable option for small, solitary intrahepatic cholangio­carcinomas in the absence of pre-transplant therapy, but the effect of neoadjuvant chemotherapy in patients with larger and multifocal tumours remains largely undefined. To our knowledge, no previous studies have prospectively assessed outcomes of liver transplantation for advanced intrahepatic cholangiocarcinoma in the setting of neo­adjuvant chemotherapy. In our experience, a subset of patients with intrahepatic cholangiocarcinoma experience sustained response to neoadjuvant therapy, and we postulated that response duration might be an appropriate surrogate marker for selection of patients for liver transplantation. Through the Methodist–MD Anderson Joint Cholangiocarcinoma Collaborative Committee (MMAJCCC), a clinical management protocol was established for liver transplantation in patients with locally advanced, unresectable intrahepatic carcinoma with a durable radiological cholangio­­ response or stability after neoadjuvant chemotherapy. We aimed to use this protocol to identify and treat patients with intrahepatic cholangiocarcinoma with reasonably favourable disease biology for liver transplantation. This report outlines the outcomes of the initial series of patients treated under this ongoing clinical management protocol.

www.thelancet.com/gastrohep Published online March 13, 2018 http://dx.doi.org/10.1016/S2468-1253(18)30045-1

Articles

Methods

Study design and participants This report is based on the clinical management of patients with intrahepatic cholangiocarcinoma at a single liver transplant centre, where the decision was made in 2010 to provide liver transplantation to qualifying patients with intrahepatic cholangio­ carcinoma under a centre-approved protocol. The clinical management protocol requires use of neo­ adjuvant therapy for intrahepatic cholangiocarcinoma and sustained biological stability or response for the selection of patients for liver transplantation. The primary objective of this report is to inform on the potential efficacy of liver transplantation in patients with biologically responsive tumours. The liver transplantation protocol for locally advanced intrahepatic cholangiocarcinoma was established by the MMAJCCC and was accepted after institutional review by the Houston Methodist Liver Transplantation Care Management and Performance Improvement governing board. Locally advanced intrahepatic cholangiocarcinoma was defined as a solitary tumour greater than 2 cm in diameter or multifocal disease confined to the liver without radiological evidence of extrahepatic, macro­ vascular, or lymph node involvement. Details of the approved protocol and the Methodist–MD Anderson Cancer Center selection criteria are shown in figure 1. Patients were eligible for protocol consideration if they had shown 6 months of disease stability or tumour regression on neoadjuvant therapy. Selection of potentially appropriate candidates was done through multi­ disciplinary case review by the MMAJCCC, including opinions of medical oncologists, hepatobiliary surgeons, radiation oncologists, transplant or hepato­ biliary surgeons, hepa­ tologists, interventional radio­ logists, pathologists, and diagnostic liver radio­logists. Selected patients then under­went a formal screening interview and detailed evaluation of medical history by tologist and transplant surgeon. an MMAJCCC hepa­ Acceptable candidates were then referred for formal liver transplantation evaluation and listing. For inclusion in the protocol, patients had to have tissue confirmation of cholangiocarcinoma via biopsy or cytology and unresectable tumours because of location or underlying liver disease after 6 months of neoadjuvant therapy. Resectability was assessed in a multidisciplinary manner by the MMAJCCC, and was the consensus opinion of hepatobiliary and liver transplant surgeons from the Houston Methodist Hospital and the University of Texas MD Anderson Cancer Center. In cases in which patients had received second-line therapy, disease must have been controlled for 6 months on that regimen. Previous surgical resection for intrahepatic cholangio­ carcinoma was allowed if surgery occurred more than 6 months before listing, and the same dur­ation of disease stability or response was required as in those without previous resection.

To qualify for liver transplantation under the clinical management protocol, patients had to have a mass that appeared malignant with imaging characteristics consistent with intrahepatic cholangiocarcinoma and either biopsy showing malignancy or serum cancer antigen (CA)19-9 concentrations greater than 100 U/mL, unresectable disease based on technical considerations or underlying liver disease, sustained response for at least 6 months with neoadjuvant therapy (assessed with CT or MRI scans), no evidence of concurrent hepatoma or mixed hepatocellular cholangiocarcinoma, no evidence of extrahepatic disease, at least 6 months of sustained response after the most recent therapy, minimisation of transperitoneal tumour aspiration or biopsy, and control of biliary sepsis. Exclusion criteria for consideration for liver transplantation under the protocol included presence of extrahepatic metastases, lymph node involvement, invasion or encasement of major hepatic vascular structures, perforation of the visceral peritoneum, invasion of extrahepatic structures, invasion of perihilar fat, or periductular invasion. In addition to compliance with oncological practices, patients had to complete the full medical and psychosocial work-up. There were no specific age or comorbidity limitations; however, only adult patients (18 years or older) were considered, and the patient had to meet medical criteria for liver transplantation. Generally, acceptable candidates did not have significant cardiac disease, cerebrovascular disease, or pulmonary disease, and had no history of extrahepatic cancer within the past

For the protocol see http://houstonmethodist.org/ liver-research

Intrahepatic cholangiocarcinoma Tumour characteristics • Biopsy-proven cholangiocarcinoma • Intrahepatic rather than periductal location • Not amenable to surgical therapy • No evidence of extrahepatic disease Diagnostic criteria • Triple-phase CT of the chest, abdomen, and pelvis • MRI bone scan • FDG-PET scan if serum cancer antigen 19-9 elevated • Endoscopic ultrasound-guided biopsy of enlarged nodes

Neoadjuvant chemotherapy • First-line platinum-based therapy and gemcitabine • Second-line chemotherapy for progression or intolerance • Addition of targeted biologics on case-by-case basis

Disease stability for at least 6 months on given regimen • Repeat imaging every 3 months • Stable or regressing disease • No extrahepatic disease

Liver transplantation • Post-transplant adjuvant therapy for 4–6 months depending on explant pathology

Figure 1: Study flow diagram Methodist–MD Anderson selection criteria for liver transplantation are shown.

www.thelancet.com/gastrohep Published online March 13, 2018 http://dx.doi.org/10.1016/S2468-1253(18)30045-1

3

Articles

5 years. Potential recipients were then reviewed and approved by the Liver Transplant Center Medical Review Boards before listing with UNOS. Patients undergoing transplant in this series were listed between Feb 11, 2010, and June 9, 2015. Listed patients received donor organs through the UNOS deceased-donor waitlist pool or through domino liver allografts from living donors. Recipients received extensive counselling from the attending surgeon during evaluation regarding the risks of using orphan livers. At transplant, recipients were consented by the transplanting surgeon with regard to the risks of transplant, including portal vein or hepatic artery thrombosis and primary non-function, with additional consent obtained in cases of donors at high risk of disease transmission. Collection of data from recipients was done in accordance with the Houston Methodist Institutional Review Board approved protocol (IRB0507-0053).

Procedures Because the availability of livers for transplantation is unpredictable, all patients received neoadjuvant chemotherapy, with or without radiotherapy, until liver transplantation. All patients were followed up at both the MD Anderson Cancer Center for chemotherapeutic compliance and oncological surveillance and the Sherrie and Alan Conover Center for Liver Disease and Transplantation at Houston Methodist Hospital for compliance with transplantation listing protocols. Neo­adjuvant chemotherapy included initial gemcitabinebased therapy, such as gemcitabine–cisplatin or gemcitabine–capecitabine. Gemcitabine–cisplatin was preferentially administered (it is considered the standard regimen according to the ABC-02 protocol24), but in cases of intolerance to this regimen or subsequent second-line regimens, the same therapy that resulted in disease control was continued until transplantation. Given the scarcity of standard second-line or third-line therapies, subsequent therapies were per institutional standards and included fluoro­pyrimidines and targeted drugs. At initial presentation, CT scans of the chest, CT or MRI scans of the abdomen and pelvis, and bone surveys were done to exclude intrahepatic and extrahepatic metastases. FDG-PET was selectively done for patients with abnormal concen­ trations of serum CA19-9 (reference range 0–35 U/mL) or suspicious lesions or nodes on cross-sectional imaging. Radiologically suspicious regional hepatic lymph nodes were biopsied via endoscopic ultrasound to exclude metastases before neoadjuvant therapy. All imaging was done at either Houston Methodist Hospital or the MD Anderson Cancer Center, and diagnostic radiologists specialising in hepatobiliary malignancies formally reviewed all images. All patients continued screening with contrast-enhanced CT or MRI of the abdomen and pelvis, CT of the chest, and bone scans every 3 months while awaiting transplant, and follow-up images were 4

reviewed by the MMAJCCC. Identification of disease progression or extrahepatic disease was grounds for case review by the Liver Transplant Center Medical Review Board and delisting from the transplant waiting list. At the beginning of the transplant operation, we did staging laparotomy and perihepatic lymph node sampling to check for extrahepatic disease. If frozen-section biopsy identified extrabiliary or extrahepatic cancer spread, the surgery was terminated and the allocated allograft was returned to the UNOS waitlist pool. The recipient surgeon initiated staging laparotomy at the time of donor liver visualisation by the procuring surgeon to minimise cold ischaemia time. After transplantation, complete histopathological analysis of the explant was done, including nextgeneration sequencing (FoundationOne; Foundation Medicine, Cambridge, MA, USA) when available. Genetic profiling was done on each tumour type present in the explanted specimen. When all individual tumours had similar histological and pathological characteristics, a repre­sentative sample was selected for profiling. All liver transplantations were done at Houston Methodist Hospital with caval anastomosis in piggyback fashion. Recipients underwent bolus steroid induction therapy during the anhepatic phase, followed by steroid taper and withdrawal by day 30. Maintenance immunotherapy included tacrolimus and myco­ phenolate mofetil for the first 30 days, followed by transition to everolimus and tacrolimus dual therapy, unless contraindicated (eg, because of severe proteinuria). Immunosuppression was minimised as per patient response. Adjuvant chemotherapy after liver transplantation was given on the basis of explant pathology. Patients showing complete pathological responses required no additional therapy, assuming they had received a total of 4–6 months of chemotherapy before the transplant. Patients with active disease on explant received adjuvant chemo­ therapy with capecitabine, gemcitabine, or both, initiated 4–6 weeks after transplant and continuing for a minimum of 4–6 months. Recurrent disease was aggressively treated with surgery, radiation, or additional chemotherapy, as indicated. Novel drugs were considered as clinical trials became available. Patients underwent intensive monitoring after liver transplantation for evidence of disease recurrence. Monitoring of serum tumour markers (α-fetoprotein, CA19-9) and repeated CT or MRI scans and bone surveys were done every 3 months for 2 years, then every 6 months up to 5 years. After 5 years, scans will be done annually. Patient data were prospectively and longitudinally evaluated from time of initial diagnosis to death or loss to follow-up. For this analysis, all data were censored on Dec 1, 2017. Multiple recipient variables (age, sex, primary liver disease diagnosis, diabetes, hypertension,

www.thelancet.com/gastrohep Published online March 13, 2018 http://dx.doi.org/10.1016/S2468-1253(18)30045-1

Articles

hyperlipidaemia, coronary artery disease, smoking, renal failure, functional status), donor and operative characteristics (donor age, sex, donor type, graft type, cold ischaemia time, warm ischaemia time), and laboratory variables (α-fetoprotein, CA19-9, MELD score) were collected. Tumour pre-transplant radiographic data (number of lesions, maximal diameter, cumulative diameter), pathological variables (number of lesions, maximal diameter, cumulative diameter, stage, grade, differentiation, vascular invasion), and treatment variables (chemotherapy, radiation, locoregional therapy, resection) were also collected. Information was collected about the incidence of recurrence after liver trans­ plantation, the site and size of recurrence, and treatment and outcomes of recurrence.

Outcomes The primary endpoints were overall survival and recurrence-free survival after liver transplantation. These outcomes address the potential efficacy of the strategy for intrahepatic cholangiocarcinoma compared with previously reported best medical or surgical therapies for intrahepatic cholangiocarcinoma and liver trans­ plantation for other hepatobiliary malignancies. Secondary endpoints were short-term and long-term morbidities and pathological and genetic correlations with favourable outcomes. Duration of hospitalisation was also assessed as a secondary outcome. Postoperative complications were evaluated and graded with the Dindo-Clavien scale.25

Statistical analysis All patients with intrahepatic cholangiocarcinoma who underwent liver transplantation according to the defined clinical protocol were included in the analysis. Because, to our knowledge, this report is the first to assess outcomes of liver transplantation for advanced intrahepatic cholangiocarcinoma in the setting of neoadjuvant chemotherapy, no preliminary published data were available to do power and sample-size calculations. Findings from this study will inform future cohort studies and clinical trials in generating hypotheses and doing power calculations. Because patients were not randomised and were treated in accordance with a centre-approved clinical policy and UNOS rules and regulations, this study did not qualify for registration as a clinical trial. Demographic, pre-transplant, and explant tumour characteristics are reported as frequencies and proportions for categorical variables and as medians (IQRs) for continuous variables. The primary endpoints were estimated with Kaplan-Meier statistics. Analyses were done with Stata MP version 15.

Role of the funding source This study did not receive any external funding. The corresponding author had full access to all of the data and the final responsibility to submit for publication.

Patients undergoing liver transplantation (n=6) Recipient demographics Age (years)

47 (31–63)

BMI

25·5 (18·3–34·0)

Sex Male

3 (50%)

Female

3 (50%)

Race (white)

6 (100%)

Primary sclerosing cholangitis

1 (17%)

Diabetes

1 (17%)

Hypertension

5 (83%)

Smoking

0 (0%)

Match MELD at transplant

8·5 (6·0–37·0)

MELD exception

0 (0%)

Pre-transplant hospitalisation Karnofsky score at transplant Duration from diagnosis (months) Duration on waitlist (days)

0 (0%) 70% (55–78) 26 (10–36) 217 (59–356)

Donor demographics Age (years)

49 (18–74)

Sex Male

5 (83%)

Female

1 (17%)

Race White

3 (50%)

African American

2 (33%)

Hispanic BMI Donation after circulatory death

1 (17%) 26·8 (22·2–28·9) 0 (0%)

Domino living liver donor

2 (33%)

Brain-dead donor

4 (67%)

Operative characteristics Venovenous bypass Major reperfusion syndrome

4 (67%) 0 (0%)

Cold ischaemia time (min)

328 (80–425)

Warm ischaemia time (min)

24 (15–75)

Packed red blood cells transfusion (units)

5·5 (0·0–10·0)

Arterial conduit

0 (0%)

Post-transplant characteristics Follow-up (months)

36 (25–74)

Postoperative stay (days)

9 (5–18)

Recurrent cholangiocarcinoma

3 (50%)

Death

1 (17%)

Postoperative morbidity Ileus

1 (17%)

Acute renal injury

1 (17%)

Hernia

1 (17%)

Data are median (IQR) or n (%). BMI=body-mass index.

Table 1: Donor and recipient characteristics for patients undergoing liver transplantation for intrahepatic cholangiocarcinoma

www.thelancet.com/gastrohep Published online March 13, 2018 http://dx.doi.org/10.1016/S2468-1253(18)30045-1

5

Articles

Overall

Recipient 1

Recipient 2

Recipient 3

Recipient 4

Recipient 5

Recipient 6

Stage II (T2bN0M0)

Stage II (T2bN0M0)

Stage II (T2aN0M0*)

Stage II (T2bN0M0*)

Stage IVB (T2bN1M0)

Stage I (T1N0M0)

Stage II (T2bN0M0)

Radiographic (pre-transplant) Stage Number of lesions

4·0 (3·0–5·8)†

3

3

1

Maximum size of largest lesion (cm)

7·0 (6·0–8·3)†

2·6

10·3

6·5

7·4

5·8

8·6

10·5 (7·0–13·5)†

4·1

14·5

10·5

10·4

5·8

18·0

Cumulative diameter (cm)

5

>5

>5

Explant Stage

Stage II (T2bN0M0)

Stage II (T2bN0M0)

Stage II (T2bN0M0*)

Number of lesions

7 (2–10)†

8

6

Maximum size of largest lesion (cm)

5·9 (4·5–8·4)†

4·2

9·0

Stage II (T2bN0M0*) 10 3·5

Stage III (T3N0M0)

Stage I (T1N0M0)

Stage II (T2bN0M0)

1

1

10

5·2

6·5

10·5

Cumulative diameter (cm)

14·2 (8·1–17·9)†

18·7

13·0

15·3

Location

NA

Bilobar

Bilobar

Bilobar

Left

5·2

Left

6·5

Bilobar

20 Poor

Differentiation

Moderate to poor

Poor

Well

Poor

Moderate

Moderate

Lymphovascular invasion

No

Yes

No

Yes

No

No

No

Perineural invasion

No

No

No

No

Yes

No

No

Microvascular invasion

No

Yes

No

Yes

No

No

No

Macrovascular invasion

No

No

No

No

No

No

No

Positive margins

No

No

Yes

No

No

Necrosis (%)

0%

No

No

0%

95%

0%

0%

0%

90%

Patient outcomes Post-transplant recurrence

NA

No

Yes

Yes

Yes

No

No

Post-transplant death

NA

No

No

Yes

No

No

No

Duration of follow-up (months)

36·3 (29·0–50·6)†

74·3

53·8

40·9

31·7

28·1

24·9

NA=not appropriate. *Retrospective radiographic analysis suggested that stable metastatic disease might have been present before liver transplantation. †Data are median (IQR).

Table 2: Radiographic and explant tumour characteristics from liver transplant recipients with intrahepatic cholangiocarcinoma

Results Between Jan 1, 2010, and Dec 1, 2017, 21 patients were evaluated for enrolment in the liver transplantation clinical manage­ment protocol for intrahepatic cholangio­ carcinoma. Of these patients, nine failed initial screening, seven because of the presence of extrahepatic disease or disease progression and two because of downstaging to resectable disease. The remaining 12 patients with intrahepatic cholangio­carcinoma fulfilled the selection criteria and were listed for liver transplantation. Six of these patients underwent successful liver transplantation and three remain actively listed and are awaiting availability of appropriate donor organs. Three patients listed for liver transplantation did not receive a transplant because of intra-abdominal adhesion severity (n=2), most likely resulting from previous external beam radiation, or resectable disease upon exploration (n=1) due to downstaging with chemotherapy (this patient underwent partial hepatectomy). None of the listed patients showed disease progression requiring delisting after the initial 6 months of stable disease. The donor and recipient characteristics for the six patients who underwent liver transplantation are shown in table 1. These six recipients had unresectable disease because of bilobar disease (n=4), diagnosis of underlying liver disease (primary sclerosing cholangitis) 6

with a MELD score of 37 (n=1), or close proximity of the tumour to the hilar structures (n=1). Donor organs were procured from living donors for two of the six recipients (as domino transplants from donors with amyloid disease). The remaining four organs were procured from brain-dead donors. One recipient received a standard-criteria liver because of decompensated cirrhosis with a physiological MELD score of 37. Three recipients received extended-criteria orphan livers that were refused by all centres on the basis of allocation MELD and would have otherwise been discarded. Of these three extended-criteria organs, one was procured from an elderly donor in a long-term care facility who had multiple medical comorbidities, including end-stage kidney disease and prolonged cold ischaemia. The second extended-criteria liver had recently been transplanted, but the recipient had herniated intraoperatively because of cerebral oedema. This recipient thus became an organ donor, and the liver was reimplanted. The final extended-criteria liver was procured from a young donor following a trauma, who had substantially elevated enzymes (aspartate amino­ transferase and alanine aminotransferase) and CT evidence of extensive hepatic lacerations extending to the porta hepatis, anterior and posterior right lobe, medial left lobe, and caudate lobe.

www.thelancet.com/gastrohep Published online March 13, 2018 http://dx.doi.org/10.1016/S2468-1253(18)30045-1

Articles

Pretreatment and explant tumour characteristics are shown in table 2, and pre-transplant images in figure 2. Pre-transplant imaging revealed multifocal disease in five patients, three of whom had T2bN0M0 disease. One patient had periportal lymph node enlargement on pretreatment imaging. In this patient, lymph nodes were not biopsied before therapy initiation, but were negative on FDG-PET and subsequently regressed substantially; lymph nodes were negative on explant. At the time of exploration, no patient undergoing transplantation was found to have extrahepatic disease precluding transplantation. On explant, four patients had stage II disease (T2bN0M0). One patient had stage III disease (T3N0M0) because of having positive perihilar fat margins, and this stage was associated with post-transplant disease recurrence. Pre-transplant or peri-transplant identification of positive perihilar fat margins through intraoperative biopsy would have precluded this patient from transplantation. On explant, disease was multifocal in four patients and bilobar in four patients. Lymphovascular invasion was present in two patients, one of whom developed recurrent disease. Perineural invasion was identified in one patient, who also developed recurrent disease. All recipients received neoadjuvant therapy before liver transplantation (table 3). The median time between diagnosis and transplantation was 26 months (IQR 17–33). All patients were treated with gemcitabine and cisplatin. Other systemic therapies included erlotinib (n=3); FOLFIRI (folinic acid, fluorouracil, and irinotecan; n=1); and fluorouracil (n=1). Changes in therapy occurred because of intolerance or disease progression. Median progression-free survival was 8 months (95% CI not available) for the first-line therapy and 3 months (95% CI not available) for the second-line therapy. Because of the reasonably long time between diagnosis and transplantation, patients received multiple therapies, but overall still qualified for transplantation because of indolent disease biology. All patients had 6 months of disease stability before listing for liver transplantation. Two of the six patients had previously undergone anatomic liver resection, but had subsequent disease recurrence, for which they received chemotherapy before transplant consideration (table 3). One of these patients underwent a second non-anatomic resection at a later time. One transplant recipient also received neoadjuvant radiotherapy with 40 Gy in five fractions to a portal vein margin that was microscopically positive after initial surgical resection. Type or duration of therapy was not associated with outcomes of liver transplantation (table 3). After transplantation, patients were discharged home after a median of 8·5 days (IQR 6·0–10·0). Postoperative complications included one patient with postoperative ileus (grade 3) and one with transient acute renal injury requiring temporary dialysis (grade 4). Late complication of a hernia occurred in one patient and required surgical

A

B

C

D

E

F

Figure 2: Pre-transplant portal venous phase, contrast-enhanced CT images for patients with intrahepatic cholangiocarcinoma undergoing liver transplantation (A) A 7·4 cm heterogeneous, hypovascular mass spanning segments IVb and V is shown (recipient 1). (B) A 10·3 cm hypovascular mass in the left-hepatic lobe is shown; multiple smaller metastatic lesions involving all segments of the right lobe were also identified, two of which are pictured (recipient 2). (C) Multifocal hypovascular lesions are shown in the remnant left lateral segment of the liver (after extended right hepatectomy), the largest of which measures 6·5 cm (recipient 3). (D) A central hypovascular mass measuring 4·9 cm superior to the liver hilum can be seen, resulting in left-sided biliary dilatation (not shown); no other hepatic lesions were present (recipient 4). (E) A 6·4 cm hypovascular mass is shown, predominantly within segment V of the liver, on a background of cirrhosis and portal hypertension, as evidenced by intra-abdominal varices and splenomegaly (recipient 5). (F) Multifocal bilobar hypovascular lesions are shown, involving all segments of the liver, the largest of which measures 8·6 cm; significant atrophy of the uninvolved portions of segments 2 and 3 is also noted (recipient 6).

repair. All patients received adjuvant therapy after liver transplantation, beginning 4–6 weeks after trans­ plantation and continuing for 4–6 months. Adjuvant therapy regimens included gemcitabine (n=4), capecitabine (n=1), and gemcitabine plus capecitabine

www.thelancet.com/gastrohep Published online March 13, 2018 http://dx.doi.org/10.1016/S2468-1253(18)30045-1

7

Articles

Recipient Time from diagnosis to outcomes transplant (months)

Neoadjuvant treatment therapies

1

2

Adjuvant therapy

3

4

5

6

7

Gemcitabine, cisplatin, and erlotinib

Gemcitabine

Erlotinib

Gemcitabine Fluorouracil Gemcitabine and capeciabine and gemcitabine

··

··

··

··

··

Gemcitabine and capecitabine

Gemcitabine, Gemcitabine Non-anatomic Folinic acid, and erlotinib cisplatin, liver resection fluorouracil, and irinotecan and erlotinib

Gemcitabine and cisplatin

··

Capecitabine

Gemcitabine

Gemcitabine and cisplatin

··

··

··

··

··

Gemcitabine

Gemcitabine and cisplatin

··

··

··

··

··

··

··

Gemcitabine

Gemcitabine and cisplatin

Gemcitabine

··

··

··

··

··

··

Gemcitabine

Anatomic Capecitabine and Gemcitabine, liver resection stereotactic body cisplatin, and erlotinib radiotherapy

1

34

Survived without recurrence

2

30

Survived with Gemcitabine and cisplatin metastatic disease

3

36

Death due to metastatic disease

4

15

Survived with Gemcitabine and cisplatin metastatic disease

5

10

Survived without recurrence

6

22

Survived without recurrence

Capecitabine Erlotinib and gemcitabine

Anatomic Gemcitabine liver resection and cisplatin

8

Table 3: Details of neoadjuvant and adjuvant therapy in liver transplant recipients with intrahepatic cholangiocarcinoma

100

Overall survival Recurrence-free survival

90

Patient survival (%)

80 70 60 50 40 30 20 10 0 Number at risk Overall survival Recurrence-free survival

0

6

12

18

24

30

36

42

48

54

60

3 2

2 1

2 1

Time after transplantation (months) 6 6

6 6

6 6

6 4

6 3

4 2

3 2

3 2

Figure 3: Cumulative overall and recurrence-free survival after liver transplantation for intrahepatic cholangiocarcinoma

(n=1; table 3). Two patients developed ascites during post-transplant adjuvant chemotherapy, which resolved upon therapy completion; for those patients, cytology from ascites was non-malignant and metastatic work-up remained negative. Recipients were followed for a median of 36 months (IQR 29–51) after transplantation (table 2). Overall survival after liver transplantation was 100% (95% CI 100–100) at 1 year, 83·3% (27·3–97·5) at 3 years, and 83·3% (27·3–97·5) at 5 years (figure 3). Recurrence-free survival was 50% (11·1–80·4) at 1, 3, and 5 years. Recurrence occurred in three of six patients at a median of 7·6 months (IQR 5·8–8·6) after transplantation. One patient died at 14·5 months after liver transplantation 8

because of recurrent cancer (diffuse metastases to the liver, lung, and bone). In a retrospective analysis of this patient’s pre-transplant images, a stable bony lesion present before transplantation might have indicated extrahepatic disease, explaining early recurrence. Recipient survival after recurrence was 66·7% (95% CI 5·4–94·7) at 1 year and 3 years. The other two patients with recurrence remain alive, with tumour burden controlled by systemic therapy at 32 months and 54 months after recurrence. Isolated pulmonary metastases were identified in one of these patients 4·1 months after transplantation. Retrospective review of pre-transplant imaging identified several stable pulmonary nodules that were too small to characterise. These nodules progressed into pulmonary metastases, indicating extrahepatic metastatic disease before trans­ plantation. This patient remains stable on targeted therapy. The second patient who remains alive with recurrence showed tumour invasion into perihilar fat on explant. An omental metastasis was noted at the time of elective hernia repair at 9·7 months after transplantation. Recurrence remains confined to the peritoneum, and the patient has undergone surgical peritoneal stripping, tumour debulking, and hyperthermic intraperitoneal chemo­therapy. Both patients remain stable and highly functional despite tumour recurrence. The remaining three patients were free of recurrent disease at 74, 28, and 25 months after transplantation. All recipients underwent genetic profiling for somatic mutations in the explant (table 4). Tumours from five recipients were analysed with next-generation sequencing, whereas one recipient underwent KRAS mutation analysis only because of availability of testing.

www.thelancet.com/gastrohep Published online March 13, 2018 http://dx.doi.org/10.1016/S2468-1253(18)30045-1

Articles

Genetic mutations were identified in all five recipients who underwent next-generation sequencing, and multiple mutations were detected in four of these recipients (range of one to 11 mutations). FGFR2 and SPTA1 were the most frequently mutated genes (in two recipients each). KRAS and BAP1 mutations were each identified in one patient, both of whom developed recurrent disease.

Discussion In this interim analysis of an initial case series, patients with stable intrahepatic cholangiocarcinoma before liver transplantation had an overall survival of 83% and a recurrence-free survival of 50% at 5 years. These findings suggest that tumour stability over time and response to therapy might serve as surrogate markers of favourable tumour biology for liver transplantation, and that the Methodist–MD Anderson selection criteria might identify subpopulations of patients with intrahepatic cholangiocarcinoma who would benefit most from liver transplantation. Only one previous report23 has suggested a potential survival advantage with pre-transplant therapy, but in patients with mostly incident tumours. In that report, 5-year recurrence-free survival was 50% in both patients with intrahepatic cholangiocarcinoma and in those with hilar cholangiocarcinoma. This finding led to renewed interest in the potential for liver transplantation as a therapy for intrahepatic cholangio­ carcinoma;11 however, patient selection criteria were largely undefined. Hilar cholangiocarcinoma is the most common biliary tract cancer in high-income countries; however, the incidence of intrahepatic cholangiocarcinoma has been increasing worldwide in the past three decades. Most cases of intrahepatic cholangiocarcinoma are detected at an advanced stage of unresectable disease, at which point median overall survival after systemic therapy is less than 1 year.26 Thus, novel approaches are needed for these patients. Liver transplantation is a widely accepted therapy for hilar cholangiocarcinoma and has shown the value of neoadjuvant therapy. However, prospective data regarding neoadjuvant therapy followed by liver transplantation for intrahepatic cholangiocarcinoma are scarce. Tumour size is a known risk factor for recurrence after liver resection for intrahepatic cholangiocarcinoma.2 Several single-centre and multicentre, international, retrospective analyses have suggested that tumour size might correlate with post-transplant recurrence. For instance, for intrahepatic cholangiocarcinoma with a cumulative radiographic diameter of less than 8 cm, a matched-cohort analysis27 in Spain reported overall survival of 78% at 1 year, 66% at 3 years, and 51% at 5 years. Similarly, in an international retrospective analysis,10 5-year overall survival was 80% for very early (<2 cm) tumours compared with 61% for intermediate tumours (2–3 cm) and 42% for advanced tumours

Recipient 1* Recipient 2

Recipient 3

Recipient 4 Recipient 5

Recipient 6

Pro302fs*33

0

0

0

0

0

0

0

0

0

0

0

Val600Glu

··

Leu329Pro

0

0

0

0

··

Rearrangement PRKN fusion intron 17

0

0

0

FGFR3

··

0

0

0

Amplification

0

FRS2

··

0

0

0

Amplification

0

IDH1

··

0

0

Arg132Cys

0

0

KDR

··

Asn141Asp

0

0

0

0

KRAS

0

0

0

Gly12Cys

0

0

MDM2

··

0

0

0

MITF

··

Tyr18Cys

0

0

0

0

MSH6

··

Arg772Gln

0

0

0

0

MYC

··

0

Amplification

0

0

0

MYST3

··

0

Amplification

0

0

0

NFKBIA

··

Thr185Met

0

0

0

0

PDK1

··

Arg238Cys

0

0

0

0

PRKAR1A

··

Glu10Lys

0

0

0

0

PTEN

··

0

0

0

Tyr76fs*2

0

SMAD4

··

0

0

0

Pro218fs*16

0

SMARCA4

··

Pro674Leu

0

0

SPTA1

··

Gln255Arg

0

0

BAP1

··

0

BLM

··

Arg15Cys

BRAF

··

FANCF FGFR2

Amplification

0 Arg891*subclonal

0

0 0

0 indicates no mutations. *This patient underwent KRAS mutation analysis only; all other patients underwent next-generation sequencing.

Table 4: Genetic mutation analysis of liver transplant recipients with intrahepatic cholangiocarcinoma

(>3 cm).10 In a Chinese series28 investigating liver transplantation for intrahepatic cholangiocarcinoma in patients treated with pre-transplant locoregional therapy or liver resection, a tumour size of greater than 5 cm was associated with disease recurrence. By contrast, a risk analysis of recurrence after transplantation for intra­hepatic cholangiocarcinoma suggested that tumour pathology and absence of neoadjuvant therapy were stronger multivariate predictors of disease recurrence after liver transplantation than was tumour size.20 Without prospective data on the efficacy of the Methodist–MD Anderson selection criteria in the setting of liver transplantation, we felt there were insufficient data to exclude patients from transplantation on the basis of tumour size. In fact, the median cumulative diameter among transplanted patients was 14·2 cm, and no patient had a cumulative tumour diameter of less than 5 cm. For the three patients with long-term recurrence-free survival, the median cumulative diameter was 18·7 cm (IQR 12·6–19·4). With such results even in patients with large tumours, a clear-cut size discrimination regarding transplant candidacy cannot be made, and a larger cohort will be necessary to establish this correlation. However, our results suggest that chemo-responsiveness serves as a better surrogate marker of tumour biology than does size for intrahepatic cholangiocarcinoma.

www.thelancet.com/gastrohep Published online March 13, 2018 http://dx.doi.org/10.1016/S2468-1253(18)30045-1

9

Articles

As per the Methodist–MD Anderson selection criteria, disease stability or regression was required for at least 6 months before transplantation. Despite all patients having locally advanced disease, 5-year overall survival in intrahepatic cholangiocarcinoma was significantly improved with transplantation compared with liver resection (83% vs 25% in a previous study2); however, 50% of patients in our study developed recurrent disease. In retrospect, several recurrences might have been avoided with more careful patient selection. Two of six patients showed stable extrahepatic lesions (bone and subcentimetre lung lesions) on imaging that were confirmed to be metastases after transplantation. More aggressive pre-transplant sampling of such lesions is warranted, particularly in patients on continuous chemotherapy. The third patient with intraperitoneal recurrence had explant tissue evidence of local tumour invasion into the perihilar fat, resulting in positive margins. Perihilar biopsy, in addition to portal lymphadenectomy, during the exploratory portion of the transplant hepa­ tectomy could potentially avoid transplantation in such patients. This initial patient series has the potential to serve as a guide for development of a larger clinical trial for analysis of liver transplantation for the treatment of locally advanced intrahepatic cholangiocarcinoma. For such a trial, careful attention should be paid to the presence of indolent disease, especially given the requirement for continued neoadjuvant therapy before liver trans­ plantation. Additionally, tumours in close proximity to the porta hepatis should be carefully considered given the potential for these tumours to locally invade perihilar structures. Ultimately, however, tumour biology, rather than size, appears to be the best indicator for potential liver transplantation recipients. Without data from a randomised controlled trial, it is difficult to know whether patients with biologically favourable disease would do equally well without liver transplantation. The ABC-02 trial24 found that the median overall survival of patients with biliary tract cancers treated with gemcitabine and cisplatin was 11 months, with 6-month progression-free survival of 57·1%. With chemotherapy alone, however, few patients were alive at 5 years. Only 20–40% of patients with intrahepatic cholangio­ carcinoma treated with liver resection are alive at 5 years, most of whom have early-stage disease. In a study29 in 933 patients undergoing surgical resection, 5-year overall survival was 41%, and 73% of patients had disease recurrence after surgical resection. Recurrence risk was greatest in patients with multiple tumours or in those with tumours greater than 5 cm in diameter.29 Based on these criteria, all patients transplanted in this series would be at high risk of recurrent disease after resection. 5-year overall survival for patients with the same disease extent as the patients in this study has rarely been more than 80%, regardless of chemotherapy or resection. 10

Our observation that no patient had a complete pathological response on explant, together with the occurrence of previous disease recurrence after resection in two patients, suggested that the disease present in these transplanted patients remained aggressive. Without transplant, patients would probably have progressed or succumbed to liver toxicity from chemotherapy. One potential consideration is the number of patients with intrahepatic cholangiocarcinoma who would potentially be eligible for this protocol. Because of the study design, neither the proportion of patients with intrahepatic cholangiocarcinoma who would be downstaged via medical therapy to become surgically resectable, nor the number of patients who would develop stable disease on medical therapy, could be discerned. However, initial observations from a clinical trial30 using gemcitabine, cisplatin, and paclitaxel for the treatment of advanced intrahepatic cholangiocarcinoma suggest that roughly 10–20% of patients have the potential for downstaging to resection and 50% have the potential for developing disease stability or responding to treatment. We hypothesise that a small number of patients who have stable disease or respond to treatment (possibly another 10–20% with favourable genetics) might benefit from liver transplantation. In view of the results of this study, liver transplantation might even be preferred to liver resection, particularly when compared with radical hepatobiliary techniques for resection of advanced intrahepatic cholangiocarcinoma, such as hepatic vein re-implantation, portal vein and hepatic artery reconstruction, in-situ cold perfusion, and ex-vivo resections. Tumour biology probably had an important part in the outcomes of our patients, given that the selected patients had sustained chemotherapy responses. Advances in next-generation sequencing and the commercial availability of comprehensive genetic profiling have substantially expanded knowledge of genetic mutations common to intrahepatic cholangiocarcinoma.31–33 In the future, profiling might identify patients with targetable mutations and favourable biology for transplant versus those with an increased likelihood of post-transplant recurrence.31,34 For example, KRAS, BAP1,34,35 and CDKN2A32 mutations occur commonly in intrahepatic cholangiocarcinoma and are associated with an aggressive phenotype. KRAS and BAP1 mutations were noted in two patients in our study, both of whom had disease recurrence. FGFR2 mutations, particularly FGFR fusions, are associated with reasonably indolent clinical courses and might indicate a favourable tumour biology for liver transplantation. Two patients in our series had FGFR2 mutations and did not have disease recurrence. However, with only five of six patients undergoing comprehensive genetic profiling, the study was insufficiently powered to definitively identify recurrence related to genetic profiles. Particular tumour characteristics might be correlated with disease progression or stability, and patterns might become evident as additional data are accrued.

www.thelancet.com/gastrohep Published online March 13, 2018 http://dx.doi.org/10.1016/S2468-1253(18)30045-1

Articles

This study has several limitations. Although the study used prospective data, the rarity of intrahepatic cholangiocarcinoma, combined with stringent require­ ments for long-term disease stability and uncertainty regarding the role of transplantation in intrahepatic cholangiocarcinoma, resulted in only nine patients being listed for liver transplantation over a 6-year period. Additionally, MELD exception was not available for these patients. Therefore, identification of organs for transplant for these patients was difficult, and discarded or extendedcriteria organs were often used. Because of the time that elapsed between diagnosis and transplantation (median of 26 months), regimens after the initial gemcitabine-based therapy were hetero­geneous among patients, reflecting an absence of standard in this setting. Moreover, because of the small sample size, statistical analysis was limited, and larger-scale prospective studies will be necessary to validate the results. Despite this, the prospective nature of our investigation allowed for collection of detailed information about treatment and pathology, which is often not available in retrospective analyses. Critics might question the use of donor livers for this disease indication given the shortage of livers for transplantation. Only one recipient had a high physiological MELD score at transplantation. All other deceased-donor recipients received marginal livers refused by all other centres, and these livers were considered too marginal for transplant into our standard-criteria patients. These recipients could accept such marginal organs because of their high underlying physiological reserve. Finally, there was potential for bias in the determination of resectability because this decision was made solely by the surgeon members of the MMAJCCC; however, the potential for bias was minimised through inclusion of both hepatobiliary surgeons and liver transplant surgeons from both institutions. To our knowledge, this study is the first prospective case-series of liver transplantation for intrahepatic cholangiocarcinoma in patients with stable disease after neoadjuvant therapy under the Methodist–MD Anderson selection criteria. Our results indicate the feasibility of this approach and highlight the need for a large-scale, multicentre, prospective trial of neoadjuvant therapy followed by liver transplantation for intrahepatic cholangiocarcinoma in the future. Contributors KEL, MJ, KH, RTS, HPM, J-NV, and RMG contributed to conception and design of the study. KEL, MJ, KH, RTS, RA-W, NG, CMM, AS, DWV, DTN, EAG, AOK, RSM, TAA, CC, XCL, AOG, J-NV, and RMG contributed to data analysis and interpretation. All authors participated in critical revision of the manuscript for important intellectual content. KEL, MJ, KH, RTS, RA-W, NG, CMM, AS, DWV, HPM, J-NV, and RMG contributed to data acquisition. KEL, MJ, and NG participated in drafting of the manuscript. All authors were responsible for final approval of the version to be published. Declaration of interests RTS reports support from Celgene, Eli Lilly, Agios, Codiak, Amgen, and Halozyme, outside the submitted work. DWV reports support from Intercept and Bristol-Myers Squibb, outside the submitted work. All other authors declare no competing interests.

Acknowledgments We thank Julie Corkrean, Julie Luczon, and Samantha Bullock (all Methodist J C Walter Jr Transplant Center, Sherrie and Alan Conover Center for Liver Disease and Transplantation, Houston Methodist Hospital, Houston, TX, USA) for their assistance with data collection and preparation, and Linda W Moore (Methodist J C Walter Jr Transplant Center, Sherrie and Alan Conover Center for Liver Disease and Transplantation, Houston Methodist Hospital, Houston, TX, USA) for her expert assistance with manuscript review. This work was presented, in part, at the 2017 Joint International Congress of the International Liver Transplantation Society, the European Liver and Intestine Transplant Association, and the Liver Intensive Care Group of Europe in Prague, Czech Republic, on May 24–27, 2017. References 1 Banales JM, Cardinale V, Carpino G, et al. Expert consensus document: cholangiocarcinoma: current knowledge and future perspectives consensus statement from the European Network for the Study of Cholangiocarcinoma (ENS-CCA). Nat Rev Gastroenterol Hepatol 2016; 13: 261–80. 2 Spolverato G, Kim Y, Ejaz A, et al. Conditional probability of long-term survival after liver resection for intrahepatic cholangiocarcinoma: a multi-institutional analysis of 535 patients. JAMA Surg 2015; 150: 538–45. 3 Wang Y, Li J, Xia Y, et al. Prognostic nomogram for intrahepatic cholangiocarcinoma after partial hepatectomy. J Clin Oncol 2013; 31: 1188–95. 4 Benson AB 3rd, Abrams TA, Ben-Josef E, et al. NCCN clinical practice guidelines in oncology: hepatobiliary cancers. J Natl Compr Canc Netw 2009; 7: 350–91. 5 Endo I, Gonen M, Yopp AC, et al. Intrahepatic cholangiocarcinoma: rising frequency, improved survival, and determinants of outcome after resection. Ann Surg 2008; 248: 84–96. 6 Hyder O, Hatzaras I, Sotiropoulos GC, et al. Recurrence after operative management of intrahepatic cholangiocarcinoma. Surgery 2013; 153: 811–18. 7 Tabrizian P, Jibara G, Hechtman JF, et al. Outcomes following resection of intrahepatic cholangiocarcinoma. HPB 2015; 17: 344–51. 8 Doussot A, Gonen M, Wiggers JK. Recurrence patterns and disease-free survival after resection of intrahepatic cholangiocarcinoma: preoperative and postoperative prognostic models. J Am Coll Surg 2016; 223: 493–505. 9 Gupta R, Gupta J. Strategies to improve survival of patients with intrahepatic cholangiocarcinoma undergoing liver transplantation. Hepatology 2017; 65: 1777–78. 10 Sapisochin G, Facciuto M, Rubbia-Brandt L, et al. Liver transplantation for “very early” intrahepatic cholangiocarcinoma: international retrospective study supporting a prospective assessment. Hepatology 2016; 64: 1178–88. 11 Rana A, Hong JC. Orthotopic liver transplantation in combination with neoadjuvant therapy: a new paradigm in the treatment of unresectable intrahepatic cholangiocarcinoma. Curr Opin Gastroenterol 2012; 28: 258–65. 12 Goldstein RM, Stone M, Tillery GW, et al. Is liver transplantation indicated for cholangiocarcinoma? Am J Surg 1993; 166: 768–71. 13 Pichlmayr R, Weimann A, Oldhafer KJ, et al. Role of liver transplantation in the treatment of unresectable liver cancer. World J Surg 1995; 19: 807–13. 14 Becker NS, Rodriguez JA, Barshes NR, O’Mahony CA, Goss JA, Aloia TA. Outcomes analysis for 280 patients with cholangiocarcinoma treated with liver transplantation over an 18-year period. J Gastrointest Surg 2008; 12: 117–22. 15 Schüle S, Altendorf-Hofmann A, Uteß F, et al. Liver transplantation for hilar cholangiocarcinoma—a single-centre experience. Langenbecks Arch Surg 2013; 398: 71–77. 16 Darwish Murad S, Kim WR, Harnois DM, et al. Efficacy of neoadjuvant chemoradiation, followed by liver transplantation, for perihilar cholangiocarcinoma at 12 US centers. Gastroenterology 2012; 143: 88–98. 17 Marchan EM, Landry JC. Neoadjuvant chemoradiation followed by orthotopic liver transplantation in cholangiocarcinomas: the emory experience. J Gastrointest Oncol 2016; 7: 248–54.

www.thelancet.com/gastrohep Published online March 13, 2018 http://dx.doi.org/10.1016/S2468-1253(18)30045-1

11

Articles

18 Rea DJ, Heimbach JK, Rosen CB, et al. Liver transplantation with neoadjuvant chemoradiation is more effective than resection for hilar cholangiocarcinoma. Ann Surg 2005; 242: 451–61. 19 Sudan D, DeRoover A, Chinnakotla S, et al. Radiochemotherapy and transplantation allow long-term survival for nonresectable hilar cholangiocarcinoma. Am J Transplant 2002; 2: 774–79. 20 Hong JC, Petrowsky H, Kaldas FM, et al. Predictive index for tumor recurrence after liver transplantation for locally advanced intrahepatic and hilar cholangiocarcinoma. J Am Coll Surg 2011; 212: 514–20. 21 Robles R, Figueras J, Turrión VS, et al. Spanish experience in liver transplantation for hilar and peripheral cholangiocarcinoma. Ann Surg 2004; 239: 265–71. 22 Meyer CG, Penn I, James L. Liver transplantation for cholangiocarcinoma: results in 207 patients. Transplantation 2000; 69: 1633–37. 23 Hong JC, Jones CM, Duffy JP, et al. Comparative analysis of resection and liver transplantation for intrahepatic and hilar cholangiocarcinoma: a 24-year experience in a single center. Arch Surg 2011; 146: 683–89. 24 Valle JW, Wasan H, Johnson P, et al. Gemcitabine alone or in combination with cisplatin in patients with advanced or metastatic cholangiocarcinomas or other biliary tract tumours: a multicentre randomised phase II study—The UK ABC-01 Study. Br J Cancer 2009; 101: 621–27. 25 Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004; 240: 205–13. 26 Patel T. Cholangiocarcinoma—controversies and challenges. Nat Rev Gastroenterol Hepatol 2011; 8: 189–200.

12

27 Sapisochin G, de Lope CR, Gastaca M, et al. Intrahepatic cholangiocarcinoma or mixed hepatocellular-cholangiocarcinoma in patients undergoing liver transplantation: a Spanish matched cohort multicenter study. Ann Surg 2014; 259: 944–52. 28 Fu BS, Zhang T, Li H, et al. The role of liver transplantation for intrahepatic cholangiocarcinoma: a single-center experience. Eur Surg Res 2011; 47: 218–21. 29 Zhang XF, Beal EW, Bagante F, et al. Early versus late recurrence of intrahepatic cholangiocarcinoma after resection with curative intent. Br J Surg 2017; published online Nov 28. DOI:10.1002/ bjs.10676. 30 Shroff RT, Borad MJ, Xiao L, et al. A phase II trial of gemcitabine (G), cisplatin (C), and nab-paclitaxel (N) in advanced biliary tract cancers (aBTCs). Proc Am Soc Clin Oncol 2017; 35: 4018. 31 Hayashi A, Misumi K, Shibahara J, et al. Distinct clinicopathologic and genetic features of 2 histologic subtypes of intrahepatic cholangiocarcinoma. Am J Surg Pathol 2016; 40: 1021–30. 32 Jain A, Kwong LN, Javle M. Genomic profiling of biliary tract cancers and implications for clinical practice. Curr Treat Options Oncol 2016; 17: 58. 33 Lee H, Ross JS. The potential role of comprehensive genomic profiling to guide targeted therapy for patients with biliary cancer. Therap Adv Gastroenterol 2017; 10: 507–20. 34 Zhu AX, Borger DR, Kim Y, et al. Genomic profiling of intrahepatic cholangiocarcinoma: refining prognosis and identifying therapeutic targets. Ann Surg Oncol 2014; 21: 3827–34. 35 Al-Shamsi HO, Anand D, Shroff RT, et al. BRCA-associated protein 1 mutant cholangiocarcinoma: an aggressive disease subtype. J Gastrointest Oncol 2016; 7: 556–61.

www.thelancet.com/gastrohep Published online March 13, 2018 http://dx.doi.org/10.1016/S2468-1253(18)30045-1