Beyond futility: The history and potential of liver transplantation in cholangiocarcinoma

Abstract

Cholangiocarcinoma (CCA) is a rare type of cancer which arises from the bile duct epithelium and carries a poor prognosis. One of the main risk factors in the Western world is primary sclerosing cholangitis. Surgical resection has traditionally been the only curative treatment but can only be offered to patients with early disease, excluding those with locally advanced disease. Despite initial poor outcomes, liver transplantation (LT) has evolved as a viable treatment for a select group of patients with CCAs that are deemed unresectable. This review aims to explore the evolution of the role of LT in patients with CCA.

Key Words: Cholangiocarcinoma; Extrahepatic cholangiocarcinoma; Intrahepatic cholangiocarcinoma; Primary sclerosing cholangitis; Liver transplantation

Core Tip: Liver transplantation (LT) had previously been contraindicated in the treatment of cholangiocarcinoma (CCA). However, recent advances in both oncological treatment and patient-selection criteria have led to a paradigm shift, with various studies showing that patients with locally advanced CCA undergoing LT have promising outcomes. Numerous centers in North America and Europe have now adopted this as standard of care. Organ scarcity is a major factor to be considered. These complex patients are best managed by a multi-disciplinary team.

INTRODUCTION

Cholangiocarcinoma (CCA) is a rare type of cancer which arises from the bile duct epithelium. CCA is the second most common primary tumour of the liver following hepatocellular carcinoma (HCC), and has an increasing global incidence and mortality[1]. It is aggressive and carries a poor prognosis with a median survival of 24 months after diagnosis despite treatment[2].

CCA is classified into two subtypes based on anatomical position, extrahepatic and intrahepatic. Extrahepatic CCA makes up 90% of all CCA and can be further divided into distal CCA (dCCA) and hilar CCA (hCCA)[2,3]. dCCA is located between the ampulla and the origin of the cystic duct, whilst hCCA is located above the origin of cystic duct up to the second order biliary ducts[4]. hCCA may also be referred to as perihilar or Klatskin tumours. Intrahepatic CCA (iCCA) is located within the liver parenchyma. Anatomical subtype is important because it dictates management strategies. The main risk factors for CCA include primary sclerosing cholangitis (PSC), cirrhosis, hepatitis B, hepatitis C, choledochal cysts, smoking, diabetes and alcohol[5]. Liver fluke is a common cause of CCA in endemic areas[6].

Surgical resection is the only curative treatment for CCA but is precluded in patients with locally advanced or metastatic disease. Anatomical subtype dictates the approach to surgical resection. dCCA is treated with excision of the extrahepatic bile duct as part of a pancreatoduodenectomy. Surgical resection in iCCA and hCCA is determined by location, radiological extension and whether the remnant liver volume carries a risk of liver failure. Innovative techniques like embolisation of the portal vein and associated liver partition with portal vein ligation for staged hepatectomy have been developed to facilitate removal of larger liver volumes while inducing hypertrophy of the unaffected liver lobe[7]. Advances in surgical techniques in patients with vascular invasion include venous and arterial reconstruction. Although these techniques have higher peri-operative mortality and morbidity, the outcomes are still better than best supportive care[8,9].

Liver transplantation (LT) has evolved as a viable treatment option for patients with CCA with promising outcomes. Despite this, CCA remains a contraindication for transplantation in most countries. This review aims to explore the evolution of the role of LT in patients with CCA, as well as those with both PSC and CCA.

HILAR OR PERIHILAR CCA (hCCA, KLATSKIN TUMOUR)

hCCA is the most common subtype of CCA and is located proximal to the cystic duct origin. It has a complex diagnostic pathway and surgical resection is challenging due to its anatomical location. Incidence varies between 0.3-3.5 per 100000/year in North America and 7.5 per 100000/year in Mediterranean regions[10]. The Bismuth-Corlette classification is used to define the macroscopic appearance to plan surgical approach[11]. One third of hCCA are not resectable at diagnosis and another third are found to have extensive disease at the time of surgical exploration. This led to consideration of LT as a treatment option in this cohort, but due to high recurrence rates and low survival initially, LT was soon contraindicated[12,13].

A study from Kentucky, United States in the late 1990s described an early experience of preoperative chemoradiation followed by resection of extrahepatic CCA. Margin-negative resections were achieved in 100% of patients who received preoperative chemoradiation, compared to 54% in patients who did not[14]. A 1999 study evaluated the results of radiotherapy with various combinations of chemotherapy and surgical resection with selective transplantation in this patient group[15], and concluded that combined modality therapy with complete surgical resection (with or without transplantation) can be curative for this patient group.

Interest in the field grew and other centres started to consider neoadjuvant therapy ahead of LT in this cohort. In 2002, Sudan et al[16] evaluated the effect of neoadjuvant chemoradiation therapy (biliary brachytherapy and intravenous 5-fluorouracil) and orthoptic LT in a selected group of patients with hCCA. Long-term tumour-free survival was achieved in 45% of transplanted patients, providing evidence that hCCA should not be an absolute contraindication for LT.

The Mayo Protocol (Figure 1) has proved to be the most successful strategy for the treatment of previously unresectable hCCA with preoperative irradiation, chemotherapy and LT[17,18]. Patients are eligible if they have a malignant-appearing stricture on cholangiography plus one of the following: Positive or strongly suspicious cytology/histopathology for CCA, CA19-9 > 100 U/mL, polysomy by fluorescence in situ hybridisation or a mass-forming lesion < 3 cm in radial diameter[18]. Patients are excluded if they have extrahepatic disease, had previous radiation to their abdomen, had a previous percutaneous transhepatic tumour biopsy or were not surgical candidates for LT[18]. Eligible patients then receive external-beam radiotherapy over 3 weeks plus an infusion of 5-FU, followed by intraluminal brachytherapy and 5-FU. Patients then undergo an exploratory laparotomy to exclude metastatic disease and have oral capecitabine until their LT.

Figure 1 Mayo clinic protocol for liver transplantation. EBRT: External beam radiotherapy; FU: Fluorouracil; FISH: Fluorescence in situ hybridisation.

In the Mayo Clinic pilot study, 19 patients were enrolled, 11 proceeded to LT and 8 were followed up for over 12 months at the time of publishing their results[17]. One patient developed disease recurrence but the remainder remained disease-free. They published further results a few years later where 28 patients had neoadjuvant therapy followed by LT, and the survival rate for this group was 82% at 5 years after transplantation[18]. They concluded that neoadjuvant chemoradiotherapy with LT achieved excellent outcomes for patients with locally advanced, regional lymph-node-negative hCCA. The promising outcomes with the Mayo protocol have been attributed to a strict adherence to the selection criteria[19].

A few years later, a multi-center retrospective study by Darwish Murad et al[20], in the United States was performed to assess the overall effectiveness of neoadjuvant therapy followed by LT in patients with hCCA[20]. Their results showed a 5-year disease-free survival rate of 65%. Another multi-center retrospective study, from 2000-2015, compared the outcomes of hCCA patients undergoing resection vs transplantation between[21]. It found that patients with a hCCA meeting criteria for transplantation (< 3 cm, lymph node negative disease) who underwent resection had decreased survival compared to those with unresectable disease who were transplanted; 5-year overall survival rate of 18% vs 64%. Of note, 61% of patients who were transplanted had PSC, compared to 2% of those who had resection[21].

A metanalysis published in 2021 Looked at the survival data following neoadjuvant chemoradiation and LT for patients with unresectable hCCA[22]. The time period covered was 2000-2019, following the publication of the Mayo protocol. The pooled 5-year survival rate following LT for patients with neoadjuvant therapy was 65.1% vs 31.6% in those without. Pooled recurrence of CCA after 3 years was 24.1% in patients with neoadjuvant therapy vs 51.7% in patients without. Patients with PSC appeared to have the most favourable outcomes.

Toronto General hospital recently developed a neoadjuvant regimen modelled on Mayo protocol[23]. The overall survival was 83.3% and 55.5% at 1 and 2 years respectively but 5-year data is not yet available[24]. Currently, CCA is not an accepted indication for LT in the United Kingdom, outside an ongoing pilot programme which is led by the Liver Advisory Group[25].

iCCA

iCCA is the second most common malignancy of liver and constitutes 10% of CCAs. Most are adenocarcinomas and only a small proportion are resectable at diagnosis[26]. The mean age of diagnosis is over 65 years and it is more common in men[27,28]. Histologically it can be divided into intraductal-infiltrating, mass-forming or periductal[29].

Liver resection is currently the only accepted curative treatment for iCCA. Liver resection is considered appropriate in iCCA patients if a free-margin resection is possible (R0) and if there is sufficient volume of the remnant liver to avoid post-operative liver failure[30], i.e. early disease. Early diagnosis in this patient group can be tricky, as most patients with iCCA will only develop symptoms such as abdominal pain, weight loss and jaundice with advanced disease; the majority of patients will present with advanced disease which is unresectable[30,31]. Even in patients undergoing a liver resection for iCCA, the median overall survival is about 40 months, with a 5-year overall survival of 25%-40%[30,32-34].

LT had been largely contraindicated in this group of patients due to poor initial results, with earlier studies showing a 2-year survival rate of only 30%[35-37]. The second ever successful transplant by Thomas Strazl in 1963 was for iCCA and the patient died on post operative day 7[38].

However, the paradigm began to change: Patient selection criteria was proving to be essential in improving the survival in early-stage HCC patients undergoing LT[30,39] and promising results were seen in patients with hCCA receiving LT. This, alongside the advancements in neoadjuvant chemo and radiotherapy, reopened the door to considering LT in iCCA.

In 2014, a multicentre retrospective cohort study reviewed cirrhotic patients who had undergone LT and had an iCCA in their liver explant[40]. The study aimed to identify risk factors for recurrence in these patient as well as assess the outcomes of patients with ‘very early’ iCCA, defined as a single tumour ≤ 2 cm. Their results identified the following as risk factors: Larger tumour volume, microscopic vascular invasion and poor degree of differentiation. The ‘very early’ iCCA group showed no tumour recurrence and achieved excellent overall survival rates (73% at 5 years), and the study recommended that validating these findings may alter the current exclusion of these patients from LT.

The same group carried out a larger retrospective multicentre study in 2016, and reviewed all patients with an iCCA in their liver explant who were transplanted for either an HCC or decompensated cirrhosis[41]. The patients were classified as ‘very early’ iCCA, defined as a single tumour ≤ 2 cm and ‘advanced’ iCCA, defined as either a single tumour > 2 cm or multi-focal disease, and their 1, 3 and 5-year recurrence and overall survival rates were compared. The study similarly showed that cirrhotic patients with ‘very early’ iCCA had significantly lower rates of recurrence and higher survival rates, and concluded that they may be considered for LT.

A retrospective study in 2011 reviewed patients transplanted for an advanced iCCA and hCCA to identify risk factors for recurrence and to develop a prognostic scoring system for risk stratification in this group[42]. Risk stratification was divided into low, intermediate and high-risk based on a scoring system. Factors that were found to be predictive for tumour recurrence included multifocal tumour, perineural and lymphovascular invasion, lack of neoadjuvant and adjuvant therapy and a history of PSC. Recurrence-free survival was significantly higher in the low-risk group, and there was a survival benefit in the intermediate group when compared to the high-risk group. The study concluded that there were grounds for expanding LT criteria in selected patients with locally advanced iCCA.

A prospective case-series in a single-center published their data in 2018, specifically looking at the efficacy of LT in patients with locally advanced iCCA who had undergone neoadjuvant therapy[40]. The group used their own protocol for neoadjuvant therapy which consisted of gemcitabine-based chemotherapy as first-line treatment. Patients with radiological response or stability following at least 6 months of neoadjuvant therapy were listed for transplantation (n = 6). Overall survival rates were 100%, 83.3% and 83.3% at 1, 3 and 5 years respectively. The study concluded that selected patients with locally advanced iCCA with stable disease on neoadjuvant therapy may benefit from LT.

Further data was published by the same group in 2022[43] when they had performed LT on 18 patients on neoadjuvant therapy. Overall survival rates were 100%, 71% and 57% at 1, 3 and 5 years respectively. The data supported the group’s hypothesis that stable disease on neoadjuvant therapy was a surrogate measure for long-term survival after liver transplantation.

Another center published their experience with LT for CCA[44] and showed that multimodal neoadjuvant treatment (chemotherapy and local therapy) was associated with improved survival in LT for locally advanced iCCA, regardless of tumour size.

IMMUNOTHERAPY AND TARGETED THERAPY IN PERI AND POST LT

It is imperative to touch upon the various advances in oncological therapy and the potential role they may play in the future of LT in CCA patients. Immunotherapy and immune checkpoint inhibitors (ICI) have revolutionised cancer management and are now being used in advanced CCA, in combination with chemotherapy[45].

The most commonly used agents are in CCA are Pembrolizumab, an anti-programmed death-1 (PD1) antibody and Durvalumab, an anti-programmed death ligand 1 (PD-L1) antibody. Chemotherapy regimens with either Pembrolizumab or Durvalumab have recently been approved by both the United States Food and Drug Administration and the European Medicines Agency as first-line combinations for advanced CCA[45-48]. The role of using targeted therapy to further enhance the effects of ICI is also being explored; for example, anti-vascular endothelial growth factor agents and tyrosine kinase inhibitors and their potential to increase tumour response to PD-L1 inhibitors and anti-PD1 agents[45,49-51]. Next generation sequencing technologies have also enabled the molecular analysis of tumours such as CCA, therefore identifying specific genetic alterations that can be targeted by small molecule inhibitors[45,52].

While these novel therapies have found their place in the management of advanced CCA, their role in peri and post-LT patients is yet to be determined[53]. The use of ICIs in patients with HCC is being explored to downstage disease to result in eligibility for a transplant, but also to control disease burden whilst a transplant is awaited[39,54]. There is also the question of ICI use in transplanted patients with tumour recurrence. The concern however is that ICIs are associated with multiple immune-related adverse events, such as colitis, hepatitis/hepatic necrosis and pneumonitis, and in the context of LT, graft-rejection[55,56]. Further data are required to understand the benefits vs risks of these treatments and to fully understand the role they may play in both the pre and post-transplantation period.

PSC AND CCA

PSC is a rare condition caused by biliary inflammation resulting in strictures and ultimately liver fibrosis[57]. PSC is more common in men[58], is strongly associated with inflammatory bowel disease and can be seen in conjunction with autoimmune hepatitis[59]. The risk of CCA in PSC is more than 100 times that of the general population and is highest in the first year following a diagnosis of PSC[60,61].

Diagnosing CCA in PSC remains a challenge. This is often due to overlap of symptoms with benign progression of disease, difficulty in differentiating benign from malignant strictures and poor sensitivity of sampling from strictures[62,63]. Although there is widespread adoption of annual imaging in this cohort, a recent prospective surveillance study found it to be ineffective in early diagnosis of CCA[64].

hCCA in PSC can be treated curatively with a liver resection if liver function can be preserved with good liver volumes after radical resection. This can be challenging in PSC patients who have parenchymal fibrosis resulting in low liver volumes and therefore high mortality and morbidity[65].

With increasingly favourable outcomes, LT is emerging as a potentially curative alternative to resection in hCCA with PSC. A multi-centre study by Ethun et al[21] reported an 18% vs 64% 5-year survival in PSC patients with hCCA undergoing liver resection vs LT, respectively.

The Mayo protocol has also been adopted in these patients with favourable results. In a highly selected group, the Mayo protocol reported a 5-year survival of 82%[18]. Various centres have adopted versions of the Mayo protocol and the success has led to an award of model for end-stage liver disease exception points for this indication[20,22]. In the Mayo protocol group, long-term data showed better long-term survival in perihilar CCA (pCCA) and PSC (74%) compared to de-novo pCCA (58%)[66]. Despite the improving 5-year survival, it needs to be recognised that PSC recurs in 15%-25% of patients in post-transplant period[67], and it is imperative to use strict diagnostic and selection criteria in this cohort.

It is also worth noting that PSC patients are often diagnosed younger, have fewer co-morbidities and are often transplanted earlier in their disease course compared to other liver conditions. This generally means that they have better post-LT outcomes compared with other liver diseases, and therefore dominate LT cohorts. But it is important to acknowledge and consider this selection bias when interpreting data and results[68].

iCCA can potentially be curable with resection if adequate liver volumes can be preserved. LT is still not established for this indication but studies have shown favourable outcomes in iCCA discovered incidentally in explants of PSC patients following transplantation[69]. Small studies have shown improved survival following adjuvant chemotherapy[40] but more trials are needed and are underway.

Novel therapies with check point inhibitors have been used in CCA with PSC with some promising results[48,70].

FUTURE DEVELOPMENTS

There are a number of ongoing clinical trials which will shed further light on the role of LT in the treatment of CCA.

The clinical trial NCT02232932 (TRANSPHIL) is a prospective randomized trial in France, which aims to compare 5-year survival and recurrence-free survival at 3 years in patients having liver resection (control group) compared to those receiving neoadjuvant chemo-radiotherapy and then a liver transplant[71]. A prospective study is being undertaken in Toronto: NCT02878473, to evaluate the effectiveness of LT as a treatment for patients with early iCCA and cirrhosis, as measured by 5-year survival and disease recurrence rates[72].

Another single-centre study in Toronto (NCT04195503) is evaluating the effectiveness of LT in patients with locally advanced, unresectable, non-metastatic iCCA in patients who have had 6-months of stable disease on neoadjuvant chemotherapy. Outcomes will again be overall survival and disease recurrence[73].

Lastly, there is a prospective trial being conducted at Oslo University in Norway: The TESLA trial NCT04556214, which looks at LT in locally advanced, unresectable non-metastatic iCCA, treated with neoadjuvant systemic therapy for at least 6 months[74].

CONCLUSION

CCAs are a challenging group of aggressive tumours which require a multi-disciplinary approach for both diagnosis and treatment. The role of LT in CCA has evolved significantly with numerous centers in North America and Europe adopting this as standard of care. Organ scarcity makes LT in this group controversial, but strict selection criteria could ensure good outcomes without disadvantaging other groups. Currently there are no standardized oncological protocols for this cohort. The future for this would be to develop personalised models for these patients to achieve the best outcomes.

In view of the substantial evidence-base that now exists, patients with CCA who cannot access assessment in LT programmes may be unfairly disadvantaged. When carefully selected, outcomes approach those of conventional indications for LT. Additionally, rapid progress in personalised oncological treatment, including immunotherapy and agents that target individual molecular and genetic characteristics, may herald an era of substantially improved post-LT survival. The progress described in this review appears to justify greater access to LT in appropriately managed programmes, thus allowing further research in what is a new era of anti-cancer therapy.