It was a privilege for me to discuss the indications for treating hilar cholangiocarcinoma (CCA) with liver transplantation and the outcomes achieved with the Mayo Clinic protocol during a general session of the 2018 Gastrointestinal Cancer Symposium.

The presentation and discussion provided an opportunity to discuss the role of liver transplantation in the treatment of this devastating disease, review indications and contraindications for transplantation, appreciate the results possible with multimodality treatment, and emphasize the need to avoid intervention during diagnosis and treatment that could preclude transplantation.

CCA is best treated with surgical resection. When a complete, potentially curative, R0 resection is possible, five year survival is approximately 30–40%. Unfortunately, few patients have cancers amenable to resection. Even when resection seems possible, an R0 resection is achieved only 75% of the time.

We developed a multimodal protocol at Mayo Clinic in 1993 designed to treat carefully selected patients with early stage, unresectable hilar cholangiocarcinoma or cholangiocarcinoma arising in the setting of primary sclerosing cholangitis (PSC). The protocol utilizes the efficacy of high dose external beam radiotherapy, chemosensitization with intravenous 5-flourouracil, and high dose intraluminal brachytherapy as neoadjuvant therapy. Capecitabine is administered as tolerated while patients await transplantation. All patients undergo operative staging prior to transplantation — either as time on the waiting list nears for deceased donor liver availability or the day before living donor transplantation. Liver transplantation is done with low division of the portal vein and bile duct, replacement of the artery during deceased donor transplantation, and biliary reconstruction with a Roux-en-Y choledochojejunostomy. Patients are followed closely during the first four months after transplantation when they are at increased risk for vascular complications (as compared to other liver transplant recipients) due to the neoadjuvant radiotherapy.

Mayo Clinic Rochester has treated 319 patients per this protocol beginning in 1993, and 199 have undergone transplantation. Approximately 20% had findings at operative staging that precluded transplantation — 16% for patients with CCA arising in PSC and 28% for those with CCA arising de novo. Five year survival rates from the start of treatment were 60% for patients with CCA/PSC and 38% for patients with de novo CCA. Five year survival rates for patients that underwent transplantation were 76% for patients with CCA/PSC and 56% for patients with de novo CCA. A total of 51 patients have developed recurrent CCA after transplantation — 17% of those with CCA/PSC and 41% of those with de novo CCA.

Side-effects and complications of neoadjuvant therapy are common. Almost all patients have at least one episode of cholangitis, and many require hospitalization for intravenous antibiotic therapy and endoscopic stent changes. Cholecystitis has required percutaneous cholecystostomy. Gastroduodenal ulceration has lead to catastrophic bleeding and/or perforation. Patients have developed nausea and vomiting, leukopenia, gastroparesis, hepatic abscesses and liver failure — especially those with underlying liver disease such as PSC.

Pathological confirmation of the diagnosis of CCA is not always possible. Mayo Clinic relies on either a pathological diagnosis from histology or cytology or a malignant appearing stricture with at least one of the following: an elevated CA-19.9 > 100 mg/mL; polysomy by fluorescent in-situ hybridization (FISH); or a mass on cross-sectional imaging at the site of the malignant appearing stricture. Outcomes data show comparable rates of residual CCA in the explanted livers and recurrence rates after transplantation for patients with and without pathological confirmation of the diagnosis prior to initiation of neoadjuvant therapy.

Clinicians must not perform a transperitoneal biopsy or fine needle aspiration of the primary tumor in order to obtain a pathological diagnosis. Our experience includes six such patients and all had tumor seeding at operative staging (5) or a recurrence after transplantation (1).

We propose that the primary treatment for patients with CCA arising in the setting of PSC should be neoadjuvant therapy and liver transplantation. CCA arising in PSC is rarely resectable, and resectability is limited by the underlying liver disease. CCA arising in PSC is often multicentric as well. Ten percent of CCA/PSC patients were found to have microscopic common bile duct involvement despite the indication for transplantation being hilar disease, and some of these patients underwent pancreatoduodenectomy at the time of transplantation. Combined liver transplantation and pancreatoduodenectomy was associated with high morbidity and postoperative mortality. Five-year survival is 33%.

We do not think that neoadjuvant therapy and transplantation should be done in lieu of resection for potentially resectable CCA. Although the results of transplantation might be a bit better than reported results for potentially curative resection (with intention-to-treat analysis), the difference is not large enough to justify use of a deceased or living donor liver since the benefit is only the small difference between results of the two treatments.

The key messages of the presentation were as follows:

• Neoadjuvant therapy and liver transplantation is effective treatment for patients with CCA arising in the setting of PSC
• Neoadjuvant therapy and liver transplantation is an effective treatment option for patients with unresectable CCA arising de novo
• Potentially resectable CCA arising de novo is best treated with resection
• Comparisons between resection and neoadjuvant therapy and transplantation should be done with an intention-to-treat analysis using operative exploration (for resection) and start of neoadjuvant therapy (for transplantation) as time zero
• Expected 5-year survival after start of neoadjuvant therapy is 60% for patients with CCA/PSC and 38% for patients with de novo CCA
• Expected 5-year survival after transplantation is 76% for CCA/PSC and 56% for de novo CCA — survival rates which are aligned with expectations for a reasonable chance for success (> 50% at 5 years)
• Success with transplantation requires strict adherence to selection criteria and the neoadjuvant treatment protocol
• Transplant teams must expect and be prepared to care for a multitude of complications resulting from the aggressive neoadjuvant therapy and the transplant operations