Adarsh Vijay , Hoonbae Jeon

Division of Hepato-Pancreato-Biliary & Abdominal Transplant Surgery; Tulane Abdominal Transplant Institute, Tulane University School of Medicine, New Orleans, LA 70112-2632, USA

Cholangiocarcinoma (CCA) constitutes a group of heterogeneous malignant tumors emerging in the biliary tree [1] . Based on their anatomical distribution, CCA is classified into perihilar cholangiocarcinoma (pCCA), distal cholangiocarcinoma (dCCA) and intrahepatic cholangiocarcinoma (iCCA) and they account for 60%-70%, 20%-30% and 5%-10% of CCA, respectively. The past decade has seen a 2-fold increase in annual percentage change in incidence of iCCA in USA when compared to the prior three decades [2] .

Liver resection remains the gold standard for the treatment of iCCA with curative intent [3] . However, 60%-88% of these patients are deemed unresectable at the time of presentation [4] . Typically,resection is not considered in presence of diffuse bilobar involvement, distant metastasis, retroperitoneal or pericoeliac node involvement, peritoneal carcinomatosis, underlying liver disease with portal hypertension, inadequate future liver remnant (FLR) or patient performance status [5] .

Surgical resection is carried out with the intent for a R0 surgical resection margin and to leave behind adequate FLR. Li et al. [6] in an international multicenter study cohort of 102 patients, describe a superior oncological benefit and survival in comparison to palliative chemotherapy in patients otherwise considered unresectable due to inadequate FLR. Selected patients with iCCA and evidence of visceral/vascular extension have been considered for resection with curative intent in high volume centers [7–9] . In a multiinstitutional analysis of 1087 patients with iCCA, major vascular resection was not associated with worse perioperative or oncologic outcomes [9] .

Lymph node metastasis has been reported in up to 45% patients with iCCA and is associated with a dismal prognosis of less than 10% five-year survival rate [ 10 , 11 ]. The survival benefit of extensive lymphadenectomy is debatable [ 12 , 13 ]. Jutric et al. [14] in a review of the National Cancer Database reported only 57% of patients that underwent curative resection to have at least one lymph node examined. The eighth edition of the American Joint Committee on Cancer (AJCC) staging system recommends a minimum of 6 nodes to be analyzed to stage patients [15] . Recent studies have advocated treating locally advanced patients (lymph node involvement and vascular invasion) with neoadjuvant chemotherapy with conversion to surgical resection rate in up to 50% [16–18] .Such tumor downstaging treatments in combination with advanced surgical procedures such as associating liver partition and portal vein ligation for staged hepatectomy (ALPPS) have increased the resectability of these tumors [5] .

In spite of advancements in surgical techniques and chemotherapy, disease-free and overall survivals in resected patients are still dismal with a reported 5-year overall survival (OS) rate of 10%-35% [ 13 , 19 , 20 ].

The evolution of liver transplantation (LT) for patients with iCCA cannot be explained without addressing the past outcomes of LT in patients with iCCA and advancements post adoption of the Mayo protocol for LT in patients with pCCA.

Initial outcomes for LT for patients with pCCA were reported by O’Grady et al. in 1988 to be dismal [21] . Of the 26 patients transplanted, none survived 4 years. This study however had patients with poor tumor biology, late stage disease and no consistent neoadjuvant therapy. Mayo Clinic in early 20 0 0’s reported substantial improvement in transplant outcomes for patients with pCCA. A 12-center consortium study in 2012 reported 69% 5-year recurrence-free survival (RFS) in patients within Mayo protocol criteria compared to 32% for those beyond criteria [22] . The Mayo protocol adopted strict inclusion criteria for pCCA with tumor mass less than 3 cm, no lymph node metastasis and with consistent neoadjuvant chemoradiotherapy. The outcomes with transplant in patients withdenovopCCA following Mayo protocol were comparable to R0 resection. In comparison primary sclerosing cholangitis(PSC) associated pCCA had superior 5-year survival rate of 77% in patients that underwent Mayo protocol. Mayo protocol does not always need a pathological diagnosis of CCA before enrollment into protocol. Forty-two percent of transplanted patients had no residual disease on explant. This finding has questioned if these patients really had CCA to begin with or if the negative explant pathology denoted complete pathological response to the neoadjuvant therapy. Detractors of the Mayo protocol owe the successful outcomes to its strict and narrow patient selection criteria and lack of initial pathological diagnosis.

Despite the progress and acceptance for LT in patients with pCCA during past two decades, patients with iCCA remains a contraindication to LT in most centers worldwide. pCCA and iCCA are separate entities with distinct cellular origin, biological markersand genomic profile [23] . Historic outcomes of LT for patients with iCCA were significantly worse when compared to the benchmark LT for patients with HCC. Several studies reported < 25% five-year survival for patients that underwent LT for iCCA [ 24 , 25 ]. Almost all available literature citing outcomes of LT in patients with iCCA are retrospective single or multicenter studies [24–33] ( Table 1 ). Most large studies have included transplanted patients who were misdiagnosed before LT and found to have iCCA on explant [ 27 , 30 , 31 , 33 ].Some of these studies are further hampered by the inclusion of patients with explant specimen containing both HCC and iCCA [ 30 , 33 ].

Table 1 Liver transplantation outcomes in patients with intrahepatic cholangiocarcinoma without neoadjuvant chemotherapy.

Initial data to support LT in cirrhotics with iCCA came from studies by Sapisochin et al. in 2014 and later substantiated in 2016 [ 27 , 31 ]. The latter study evaluated data from 17 multinational centers with iCCA on explant as an incidental finding/misdiagnosis. Most patients in the cohort received locoregional therapy and none received neoadjuvant chemotherapy. Patients with “very early” iCCA (single tumor ≤2 cm) had a 5-year survival rate of 65% compared to a mere 45% in those with “advanced” disease (single tumor > 2 cm or multifocal disease). De Martin et al. in 2020 in a small series of patients with iCCA reported a 5-year survival rate of 69% in tumors ≤2 cm [33] . These rates were comparable with the 5-year overall survival rates for patients with pCCA undergoing LT. The concept of appropriate patient selection for LT is being prospectively studied in patients with“very early” iCCA (single tumor ≤2 cm). This experimental study(NCT02878473) is currently in the enrollment phase. Imaging and biopsy proven iCCA ≤2 cm in the absence of HCC in a cirrhotic nodular liver is a practical and resource intensive challenge. The stringent inclusion criteria with a high drop-out rate due to disease progression/failure to thrive as witnessed in prior studies with pCCA may further jeopardize this study from aggregating enough sample size to realize its potential.

Hong et al. in 2011, retrospectively evaluated 38 patients who underwent LT with explant pathology showing 25 iCCA and 13 pCCA. Patients in the study with locally advanced cholangiocarcinoma received stereotactic body radiotherapy followed by neoadjuvant chemotherapy until time of transplant. Intrahepatic location and pretransplant neoadjuvant chemoradiotherapy were associated with improved outcomes [20] . Wong et al. evaluated the efficacy of neoadjuvant therapy in downstaging locally advanced CCA (intrahepatic ≤8 cm, and regional lymphadenopathy but without distant metastasis) prior to LT, and reported acceptable short-term recurrence-free survival [34] .

Lunsford et al. in 2018 further explored the use of neoadjuvant chemotherapy in a prospective series of patients who had locally advanced unresectable iCCA [35] . Patients included had no evidence of extrahepatic disease. However, there were no limitations on tumor size or multifocality. A minimum wait period of 6 months of radiographic response or stability while on neoadjuvant therapy before LT was used to screen out patients with aggressive tumor biology. This small case series of 6 LT reported a 5-year OS and RFS rates of 83.3% and 50%, respectively. Mean duration from diagnosis to transplant was 26 months and drop-out rate was 50%.

The future of LT in patients with iCCA will depend on studies modeled on interplay of patient selection, neoadjuvant therapy and targeted immunotherapy.

The Mayo protocol for pCCA over the past two decades has reinstated the importance of strict patient selection and neoadjuvant chemoradiotherapy ( Table 2 ). Addition of Yttrium-90 radio embolization (Y90-RE) to the existing locoregional neoadjuvant therapies has proven to successfully downstage locally advanced iCCA [36] . More recent retrospective series that have not utilized neoadjuvant chemotherapy have stressed on patient selection with tumor size less than 2 cm yielding significantly better outcomes in patients with iCCA that underwent LT [ 31 , 33 ].The retrospective series by Hong et al. and prospective series by Lunsford et al. provided proof for superior outcomes in LT patients with iCCA post neoadjuvant therapy irrespective of tumor size/multifocality [ 20 , 35 ]. The cohorts in the studies of Hong et al.and Wong et al. included patients with regional lymph nodes involvement [ 20 , 34 ]. Given the evidence, we believe that tumor size/multifocality should not be an exclusion criterion in prospective trials in the future that incorporates neoadjuvant therapy. The inclusion of those with regional lymph nodes involvement may be considered if tumor shows appropriate response to neoadjuvant therapy.

The response of targeted therapies in the past have been hampered by the generous genetic heterogeneity displayed by iCCA [ 37 , 38 ]. Specific targetable genetic alterations under study include fibroblast growth factor receptor (FGFR) fusion inhibitors(NCT03773302, NCT04093362) and isocitrate dehydrogenase (IDH-1, IDH-2) inhibitors (NCT036 84 811, NCT02381886, NCT02989857).In April 2020, The United States Food and Drug Administration

Table 2 Comparison of liver transplant protocols that utilized neoadjuvant chemotherapy for pCCA and iCCA.

(FDA) approved an FGFR inhibitor (pemigatinib) for the treatment of patients with advanced iCCA [39] .

Based on encouraging results from programmed death-ligand 1 (PD-L1) blockade, FDA approved pembrolizumab for microsatellite instability high/mismatch repair deficient tumors in 2017 [40] . Limited success with immune checkpoint blockade(ICB) monotherapy has led to more trials testing combination immunotherapy strategies that include dual ICB as well as ICB combined with another immunomodulatory agent, molecular targeted therapy, cytotoxic chemotherapy, or local therapy (NCT03898895,NCT03820310).

There is enough evidence in support of LT in patients with unresectable very early iCCA. However, large scale prospective studies that incorporate neoadjuvant therapy and targeted immunotherapy with LT are needed to justify the use of scare donor organs in this subpopulation. From existing evidence, selection criteria are imperative to LT outcomes. However, the best inclusion criteria are yet to be determined. Better understanding of the biological behavior,tumor genetic profile, biomarkers and advancements in targeted therapy and immunotherapy is pivotal to better post-LT outcomes in patients with iCCA.

Acknowledgments

None.

CRediT authorship contribution statement

Adarsh Vijay: Conceptualization, Writing - original draft, Writing - review & editing. Hoonbae Jeon: Writing - review & editing.

Funding

None.

Ethical approval

Not needed.

Competing interest

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.