INTRODUCTION
Cholangiocarcinoma (CCA) comprises a biologically heterogeneous group of biliary tract tumours, traditionally classified into intrahepatic (iCCA), perihilar (pCCA), and distal (dCCA) subtypes based on anatomical location 1. Beyond this anatomical classification, additional histomorphological and molecular classification systems have been proposed in recent years. Kendall et al. 2, described a refined framework in which intrahepatic cholangiocarcinoma is subdivided into distinct entities based on histological features and molecular signatures, including cholangiolocellular and progenitor cell-derived tumours (small-duct type and large-duct type iCCA). This classification reflects underlying biological heterogeneity and has potential implications for prognosis and therapeutic strategies. Despite advances in systemic therapy and diagnostics, prognosis remains poor due to late diagnosis and the advanced stage at presentation. Surgical resection offers the best chance of cure, yet resectability rates are low, especially for pCCA, where proximity to vascular and biliary structures complicates surgical intervention 3.
The early experience with liver transplantation (LT) for unresectable CCA demonstrated high perioperative mortality and poor early oncological outcomes 4. Among the six patients transplanted for unresectable CCA who survived beyond one month postoperatively, only three (50%) were recurrence-free at one year. All patients who experienced recurrence died within the first year following LT. In contrast, patients without recurrence at one year achieved survival ranging from 20 to 54 months 5. Overall, outcomes for LT in CCA remained poor throughout the 1980s and early 1990s, with reported 5-year overall survival rates below 17% and recurrence rates exceeding 50% in most published series 6,7.
These unfavourable results were not primarily attributable to limitations of the transplant procedure itself, which had already undergone significant optimization, but rather to two major factors. First, early series were characterized by inadequate patient selection, with transplantation offered to patients with advanced disease, extrahepatic spread, and across all anatomical subtypes of CCA. Second, no neoadjuvant or adjuvant oncological therapies were employed to control microscopic disease and reduce the risk of post-transplant recurrence.
Recognition of these limitations prompted a paradigm shift in the late 1980s and early 1990s, leading to the development of structured neoadjuvant chemoradiotherapy protocols combined with rigorous pre-transplant staging. Pioneering work by the University of Nebraska and the Mayo Clinic resulted in the establishment of standardized protocols specifically for pCCA in 1988 and 1993, respectively 8. The Mayo Clinic protocol represented a turning point in the field and has since become the reference strategy for liver transplantation in unresectable pCCA.
Initial and subsequent series consistently reported 5-year overall survival rates exceeding 65-70% in carefully selected patients, establishing liver transplantation as a potentially curative option for unresectable pCCA 9,10.
On the basis of these landmark results, the Mayo Clinic protocol has been increasingly adopted and refined worldwide 11.
LIVER TRANSPLANTATION FOR PERIHILAR CHOLANGIOCARCINOMA
pCCA represents the most common subtype of extrahepatic cholangiocarcinoma, accounting for approximately 50–70% of cases 12. Because of its frequently indolent and initially asymptomatic clinical course, combined with its intrinsically aggressive biological behaviour, pCCA is often diagnosed at an advanced stage, when curative treatment options are limited. At present, complete surgical resection with negative margins (R0) remains the cornerstone of curative-intent therapy for resectable disease. Nevertheless, even in high-volume centres and after R0 resection, long-term outcomes remain suboptimal, with reported 5-year OS rates ranging from 15 to 35% 3.
In this context, LT has emerged as the only potentially curative therapeutic option for patients with unresectable pCCA. This strategy is particularly relevant in patients with underlying chronic liver disease, most notably primary sclerosing cholangitis (PSC), a well-established risk factor for cholangiocarcinoma, with up to 30% of PSC patients developing CCA during their lifetime. In patients with pCCA arising in the setting of PSC, LT offers the theoretical advantage of complete oncologic clearance while simultaneously treating the underlying liver disease 13.
As previously reported, the Mayo Clinic protocol revolutionized the approach to unresectable pCCA through the integration of external-beam radiation therapy, intraluminal brachytherapy, radiosensitizing chemotherapy, prolonged capecitabine administration, and mandatory staging surgery to exclude metastatic disease. This approach not only reduces local tumour burden, but also serves as a biological stress test by eliminating patients whose disease progresses during therapy 8.
Outcomes following this approach have been consistently superior to historical LT and resection results. Early Mayo series reported 5-year overall survival rates of 65-70% and recurrence-free survival near 60%.
Selection criteria for LT typically include unresectable tumours due to vascular involvement, tumour size ≤ 3 cm, no prior transperitoneal biopsy or surgical manipulation and no evidence of metastatic disease or lymph node involvement – who are deemed technically unresectable undergo meticulous pre-transplant staging. This includes endoscopic ultrasound with fine-needle aspiration of regional lymph nodes. Patients with negative nodal assessment proceed to neoadjuvant chemoradiotherapy, followed by laparoscopic or open surgical exploration immediately prior to LT to definitively exclude occult metastatic disease and reassess regional lymph nodes 10,11.
Strict selection results in dropout rates of 30-40%, predominantly due to disease progression during neoadjuvant therapy. While this may seem a limitation, dropout effectively enriches the final transplant cohort with biologically favourable tumours.
Subsequent updates from the same centre confirmed these outcomes, reporting a 5-year survival of 71% in a larger cohort 14. These findings were later validated in a North American multicentre study involving 12 transplant centres (n = 287), which demonstrated a 5-year disease-free OS of 65% among patients treated according to the Mayo protocol 15. European data from the European Liver Transplant Registry (ELTR) further corroborated these results, with Mantel et al., reporting a 5-year OS of 59% in patients transplanted under Mayo protocol criteria 16. Collectively, these studies underscore the pivotal role of careful patient selection and neoadjuvant therapy in achieving acceptable oncological outcomes following LT for pCCA.
Importantly, the Mayo protocol has been applied to both PSC-associated pCCA and unresectable de novo pCCA. PSC patients are often unsuitable candidates for surgical resection due to multifocal biliary involvement or advanced underlying liver disease, making LT the preferred therapeutic option in this subgroup. Moreover, post-transplant survival outcomes appear to be superior in PSC-associated pCCA compared with de novo tumours, likely reflecting earlier diagnosis and more favourable tumour biology 17.
In recognition of the growing body of evidence supporting LT for unresectable pCCA, standardized Model for End-Stage Liver Disease (MELD) exception points were approved by UNOS/OPTN in 2009, allowing eligible pCCA patients to receive priority comparable to that of patients with HCC 18. Despite this policy, patients with pCCA continue to experience lower access to LT compared with other indications, largely due to significant pre-transplant dropout rates. Dropout may occur as a result of disease progression, intolerance to neoadjuvant therapy, positive staging, or death. Rea et al. reported a 5-year OS of 82% among transplanted patients; however, this decreased to 58% in an intention-to-treat analysis due to a 46% dropout rate prior to LT 19. Conversely, Ethun et al., reported a substantially lower dropout rate of 25% in a multicentre cohort, highlighting the variability across centres and protocols 20.
One of the major challenges in implementing LT for pCCA is establishing a definitive pre-transplant diagnosis. Histological confirmation can be difficult, particularly in PSC patients, where benign inflammatory strictures may closely mimic malignancy 21. Endoscopic biopsy and biliary brush cytology yield positive results in only approximately 30% of cases. Furthermore, transabdominal biopsy is contraindicated in this setting, as it automatically excludes patients from MELD exception eligibility due to the risk of tumour seeding.
A recent meta-analysis including 20 studies confirmed the critical role of neoadjuvant chemoradiation, demonstrating a 5-year OS exceeding 50% in patients who completed neoadjuvant therapy prior to LT, compared with only 31.6% in those who underwent upfront transplantation 22. Neoadjuvant treatment was also associated with a significantly lower recurrence rate (24.1 vs 51.7%) 22.
The potential role of LT in patients with initially resectable pCCA remains controversial. Several retrospective studies have explored this issue with conflicting results. Croome et al., compared resection versus neoadjuvant therapy followed by LT in patients with de novo pCCA and reported superior 1-, 3-, and 5-year OS in the LT group (90, 71, and 59%, respectively) compared with resection (81, 53, and 36%) 24. However, when the analysis was restricted to patients undergoing R0 resection with N0 disease, survival outcomes were comparable, leading the authors to recommend surgical resection as the preferred option for resectable de novo pCCA.
Conversely, Ethun et al. 20, analysing data from the US Extrahepatic Biliary Malignancy Consortium, demonstrated significantly improved 5-year OS in patients with unresectable pCCA treated with neoadjuvant therapy and LT compared with those undergoing resection alone (64 vs 18%), even after adjustment for tumour size, nodal status, and PSC. These findings further emphasize the negative prognostic impact of lymph node involvement and the fundamental role of neoadjuvant therapy.
Notably, excellent long-term outcomes following resection have been reported by high-volume Asian centres. Nagino et al., and Ebata et al., described 5-year OS rates exceeding 60% in selected patients with Bismuth type IV pCCA without lymph node involvement 25,26. Given that Bismuth type IV pCCA is considered unresectable in most Western centres, these findings highlight ongoing differences in surgical philosophy and patient selection. Consequently, the optimal treatment strategy for both resectable and unresectable pCCA remains debated.
LIVER TRANSPLANTATION FOR INTRAHEPATIC CHOLANGIOCARCINOMA (ICCA)
iCCA has historically been regarded as a contraindication to LT because of its aggressive biological behaviour and the high risk of post-transplant recurrence. Early experiences with LT for iCCA reported extremely poor outcomes, with 5-year overall survival (OS) rates ranging from 10% to 18% 27.
More recent evidence has challenged this paradigm, particularly in highly selected patients with very early-stage disease. In 2014, a Spanish multicentre study evaluated cirrhotic patients who underwent LT for presumed HCC and were subsequently found to have mixed HCC–CCA or iCCA on explant pathology 28. Notably, no significant differences in survival were observed between patients with a single iCCA ≤ 2 cm and those transplanted for HCC, with a 5-year OS of 73%.
Subsequently, Facciuto et al. 29, reported outcomes in a cohort of 32 cirrhotic patients with iCCA identified on explant specimens. Among patients whose tumours fulfilled the Milan Criteria, the authors observed a 5-year survival rate of 78% and a recurrence rate of only 10%, outcomes comparable to those achieved in HCC patients selected according to the same criteria.
Further support for a potential role of LT in selected iCCA cases came from the study by Lunsford et al. 30, who reported a series of six patients with unresectable iCCA treated with gemcitabine-based neoadjuvant chemotherapy followed by LT. Overall survival was 100% at 1 year and 83.3% at 3 and 5 years, although three patients developed tumour recurrence after a median of 7.6 months post-transplantation.
More recently, Gruttadauria et al. 31, described an Italian experience including 14 LTs performed in patients with iCCA. In 12 cases, the diagnosis was made incidentally on histological examination of the explanted liver, whereas two patients with unresectable iCCA underwent LT after neoadjuvant selective internal radiation therapy (SIRT) and a period of clinical observation. Both patients were alive at 19 and 2 months of follow-up, respectively.
Despite these promising developments, it is important to underscore that LT for iCCA remains investigational. Evidence is still based on small cohorts and heterogeneous protocols, and standardized neoadjuvant regimens have not yet been established. For this reason, LT for iCCA should presently be reserved for clinical trials or highly selected cases evaluated within expert multidisciplinary teams. Nevertheless, as molecular profiling, imaging analytics, and biological assessment methods continue to evolve, the potential for LT to play a meaningful role in selected iCCA patients is likely to expand.
COMBINED HEPATOCELLULAR-CHOLANGIOCARCINOMA
Combined hepatocellular-cholangiocarcinoma (cHCC-CCA) is a rare tumour displaying histological features of both HCC and CCA. Preoperative diagnosis is challenging, as imaging often mimics typical HCC. Outcomes after LT are generally inferior to HCC, with high recurrence rates and variable survival 32. Recent studies have suggested that LT may benefit carefully selected patients, particularly those with early-stage tumours or cirrhosis. However, there is no consensus on the criteria for LT in this patients 33.
EMERGING TECHNOLOGIES IN TRANSPLANT ONCOLOGY
Technological innovation is reshaping the landscape of transplant oncology, offering new tools to better characterize tumour biology, optimize candidate selection, and reduce post-transplant recurrence. Among these innovations, radiomics and artificial intelligence (AI) are particularly promising. Radiomics allows extraction of quantitative imaging features beyond the capacity of visual interpretation, enabling detailed characterization of tumour texture, heterogeneity, and internal architecture. When these high-dimensional datasets are integrated into AI-based predictive models, they can provide insights into tumour aggressiveness, likelihood of microvascular invasion, and expected response to neoadjuvant therapy. Early studies suggest that these models outperform traditional imaging interpretation, offering a more accurate assessment of biological behaviour and potentially aiding in the identification of suitable LT candidates 34.
Complementing these imaging advances is the increasing use of circulating tumour DNA (ctDNA), which has emerged as a powerful biomarker for minimal residual disease (MRD). ctDNA detection can reveal occult metastases not visible on imaging and can identify molecular signatures associated with aggressive disease, sometimes months before clinical recurrence becomes apparent. In the transplant setting, ctDNA may prove invaluable in determining which patients are at heightened risk of early post-transplant recurrence and therefore unlikely to benefit from LT. Detection of high-risk variants, including TP53 and KRAS mutations, in ctDNA samples could therefore serve as a non-invasive exclusion criterion in future selection algorithms 35.
Emerging advances are not limited to diagnostics. Machine perfusion (MP) represented the significant innovation in graft preservation of last years. By maintaining the liver in a metabolically active state during ex vivo perfusion, this technology allows clinicians to assess organ viability and function prior to implantation. This not only expands the donor pool by enabling safe use of marginal grafts, but may also mitigate ischemia-reperfusion injury, a factor potentially linked to tumour recurrence. Although evidence specific to CCA recipients is still limited, the use of perfusion technologies may become particularly impactful in oncology-related transplant indications, where suitable donor organs are constrained 36,37.
On the therapeutic front, targeted therapies directed against FGFR2 fusions, IDH1/2 mutations, and BRAF alterations are increasingly used in patients with advanced CCA, raising the question of whether they might serve in future as adjuncts to neoadjuvant therapy in patients considered for LT. The interaction of these therapies with post-transplant immunosuppression remains a significant challenge, however, and requires cautious evaluation in controlled studies. Even greater caution is required with immune checkpoint inhibitors, which – despite their efficacy in advanced CCA – carry a high risk of fulminant rejection when used near the time of transplantation. For now, their use in LT candidates remains limited to highly selected clinical trial settings 38.
POST-TRANSPLANT MANAGEMENT
Immunosuppression strategies may influence recurrence risk. Minimization of calcineurin inhibitors and judicious use of mTOR inhibitors have been proposed to reduce recurrence, although prospective evidence is limited. Surveillance protocols typically include cross-sectional imaging every 3-6 months during the first two years, followed by annual or semi-annual evaluations 39.
CONCLUSIONS AND FUTURE PERSPECTIVES
Over the past decade, transplant oncology has undergone substantial evolution, leading to a renewed interest in LT as a potential therapeutic strategy for selected patients with CCA. Improvements in neoadjuvant and adjuvant treatment protocols, together with a deeper understanding of tumour biology and recurrence risk, have progressively challenged the historical contraindication of LT in this setting.
Among CCA subtypes, the strongest evidence currently supports the role of LT for patients with locally advanced, unresectable pCCA, treated within structured neoadjuvant chemoradiation protocols. In highly selected patients, this approach has been associated with long-term survival outcomes comparable to those achieved for established transplant indications, particularly when stringent selection criteria and mandatory operative staging are applied. These results underscore the critical importance of careful patient selection, which remains the cornerstone of successful LT in this scenario.
Although the Mayo Clinic protocol has represented a major breakthrough in the treatment of unresectable pCCA, the rapidly evolving therapeutic landscape of biliary tract cancers and the maturation of transplant oncology as a discipline call for innovative strategies aimed at overcoming some of its intrinsic limitations. In particular, advances in systemic chemotherapy, immunotherapy, and imaging modalities have created the opportunity to rethink neoadjuvant approaches while preserving the strict selection principles that underpin the success of transplantation in this setting.
In contrast, iCCA continues to be considered a contraindication to LT in most transplant centres. Nevertheless, emerging data suggest that LT may be a viable option in carefully selected patients with very early-stage iCCA, particularly in the presence of cirrhosis or when liver resection is not feasible. Moreover, preliminary experiences integrating neoadjuvant therapies prior to LT have demonstrated encouraging oncological outcomes, supporting further exploration of this strategy in prospective studies.
Despite these advances, several critical gaps remain. Significant heterogeneity persists with regard to pre-transplant disease characteristics – including tumour size, nodal status, biological aggressiveness, and response to neoadjuvant therapy – as well as in the availability and standardization of pre-LT treatment protocols, particularly for iCCA. Future efforts should focus on refining selection criteria through the integration of biological markers, radiological response assessment, and molecular profiling to better identify patients most likely to benefit from LT.
Looking forward, well-designed prospective and multicentre clinical trials are urgently needed to clarify the comparative effectiveness of LT versus resection and non-surgical therapies, evaluate standardized neoadjuvant and immunosuppressive regimens, and assess long-term oncological and quality-of-life outcomes. Advances in diagnostic accuracy, treatment personalization, and organ allocation policies will be essential to further establish LT as a rational and ethically sound option in selected patients with cholangiocarcinoma.
Within this framework, a nationwide study protocols, named LITALHICA (LIver TrAnspLantation for Non-resectable Peri-HIlar Cholangiocarcinoma – NCT06125769), and LIRICA (LIver Transplantation for Non-Resectable Intrahepatic CholAngiocarcinoma – NCT06098547, are being implemented in Italy 40.
The LITALHICA protocol is conceptually rooted in the Mayo Clinic experience, as it maintains identical inclusion and exclusion criteria and preserves the overall structure of patient selection and staging. However, it introduces a relevant innovation in the neoadjuvant treatment strategy (Tab. I).
Unlike the Mayo protocol, radiotherapy is no longer included in the neoadjuvant regimen. Instead, patients receive systemic chemotherapy with the current standard-of-care combination of gemcitabine, cisplatin, and durvalumab. Neoadjuvant treatment is administered for six months, after which patients undergo comprehensive restaging using contrast-enhanced CT and PET–MRI, in addition to standard pre-transplant staging laparoscopy. Only patients demonstrating disease stability or partial response are prioritized for transplantation, whereas those with radiological or metabolic disease progression are excluded from LT and referred to second-line systemic therapy.
In the LIRICA protocol, patients enrolled at least six months of chemotherapy, achieving disease stability or partial response (according to RECIST version 1.1) before listing for LT (Tab. II).
In this context, international cooperative efforts are essential. The implementation of a unified, prospective protocol represents a critical step toward standardizing clinical practice, optimizing patient selection, and systematically collecting outcome data. Given the rarity of CCA and the technical and organizational complexity of LT in this setting, only a coordinated, multidisciplinary, and multicentre approach will allow the transplant community to adequately assess the true oncological value of LT for CCA and to define its role within modern treatment algorithms.
LT has evolved into an effective and potentially curative therapy for selected patients with early-stage, unresectable iCCA and pCCA treated within strict neoadjuvant protocols. Advances in molecular profiling, radiomics, ctDNA, machine learning, and organ-preservation technologies are poised to transform transplant oncology, enabling more refined patient selection and improved long-term outcomes. The next decade will likely see expanding indications for LT in CCA, guided by principles of precision oncology and responsible organ allocation.
Conflict of interest statement
The authors declare no conflict of interest.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Author contributions
CM: conceived the overall structure of the manuscript, performed the literature review, and drafted the initial version; FM: critically revised the text, supervised the preparation of the final version, and approved it for sub-mission as corresponding author. All authors have read and approved the final manuscript.
Ethical consideration
Not applicable.
History
Received: January 3, 2026
Accepted: January 20, 2026
Figures and tables
| Step | Mayo clinic protocol | LITALHICA protocol |
|---|---|---|
| Indication | Unresectable pCCA | Unresectable pCCA |
| Tumor criteria | ≤ 3 cm | ≤ 3 cm |
| Metastases/LN | Absent | Absent |
| Baseline staging | CT/MRI | CT + PET–MRI |
| LN assessment | Staging laparoscopy/laparotomy | EUS-guided FNA of regional lymph nodes and/or cross-sectional imaging (CT, PET–MRI) |
| Targeted lymphadenectomy | ||
| Neoadjuvant therapy | Chemoradiotherapy | Systemic therapy |
| External beam RT | Gemcitabine | |
| 5-FU | Cisplatin | |
| Brachytherapy | Durvalumab | |
| Capecitabine | ||
| Pre-LT assessment | Operative staging | Radiologic restaging |
| Mandatory LN sampling | CT + PET–MRI | |
| Exclusion of metastases | Response assessment | |
| Eligibility for LT | Negative staging | Stable disease/PR |
| Exclusion criteria | Positive LN/metastases | Progressive disease |
| Post-LT management | Standard immunosuppression | Tailored immunosuppression |
| mTORi-based strategies | mTORi-based strategies | |
| pCCA: peri-hilar cholangiocarcinoma; LN: lymph nodes; CT: computed tomography; MRI: magnetic resonance imaging; PET: positron emission tomography; EUS: endoscopic ultrasound; FNA: fine needle aspiration; RT: radiotherapy; 5-FU: 5-fluorouracile; PR: partial response; LT: liver transplantation; m-TORi: mechanistic Target of rapamycin inhibitor. | ||
| Step | LIRICA protocol |
|---|---|
| Indication | iCCA non resectable |
| Liver status | Cirrhosis or liver disease precluding safe hepatic resection |
| Tumour burden | Very early iCCA (single lesion, limited size) * |
| Extrahepatic disease | Absent |
| Lymph node involvement | Absent |
| Baseline staging | CT/MRI ± PET |
| Neoadjuvant therapy | Systemic chemotherapy |
| (gemcitabine-based regimens) | |
| Treatment duration | Defined period with radiologic reassessment |
| Restaging | CT/MRI ± PET |
| Eligibility for LT | Stable disease or partial response |
| Exclusion criteria | Disease progression |
| Liver transplantation | Within dedicated allocation policy |
| Post-LT management | Tailored immunosuppression |
| Preference for mTORi-based strategies | |
| Follow-up | Strict oncologic surveillance |
| *Exact tumour size and number defined by protocol-specific criteria; iCCA: intrahepatic cholangiocarcinoma; CT: computed tomography; MRI: magnetic resonance imaging; PET: positron emission tomography; LT: liver transplantation; m-TORi: mechanistic target of rapamycinn inhibitor. | |
