Inclusion criteria for liver transplantation in patients with colorectal liver metastases: How to make the best selection?

Abstract

The selection of patients with colorectal cancer liver metastases (CRLM) for liver transplantation (LT) represents a significant challenge, requiring a balance between oncological outcomes and organ scarcity. Recent advancements in transplantation outcomes for CRLM have prompted the establishment of rigorous selection criteria to optimize patient survival and graft utilization. This review examines the key criteria used to select candidates for LT in this setting, with a focus on oncological factors, patient characteristics, and response to therapy. Eligible candidates are typically those with non-resectable liver-only CRLM, demonstrating controlled primary tumor disease. Tumor biology is a critical determinant, excluding patients exhibiting high-risk molecular features such as BRAF or RAS mutations. Furthermore, candidates must show a favorable response to systemic chemotherapy, with either tumor stability or reduction in size and no extrahepatic progression during a defined treatment period. Specific tumor burden scores, such as the Oslo score or criteria based on the number and size of lesions, aid in stratifying candidates with acceptable recurrence risks. Other factors, including age, performance status, and absence of significant comorbidities, are also pivotal. Long-term follow-up data highlight the importance of stringent patient selection, showing superior 5-year survival in patients meeting these criteria compared to those who do not. In conclusion, strict selection criteria based on tumor biology, systemic disease control, and patient-specific factors ensure optimized outcomes for LT in CRLM patients, marking a pivotal step toward broader clinical acceptance of this novel approach.

Key Words: Colorectal carcinoma; Liver metastases; Liver transplantation; Donor characteristics; Recipient characteristics; Outcomes

Core Tip: Colorectal cancer liver metastases (CRLM) strongly impact patients’ survival despite available chemotherapy, surgical removal of liver metastases, and locoregional treatment alone or in combination with systemic chemotherapy. In the last few decades, liver transplantation (LT) in a selected patient population has significantly improved survival, nearing 83% during a 5-year follow-up. The selection of patients with CRLM for LT represents a significant challenge, requiring a balance between oncological outcomes and organ scarcity. Stringent criteria for eligible patient selection are key to long-term patient survival and can be summarized into those related to tumor biology and oncological factors, patient and donor characteristics, and previous response to therapy. Until more generalizable data are available, we explore and compare the inclusion and exclusion criteria used in various studies registered on ClinicalTrials.gov, highlighting their similarities, differences, and implications.

INTRODUCTION

Colorectal cancer (CRC) is the third most diagnosed cancer in the world, with an increasing burden of new cases occurring in patients aged ≤ 50 years and modest 5-year survival among those with metastatic disease treated with conventional therapies. CRC-related metastases in more than 50% of patients develop in the liver, strongly impacting overall survival (OS), with a median 5-year OS less than 14% with palliative chemotherapy, somewhat improved survival in case of surgical removal of liver metastases, or locoregional treatment alone or in combination with systemic chemotherapy, but unfortunately with a high level of relapses[1-4].

Early attempts for liver transplantation (LT) for colorectal cancer liver metastases (CRLM) were quite unsuccessful, correlated with high levels of complications and poor survival, and were therefore almost abandoned[5]. In the last two decades, due to improvements in surgical techniques, immunosuppressive and chemotherapy agents, and a better understanding of tumor biology, LT in a selected patient population has significantly improved survival, nearing 83% during a 5-year follow-up[6-8]. Several small, primarily unicentric studies have tried to extract factors influencing patient prognosis. One of the most prominent was the Norvegian study secondary cancer (SECA) I, which brought up the Oslo prognostic score considering lesion diameter (largest tumor diameter ≥ 5.5 cm), < 2 years since primary tumor resection, elevated carcinoembryonic antigen (CEA) levels (> 80 ng/mL), and progressive disease at the time of LT, and categorizes patients in three subgroups: 0-1 factors, 2-3 factors, and 4 factors, with different survival outcomes[9]. Further on, in the SECA-II trial, even more stringent inclusion criteria for liver-only metastatic CRC patients were implied, resulting in 100%, 83%, and 83% OS at 1, 3, and 5 years, respectively. At the same time, recurrence was mainly seen as slow-growing pulmonary metastases, which were potentially treatable by curative resection[8]. Solheim et al[10] published the long-term results of 23 transplanted patients from the initial Norwegian experience, recruiting patients transplanted from 2006 to 2012, showing excellent OS compared to available oncological treatments.

Recently, a multicentric prospective study (TRANSMET) investigated the intention-to-treat survival rates of 94 patients receiving LT plus chemotherapy vs chemotherapy alone. In the intention-to-treat population, 5-year OS was 56.6% vs 12.6% (hazard ratio = 0.37; P = 0.0003), while the survival rates in the per-protocol population were 73.3% vs 9.3%, respectively[11].

Therefore, stringent criteria for eligible patient selection are key to long-term patient survival and can be summarized into those related to tumor biology and oncological factors, patient and donor characteristics, and previous response to therapy. However, some inclusion/exclusion criteria are still based on the transplant team’s discretional right, mainly related to the perceived inability of a potential candidate for LT to follow the protocol. In this document, we explore and compare the inclusion and exclusion criteria used in various studies registered on ClinicalTrials.gov, highlighting their similarities, differences, and implications.

COMPARATIVE ANALYSIS OF CLINICAL TRIAL OUTCOMES

Exploring the outcomes of the key studies already published, comparable outcome measures have been reported that merit detailed evaluation. The SECA-II study, for instance, demonstrated 5-year OS of 83% in highly selected patients, with recurrence primarily limited to treatable lung metastases[8]. In contrast, the recent TRANSMET randomized trial revealed a 5-year OS of 73.3% in the per-protocol group undergoing LT plus chemotherapy vs 9.3% in the chemotherapy-only arm, underscoring the potential benefit of transplantation even in a broader patient population[11]. Meanwhile, the SECA-III study, which included patients with more extensive disease, reported lower OS, highlighting the trade-off between expanding indications and preserving outcomes. Recurrence patterns also vary; while SECA studies often report pulmonary relapse, these are frequently indolent and amenable to local treatment, potentially maintaining long-term survival. Comparative data on disease-free survival (DFS) are less consistently reported but remain critical for guiding patient selection.

INCLUSION CRITERIA OF ONGOING TRIALS

Among the inclusion criteria, different categories should be identified. In detail, the general characteristics of the patient, the features of the primitive and metastatic disease, the tumor response to therapies, and the timing from primitive tumor removal to LT have been considered as principal inclusion features for LT (Table 1).

Table 1 Inclusion criteria reported in the studies registered in ClinicalTrials.gov focused on liver transplantation for colorectal cancer liver metastases.

NCT	Location	Criteria
Demographic parameters
NCT02215889	Oslo, Norway	ECOG 0-1
NCT05186116	Modena, Italy	Age ≥ 18, ECOG 0-2
NCT05750329	RenJi Hospital, Shanghai, China	Age 18-75
NCT04870879	Padua, Italy	Age ≥ 18 and < 70, ECOG 0-1
NCT05248581	Rochester, NY, United States	Age ≥ 18
NCT03803436	Milan, Italy	ECOG 0
NCT03488953	Jena, Germany	Age ≥ 18
NCT05398380	Barcelona, Spain	Age 18-70, ECOG 0-1
NCT06069960	RenJi Hospital, Shanghai, China	Age 18-75, ECOG 0-1
NCT04742621	NY, United States	Age 18-65, ECOG 0-1
NCT05185245	Bologna, Italy	ECOG 0-1
NCT03494946	Oslo, Norway	ECOG 0-1
NCT01479608	Oslo, Norway	ECOG 0-1
NCT04616495	Valencia, Spain	Age ≥ 18, ECOG 0-1
NCT02864485	Toronto, Canada	ECOG 0-1
NCT02597348	Villejuif, France	Age ≥ 18 and ≤ 65, ECOG 0-1
NCT04865471	Padua, Italy	Age ≥ 18 and < 70, ECOG 0-1
NCT04161092	Gothenburg, Sweden	Age ≥ 18, ECOG 0-1
Biochemical markers
NCT02215889	Oslo, Norway	Hb > 10 g/dL, neutrophils > 1.0, TRC > 75, bilirubin < 2 × ULN, AST/ALT < 5 × ULN, creatinine < 1.25 × ULN, albumin > LLN
NCT04870879	Padua, Italy	Hb > 10 g/dL, neutrophils > 1.0, bilirubin < 2 × ULN, AST/ALT < 5 × ULN, creatinine and albumin normal
NCT03803436	Milan, Italy	Hb > 10 g/dL, neutrophils > 1.0, TRC > 75, bilirubin < 2 × ULN, AST/ALT < 5 × ULN, creatinine < 1.25 × ULN
NCT05398380	Barcelona, Spain	Creatinine ≤ 1.25 × ULN or eGFR ≥ 60, platelets ≥ 80 × 109/L, neutrophils ≥ 2.5 × 109/L
NCT04742621	NY, United States	Hb > 10 g/dL, ANC > 1000/μL, platelets > 100000/μL, bilirubin < 2 × ULN, AST/ALT < 5 × ULN, creatinine < 1.25 × ULN, albumin > LLN
NCT05185245	Bologna, Italy	Neutrophils > 1.0
NCT03494946	Oslo, Norway	Hb > 10 g/dL, neutrophils > 1.0, TRC > 75, bilirubin < 1.5 × ULN, AST/ALT < 5 × ULN, creatinine < 1.25 × ULN, albumin > LLN
NCT01479608	Oslo, Norway	Hb > 10 g/dL, neutrophils > 1.0, TRC > 75, bilirubin < 2 × ULN, AST/ALT < 5 × ULN, creatinine < 1.25 × ULN, albumin > LLN
NCT02597348	Villejuif, France	Platelet count > 80000/mm³, WBC > 2500/mm³, normal renal function
NCT04865471	Padua, Italy	Creatinine normal, platelets > 60000/mm³, WBC > 2500/mm³
NCT04161092	Gothenburg, Sweden	Hb ≥ 90 g/L, WBC > 3.0 ×109/L, ANC ≥ 1.5 ×109/L, PLT > 75, bilirubin < 2 × ULN, AST/ALT < 5 × ULN, creatinine clearance ≥ 50 mL/minute
CEA
NCT04870879	Padua, Italy	CEA < 100 ng/mL
NCT03803436	Milan, Italy	CEA < 50 ng/mL
NCT05398380	Barcelona, Spain	CEA ≤ 80 μg/L
NCT06069960	RenJi Hospital, Shanghai, China	CEA ≤ 80 μg/L or ≥ 50% reduction after treatment
NCT04742621	NY, United States	CEA < 200 μg/L
NCT05185245	Bologna, Italy	CEA < 80 μg/L or ≥ 50% reduction
NCT01479608	Oslo, Norway	CEA < 100 ng/mL (subset)
NCT02597348	Villejuif, France	CEA < 80 μg/L or ≥ 50% decrease
NCT04865471	Padua, Italy	CEA stable or decreasing
Tumor characteristics
NCT02215889	Oslo, Norway	Unresectable CRLM, no extrahepatic disease except 1-3 resectable lung lesions < 15 mm, no local recurrence (confirmed by MR and colonoscopy)
NCT05186116	Modena, Italy	Unresectable CRLM, pT1-3, pN0/pN1 (< 4 nodes), no mucinous > 50%, R0 resection, BRAF wild-type
NCT05750329	RenJi Hospital, Shanghai, China	Unresectable HCC or CRLM, tumor shrinkage or stable after chemo, no abdominal metastases or ≤ 3 resectable lung metastases
NCT04870879	Padua, Italy	Unresectable CRLM, no extrahepatic disease, no lesion > 10 cm before chemo, < 10% chemo response accepted if 20% response after TACE/90Y
NCT03803436	Milan, Italy	Non-mucinous colon adenocarcinoma, pT1-3 pN0/1 (< 4 nodes), R0 resection, BRAF and RAS wild-type, MSS
NCT05398380	Barcelona, Spain	Bilateral, liver-limited unresectable CRLM, R0 resection of primary, stage ≤ T3N1 (or T4 if ≥ 2 years interval)
NCT06069960	RenJi Hospital, Shanghai, China	CRLM limited to liver or bilateral, unresectable, T ≤ T3N1 or T4N0/T4N2 if ≥ 2 years
NCT04742621	NY, United States	No extrahepatic disease or local recurrence, liver metastases stable/regressed for ≥ 6 months
NCT05185245	Bologna, Italy	Unresectable CRLM, p ≤ T4a, R0 resection, no extrahepatic disease
NCT03494946	Oslo, Norway	Liver metastases not resectable, No extrahepatic disease except resectable lung lesions < 15 mm
NCT01479608	Oslo, Norway	Histologically confirmed CRC, no extrahepatic disease or recurrence, ≥ 6 liver metastases (part A), pN0 (part B), metachronous liver metastases
NCT02597348	Villejuif, France	Confirmed unresectable CRLM, BRAF wild-type, no local recurrence
NCT04865471	Padua, Italy	Unresectable CRLM, no extrahepatic disease except resectable lung/hilar metastases
NCT04161092	Gothenburg, Sweden	Unresectable CRLM, R0 resection of primary, no extrahepatic disease, measurable liver metastases
Treatment history
NCT02215889	Oslo, Norway	≥ 8 weeks of chemotherapy
NCT05186116	Modena, Italy	Objective response to 1st-line (≥ 4 months) or stable disease during 2nd-line (≥ 4 months)
NCT05750329	RenJi Hospital, Shanghai, China	6-8 weeks of 1st-line chemotherapy
NCT04870879	Padua, Italy	≥ 3 months chemotherapy, RECIST response or SD, response after TACE/90Y if poor initial response
NCT03803436	Milan, Italy	1st or 2nd-line chemo response ≥ 4 months, max 2 lines
NCT03488953	Jena, Germany	≥ 8 weeks of systemic chemotherapy, SD or regression
NCT05398380	Barcelona, Spain	≥ 3 months chemotherapy, max 2 lines, response per RECIST within 3 months before screening
NCT06069960	RenJi Hospital, Shanghai, China	6-8 weeks 1st-line chemo, stable or partial regression
NCT04742621	NY, United States	≥ 6 months chemotherapy
NCT05185245	Bologna, Italy	≥ 3 months of at least 1 line, PR or SD per mRECIST
NCT03494946	Oslo, Norway	Progressive disease or intolerance to 1st-line, randomized before evaluation 8-12 weeks after 2nd-line
NCT01479608	Oslo, Norway	≥ 3 cycles chemotherapy (6 weeks), 10% response to chemo before progression (varies by subgroup)
NCT02597348	Villejuif, France	≥ 3 months tumor control on last chemo line, ≤ 3 lines total
NCT02864485	Toronto, Canada	≥ 3 months chemo with FOLFOX/FOLFIRI +/- bevacizumab, stable or regressing LM
NCT04865471	Padua, Italy	≥ 3 months chemotherapy, ≥ 8 weeks SD or PR per RECIST 11
NCT04161092	Gothenburg, Sweden	≥ 2 months chemotherapy, no progression at last RECIST evaluation
Other criteria
NCT02215889	Oslo, Norway	Imaging within 4 weeks prior to LT meeting, colonoscopy/CT colography ≤ 12 months
NCT05186116	Modena, Italy	Confirmed R0 resection, multi-modal imaging (CT + MRI + PET)
NCT04870879	Padua, Italy	≥ 10 months from CRC resection to LT listing
NCT05398380	Barcelona, Spain	≥ 12 months from primary CRC resection to transplant, TNM staging required, no contraindications from prior hepatic resection
NCT06069960	RenJi Hospital, Shanghai, China	≥ 3 months from CRC resection to transplant, MDT confirmation, imaging via PET/CT + MRI
NCT04742621	NY, United States	≥ 1 year from CRC diagnosis and ≥ 6 months from resection to LT
NCT01479608	Oslo, Norway	Signed informed consent, standard oncologic surgery with CRM ≥ 2 mm, study includes special subgroups (A-D)
NCT02597348	Villejuif, France	R0 resection, no extrahepatic localization (CT/PET), normal renal function, nephrologist evaluation
NCT02864485	Toronto, Canada	Living donor identified, ABO compatible, no major vascular invasion
NCT04865471	Padua, Italy	≥ 6 months from CRC resection to LT list, validation committee approval
NCT04161092	Gothenburg, Sweden	≥ 1 year from CRC diagnosis to inclusion, imaging within 4 weeks, colonoscopy < 12 months

ECOG: Eastern Cooperative Oncology Group; Hb: Hemoglobin; TRC: Tanned red cell; ULN: Upper limit of normal; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; LLN: Lower limit of normal; ANC: Absolute neutrophil count; WBC: White blood cell; CEA: Carcinoembryonic antigen; CRLM: Colorectal cancer liver metastases; MR: Magnetic resonance; HCC: Hepatocellular carcinoma; TACE: Transcatheter arterial chemoembolization; MSS: Micro satellite stability; CRC: Colorectal cancer; PR: Partial response; SD: Stable disease; LT: Liver transplantation; CT: Computed tomography; MRI: Magnetic resonance imaging; PET: positron emission tomography; TNM: Tumor node metastasis; MDT: Multi-disciplinary treatment; CRM: Circumferential resection margin; eGFR: Estimated glomerular filtration rate.

General patient characteristics

A common requirement across most studies is that candidates be within a specified age range, typically between 18 and 75 years, though some studies impose stricter limits, such as 18-65 years. For instance, the study conducted in Valencia (No. NCT04616495) restricts inclusion to patients aged ≤ 65 years, whereas RenJi Hospital (No. NCT05750329) allows patients up to 75 years old. This discrepancy likely reflects differences in institutional perspectives on the impact of age on post-transplant outcomes.

Another crucial factor is the Eastern Cooperative Oncology Group (ECOG) Performance Status, with nearly all studies requiring a baseline ECOG score of 0-1, reflecting a good functional status. Few trials, such as Modena (No. NCT0518 6116), extend eligibility to patients with ECOG 2.

Body mass index (BMI) is another area of divergence. Many protocols exclude patients with BMI > 30 kg/m², reflecting concerns about post-surgical complications and long-term survival. However, some trials do not explicitly mention BMI restrictions, possibly indicating a case-by-case assessment.

Primary tumor characteristics

All studies require histologically confirmed colorectal adenocarcinoma, but they differ in their molecular and histological restrictions.

A number of trials exclude patients with BRAF mutations or microsatellite instability-high tumors, given their association with poor prognosis and higher recurrence rates. For example, the Barcelona study (No. NCT05398380) explicitly excludes patients with BRAF mutations, whereas Padova (No. NCT04870879) and RenJi Hospital (No. NCT 06069960) groups require BRAF wild-type status for inclusion. COLT trial requires both BRAF and RAS wild-type.

Tumor staging requirements also vary. While many studies accept patients up to T3N1 tumors, others impose stricter conditions. The Toronto study (No. NCT02864485) allows T4a cases but mandates a minimum six-month interval between primary tumor resection and transplantation, ensuring disease stability. In the case of lymphatic involvement, often, a limitation is reported in terms of the numerosity of metastatic nodes. As an example, Modena (No. NCT05186116) and the COLT (No. NCT03803436) trials reported a limit of 4 nodes.

Metastatic disease characteristics

All studies focus on the diagnosis of unresectable, limited to liver disease, while Valencia (No. NCT04616495) and Barcelona (No. NCT05398380) trials impose additional conditions such as lesion size restrictions (≤ 5.5 cm in some cases). Interestingly, some protocols allow limited lung metastases if they are resectable. The Oslo (No. NCT03494946 and No. NCT02215889) and Padova (No. NCT04865471) trials permit 1-3 lung lesions.

Response to systemic therapy

Pre-transplant chemotherapy response is a widely accepted inclusion criterion, but study requirements differ significantly. Most trials demand a minimum of 3-6 months of systemic chemotherapy, with RECIST 1.1 criteria used to evaluate tumor stability or regression. Some trials, such as Barcelona (No. NCT05398380), require CEA levels to remain below 80 μg/L or show a ≥ 50% reduction, while others, such as Padova (No. NCT04870879), include patients if CEA remains stable or decreases over time (allows patients with CEA up to 100 μg/L). These variations reflect differing institutional perspectives on the role of biomarker dynamics in predicting recurrence risk post-transplant.

Time from primary tumor resection to transplantation

A minimum waiting period between primary tumor resection and transplantation is often imposed to assess disease stability. Some studies require at least 6 months, while Barcelona (No. NCT05398380) extends this period to 12 months. The rationale for this restriction is to ensure that patients do not develop rapid disease progression, which would contraindicate transplantation.

EXCLUSION CRITERIA OF ONGOING TRIALS

Also, in the case of exclusion criteria, different approaches have been identified (Table 2).

Table 2 Exclusion criteria reported in the studies registered in ClinicalTrials.gov focused on liver transplantation for colorectal cancer liver metastases.

NCT	Location	Criteria
Demographic parameters
NCT02215889	Oslo, Norway	Weight loss > 10% in the last 6 months; BMI > 30
NCT04870879	Padua, Italy	Weight loss > 10% in the last 6 months; BMI > 30
NCT03494946	Oslo, Norway	Weight loss > 10% in the last 6 months; BMI > 30
NCT01479608	Oslo, Norway	Weight loss > 10% in the last 6 months; BMI > 30
NCT04616495	Valencia, Spain	BMI ≥ 30; pregnancy at time of inclusion; 10% weight loss
NCT04865471	Padua, Italy	Weight loss > 10% in the last 6 months; BMI > 30; pregnancy or breastfeeding
NCT04161092	Gothenburg, Sweden	Weight loss > 10% in the last 6 months; pregnancy or breastfeeding
Biochemical markers
NCT04616495	Valencia, Spain	Creatinine clearance < 50 mL/minute
NCT02864485	Toronto, Canada	Renal dysfunction with creatinine clearance < 50 mL/minute
CEA
NCT03494946	Oslo, Norway	CEA > 80 ng/mL with 2 other negative prognostic factors
NCT04616495	Valencia, Spain	CEA > 80 ng/mL at time of enrolment
Tumor characteristics
NCT02215889	Oslo, Norway	Bone or CNS metastases; prior breast cancer or malignant melanoma
NCT05186116	Modena, Italy	Prior extrahepatic disease or primary tumor relapse
NCT05750329	RenJi Hospital, Shanghai, China	Extrahepatic tumor burden (except resectable lung metastases); macrovascular infiltration
NCT04870879	Padua, Italy	Prior extrahepatic metastases or local relapse
NCT03803436	Milan, Italy	Prior extrahepatic metastases or primary tumor relapse; extraperitoneal tumors
NCT05398380	Barcelona, Spain	Lesion > 5.5 cm; primary tumor recurrence; Lynch syndrome; BRAF/MSI tumors
NCT06069960	RenJi Hospital, Shanghai, China	Extrahepatic burden or large vessel invasion; lesion > 5.5 cm; BRAF/MSI
NCT04742621	NY, United States	Extrahepatic disease; MSI-H/dMMR or BRAF mutation; prior lung metastasectomy
NCT05185245	Bologna, Italy	Extrahepatic metastases; local recurrence
NCT03494946	Oslo, Norway	Local relapse; non-hepatic metastasis; thoracic/abdominal lymph nodes; lesion > 10 cm; > 5.5 cm with other risks
NCT01479608	Oslo, Norway	Extrahepatic metastases or local relapse
NCT04616495	Valencia, Spain	Extrahepatic metastases; lesion > 5 cm; BRAF mutation
NCT04161092	Gothenburg, Sweden	Extrahepatic disease; lesion > 10 cm; abdominal lymphadenopathy; BRAF mutation
Treatment history
NCT05186116	Modena, Italy	Disease progression
NCT05750329	RenJi Hospital, Shanghai, China	Tumor progression during chemotherapy
NCT03488953	Jena, Germany	Progression during chemotherapy
NCT05398380	Barcelona, Spain	Tumor recurrence in the last 12 months
NCT06069960	RenJi Hospital, Shanghai, China	Tumor progression during chemotherapy
NCT04874259	Seoul, Korea	Progression of liver metastases at any time point
NCT04616495	Valencia, Spain	No neoadjuvant chemotherapy
NCT01479608	Oslo, Norway	Not received standard CRC treatment
NCT02597348	Villejuif, France	Not received standard CRC treatment
Other criteria
NCT05186116	Modena, Italy	HIV, psychiatric disorders, active substance abuse, low compliance
NCT05750329	RenJi Hospital, Shanghai, China	AIDS, uncorrectable cardiopulmonary disease, anatomical abnormalities
NCT03803436	Milan, Italy	HIV, substance abuse
NCT05398380	Barcelona, Spain	Substance abuse, psychological/social issues, cardiac/pulmonary disease, infection
NCT06069960	RenJi Hospital, Shanghai, China	Cardiopulmonary disease, anatomical abnormalities, substance abuse, AIDS
NCT04874259	Seoul, Korea	Hemodynamic instability, peptic ulcer, HIV, pregnancy
NCT04616495	Valencia, Spain	HIV or HCV, pregnancy, general contraindication to LT, insurance issues
NCT02864485	Toronto, Canada	HIV, HBV/HCV, cardiac or pulmonary insufficiency, debilitating neuropathy
NCT02597348	Villejuif, France	Severe comorbidities, active infection, alcohol abuse, lack of compliance/support
NCT04865471	Padua, Italy	General contraindication to LT, refusal, pregnancy/breastfeeding
NCT04161092	Gothenburg, Sweden	Pregnancy, previous organ transplant

LT: Liver transplantation; BMI: Body mass index; CEA: Carcinoembryonic antigen; CNS: Central nervous system; MSI: Microsatellite instability; MSI-H: Microsatellite instability-high; dMMR: Mismatch repair-deficient; CRC: Colorectal cancer; HIV: Human immunodeficiency virus; AIDS: Acquired immune deficiency syndrome; HBV: Hepatitis B virus; HCV: Hepatitis C virus.

Oncological factors

A major exclusion criterion across all studies is progressive disease despite chemotherapy, as this suggests aggressive tumor biology. As previously reported, the presence of extrahepatic metastatic disease beyond resectable lung metastases is a near-universal exclusion factor. Node dimensions is an exclusion factor in some trials, such as Barcelona (No. NCT05398380) and RenJi Hospital (No. NCT06069960).

Patient-related factors

Beyond oncological considerations, studies impose exclusions based on poor general health. Weight loss exceeding 10% in six months is a frequent exclusion criterion, seen in trials like Oslo (No. NCT03494946) and Padova (No. NCT048 65471). Obesity (BMI > 30 kg/m²) is another common exclusion, though some studies assess this on a case-by-case basis.

Patients with multiple and severe comorbidities-including uncontrolled cardiac or pulmonary disease, renal dysfunction (estimated glomerular filtration rate < 50 mL/minute), or active systemic infections (human immunodeficiency virus, hepatitis B/C)-are excluded across all studies.

Prior malignancies

A history of another malignancy within the past 5 years is a frequent exclusion criterion. However, exceptions are sometimes made for basal cell carcinoma or in situ cervical cancer.

DISCUSSION

General considerations

While there are strong commonalities in patient selection criteria across these studies, key differences exist in age limits, tumor burden thresholds, chemotherapy requirements, and allowance for extrahepatic disease. The Oslo RAPID and SECA-III protocols, along with the Padova RAPID trial, adopt more permissive approaches, allowing limited lung metastases and higher CEA cutoffs. In contrast, Valencia and Barcelona enforce stricter criteria, excluding even resectable extrahepatic disease and imposing narrower tumor burden limits. Patient selection varies considerably depending on institutional philosophies, clinical trial designs, and evolving transplantation guidelines. This variability complicates outcome comparisons across different studies and hinders the development of universally accepted selection criteria. As previously reported, some trials have developed scores to identify low-risk cases, such as the Oslo and Fong Scores[9,12]. These scores incorporate internationally recognized risk factors for tumor relapse after LT, including lesion diameter, number of liver lesions, CEA levels, lymph node positivity in the primary tumor, the time elapsed between primary tumor removal and CRLM discovery or waiting list inscription, and radiological response to pre-transplant treatments (Table 3). These variables are integrated into protocols with varying degrees of stringency, as identified in ClinicalTrials.gov. In recent years, additional prognostic factors have emerged as potentially relevant, including tumor sidedness, suspected pulmonary metastases, and positron emission tomography (PET) liver uptake.

Table 3 Characteristics of the Oslo, Fong, and Eastern Cooperative Oncology Group scoring systems.

Score	Component	Scoring/criteria	Cutoff	Clinical interpretation
Oslo	Tumor size	> 5.5 cm = 1 point	≥ 2 points	Higher recurrence risk post-LT
	CEA level	> 80 ng/mL = 1 point
	Time from primary to LT	< 2 years = 1 point
	Response to chemotherapy	< 10% shrinkage = 1 point
Fong	Nodal status of primary	Positive = 1 point	≥ 3 points	Worse prognosis post-resection
	Disease-free interval	< 12 months = 1 point
	Number of liver metastases	> 1 = 1 point
	CEA level	> 200 ng/mL = 1 point
	Largest tumor	> 5 cm = 1 point
ECOG	Performance status	0 = fully active; 1 = symptomatic but completely ambulatory; 2 = symptomatic, < 50% in bed during the day; 3 = symptomatic, > 50% in bed, but not bedbound; 4 = bedridden	≥ 2 = ineligible	Stratifies fitness for treatment or surgery

ECOG: Eastern Cooperative Oncology Group; LT: Liver transplantation; CEA: Carcinoembryonic antigen.

Sidedness

Regarding sidedness, non-transplant studies have reported that patients with left-sided tumors have better five-year OS compared to those with right-sided CRLM, likely due to more aggressive biological and molecular features[13,14]. An Oslo study merging SECA-I and SECA-II transplant patients (n = 19) found that right-sided primary tumors correlated with significantly reduced DFS (median 4 vs 13 months; P = 0.044) and inferior five-year OS (P = 0.001). Right-sided tumors also showed shorter time-to-relapse (P = 0.016) and a borderline higher rate of KRAS mutations (P = 0.079)[15]. Another Oslo trial compared 53 patients with CRLM treated with portal vein embolization (PVE) followed by resection vs 50 patients enrolled for LT. Among LT-enlisted patients, five-year OS rates were 45.3% for left-sided and 0% for right-sided tumors, while PVE + resection patients had a five-year OS rate of 12.5%[16]. A third Norwegian study (n = 61) further demonstrated that right-sided tumors had worse outcomes (five-year OS: 10.0% vs 60.1%; P < 0.001)[17].

Lung metastatic involvement

Regarding lung metastases, a study on six SECA patients whose disease had progressed despite standard chemotherapy revealed that half experienced pulmonary relapse, with a short median DFS (3.3 months). However, despite this, the median OS was remarkably long (41 months; five-year OS: 44%), suggesting that lung metastases may progress slowly, even in the absence of treatment[18]. Another Norwegian study enrolled 10 patients in the SECA-II arm D study, which included the patients ineligible for arms A, B, and C and allowed for resectable pulmonary metastases (n = 2). Post-LT, 60% experienced pulmonary relapse, with a median DFS and OS of 4 and 18 months, respectively[19]. In the specific setting of pulmonary metastatic disease, the opportunity to consider LT not as a curative but a palliative approach remains obviously controversial. However, while survival remains the primary endpoint in the majority of the “transplant oncology” studies, long-term quality of life outcomes are underreported and warrant more attention. In fact, a LT patient presenting a manageable recurrence generally reports a favorable long-term high-quality life, substantially superior to the one predictable remaining only under palliative oncological therapies. Further prospective data are needed to better understand the long-term physical, emotional, and functional wellbeing of these patients.

PET scan

Concerning PET, the Oslo trial merging SECA-I and SECA-II patients (n = 19) demonstrated that a metabolic tumor volume (MTV) < 70 cm³ effectively stratified patients into risk categories, with a five-year OS of 78% vs 0% for those meeting or exceeding this threshold[15]. Another Oslo trial (n = 61) reported a five-year survival of 66.7% for patients with MTV < 70 cm³ vs 23.3% for those exceeding it (P < 0.001)[17]. A multicenter United States study (n = 26) found that MTV > 70 cm³ was associated with a hazard ratio of 2.42 [95% confidence interval (CI): 2.2-62.2, P = 0.006] for recurrence risk and an area under the curve of 0.771 (95%CI: 0.560-0.981, P = 0.030) for predicting recurrence[20].

Challenges and limitations

The reported results highlight the challenges in defining definitive selection parameters for CRLM patients. A key limitation is that current prognostic scores are based on relatively small patient cohorts, reducing their robustness and generalizability. Additionally, these scores often reflect specific institutional practices, making them difficult to apply universally. Notably, the majority of trials originate from Oslo, limiting the external validity of findings. Small study populations also reduce statistical power, making it difficult to draw strong conclusions about survival benefits and recurrence risks. Another major limitation is the lack of large-scale, multicenter studies that integrate various selection criteria into a unified framework. Most research remains single-center, restricting the feasibility of inferential statistical analyses such as multivariate models that account for confounding variables. As a result, the exact impact of sidedness, pre-existing lung metastases, or PET remains speculative. Without broader studies compiling data across institutions, establishing a definitive, evidence-based standard for patient selection remains challenging[21,22].

An additional layer of complexity arises from the ethical considerations surrounding organ allocation. Given the persistent scarcity of donor livers, especially in regions with limited resources, the use of grafts for patients with CRC metastases-particularly those with previous or suspected extrahepatic disease-raises concerns about potential organ waste. While early outcomes in strictly selected patients have been encouraging, further discourse is needed to assess whether such allocation aligns with broader principles of justice and utility in transplantation. Future studies should incorporate ethical frameworks and cost-benefit analyses across diverse healthcare settings to ensure equitable and efficient use of organs.

The present study has certain limitations. First, it is based on a synthesis of available clinical trials and published data, meaning that unpublished or ongoing research may introduce additional variables not accounted for in this analysis. Second, inherent variability exists in patient selection, treatment response, and transplant outcomes, which are not always standardized across studies. Finally, while this study compares different inclusion and exclusion criteria, the absence of a universally accepted benchmark complicates definitive conclusions. Addressing these challenges will require international collaboration, larger prospective studies, and the harmonization of selection criteria to ensure that LT for CRLM is both effective and equitably applied.

CONCLUSION

The landscape of LT for CRLM is continuously evolving, with different institutions tailoring criteria based on local expertise, patient selection philosophy, and emerging clinical data. The present study showed that relevant differences in inclusion and exclusion criteria exist among different studies, suggesting to work in a unified screening criteria framework for reference in future studies. While stringent selection is crucial for optimizing survival and minimizing recurrence, further research is needed to refine these criteria and expand transplantation access to a broader patient cohort. Beyond refining selection criteria, future research should also focus on comparative ethical analyses of graft allocation, regional disparities in access, and the development of prognostic models that include quality-of-life indicators. Multidisciplinary collaboration will be key to shaping a responsible and patient-centered expansion of LT for CRLM.