Liver transplantation as an alternative for the treatment of non-resectable liver colorectal cancer: Advancing the therapeutic algorithm

Bdi Rwshdeh ,Richrd Bell ,Adul Hkeem ,Rj Prsd ,*

a Division of Transplant Surgery, Department of General Surgery, Medical College of Wisconsin, Milwaukee, WI, 53226, USA

b Leeds Teaching Hospital, Leeds, UK

Keywords: Liver transplantation Colorectal cancer liver metastases Non-resectable liver metastases

ABSTRACT Colorectal cancer is a leading cause of cancer-related mortality,with nearly half of the affected patients developing liver metastases.For three decades,liver resection (LR) has been the primary curative strategy,yet its applicability is limited to about 20% of cases.Liver transplantation (LT) for unresectable metastases was attempted unsuccessfully in the 1990s,with high rates of perioperative death and recurrence.There is now more interest in this strategy due to improvements in systemic therapies and surgical techniques.A significant study conducted by the Oslo group showed that patients receiving liver transplants had a 60% chance of survival after five years.Significantly better results have been achieved by using advanced imaging for risk stratification and further refining selection criteria,especially in the Norvegian SECA trials.This review carefully charts the development and history of LT as a treatment option for colorectal cancer liver metastases.The revolutionary path from the early days of exploratory surgery to the current situation of cautious optimism is traced,highlighting the critical clinical developments and improved patient selection standards that have made LT a potentially curative treatment for such challenging very well selected cases.

Introduction

The emergence of the term ＂transplant oncology＂ can be traced back to its introduction in medical literature,which was prompted by the pressing demand for efficacious therapeutic interventions targeting hepatocellular carcinoma [1,2].Over the course of time,the scope of the application of transplant oncology has expanded,indicating a notable expansion in its utilization for the treatment of colorectal liver metastases (CRLM) [3-5].

Colorectal cancer is one of the most common cancers globally [6].Nearly half of patients develop liver metastases at some stage in their natural history [6-8].Over the last 30 years,liver resection (LR) has been established together with adequate chemotherapies as the primary potentially curative option for CRLM [9,10].Contrary to the evolution of resection,liver transplantation (LT) for unresectable colorectal metastases attempted during the 1960’s through the 1990’s resulted in dismal outcomes and a consequent lack of progress [5,11-13].The European Liver Transplant Registry (ELTR) data showed a 1-and 5-year survival of 60%and 18%,with significant perioperative mortality and high cancer recurrence [14,15].Many deaths were however due to major surgical and infectious complications [16].

Significant improvements in surgical techniques,imaging modalities for staging,and systemic therapies resulted in an expansion of criteria for resection [17].Two-stage hepatectomies,the liver first approach,preoperative portal vein embolization,associating liver partition and portal vein ligation for staged hepatectomy (ALPPS),down-staging systemic therapies,repeat resections,laparoscopic and robotic approaches to liver surgery have advanced the role of LR in CRLM [10,18-20].Despite all technical advances LR is only possible in about 20% of CRLM patients [10,21].A 5-year survival of 30%-58% is achieved in this group of patients whose outcome otherwise with systemic therapy would result in a less than 10% 5-year survival [7,22].As boundaries for LR have been pushed with extreme techniques,it also became apparent that tumor biology and burden both play a significant role in determining the short-and long-term outcomes [7,23].

It has been shown that LT has emerged as a valid treatment option for carefully selected cases of metastatic neuroendocrine and gastrointestinal stromal tumor (GIST).When performed under restrictive criteria,LT can indeed provide excellent long-term,even disease-free outcomes [24-26].

Within the dynamic realm of CRLM treatment,non-randomized studies have begun to illuminate the feasibility of LT as a viable therapeutic alternative for a specific subset of patients [27-30].The resurgence of interest in LT as a treatment for CRLM started with the landmark paper by Hagness et al.from the Oslo group in 2013 [31].At a time when most health care systems had significant mortality on the LT waiting list due to the gap between demand and supply,Norway was in an enviable position of organ surplus.This allowed the Oslo group to explore the role of LT for CRLM.With broad entry criteria of radical excision of primary tumor,ECOG performance status of 0-1,absence of extrahepatic disease,and at least 6 weeks of systemic chemotherapy,25 patients were recruited to the study (SECA,NCT01311453).Twenty-one of these patients underwent LT.The overall survival (OS) at 1,3 and 5 years was 95%,68% and 60%,respectively.Nineteen of the 21 developed recurrent disease,but with aggressive management of metastatic,mainly pulmonary,disease,33% were rendered free of disease at a median follow-up of 27 months.

Evidence of LT for CRLM

Recent publications in the literature supporting LT for CRLM were predominantly from Europe,with a single multicenter report from North America.Most of the European experience is from the Oslo program.

SECA 1 study

The original report from Oslo,with few selection criteria,demonstrated the potential role of LT in the management of CRLM.With a high risk of recurrent disease and aggressive management of such recurrences,a 60% 5-year OS was a significant improvement from the previously published literature [31].The report highlighted the importance of selection criteria in predicting disease recurrence and OS.Prognostic factors shown to be important in determining outcomes after LR like nodal status of primary cancer,tumor burden in liver,carcinoembryonic antigen(CEA) and carbohydrate antigen 19-9 levels,metachronous versus synchronous metastases,delay between surgery of the primary tumor and treatment of metastases,and response to chemotherapy were evaluated.Maximum tumor diameter of＞5.5 cm,progression on chemotherapy,disease-free interval of＜2 years from primary surgery and LT and CEA levels＞80 ng/mL were predictors of adverse outcome.

The Oslo score was used as a prognostic measure for CRLM patients receiving LT in the SECA 1 study.The length of time between the initial cancer diagnosis and the development of liver metastases,the size and number of liver metastases,CEA levels,and the patient’s response to chemotherapy are all considered in this scoring system.Together,these factors help predict survival rates and the chance of recurrence after transplantation,which helps with patient selection and management plans when it comes to LT for CRLM.

The Oslo score (0 to 4 points) was calculated with 1 point for each of the following: largest lesion＞5.5 cm,CEA＞80 ng/mL,time from surgery of primary tumor to LT of less than 2 years and progressive disease on chemotherapy at the time of LT.

All patients were maintained on the mechanistic target of rapamycin (mTOR) based immunosuppression.This landmark study not only showed feasibility but allowed development of future,more selective criteria to evaluate the role of LT for this indication.

SECA 2 study

The follow-up study following SECA 1 study from the Oslo group was SECA 2 study in 2020 [3]which included 15 patients with a more restrictive set of criteria.These included time from diagnosis to transplant of at least 1 year and at least 10% response to chemotherapy.All patients had preoperative positron emission tomography/computed tomography (PET/CT) and in addition to standard parameters,metabolic tumor volume (MTV) and total lesions glycolysis (TLG) were calculated.Patients were stratified according to the Fong clinical risk score (FCRS) and Oslo scores.

At a median follow-up of 36 months,the 1-,3-and 5-year OS now reached 100%,83% and 83%,respectively.Patients with FCRS score of 1-2 had a significantly better disease-free survival (DFS)compared to those with FCRS score of 3-4.Six of the eight patients with recurrent disease had metastases in lungs,either only or as first site of recurrence.A total of 11 patients had no evidence of disease at the time of last follow-up of 60 months.This study clearly represents a further evolution in exploring the role of LT for CRLM.This demonstrated a significant improvement in OS and outcome compared to SECA 1 study by implementing stricter selection criteria.This also demonstrated the importance of FCRS score in risk stratification and selecting patients for LT.These two studies (SECA 1 and SECA 2) laid groundwork for using various parameters available with PET/CT like MTV for risk stratification and tumor biology prediction.

Compagnons hepato-biliaires survey

An interrogatory between members of this group identified 12 patients that were transplanted at 4 European centers between October 1995 and October 2015 [32].There were in fact no selection criteria as the indication for LT was mainly based on “emotional grounds”.Nine patients had liver metastases detected within 12 months of diagnosis of primary tumor.Pre-LT chemotherapy was administered to the majority of patients-11 out of 12-all of whom responded to it and none of whom had any progression at the time of transplantation.Four patients had adjuvant chemotherapy.Interestingly,10 patients had prior LR;5 of them had more than one LR.Median time to transplant was 41 months from primary tumor surgery.After a median follow-up of 26 months,the 1-,3-and 5-year OS rates were 83%,62% and 50%,and DFS 38% at 5 years.Despite the heterogeneity of this small cohort,this paper also demonstrated that DFS following LT is achievable in patients that had a long treatment period from the diagnosis of CRLM,with 5 of the 12 patients having repeated LR before LT suggesting that the tumor biology in these patients had withstood the test of time with liver limited disease.

North American experience

Three North American centers reported their results from a prospective cohort study including ten patients undergoing living donor LT (LDLT) [33].Only one patient had metachronous disease,and three had previous LR.At a median follow-up of 18 months,3 patients developed recurrent disease despite having low Oslo and Fong scores.The 18-month OS and DFS were 100% and 62%,respectively.This study demonstrates practical challenges with availability of grafts in a health care system with overwhelming demands on cadaveric grafts and ethical dilemma of risk for a living donor and long-term recipient benefit.

Prognostic factors of LT for CRLM

Whilst several centers described prognostic factors and scoring systems that influence outcomes following resection for CRLM with a mix of contribution from different healthcare systems and socioeconomic demographics,majority of the data on prognostic factors following LT are from the single Oslo center.In a recent study,Dueland et al.[34]included all patients who underwent LT for CRLM at Oslo,analyzing the prognostic factors that impacted survival.Sixty-one patients with a follow-up of 16 to 165 months were included in the study.Out of the 60 patients who survived LT,47 experienced a recurrence.After a median period of 9 months,median DFS,OS,and survival after relapse were 11.8,60.3,and 37.1 months,respectively.

Factors that significantly influenced poor OS were largest tumor size＞5.5 cm (P ＜0.001),disease progression on chemotherapy(P=0.02),CEA＞80 ng/mL (P ＜0.001),MTV＞70 cm3(P ＜0.001),right colon primary (P ＜0.001),tumor burden score of 9 or higher (P=0.02) and 9 or more liver metastases (P=0.02).The tumor burden score measures a patient’s cancer impact based on factors like tumor number,size,location,growth rate,and organ function.This measurement helps determine disease severity,treatment,and prognosis in tumors.High tumor burden scores,especially in CRLM,indicate advanced disease stage and affect transplant suitability and outcomes [35].The Oslo score was significant in predicting outcomes.Ten patients with Oslo score of 0 had a median OS of 151.6 months and both 5-and 10-year OS of 88.9%.The median OS of patients with Oslo score of 0 to 2 was 92.0 months with 5-and 10-year OS of 63.4% and 45.7%,compared to median OS of 24.8 months and 5-and 10-year OS of 8.3% and 0%for patients with Oslo score of 3 to 4 (P ＜0.001).FCRS was equally important in predicting outcomes.Patients with FCRS score of 1,2,3,4,and 5 had a median OS of 164.9,90.5,59.9,32.8,and 25.3 months,respectively (P ＜0.001).Patients with FCRS score of 0 to 2 had a significantly better OS compared to patients with FCRS score of 3 to 5 (FCRS score of 0-2 median OS of 151.6 months,5-year OS of 75.4%;FCRS score of 3-5 median OS of 50.8 months,5-year OS of 39.7%;P=0.01).Only 5 patients had metachronous disease as defined by diagnosis of liver metastases＞12 months from primary diagnosis.All five patients were alive at 5 years.

This study provides further clarity on the impact of tumor biology on influencing post-transplant overall and disease-free outcomes.The individual factors contributing to adverse tumor biology are looking increasingly similar to the factors that affect outcome after LR for colorectal metastases.

Transplantation vs.chemotherapy

Any new treatment for unresectable liver metastases needs to be compared to systemic therapy which is the current standard of care.Outcomes from systemic therapies have improved significantly in the last twenty years with the introduction of newer drugs and regimens.The only study that retrospectively compared systemic therapy in patients with comparable liver disease burden to LT (NORDIC VII vs.SECA 1) showed benefit with LT and a significantly improved 5-year OS [36].The chemotherapy used was 5-fluorouracil+leucovorin+oxaliplatin (FLOX) with or without cetuximab.In this study,the 5-year OS following LT was 56%compared to 9% with first-line chemotherapy despite the diseasefree/progression-free survival in both groups being comparable (8-10 months).The reasons postulated for this different OS was the relatively insidious nature of lung metastases seen after LT compared to far more aggressive liver disease in the chemotherapy group.Three randomised controlled trials (RCTs) are currently at different stages of progress compares LT to chemotherapy.A multicenter RCT TRANSMET (NCT02597348) in France comparing LT with best practice chemotherapy;the inclusion of 80 patients has been finalized,and the analysis will be available at the beginning of 2024 [17].The SECA 3 study (NCT034 94 946) compares LT with systemic therapy,transarterial embolization (TACE),selective internal radiation therapy (SIRT) and other treatments [17].EXCALIBUR 1+2 (NCT04898504) [17]is a three-arm randomised trial comparing LT,second-line chemotherapy or second-line chemotherapy with hepatic artery infusion.Needless to state,the results from these studies are eagerly awaited to firmly establish the role of LT in unresectable colorectal metastases.

In the MELODIC trial (NCT04870879),patients with unresectable liver-only metastases from colorectal cancer (CRC) are compared to chemotherapy and LT and its efficacy.Along with secondary outcomes like progression-free survival,dropout rates,and post-transplant complications,its main objective is to measure OS at 3 and 5 years.Eighteen chosen patients,aged 18 to 70 years,who fulfill the following requirements were included in the trial:ECOG performance status of 0-1,confirmed colon or rectum adenocarcinoma,ineligibility for curative liver resection,and absence of extrahepatic metastases.They had to have received chemotherapy and standard oncological resection,with certain requirements for response.Significant weight loss,a high BMI,prior malignancy,and general contraindications to LT are examples of exclusion factors.The purpose of this research is to shed light on whether LT is a feasible course of treatment for this particular patient population.

Another phase II trial is looking at the safety and efficacy of LDLT after standard chemotherapy for patients with non-resectable liver-only metastases from CRC (NCT02864485) [28].Eligibility for the study hinges on favorable Oslo scores and evidence of disease control after 3 months of chemotherapy.Those who are successful in finding a suitable donor take a 4-6 week break from chemotherapy prior to their preoperative evaluations.Those who do not have extrahepatic illness move on to transplantation.On the other hand,exclusion results from peritoneal disease or metastatic involvement in portal lymph nodes.The trial will recruit participants in two stages.First,a pilot group of five patients will be recruited;if this group is successful,20 more participants will be recruited.This research provides valuable insights regarding the potential of LDLT as a substitute for ongoing chemotherapy in a subset of patients diagnosed with CRC.

Transplantation vs.resection

LR for resectable tumors remains the standard of care.As surgical techniques evolved with two stage resection,remnant liver augmentation with portal vein embolization (PVE) and ALPPS,patients with high tumor burden have been offered the prospect of resection.Data from multiple LR studies demonstrate that patients with high tumor burden,and multiple bilobar metastases,(especially＞8) had outcomes inferior to those with lower tumor burden.It is highly likely that tumor burden is a surrogate marker for tumor biology.These studies demonstrate that there is a subgroup of patients with “technically resectable” disease which at the same time is “biologically unresectable” [14,37].Whilst transplantation is gaining ground in the management of unresectable liver metastases,the excellent OS prompted comparison to outcomes following LR in ‘technically resectable’ but “biologically unresectable” patients.

Dueland et al.[38]compared 50 patients who underwent LT to 53 patients with no extrahepatic disease who underwent PVE and resection.High tumor load (HTL) was defined as the presence of 9 or more metastases,or the maximum diameter of the largest tumor＞5.5 cm.The remaining patients were considered to have low tumor load (LTL).Fifteen (28.3%) of the 53 patients who had PVE did not progress to LR due to insufficient remnant liver hypertrophy or tumor progression.The median OS in this group was 10.9 months.The rest of the 38 patients who underwent resection following PVE had a median OS of 43.9 months and 5-year survival of 44.6%.In both LT and PVE groups,patients with LTL had a better survival compared to those with HTL.Thirty patients with an LTL in the PVE group had a 5-year OS of 53.1% and in the 23 who had resection,the survival was better at 69.3%.The OS following LT in LTL was 72.4% at 5 years.This is similar to the outcomes following LR in the LTL+PVE group (69.3%) and the whole LTL+PVE group(53.1%,P=0.08).In patients with HTL,there was a significant increase (P=0.007) in OS amongst patients undergoing LT compared to patients receiving PVE,with a median OS of 40.5 months in the LT group and 19.2 months in the PVE group.Twenty-one patients in the HTL-LT group with a left-sided primary tumor had significantly longer OS compared with 8 patients in the HTL-PVELR group (P=0.04) with a median OS of 59.9 months in the LT group compared to 29.8 months in the PVE-LR group.The 5-year OS was 45.3% (HTL-LT group) vs.12.5% (HTL-PVE-LR group).This study raises important questions on the role of LT in patients with“technically resectable” but “biologically unresectable” disease.In a similar study using LR patients with HTL,Lanari et al.[14]showed superior outcomes with LT over LR in a select group of patients with low Oslo score and HTL.

These two studies raise important questions on the possible superiority of LT over extended resections such as two stage hepatectomy and ALPPS in a selected group of patients where tumor biology favors liver replacement rather than inadequate microscopic clearance achieved by LR.A recent paper by Chávez-Villa et al.evaluated explant livers following transplantation for CRLM and found that 9 out of 14 patients had undiagnosed tumor at explant pathology [39].Whilst this may be an overestimate as the study was not clear about how many of these patients had state of art radiological imaging for CRLM in form of a magnetic resonance imaging (MRI) with Primovist/Eovist,it does provide some evidence to the presence of occult metastases in patients with high tumor burden.This further lends support for LT over complex and extreme liver resections in a selected group of patients.

RAPID procedure and CRLM

RAPID technique (Resection and Partial Liver Auxiliary Transplantation with Delayed Completion Hepatectomy) has been developed as a way to circumvent graft shortage by using left lateral grafts and employing the knowledge gained from remnant liver augmentation by portal flow modulation [18,40,41].

This technique is currently being evaluated in three studies(RAPID,LIVER-TWO-HEAL,and RAPID-PADOVA).Several technical and functional aspects in addition to oncological outcomes of this technique need further evaluation.It is important for institutions pursuing novel techniques to establish inter-institutional collaborations and evaluate along the lines of well-known frameworks for surgical innovation.

Organ allocation and LDLT

Resurgence of interest in LT for CRLM started with the Oslo studies made possible due to the organ surplus in Norway.Most countries continue to struggle with the gap in supply and demand for liver grafts.This gap resulted in waiting list mortality for patients with well-established indications for LT.The tremendous interest generated by the work from Oslo resulted in transplant programs and healthcare systems in different parts of the world exploring alternative graft options and surgical techniques to enable evaluation of transplantation for CRLM.Living donor grafts have been proposed as a solution to address the ethical conflicts in diverting cadaveric grafts to this evolving indication.There are current ongoing studies in North America and Europe using living donor partial grafts for single institutional or multicenter evaluations (NCT03488953 and NCT02864485) [17,28,42-45].The question of donor safety and double equipoise must evidently be thoughtfully considered but remains to be addressed from the outcomes of these studies.

Conclusions

Transplant oncology remains an emerging field.LT in wellselected patients presenting unresectable CRLM appears to offer an excellent 5-year survival when compared to conventional treatments.

The entire published experience on LT for colorectal metastases since the renewed interest in this treatment concerns less than 100 patients.The experience from Norway provides several pointers on prognostic factors and scores that can predict tumor biology and post-transplant outcomes.It is natural that given the enthusiasm and need to develop further evidence,that several single center and multicenter trials are currently ongoing.

Future studies will further refine selection criteria for patients likely to benefit from this approach,in particular,those with‘technically resectable’ but “biologically unresectable” disease.How these patients will be allocated grafts particularly in the context of systems with an organ deficit also requires evaluation although LDLT is one good proposed solution.

The future role of LT will be determined by the outcomes from these ongoing studies (Table 1) [43-58].

Table 1Completed and ongoing studies -liver transplant for colorectal liver metastases.

Acknowledgments

None.

CRediT authorship contribution statement

Badi Rawashdeh:Formal analysis,Writing -original draft.Richard Bell:Investigation,Resources.Abdul Hakeem:Investigation,Resources.Raj Prasad:Conceptualization,Supervision,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.