Liver transplantation and liver resection as alternative treatments for primary hepatobiliary and secondary liver tumors: Competitors or allies?

Jan Lerut

Institute for Experimental and Clinical Research (IREC), Université catholique Louvain (UCL), Avenue Hippocrate 56, 1200 Woluwe Saint Pierre, Brussels,Belgium

Although Starzl designed in the 1960’s liver transplantation (LT)to treat unresectable primary and also secondary liver tumors,transplantation still occupies a (too) small place in the respective therapeutic algorithms [1].Due to the lack of (any) selection criteria,the concept of transplantation became rapidly challenged because of the prohibitively high incidence of tumor recurrence [1-3].Not surprisingly,the “oncological pendulum” reversed in the nineties and moved the indication for LT from large multifocal lesions to a more limited tumor burden.In fact the paradigm changed towards transplanting the resectable and resecting the“untransplantable” cancer [4].More recently the pendulum is going again in the opposite direction based on major progresses made in the medico-surgical treatment of these different tumors [5-14].Seen the great potential of LT in the treatment of liver tumors,our team introduced in 2015 in the medical literature the term “transplant oncology” in clinical practice [15].

In this special issue ofHepatobiliary&Pancreatic Diseases In

ternational,expert hepatobiliary oncologic surgeons look at the respective roles of liver resection (LR) and LT in the treatment of primary and secondary hepatobiliary tumors.The aim of this issue is to convince hepatologists,oncologists,interventional radiologists,radiotherapists,“l(fā)iver resectors” and “l(fā)iver transplanters”,that LR and LT have to be seen as allies and not as competitors but also that LT should be considered more frequently to well selected cancer patients,especially when considering long-term disease-free survival (DFS).

Hepatocellular cancer (HCC) arising in a diseased liver evidently dominates the oncologic liver scene.In the 2022,eighth version of the “Barcelona therapeutic algorithm” (adopted also by the major Western HCC communities),the place attributed to LR and LT is still (too) small [16].The inclusion criteria of HCC patients for LT are since decades,curiously enough,based on morphologic tumor characteristics only,as described in two retrospective studies including 119 liver recipients only! Access to LT is indeed still merely restricted to patients presenting a tumor burden within the Milan criteria (one tumor ≤5 cm or 3 tumors ≤3 cm) or somewhat extended,the San Francisco criteria (one tumor ≤6.5 cm or maximum tumor burden ≤8.5 cm) [17,18].Three years before(1993),the Bismuth group had already reported an almost similar criterium nl.“the rule of three” (up to 3 lesions,up to 3 cm) [4].These restrictive Milan and San Francisco criteria,introduced in clinical practice in 1996 and 2001,denied and still deny access to a,potentially curative,LT for many,many patients worldwide.The reasons to implement these restrictive criteria were two-fold:giving maximal transplant chances to patients presenting a (prognostically more favorable) decompensated benign liver disease and avoiding erosion of the scarce post-mortem organ pool by allocating grafts to (prognostically less favorable) cancer patients.This argumentation is illogic for three reasons: tumor behavior can never be judged on morphologic aspects only,cancer patients merit an equal chance to cure and,last but not least,modern medicosurgical developments markedly change the outlook of these patients.The more than 60 (!) scores that have been developed since 2001,clearly point to the dissatisfaction of the transplant community with the widely used restricted,allograft allocation practices [19].Many scores generated both very good overall survival(OS) and DFS rates;many clearly overruled the Milan criteria or the San Francisco criteria.The main reason for such evolution were the completion of the static,“morphologic only” (nl.tumor number and diameter) by the dynamic,“morphologic-biologic” (nl.biomarker and radiologic responses) tumor characteristics.This dynamic combination is the gold standard in modern oncology because it allows to judge the response of a given tumor to a given(neo-adjuvant) treatment [7,8,10].Nowadays 80% of HCC patients have some surgical and/or non-surgical locoregional (LRT) and/or,more recently,also systemic therapies [7-14,20-25].It is very important to have all readily available serologic and non-serologic tumor markers [e.g.alpha-fetoprotein (AFP),AFP bound to Lens culinaris agglutinin (AFP-L3),des-gamma-carboxy prothrombin (DCP)or prothrombin induced by vitamin K deficiency or antagonist-II (PIVKA-II),inflammatory markers and tumor uptake at (18)Ffluorodeoxyglucose positron emission tomography (FDG-PET)-scan]as well as all modern imaging techniques (e.g.functional magnetic resonance and radiomics imaging;evaluation using Li-RADs and m-RECIST criteria) at hand in order to allow a correct judgment about the respective downstaging or stabilizing effects of all these different treatments on the tumor load [21-37].This information is of importance not only for OS and DFS rates but also to avoid futile LTs [14,22-25,38,39].The “allowance” in the recent BCLC algorithm to migrate from one HCC stage to another (e.g.from advanced and extended to early HCC) whilst applying a locoregional and/or systemic therapy is a major step forward to widen the access of HCC patients to LR and especially LT [16].

HCC is now the primary indication for LT.The recent,more deliberate,“transplant approach” to HCC is mainly based on one epidemiological and two “technical” factors.In the Western world,the disappearance of the viral B,C and D liver disease tsunamis led to a melting away of the waiting lists.As of a sudden HCC patients became also considered to be sick and so were more than welcomed to fill up the lists [40].In the Eastern world,the spectacular development of both advanced liver resection surgery and living donor liver transplantation (LDLT) not only allowed to eliminate interference with the post-mortem liver donor pool but,most of all,allowed to explore the boundaries of the extension of inclusion criteria and to transform LR and/or LT into real oncologic interventions,which means scheduled in function of the result of the neo-adjuvant treatment(s) [41-46].The large and more aggressive attitude towards LR for,also advanced,HCC played without any doubt a major role in this development [47,48].It is to be expected,seen the evolution form extended to advanced inclusion criteria,that many liver cancer patients will also need an adjuvant therapy post-LT [6,7,16,22,25,49].Indeed the 2022 BCLC algorithm is already challenged by recent developments.Several types of bi-and tri-modal combinations including interventional radiologic procedures (e.g.transarterial chemo-or radio-embolisation and radiofrequency),tyrokinase inhibition,chemo-and immunotherapy [38,39]and more precise external stereotactic (proton) radiotherapy opened already the door for transplanting patients presenting macrovascular (portal and/or hepatic venous) tumor invasion [48-54].Although already reported by the Starzl-team in 1999 (!),it has now been shown unequivocally that LT is superior to LR,the longer the follow-up,the more the results favor LT [55-58].Five-and ten-year recurrence rates reach 70% and 80%in LR patients,whereas this incidence is around 12% and 20% in LT patients [4].In the context of LT,it is important to make the differential diagnosis with mixed HCC-cholangiocellular cancer (CCC).This particular group of liver cancer is nowadays more frequently diagnosed based on the more extended use of refined immunohistochemical examinations [59,60].Modern imaging techniques,based on functional MRI and radiomics,allow to improve pre-LT tumor staging beyond size and number (the Milan or UCSF criteria) thereby increasing the yield of diagnosing “targetoid” (and thus not “pure” HCC) tumor lesions [35-37].Surgical refinements obtained from LDLT practice will for sure lead to expand advanced LR surgery including the hybridex-vivoliver resection and autotransplantation (ELRA) and of,the still largely underused,split LT [46,61,62].The liver graft will finally become a dual organ allowing to treat simultaneously two cancer patients when a suitable post-mortem allograft becomes available.

Parallel to the developments in the field of HCC,LT becomes more frequently proposed as a therapeutic option for perihilar (Klatskin) (phCCC) and intrahepatic cholangiocellular cancers(IHCCC).The Mayo Clinic team showed that excellent long-term outcome (50% to 70% 5-year DFS) can be obtained for irresectable phCCC by adhering to a,well thought and progressively improved,strict plan combining neo-adjuvant (local and external) radiochemotherapy and LT [61-66].This protocol,initiated in 1993,derived from in depth pathologic examination of large liver resection specimen [63,67].Previous abdominal radiotherapy,transperitoneal or trans-gastric biopsy,previous attempted resection,lymph node invasion as well as a tumor mass having a radial diameter ≥3 cm are all formal contraindications for LT.As a consequence of a better screening,intention-to-treat (ITT) and post-LT outcomes are 20% better in case of primary sclerosing cholangitis (PSC) associated phCCC when compared tode novophCCC.ITT and post-LT 5-year OS rates now reach 60% and 77% in phCCC-PSC patients versus 37% and 56% inde novophCCC patients.Presence of residual tumor in the total hepatectomy specimen remains the most important negative prognostic factor reducing 5-year OS to around 45% [63].LDLT gains a more and more important place as it allows a better planning of the procedure.Vascular modifications due to radiation injury are frequent;they are responsible for arterial and (delayed) portal venous complications in 24.3% and 37.8%of recipients.These problems are now overcome by the readily use of free interposition vascular,venous and arterial,grafts [64,65].It has to be anticipated that the more deliberate use of LDLT will overrule in a very near future complex and risky hepatobiliary resections,even in case of resectable phCCC.For ethical reasons it will be impossible to set up randomized controlled trials to compare LR and LT,but earlier experiences have shown that these patients frequently end up,in contrast to total hepatectomy,with a higher risk for non R0-resection [63,67].LT is on the way to become the new standard for the treatment of phCCC [68].

More recently there is also a renewed interest for LT as an alternative treatment of IHCCC [67-72].Here also a neo-adjuvant radio-chemotherapy protocol has been implemented to improve outcome.The initial experiences,especially in early tumor stages,are encouraging [69-71].Smaller series allowed to obtain 5-year OS of 83% with a DFS of 50% for irresectable IHCCC,numbers that exceed the results obtained after LR (5-year OS and DFS of 25%-40% and 30%-50%) [71,73].The introduction of molecular biology and related markers will very probably allow to further refine both,the chemotherapeutic approach and the patient selection and so hopefully to further improve results of LT [72-74].

All therapeutic innovations made in the treatment of HCC and CCC,fostered the interest for LT as part of the armamentarium dealing with secondary liver tumors originating from colorectal(CRLM),neuroendocrine (NETLM) and,more exceptionally,GIST liver secondaries [75-78].Although remarkable results have been reported repeatedly in well-selected patients,LT did not gain yet sufficient interest from the endocrinological,colorectal,surgical and oncologic communities [79,80].Similar to the “transplant history” of primary liver tumors,initial experiences were disappointing again due to the lack of adequate selection procedures and of efficacious neo-adjuvant and adjuvant therapies.Time has now come to switch gears and to take LT into consideration as a valuable alternative to complex regenerative LRs and long standing,costly,medical therapies [6,7,10-13,39,79-83].Even in case of extended LRs,many patients present small missed or disappearing metastases,explaining the high incidence of post-surgical “recurrences” [84,85].

The NETLM Milan criteria,introduced in 2016,showed that LT allows to obtain excellent results when adhering to well-defined selection criteria [86-88].Low grade tumor histology and biology(G1 and G2 gastro-entero-pancreatic tumors),Ki67 level ≤10% (or even ≤20%),a primary tumor located within the portal drainage territory,a tumor mass＜50% of the liver volume,response to medical therapy (using long-acting somatostatin and/or mTOR inhibitors) and disease regression or stability for a period of 6 to 12 months allow to reach excellent results.The transplant group showed a significant advantage over non-transplant treatments at 5 and 10 years in survival (97.2% and 88.8% vs.50.9% and 22.4%)and time to-progression rates (13.1% and 13.1% vs.83.5% and 89%).The transplant-related 10 years survival benefit reached more than 3 years [86].The hesitation to value LT for this indication relates to the fact that NETLM remains for a longer time within the liver,grows slowly and that an important prolongation of life can be obtained using different,modern,systemic therapies [88,89].To be accepted by the medical community LT must therefore generate results which overrule those obtained by these different systemic treatments [86,88].LT is nevertheless,compared to all other treatments,the only one that may offer to well-selected patients a long-term DFS [79,90,91].Multimodal therapies integrating the knowledge of molecular biology also will allow to further advance in this vast field of oncology [88,90,92].

The same reasoning holds in relation to the debate about the place of LT in the therapeutic algorithm of CRLM;here also the combination of chemo-and immunotherapy alone is able to prolong survival up to 3 years and repetitive,regenerative liver surgeries become very effective and allow to further expand overall survival rates [8,11-13,93].Since 10 years,the Oslo group draw again the attention to the place of LT in the treatment of nonresectable CRLM.The 2013 SECA 1 and the 2020 SECA 2 studies,which were made possible due to a surplus in liver allografts in Norway,showed that 5-year OS and DFS rates can be significantly improved by refining selection LT-criteria.Indeed 5-year OS and DFS rates raised from 60% and 33% in the SECA 1 study to 83% and 73% in the SECA 2 study [94,95].The contrast with the 5-year OS of only 9% obtained in a similar CRLM cohort having chemotherapy only is more than striking [96].Of note the fact that,especially recipients presenting a pulmonary recurrence,could be made disease free in both studies with redo-surgery or radiofrequency application.A diameter of the largest lesion＞5.5 cm,a carcinoembryonic antigen (CEA) level＞80 ng/mL,a delay between colorectal surgery and LT＜24 months,progressive disease on chemotherapy,a mean metabolic tumor volume of＜70 cm 3 (at18F-FDGPET-scan),BRAF mutation and LM deriving from right sided tumors were all identified as negative prognostic factors [96].The four former factors were grouped into the OSLO score;low-risk patients (defined as having a 0-2 score) had very good OS [97].Their 5-year OS rate following LT was similar to that of Milanin HCC patients [98].If classified following the Fong clinical risk score (meaning one point for each of the following: surgery＜12 months from diagnosis,lymph node-positive primary,number＞1 lesion,size＞5 cm,and CEA＞200 ng/mL),recipients with a 0-2 score even reached a 100% 5-year OS [93].It is important to underline that the Norwegian group showed that growth rates of pulmonary LM in non-and immunosuppressed patients were similar and that recurring patients tolerated well adjuvant chemotherapy [99,100].LT was shown not only to be a cost-effective treatment but also one permitting to obtain a good quality of life [101,102].LT also offered better results than those obtained after advanced LRs [11-12,80].These excellent results were obtained after applying lessons learned from the initial trials leading to a refinement of the selection criteria [3,80,94,96,97].LT for CRLM remains still a difficult balance between (prolonged) patient survival and recurrence [97,103].Advances in other fields of surgery,e.g.,thoracic surgery and oncology will be important to make further progresses [99,100,103].It becomes clear that well-selected patients will benefit more from LT than from LR and/or heavy systemic treatment modalities [97,104,105]

The more inclusion criteria for LT (and LR) are widened,the higher the risk for recurrence.Tight post-surgical follow-up is therefore utmost important for both primary and secondary liver tumors.Indeed if possible,surgery for early,localized and limited,recurrences is not only beneficial but may even make the recipient disease free again [106-112].It has now also been shown in several experiences that the handling of simultaneous immunosuppression,chemotherapy and immunotherapy becomes not only better but also safer [100,113,114].As an example,better knowledge of the wash-out of immune checkpoint inhibitors will allow to reduce the incidence of (serious) rejection episodes reported in up to 30% of these oncologic patients [115].

As immunosuppressants are potent pro-oncogenic drugs,it is also logical to reduce as much as possible the immunosuppressive load in these transplant indications [116,117].Minimization and even tolerogenic regimen should be opted for as much as possible [118,119].

In conclusion,the therapeutic plan of every patient presenting with a primary hepatobiliary or secondary liver tumor must be discussed in a “real” multidisciplinary meeting taking into consideration all medical and surgical therapeutical options including LT.Still too much patients presenting a primary hepatobiliary and secondary hepatic tumors,localized in both a diseased or (also!) normal liver parenchyma,are resected or submitted to interventional radiologic procedures instead of taking into consideration the transplant option.LT offers indeed in many,wellselected patients,the best chance to obtain a long-term DFS.Indeed the medical community should give much more attention to long-term DFS than to OS,the former being the most important (psychologic) fact for the patient.The fundamental question for every patient “Doctor,can I be cured from my liver cancer?”must hereby more than ever be taken into consideration.Treatment options must furthermore be individualized by integrating all available and reliable tumor and molecular markers (to better understand the tumor) as well as artificial intelligence (to better sort out which are the best candidates for liver surgery in form of,both,a partial or total hepatectomy) [71,74,87,120-122].When considering LT,one has to incorporate theITT transplant benefitlooking thereby at the long-term outcome of the patient from the moment the registration on the waiting list took place [123-125].When looking at survival benefit with and without transplantation,one should,especially when dealing with cancer patients,extend the analyses beyond the 5-year time horizon as it has been repeatedly shown that the longer (＞5 years) the follow-up,the higher the benefit of LT when compared to all other alternative treatments [51,55-58,79-80,96].Moreover benefit analyses must also incorporate the heavy medical,psychologic,quality of life and economic burdens caused by the very high number of recurrences occurring after locoregional (including LR) and systemic cancer treatments [101,102].It is therefore not valid (anymore) to simply argue that patients without liver cancer (having a higher MELD score) disserve more access to transplantation than patients with liver malignancies because having more (5 years) survival benefit from LT [126].Cancer patients are equally sick compared to liver diseased ones.Leaving a “small place” (up to 5%of liver allograft resources,such as proposed by the Italian college of liver transplant surgeons and hepatologists) on the waiting list to be used for innovative transplant indicationsis a very sound idea because allowing room to explore further the value of LT for,not yet uniformly accepted,oncologic indications such as discussed in this special issue ofHepatobiliary&Pancreatic Diseases

International[127].This is important to judge if the LT is indeed the better or best alternative treatment for these cancer patients.From now onwards,LR and LT (eventually completed with different types of locoregional interventions) have to be seen,more than ever before,not anymore as competitors but as allies.Integrating all alternative treatment modalities in an ITT transplant,oncologic therapeutic algorithm will offer the best chances for these cancer patients on the condition that adherence to strict selection criteria,based on progressing,evidence-based,knowledge,is respected [16,19,46,63,71,74,86,97,128,129].Cross-fertilization between advanced LR and LDLT procedures (and their derivatives such as auto-transplantation and split LT) and between medical oncologic experiences in non-and in immunosuppressed patients,will further advance this fields of liver and transplant oncology[12,13,42,44,82,83].By applying all this knowledge in a context of real multi-modal,multi-disciplinarity (meaning involvement of all respective experts in a given team or center),the access to a potentially,curative oncologic treatment will undoubtedly increase substantially for a large number of liver cancer patients.

Acknowledgments

None.

CRediT authorship contribution statement

Jan Lerut:Conceptualization,Writing -original draft,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.