Hepatocellular Carcinoma and Liver Transplantation

Abstract

Hepatocellular carcinoma (HCC) is the sixth most common cancer worldwide. It has the third most common cancer-related death. Here, there are different treatment options for HCC that develops on cirrhosis background. These include liver resection, liver transplantation (LT), locoregional therapy (LRT), and systemic therapy. LT is an effective treatment choice for eligible patients who provide excellent posttransplant outcomes with a low risk of HCC recurrence, especially when strict patient selection criteria and appropriate posttransplant management are applied. Different selection criteria have been proposed for LT candidates in recent years. The use of these criteria also allows for recurrence rates that can be achieved at acceptable rates. These models continue to evolve and incorporate features such as tumor biology in addition to the response to LRT as efforts to identify patient populations that may benefit more from LT by expanding access to it. Milan criteria were considered the gold standard for LT. Post-LT HCC recurrence is among the leading causes of death in patients transplanted for this indication. Posttransplant HCC surveillance is important in this regard. Early diagnosis and aggressive treatment have been proven to improve survival outcomes.

1. Introduction

The regenerative capacity of the liver was first described in Greek mythology in the story of Prometheus. A thorough understanding of the segmental anatomy of the liver, on the other hand, was first introduced by Couinaud [1] and then by Bismuth [2]. Although the mechanism of liver regeneration is not fully understood, it has played an important role in the development of liver surgery and living donor liver transplantation (LDLT) as a result of these processes. Following the Hannover team’s first demonstration of the feasibility of the innovative split-liver transplantation (SLT) in 1988, reduced-size liver transplantation and SLT were performed together in a child as LDLT in Australia by Strong using segments 2 and 3 from the mother’s liver [3, 4]. OLT is the recognized optimal treatment for patients with the low-volume unresectable disease.

The first successful adult recipient LDLT was performed in Japan as left lobe and then right lobe transplantation [5, 6]. LDLT outcomes have been improving more in recent years. The concomitant organ shortage, along with concerns related to faith and ethical issues, the decrease in deceased organ donation, and increased demand for liver transplantation, has caused the need for LDLT to increase even more. There has been a continuous rise in the incidence of liver cancer and liver cancer-associated mortality and in recent decades. In 2020, it was the sixth most common cancer worldwide with a total of 905,677 new cases worldwide. It had the third most common cancer-related death with 830,180 deaths [7]. As many as 20–40% of worldwide liver transplants are for the treatment of HCC [8].

Hepatocellular carcinoma (HCC) patients are widely distributed, especially in East Asia and Africa. However, incidence and mortality have recently increased considerably in North America and Europe [9, 10]. The epidemiology of HCC has changed markedly in recent decades. While the relationship with viral hepatitis is decreasing in certain parts of the world, the relationship with alcohol and nonalcoholic fatty liver disease (NAFLD) is increasing, especially in the Western World. The median age of diagnosis in the United States is usually the sixth decade of life (60–65 in men and 65–69 in women).

HCC prevalence varies by gender and is more prevalent in men. Likewise, HCC incidence and mortality vary significantly with race and ethnicity. A higher incidence is observed in the African American and Hispanic populations [11]. Approximately 90% of HCC cases occur in the background of cirrhosis. The most common causes of cirrhosis leading to HCC in the United States are alcohol-related liver disease (ALD), NAFLD or nonalcoholic steatohepatitis (NASH), viral hepatitis B [HBV], and hepatitis C [HCV]. However, in the absence of cirrhosis, the most common cause of HCC is HBV although there has been a recent increase in HCC due to non-cirrhotic NAFLD [11]. Liver function is defined as compensated or decompensated. Decompensated liver disease is defined as jaundice, ascites, or hepatic encephalopathy regardless of Model for End-Stage Liver Disease (MELD). Stage migration describes the patient concept for whom it is necessary to make reclassification as a more advanced stage due to the changing factors even if TNM classification, the degree of liver dysfunction, and PS remain unchanged. Detection of stage migration includes progression pattern, growing or newly emerging extra or intrahepatic lesions, and newly detected vascular invasion. Liver transplantation (LT) is the treatment of choice in patients with clinically significant portal hypertension (CSPH) or significant liver dysfunction and cirrhosis if there are no medical or psychosocial contraindications.

LT is a treatment that is intended as the last option in HCC patients who still cannot be treated surgically and in impaired liver function or advanced HCC. Recently, 1100 patients per year in Europe have undergone LT due to HCC. And the annual number of new HCC cases is reported to be as high as 65,000 [12]. The introduction of direct acting antiviral therapies (DAAs) for the treatment of HCV has provided some promise in the areas of transplantation and infectious disease [13]. As a result, the number of patients listed for LT due to HCV continues to decrease. The LT perspective for HCC is parallel to the change in patients transplanted for primary liver disease. Between 2014 and 2019, the percentage of patients listed for LT due to HCV-associated HCC decreased by 22%, while the percentage of patients listed for LT due to NASH-associated HCC increased by 14.5%. NASH was the most common etiology of HCC in women listed for LT in 2019. However, although it tends to decrease in men, the most common indication for LT listing is HCV-associated HCC [14]. There is a distinct racial difference here. Compared to Caucasians and Hispanics, Black patients have high HCV associated with HCC load, which constitutes a potential obstacle for HCV treatment. On the other hand, hepatitis B virus (HBV) is the most common etiology in HCC-associated LT list among Asian patients.

After the first successful LT in 1967, the National Organ Transplant Movement established the beginning of transplantation in 1984 and organized United Network for Organ Sharing (UNOS) with a registry system to track transplant results and data [15]. LT UNOS distribution has recently taken into account multiple factors including distance of recent donor to hospital, blood type, waiting time, and severity of liver disease.

Distribution policies help reduce bias and improve post-LT outcomes. To begin with, the waiting list implemented the Child-Pugh classification system, which better emphasized the disease severity.

In 2002, the UNOS Model for End Stage Liver Disease (MELD) was launched. In 2016, the MELD-Na scoring system was introduced, which took serum sodium level into consideration. Finally, in 2023, the MELD model was corrected again by including gender and albumin level [16]. Recently, UNOS T1 (a tumor up to 2 cm) or UNOS T2 (a tumor up to 5 cm or three tumors up to 3 cm) were identified as eligible patients with non-MELD scores after a mandatory waiting period of 6 months. Here, UNOS could select the population with much better post-LT prospects among the patients with tumors that respond to treatment. To do this, UNOS is now combining alpha-fetoprotein (AFP) changes with DS data to provide exception scores to the most suitable candidates. Here, there are different treatment options for HCC that develops on cirrhosis background. These include liver resection, liver transplantation (LT), locoregional therapy (LRT), and systemic therapy. The choice of treatment or combination of therapies depends on the extent of the tumor, extrahepatic spread, macrovascular invasion, preserved liver function, and performance status (PS) [17]. LT is the only feasible curative approach in HCC patients with considerably impaired liver function. A prerequisite for the application of LT as a treatment is that the neoplasm is localized especially in the liver.

The Barcelona Clinic Liver Cancer (BCLC) staging system is recommended by American Association for the Study of Liver Diseases (AASLD) and the European Association for the Study of the Liver (EASL). This was updated in 2022 to better define the prognosis and most appropriate treatment option [17]. The BCLC staging system, which is used to determine the target population in clinical research studies, also regulates the inclusion/exclusion criteria to determine the patient profile. Clinical practice guidelines and algorithms, such as the BCLC model, indicate the most up-to-date state of scientific evidence and knowledge appropriate for each intervention. Major hepatectomy carries extreme risk in patients with cirrhosis. Specific tumor localizations can also sometimes be an obstacle to resection [18]. Large tumors are often associated with cancer-related symptoms (pain, etc.), which indicates poor outcome after resection. Upon listing for LT, if the expected waiting time exceeds 6 months, a number of treatment options are considered to prevent the tumor from progressing and being excluded from LT. Ablation, transarterial chemoembolization (TACE), and transarterial radioembolization (TARE) are the most commonly used methods for this purpose. In patients with peripheral tumor localization who were initially selected for ablation, this approach may be contraindicated due to the risk of tract seeding or adjacent organ damage (if puncture was performed without a non-tumor liver rim) [19]. The BCLC 2022 staging system version does not recommend resection for multinodular HCC within the Milan criteria. Although resection cohort studies report promising survival outcomes, prospective data are necessary to determine its effectiveness [20]. TACE should be preferred when the first-line treatment approaches for HCC (ablation or LT) are not appropriate. However, it is known that tumors exceeding 8–10 cm are associated with poor prognosis after TACE [21, 22]. This is potentially related to disruption of portal venous flow due to compression or invasion rather than the effect of major tumor necrosis. Moreover, large tumors are rarely asymptomatic. Patients with symptoms (performance status 1) have lower survival after TACE compared to asymptomatic patients [23]. In the 2022 BCLC version, BCLC stage B was divided into three groups according to tumor burden and liver function. This stage is determined as multifocal HCC that exceeds BCLC-A criteria. Here, liver function is preserved, and there are no cancer-related symptoms (PS-0) and no vascular invasion or extrahepatic spread. At this stage, the size of the tumor burden may be heterogeneous, and the prognosis is also affected by AFP levels [24]. For BCLC-0 and A, ablation should be the preferred approach if the patients are not candidates for LT. But there is not a strong cutoff value for this. An individualized patient profile can determine whether the patient would benefit from LT, systemic therapy, or TACE [22, 25]. BCLC-B stage group 1 covers well-defined HCC nodules. These patients are candidates for LT if they meet the Extended Liver Transplant criteria according to the center criteria [26, 27, 28, 29, 30, 31, 32]. A slight increase in tumor size and number provides acceptable survival outcomes. However, this may lead to a slight increase in recurrence and a decrease in long-term survival [33]. Competitive survival can be achieved with patients within Milan criteria, taking into account the reduced survival and increased risk of recurrence in patients who are allowed to progress to asymptomatic limited progression exceeding the Milan criteria [34]. Accordingly, the decision to adopt the expanded criteria is determined by the factors such as the prevalence of access to LT for other indications, and this should be comparable to the minimal results achieved for other indications.

A cutoff value of 1000 ng/dl has recently been applied as an exclusion criterion. Even if downstaging treatment induces a decrease in AFP level, there is no strong data on the reduction required or the time period to wait before LT planning. Non-MC patients with low AFP levels have better post-LT survival than those within MC but with high AFP values. AFP level intervals and UNOS exception scores are now included in the post-LT prognostic model [35]. However, patients in BCLC-B group 2 include those with preserved portal flow and defined tumor burden, and patients whose feasibility of selective access to the tumor-feeding artery was evaluated, besides the LT option. These patients are candidates for TACE. If these patients are not within the scope of either TACE or extended liver transplant criteria, systemic treatment should be recommended to ensure optimal results [36, 37]. BCLC-B group 3 patients include diffuse, extended, and infiltrative HCC covering the liver. They do not benefit from LT and TACE. Systemic treatment options should be recommended [38]. BCLC-C stage patients include the ones who are relatively fit with extrahepatic spread or vascular invasion. These patients have PS ≤ 2 and preserved liver function. They should be evaluated for systemic therapy [39]. BCLC stage D includes patients with impaired liver function and/or cancer-related symptoms (PS > 2) who do not have an LT option due to HCC burden or non-HCC factors. However, LT may be an option for patients whose tumor burden does not exceed the specified criteria.

1.1 Selection criteria for LT

Pre-LT imaging techniques identify macroscopic metastases but do not guarantee the exclusion of microscopic ones. In order to reduce the risk of unidentified micro-metastasis to prevent post-LT recurrence, different selection criteria have been proposed for LT candidates in recent years. The use of these criteria also allows for recurrence rates that can be achieved at acceptable rates. The molecular profile still cannot predict the risk of recurrence, the best treatment option, and patient outcomes after successful ablation or surgery [40].

LT is an effective treatment choice for eligible patients who provide excellent posttransplant outcomes with a low risk of HCC recurrence, especially when strict patient selection criteria and appropriate posttransplant management are applied. Although cirrhotic patients who underwent liver transplantation due to HCC have the longest survival, high recurrence and short survival are often observed after early intervention. Therefore, the threshold value of the tumor burden must be well-defined [41]. Due to the strong effect of HCC recurrence on posttransplant prognosis, many criteria and scoring systems have been established for the selection of transplant candidates.

More than two decades ago, in 1996, Mazzaferro published the results referred to as the Milan criteria. They were considered the gold standard for LT. They have become the cornerstone of selection policies in HCC patients waiting for LT globally. Of course, simple and easy-to-use criteria like these have had a profound impact on survival, dramatically increasing cure rates in HCC patients receiving LT worldwide. Accordingly, there should be ≤5 cm single tumor or a maximum of three ≤3 cm tumors, and there should be no macrovascular invasion or extrahepatic spread. Here, after LT, a 4-year overall survival rate of 75% and a recurrence-free survival rate of 83% were achieved [26]. However, despite its usefulness and reproducibility advantage, it has major limitations. More importantly, there is a lack of biomarkers of the tumor that would affect the best oncological outcomes after transplantation. By including biological tumor indices, selection models go beyond the simple tumor burden.

These models continue to evolve and incorporate features such as tumor biology in addition to the response to LRT as efforts to identify patient populations that may benefit more from LT by expanding access to it [42]. The use of downstaging for this purpose has led to LT expansion efforts for HCC that exceeds the Milan criteria. Although UNOS-DS (United Network for Organ Sharing-Downstaging) has expanded LT indications for HCC patients beyond traditional MC, the outcomes of patients with tumor burden beyond UNOS-DS are still a matter of interest. Here, the upper limit of the tumor is yet to be determined [43]. Many groups recommend transplantation with results comparable to Milan for patients with larger numbers and larger tumors. The vast majority of them use radiological morphometric criteria. They represent the explant pathology that predicts the outcomes. Because of the close relationship between morphometrics and outcomes, they have become the standard treatment in HCC patients who exceed the Milan criteria. The combination of dynamic markers for wait time and response to LRT and important tumor markers allows for the successful selection of broad criteria without risking expected outcomes. After the adoption of the Milan criteria around the world, some groups emphasized the extreme restrictiveness of these criteria. The wide benefits of the original morphometric criteria also indicated excellent posttransplant outcomes. In the United States, Yao et al. proposed the University of Southern California San Francisco (UCSF) criteria in 2007 showing that expansion of the upper limit of tumor size and addition of total tumor diameter did not adversely affect outcomes, with an 80% 5-year survival for those within UCSF (single tumor ≤6.5 cm, up to three tumors ≤4.5 cm, total tumor diameter ≤ 8 cm) and poor outcomes in those outside the UCSF criteria. These criteria led to a favorable 5-year patient survival rate of 75.2% and a post-LT recurrence rate of 11.4% [44]. Here, the 4-year survival was 70% for patients in Milan criteria and 64% for patients in UCSF but outside Milan. Survival was poor in transplant recipients who did not meet both criteria. For them, the 5-year survival was only 41%. Later, in 2009, Mazzaferro et al. proposed their “Up-To-Seven Criteria” based on patients who exceeded the Milan criteria in explant pathology with a multi-inductive collaboration. Subjects are considered for OLT (orthotopic liver transplantation). These criteria include non-Milan patients without microvascular invasion (MVI). The number of nodules should not be more than seven, and the diameter of the largest nodule does not exceed 7 cm. Five-year survival was similar to patients within the Milan criteria (73.3% within Milan, 71.2% Outside Milan within Up-To-Seven), and this was referred to as the Metroticket model [45]. A morphometric selection model was also adopted by some Canadian centers. This was introduced by Toso et al. as total tumor volume (TTV). Here, HCC patient selection via TTV should not exceed 115 cm³. Compared to Milan and UCSF, it further increases total inclusivity (28–53% patient increase compared to Milan criteria, and 16–26% patient increase compared to UCSF criteria). It had similar patient outcomes as Milan (beyond Milan within TTV vs. within Milan – 5-year overall survival (OS) 74 vs. 79%, 5-year recurrence-free survival (RFS) 78 vs. 80%). Broad morphometric criteria other than the Milan criteria always include more patients who would have LT and ensure survival with acceptable risk of recurrence. However, unless combined with tumor biology markers, tumor number and size alone cannot accurately predict tumor biology. Deficiencies in tumor biology limit the ability to perform risk stratification due to the gap between pathology and radiology. This limits the use of morphometric criteria alone [46, 47]. These criteria are based on the morphological characteristics of HCC. Liver imaging techniques such as contrast-enhanced computed tomography or magnetic resonance imaging are configured according to the pre-LT precision. However, they are affected by suboptimal sensitivity and heterogeneity [48]. The efficacy of these criteria in predicting acceptable disease recurrence rate and post-LT survival is based on the principle of tumor burden. The size and number of the nodules are considered to be a marker representing the presence of microvascular invasion and/or poor differentiation, which is an independent predictor of HCC recurrence. Macroscopic vascular invasion is considered to be an unacceptable risk factor for post LT. In recent years, it has been understood how microvascular invasion is associated with a high recurrence rate [49]. The diagnosis of HCC in cirrhotic individuals is often based on radiological criteria in accordance with the latest guidelines. Precise information about microvascular invasion before transplantation cannot be provided by these methods. Liver biopsy is stored in cases with nontypical imaging [37, 50]. However, although not performed routinely, histological examination provides a direct perspective on tumor grade. The expanded Toronto criteria suggested the idea of considering preoperative histology. Accordingly, LT was performed in the group of patients with no poorly differentiated neoplasm in preoperative histology from the largest nodule, no extrahepatic spread including macrovascular invasion, and no cancer-related symptoms with tumor nodules of any size and number. These expanded Toronto criteria were later confirmed by a prospective ITT study. In the Toronto study, high levels of AFP at the time of transplantation and listing were associated with poor post-LT outcomes in patients transplanted according to both within Milan and extended Toronto criteria. AFP levels and slopes appear to be appropriate prognostic indicators, and a cutoff level of 500 ng/ml was suggested as a poor prognostic factor. Sapisochin et al. showed an excellent 5-year OS outcome of 68% in patients who had transplants using AFP and tumor differentiation of any size and number beyond group criteria [51]. A number of tumor biomarkers have been proposed and validated to improve prediction of survival and recurrence outcomes. These are serum tumor and inflammatory markers [46, 47, 52, 53].

In 2018, the model known as Metroticket 2.0 was proposed. These morphological characteristics and AFP as a representative marker of microvascular invasion were created to predict HCC-specific death [30]. Dynamic changes in AFP levels are also noteworthy. In 2009, in France, it was determined that a 15 μg/l monthly increase in AFP was a poor prognostic factor in transplant recipients, and then a 27% difference in 5-year RFS was demonstrated with an upward progression of AFP values every month before LT. Survival improved in patients with lower AFP progression. However, there was no specific regulation for non-Milan patients. Nevertheless, this factor was not significant in multivariate analysis, which showed that progression of AFP was more important than morphometrics [54]. This study was subsequently confirmed by intention-to-treat (ITT). The AFP curve was crucial in predicting post-LT and waiting list in patients outside and within criteria. The 5-year ITT OS and RFS for patients with AFP < and > 15 ng/ml/month were 66.0 versus 36.7% and 92.3 versus 53.8%, respectively [55]. The ideal AFP cutoff value for AFP is <10 ng/ml, and this is an independent predictor of post-LT HCC recurrence. This rate is 50% for AFP > 10 ng/ml compared to 11% recurrence rate for AFP < 10 ng/ml.

Explant findings that considerably increase the risk of HCC recurrence include the presence of viable HCC, number of viable tumors, satellite nodules, and exceeding the Milan criteria [56]. Various AFP levels are also recommended as cutoffs for patient selection. It was stated that in Milan transplantation, AFP value >400 IU/ml was an independent predictor for survival, regardless of tumor size and number. But here, non-Milan patients were 36% [57]. It was also proven that there is a high risk of dropout from the waiting list after LRT, with the highest dropout probability at high AFP values ranging from 20 to 500 [58]. It is known that extremely high AFP levels (> 1000 ng/dl) are associated with poor outcomes in both patients undergoing LT and in listed patients. In the UCSF cohort, patients with AFP > 1000 ng/dl were shown to be strongly associated with MVI. It was emphasized that AFP > 1000 ng/dl was a strong predictor for OS and RFS. It is also known that AFP > 1000 is consistently associated with poor post-LT OS and DFS, as well as poor explant pathological features [59, 60]. Combining AFP with morphometric tumor markers is also important in selected eligible patients who exceed the criteria for LT. Toso used the combination of AFP and TTV to predict post-LT outcomes. It was later proven that patients within the TTV/AFP criteria had the same overall survival and disease-free survival rates as those within the Milan criteria [61, 62]. At the increase levels of AFP > 100 ng/dl, there is a progressive deterioration in OS and RFS [29]. After this, the AFP model combining different AFP levels and morphometric criteria was introduced. This model independently predicted well the risk of tumor recurrence and vascular invasion. The model combined tumor size (3 cm, 3–6 cm, and > 6 cm), tumor number, and AFP and had excellent ability to predict outcomes in non-Milan patients. While the model showed acceptable results in non-MC patients with AFP ≤100 ng/mL, it had a better ability to predict HCC recurrence than MC alone. In non-Milan patients with low AFP scores, the 5-year recurrence rate was 7.7% in the training cohort and 14.4% in the validation cohort. On the contrary, in non-Milan patients with high AFP scores, the rates were 46.3% and 47.55, respectively [29]. Mazzaferro up-to-7 for AFP < 200, up-to-5 for AFP 200–400, and up-to-4 for AFP between 400 and 1000 indicated that the perfect result was achieved at the lowest tumor size. This was later externally validated and accepted for the liver delivery system combining the AFP model and Metroticket 2.0 [30]. Only static AFP levels were used here. A new scoring system was subsequently developed that used dynamic AFP to predict RFS. The difference between maximum and final AFP was found to predict post-LT outcomes. It was proven that LRT in patients with max AFP > 1000 ng/dl has similar results as those with max AFP < 1000 and 50% response, as in those with max AFP between 200 and 1000 and those with latest AFP < 200 ng/mL. It was proven that LRT in patients with max AFP > 1000 ng/dl has similar results to those with max AFP < 1000 and 50% response as those with max AFP between 200 and 1000 and those with latest AFP < 200 ng/mL. Pretransplant rapid AFP increase was reported to be a poor prognostic indicator in HCC patients undergoing LT [63].

The combination of radiologic tumor burden with AFP is important in improving the criterion decision in patients other than traditional MC. The New York/California (NYCA) scoring system, which combines AFP response with tumor size and number, demonstrated superiority to predict HCC recurrence compared to the Milan and the French AFP scoring systems. This model validated the inclusion of AFP response within the selection criteria, allowing to expand the non-Milan criteria to offer LT to patients otherwise denied LT. Importantly, this scoring system correlated the degree of AFP response with tumor differentiation and vascular invasion. This system also showed that over 85% of patients in Milan could be recategorized into low or acceptable NYCA with a 5-year RFS > 70%.

The most commonly used tumor marker is alpha-fetoprotein (AFP). It was first noticed as a diagnostic and confirmatory marker in the late 1960s and early 1970s. Disagreement over appropriate AFP increase levels resulted in a tumultuous history that could have brought an end to the use of AFP in 2001 [64]. However, the evidence regarding the benefit of AFP has recently supported its revival [65]. As a result, AFP has been shown to be useful as a prognostic marker.

AFP is strongly associated with reduced survival and pathological characteristics in patients with HCC. Serum AFP levels are also strongly associated with pathological characteristics of the tumor, such as vascular invasion and poor differentiation, and with OS and RFS [60, 66, 67]. Especially when used in conjunction with morphological criteria, the AFP level is a useful tool in identifying patients with high risk of recurrence before LT [68]. There are other biological tumor markers that are useful in non-Milan HCC patient selection. Vitamin K absence-induced protein (PIVKA-II) is one of them. It is also known as des-gamma-carboxy prothrombin. In the absence of vitamin K, it is released into the blood as abnormal prothrombin. It was identified as an HCC biomarker by Liebman et al. in 1984. It was found that patients with biopsy-confirmed HCC had a 91% increase in liver disease compared to those with metastatic cancer or chronic active hepatitis [69]. High PIVKA-II level predicts poorly differentiated tumor and MVI. PIVKA II is widely used in Asia to predict outcomes after LT, especially in the living donor liver transplant arrangement (LDLT). A multicenter Japanese study showed a 5-year RFS of 83.5% after LDLT in non-Milan patients with PIVKA II < 100 and AFP < 200, while a multicenter Korean study combined AFP with PIVKA II. It showed good results in non-Milan patients, especially in the advanced HCC subgroup and at low AFP/PIVKA II level. It was shown that the 5-year RFS in advanced HCC patients with AFP + PIVKA II < 300 was 53.4% compared to 10.1% in patients with AFP + PIVKA II > 300 [70].

Another marker is neutrophil-lymphocyte ratio (NLR). Post-LT recurrence is associated with decreased survival in HCC patients with NLR ≥5. A high recurrence rate of 62% was observed in patients with high NLR, whereas a recurrence rate of 13% was found in patients with normal NLR. Patients within Milan NLR ≥5 had a worse 5-year RFS rate than within Milan patients with normal NLR (30 vs. 81%, respectively). A much better survival rate was found in non-Milan patients with low NLR than in-Milan patients with high NLR (61 vs. 30%, respectively) [71, 72, 73, 74].

1.2 Bridge therapies and downstaging for liver transplantation: ablation, TACE, and TARE

In patients with HCC, LDLT offers an attractive alternative to patients awaiting suitable deceased donor liver transplant (DDLT). For advanced HCC beyond MC, transplantation is more appropriate for LDLT as opposed to DDLT. The scarcity of organ distribution required for those with optimal expected survival in DDLT is a major problem.

Downstaging therapy is defined as tumor downstaging after ablation/resection or treatment strategies that reduce tumor burden for risk reduction and radical surgery including LT. Thus, the patient receives greater survival benefits from these definitive treatments. A number of studies reported similar survival rates between patients who were within the Milan criteria and those who exceeded the Milan criteria at diagnosis but were successfully downstaged after LRT. Downstaging protocols require an observational period with continued disease stability. Although the tumor upper limit suitable for downstaging is still controversial, the Milan criteria are frequently used as the final decision parameter in downstaging protocols [12]. It is unclear whether LRT is necessary for HCC and cirrhotic patients who are already within the Milan criteria, as a bridge to LT to prevent tumor progression, even though they are on the waiting list.

There are different types of LRT performed for HCC: percutaneous techniques, transarterial techniques, and stereotactic body radiation therapy (SBRT). Percutaneous techniques are generally applied to up to three tumors, each measuring up to 3 cm. Here, preserved liver functions (Child-Pugh A) and the absence of untreated coagulopathy are required. These techniques include radiofrequency ablation (RFA), microwave ablation (MWA), and percutaneous ethanol injection (PEI).

Although the most commonly used technique is RFA, PEI is preferred when the lesion is very close to the main bile duct and hollow organs such as the stomach or intestine. On the other hand, if the lesion is very close to major vessels such as the inferior Vena Cava and the main hepatic vein, there is a heat-sink effect. In this case, MWA is also applied with less sensitivity to the heat-sink effect even in slightly larger lesions.

In recent years, the use of PEI has been decreasing considerably due to the greater effectiveness of RFA [75, 76]. Cirrhotic HCC patients who are initially outside the listing criteria benefit from downstaging treatment. Approximately 40–60% of cirrhotic HCC patients who are outside the Milan criteria at diagnosis eventually meet the Milan criteria with their disease downstaged after receiving LRT [77, 78, 79]. Although there is no consensus in the literature as to whether post-LT survival rates in patients who have undergone successful downstaging are similar to those in patients who are within the Milan criteria at diagnosis, results depend on the extent of disease before downstaging treatment.

Especially if the patients who are outside the Milan criteria but within the University of California at San Francisco (UCSF-DS) (1 lesion >5 cm and ≤ 8 cm, or 2–3 lesions with at least 1 > 3 cm and ≤ 5 cm, or 4 to patients with 5 lesions, none of them is >3 cm and total tumor diameter ≤ 8 cm) during diagnosis undergo successful downstaging, they have post-LT survival similar to those within the Milan criteria at diagnosis. However, even if successful DS is performed in cases beyond the UCSF-DS criteria, they have a lower survival rate than patients within the Milan criteria [79, 80]. Moreover, LRT is also used as bridging in cirrhotic HCC patients who are on the waiting list within the Milan criteria to prevent dropping off the list or disease progression. Non-comparative studies report the dropout rate in patients without bridging treatment as 25–40% and the rate in patients without bridging treatment as 10–20%. Here, the time spent on the waiting list was determined to be 6–12 months. This is not clear for transplant waiting times of less than 6 months [41].

The most commonly used method for LRT is TACE, followed by RFA and MWA. While radiological complete response is observed in 30% of cases, only 20% of cases have pathological complete response.

LRT bridging treatment in cirrhotic HCC patients within the Milan criteria reduces neither the risk of disease recurrence after LT nor the risk of dropout from the waiting list. While a lower risk of death is observed in the first year after LT, this advantage decreases later on. Here, within the Milan criteria, there is no difference between bridging or non-bridging for 3- or 5-year overall survival after LT [81]. It is also necessary to pay attention to the management of HCC patients during the waiting period before LT. Bridging and monitoring to ensure patients remain within transplant criteria is based on HCC and underlying liver disease and the patient’s fitness status. On the other hand, similar treatment is recommended for the purpose of downstaging in the subgroup of patients who are outside the transplant criteria. Indications and treatment procedures vary in centers, depending on their experience.

Frequently, TARE treatment is described to include the procedures described. In the TARE technique, radioisotope yttrium-90 absorbed into microspheres is administered percutaneously or selectively through the hepatic artery feeding the targeted tumor. Y-90 is a substance that emits short-lived beta rays and has limited tissue penetration. Microspheres in glass or resin have different sizes and numbers of beads and different specific bead activities. Before treatment, tumor mapping is performed with cone-beam CT-guided angiography and tumor stimulation is carried out with 150 MBq technetium-99 m macro-aggregated albumin. After treatment, SPECT/CT or PET/CT is performed to evaluate microsphere distribution and target mean dose in the liver and lung.

TARE methods can involve the entire liver by distributing the microspheres from the appropriate hepatic artery to the entire liver, or they can be performed bilobarly with a catheter placed in both lobes simultaneously, or segmentally or sub-segmentally in a targeted lobe or less. Apart from these, two-phase TARE is also applied to the liver lobe responsible for the disease in which partial liver atrophy and growth of the healthy liver lobe are achieved with segmental TARE. In a study involving 18 patients with ipsilateral PVTT, an 83% overall response rate was observed radiologically after TARE-DS, while 72% of the patients showed response to both tumor and thrombus (complete or partial). Partial or complete opening was observed in PVTT in 11 patients [82]. In a series of 349 patients in Mount Sinai Hospital, transplantation criteria were UNOS T2. Pathological and radiological complete response rates were 34 and 56%, respectively. Of these, 22 patients underwent LT [83]. In the study of the University of Bologna, two patients out of 63 consecutive patients experienced downstaging after TARE and underwent LT. Complete and partial response was 73%. TARE and TACE were compared in a multicenter large study. In the TARE group, progression was observed in 12.9% of patients, whereas LT was performed in 22.6% of patients. While 23.1% of them were non-Milan, complete tumor necrosis was observed in 30.8% of the patients [84]. In a randomized controlled study, TARE and TACE were compared in nonsurgical BCLC stage A and B HCC patients. Nine of the 33 patients who underwent TARE had DS, and LT was performed [85]. There are three studies on LT after bridging with TARE. Two of them reported patient outcomes after LT. One study provided details of dropout from the waiting list. In two studies, median survival after LT ranged from 46 to 69 months. The 3-year survival was 92.9%. TARE experience is limited to a few centers. The whole liver TARE procedure was rarely performed. For this purpose, repeated attempts can be made safely. Despite the limited number of reports, tumor control and response rates are high, and oncological long-term survival outcomes after LT are available. The safety profile is better and acceptable compared to systemic and other local treatments. Procedure-related complications did not lead to dropout from the waiting list.

Bridging procedures are neoadjuvant therapeutic options used to prevent patients dropping out of the waiting list and disease progression during the waiting period. Downstaging, if there is disease spread outside the Milan criteria, already shows similar LT results with tumors within the Milan criteria with this option. Classically, TACE is the treatment of choice for downstaging and bridging in HCC. However, although there is a growing trend supporting TARE, only unresectable or intermediate stage HCC is valid due to the number and size of this tumor [86, 87].

Bridge treatments are adjusted depending on the relationship of the tumor with the biliary and vascular structures, the location of the tumor, and the specific tumor number. Regarding the waiting time, the period from HCC diagnosis to LT has recently been elucidated. Moving quickly to transplant may result in transplantation in patients with more aggressive tumors. There is a risk here that the biological characteristics of tumors that are more likely to recur are not adequately tested. However, prolonged waiting time is associated with dropout from the LT waiting list, and this was reported to be 18–20% per year.

In patients undergoing LT, short waiting time (median 1.6 months) was an independent predictor of poor OS. A very short waiting period (<6 months) or a very long waiting time (>18 months) was associated with an increased risk of HCC recurrence [88, 89]. But there is a time lapse as to when patients are diagnosed, referred, and then listed at a transplant center.

Data show that, regardless of the choice of LRT, complete pathological response (cPR) and degree of tumor necrosis on explant examination are associated with improved survival and reduced recurrence [90]. The likelihood of achieving cPR is less in patients who have a short waiting time and receive more than three LRTs. The first criterion to evaluate response to treatment was the response evaluation criteria in solid tumors (RECIST) where a complete response was the disappearance of all target lesions [91]. The RECIST criteria were then “modified” to mRECIST where a complete response is the loss of any intra-tumoral arterial enhancement [92].

Milan criteria (MC) are used successfully in selecting suitable candidates for orthotopic liver transplantation (OLT). Downstaging (DS) to MC concept was first published by the University of California, San Francisco (UCSF) in 2008. Additional data have emerged supporting DS as a current approach to access OLT with excellent post-LT outcomes [80]. Traditionally, the prior DS was a prediction of gain versus biological activity of tumors exceeding MC based on response to LRT. A key component of the UCSF-DS protocol is an upper limit to tumor burden.

Management of patient with HCC on the waiting list aims to avoid disease progression. The main reason for this is to reduce the risk of developing new HCC in the remnant liver and to simultaneously treat the underlying liver cirrhosis and HCC, making LT the best treatment. However, one of the concerns in HCC is the prolonged time period between diagnosis and LT, as well as overcoming the inconsistency in the selection of LT candidates for patients requiring LT and the appropriate liver graft to identify possible patients who may benefit from this procedure, as well as LT for reasons other than HCC. There is a lack of randomized controlled trials (RCTs) evaluating neoadjuvant treatment recommendations to reduce the risk of dropout due to tumor progression. If there is no appropriate LT, the criteria will be exceeded as a result of HCC progression and the chance of treatment will be lost. If organ waiting times exceeding 6 months are anticipated, locoregional treatments (LRTs) are recommended for HCC patients for LT. In 2012, the international consensus conference UNOS reported that bridge therapy would not be beneficial for patients with T1 tumors. However, UNOS T2 lesions may benefit if the expected waiting time is 6 months or more [93]. However, most patients who receive LRT for the first time before transplantation are referred by a multidisciplinary tumor board. There are many options including trans-arterial chemoembolization, trans-arterial radioembolization, thermal ablation, resection, systemic therapy, radiation, and immunotherapy.

While 42% of LT candidates who received an HCC exception score received at least one LRT in 2003, this rate was 92% in 2018. Here, TACE is still the most utilized treatment (50% in 2018), while thermal ablation is in second place with 22%. Moreover, the use of y90 and TARE in bridge treatment has increased considerably over the time. While it was used less than 5% in 2013, this rate was 19% in 2018 [94]. Accordingly, to prevent the dropout of patients from the LT waiting list, slowing down the progression of the tumor or therapeutic options are applied as a bridge to LT. Preoperative radiological evaluations and serum alpha-fetoprotein (AFP) changes are evaluated in response to this treatment [95, 96].

2. Portal vein tumor trombosis and transplantation

Approximately 44–62.2% of HCCs have advanced vascular invasion, mainly as a form of tumor thrombosis in the portal vein [97, 98]. Once portal vein tumor thrombosis (PVTT) is observed, intrahepatic and extrahepatic metastasis, portal hypertension, jaundice, and ascites are observed in a short time. After that, the median overall survival time is only 2.7 months [99].

Three-year survival is around 5% in untreated patients or those who do not receive palliative treatment [100]. Systemic treatment is generally recommended for these patients. However, treatment including ¹²⁵I seeding combined with transarterial chemoembolization (TACE) has also been shown to improve overall survival in HCC patients with PVTT [101, 102]. Downstaging options for patients with PVTT and HCC include radiotherapy, systemic therapy, and multimodal treatments. There are also disadvantages here such as extensive side effects and low downstaging rate [103, 104, 105]. PVTT is found in 10–40% of HCC patients at diagnosis. Its presence indicates a poor prognosis. The presence of HCC-PVTT is considered contraindicated for LT [106, 107, 108, 109]. Only palliative treatment is recommended. Based on the BCLC prognosis and treatment strategy algorithm, tyrosine kinase inhibitors, and immunotherapy recommendation, the estimated median overall survival is 5–19 months [17]. Three- and five-year survival rates are better than patients with segmental PVTT who receive only systemic or palliative treatment, especially those with PVTT in the first and second order portals and branches [110]. These patients also have low AFP value and low SUV-Max value by fluorodeoxyglucose-18 (FDG-18) positron emission tomography (PET). They are also patients with small tumors [111, 112]. These results demonstrate long-term survival in patients, supporting the concept of favorable tumor biology.

As a result of the dual balance in LDLT with the latest downstaging treatments and the latest successes in downstaging treatment ensuring valid benefit and equitable distribution, especially in the deceased donor scenario, there have been attempts to improve post-LT survival in HCC-PVTT with upfront downstaging. In this context, full/partial response to PVTT can be achieved with radiotherapy (stereotactic body radiation therapy [SBRT]/external beam radiation therapy [EBRT]/selective internal radiation therapy [SIRT]), transarterial chemoembolization (TACE), proton beam therapy, or systemic therapy. This also improved progression-free survival and OS. In planning for DDLT and to a lesser degree planning for LDLT, patients with HCC-PVTT may also be considered for DS before LT [113, 114, 115, 116, 117, 118, 119]. In downstaging to meet the criteria, evaluations of response to DS treatment and tumor biology features are combined [120, 121, 122]. In patients treated with TARE or TARE and TACE together or with stereotactic radiotherapy alone in the DS protocol, the 5-year OS and RFS rates are 53 and 52%, respectively. In patients who remain stable during the waiting period of the disease, the absence of loss of FDG-18 uptake on PET scan and enhancement in PVTT on CT scan is a successful indicator of DS. Advanced tumor grade (grade 3 and 4) and AFP > 400 ng/ml at diagnosis are poor prognosis factors for DS before LT [120]. The benefit of transplantation is determined as the expected survival gain with LT versus the most appropriate alternative treatment. Accordingly, the highest survival with LT is in HCC patients with advanced cirrhosis and those without extrahepatic disease (BCLC stages B, C) [123, 124]. When comparing survival in patients with HCC-PVTT undergoing LDLT, transplant benefit after downstaging with sorafenib or palliative radiotherapy alone was reported to be 23 months in 3-year survival and 34 months in 5-year survival [125]. Likewise, in two separate reports, a 5-year OS of 60% was reported after DDLT in patients with HCC-PVTT who received yttrium-90 TARE for downstaging and had a complete and ongoing radiological response with successful treatment. In contrast, this rate was 0% in non-transplant patients. The 5-year OS in transplanted HCC patients after complete radiological regression of vascular invasion with locoregional therapy was 60% [56, 126].

Since 2007, palliative treatment options have also been introduced for patients with advanced stage HCC (BCLC C). Patients with preserved liver functions (Child-Pugh A and B) and ECOG performance status (PS) 0–2 are candidates for systemic treatment. In addition, systemic treatment is recommended for cancers that show tumor progression upon locoregional therapy (LRT) or are not amenable to LRT [106, 127]. The SHARP phase 3 trial, which first established the tyrosine kinase inhibitor sorafenib as first-line treatment, changed the view of inappropriate systemic treatment in advanced HCC patients showing obvious resistance to classical chemotherapeutic agents [128]. Recently, immunotherapy has been emerging as the mainstay of systemic oncological treatment for unresectable HCC. Here, together with the establishment of evidence regarding the underlying etiology that will support hepatocarcinogenesis, which may be associated with the effectiveness of immune-based therapeutic approaches, systemic treatment may play a key role, especially for NAFLD-HCC, which is often diagnosed at an advanced stage. However, there is a need for appropriate biomarkers to predict the response to immunotherapy [129]. IMbrave 150 phase 3 study identified the combination of atezolizumab and the anti-vascular endothelial growth factor (VEGF)-A antibody bevacizumab as a new first-line treatment in advanced HCC patients who had not previously received systemic treatment. Here, a 5.8 months of improvement in median OS and a 2.6 months of improvement in PFS were observed compared to sorafenib. The OS was 19.2 months better here. One of the important features showed better preserved health quality index. Here, 69.5% of the patients were viral HCC patients. A subsequent subgroup study including nonviral HCC patients did not show any difference [130]. The KEYNOTE-240 phase III trial and the CheckMate-459 phase III trial also failed to show a survival advantage in OS and PFS [131, 132]. In fact, the latest meta-analysis of phase 3 studies showed whether OS improved in patients with unresectable HCC and those with viral HCC etiology, while noninvasive disease improved OS in patients with viral HCC etiology did not show a survival benefit in viral HCCs [133].

Ideal cutoff for AFP is independently predictive of post-LT HCC recurrence with <10 ng/ml. Compared to the 11% recurrence rate for AFP < 10 ng/ml, this rate is 50% for AFP > 10 ng/ml. Explant findings that considerably increase the risk of HCC recurrence include the presence of viable HCC, number of viable tumors, satellite nodules, and criteria greater than the Milan criteria [56].

The field of transplant oncology is rapidly developing in terms of treatment options. Immunotherapy is especially becoming widespread. For a while, ICI use before LT was considered a contraindication due to concerns about graft dysfunction or loss. However, graft loss was not noted in subsequent small retrospective studies [134, 135]. Moreover, UNOS took this further and stated that taking ICI alone, along with data on the use of immunotherapy for DS or bridge before LT, is not a contraindication for the HCC MELD update. However, despite all these data, it is of great importance to determine whether ICI and TKI or radiotherapy combination treatments would reduce post-LT HCC recurrence in patients exceeding MC, whether there would be an increase in the risk of graft dysfunction with ICI, what is the ICI termination time before LT, and what are the biomarkers in the evaluation of tumor response to treatment. In patients with advanced HCC, palliative treatments with TARE or TACE alone showed a 3-year OS of 20 and 10%, respectively [136, 137].

The REFLECT study, which compared lenvatinib with sorafenib in patients with unresectable BCLC stage B or C HCC, showed similar survival results with 13.6 and 12.3 months, respectively [138]. However, in the SARAH study, only 8 months of median OS was observed with selective internal radiotherapy (SIRT) in BCLC stage C HCC patients with locally advanced tumors [139]. The use of SIRT combined with sorafenib did not provide a statistically significant advantage in OS (14 vs. 11 months) [140]. The abovementioned palliative treatments, from downstaging to LT, achieved different survival figures in HCC patients with HCC-PVTT. Patients with locally advanced HCC and patients with PVTT without extrahepatic spread should not be categorized as having systemic disease. Targeted radiotherapy at appropriate doses results in tumor regression within the portal vein. Many patients were successfully downstaged and met the transplant criteria with this method. Although the optimal period between successful downstaging and LT is not clear, LT can be performed after 3–4 months with acceptable long-term results. In patients with suitable living donor, in LDLT regulation after downstaging, levnatinib treatment is superior to alternative treatments such as immunotherapy or TARE/SBRT alone. LT may be recommended as the only way to hope for long-term survival in patients with HCC-PVTT, despite the majority having locally rather than systemically advanced disease. These were traditionally considered suitable for palliative treatment only.

HCC may be associated with PVTT 16–30% [141]. PVTT is generally classified according to the level of portal vein involvement. There are two most commonly used classifications. In the Japanese Vp classification, Vp 0: There is no PVTT. Vp1: There is segmental tumor thrombus in the segmental portal vein. Vp2: There is tumor thrombus in the secondary order branches of the portal vein. Vp3: There is tumor thrombus in the first order branches. Vp4: There is tumor thrombus in the main portal vein and/or contralateral portal vein branch [142]. In Cheng’s classification, PVTT is divided into four categories. Type I: Tumor thrombus involving segmental or sectoral portal vein branches or the part above it. Type II: Tumor thrombus involving the left/right portal vein branch. Type III: Tumor thrombus involving the main portal vein. Type IV: Tumor thrombus involving the superior mesenteric vein [143]. However, the BCLC staging system and clinical guidelines do not mention these classifications. Patients with PVTT and HCC are considered to have BCLC stage C disease. This corresponds to advanced-stage disease for which systemic treatment is usually recommended. Locoregional-based downstaging treatments play an increasingly important role for HCC patients. However, there has been no approved adjuvant treatment protocol for HCC patients with PVTT so far. Prospective multicenter controlled randomized studies are required to understand the role of LT in the intent to treatment (ITT) strategy in this group of patients. Locoregional downstaging treatments, adjuvant treatments, and their standardization are important to expand the inclusion criteria for LT. Recently, the use of immunotherapy in this field has significantly changed the treatment approach, especially in HCC patients with advanced disease. Tyrosine kinase inhibitors, vascular endothelial growth factor inhibitors, and immune checkpoint inhibitors are becoming increasingly important for the treatment of HCC [144, 145]. There is no study evaluating specific LT in patients with PVTT after downstaging treatment with immunotherapy. The use of immunotherapy in cancer patients undergoing transplantation has not been explicitly approved. Here, data regarding oncological outcomes and graft rejection outcomes associated with immunosuppressive therapy, especially after LT, are not sufficient [146, 147]. Recently, there have been a number of attempts to expand LT inclusion criteria in HCC patients, including PVTT patients. Expanded criteria are primarily accepted in the regulation made in the context of LDLT. Expanded criteria including PVTT patients is retrospective in nature. Along with the large heterogeneous patient population, there are also major deficiencies in important points such as neoadjuvant downstaging or systemic treatments. Studies with LDLT reported a lower survival rate in patients with PVTT, as well as 5-year OS rates of 57 and 48% in VP 1 and 2 patients with and without prior downstaging, respectively. Therefore, there is no consensus for downstaging strategies in patients with PVTT [112, 120, 148].

Evidence regarding the benefit of LT in patients with PVTT and HCC is limited due to the short follow-up period, small sample size, and retrospective design of the studies on the subject. However, the common results are that the prognosis after LT is poor in HCC patients with lobar PVTT. PVTT is a factor that increases the risk of mortality and recurrence after LT. HCC patients with segmental PVTT with low biological aggressiveness (such as low AFP, early grade tumor) have better results in terms of DFS and OS after LT. Although LT appears to be superior to palliative systemic treatments in patients with non-extended PVTT (Cheng’s type I or Vp 1-2) and HCC, OS rates are within the recommended criteria and below 70% in patients with HCC or nonmalignant LT. This must be taken into account, especially in the context of falling off waiting lists and organ shortages around the world.

A recent study has described the experience of LDLT in HCC patients who underwent DS for PVTT with an otherwise dismal prognosis. In patients successfully downstaged after stereotactic body radiation (SBRT) and transarterial chemo- or radio-embolization (TACE or TARE) treatments, the 5-year OS and RFS were 57 and 51%, respectively. This finding was comparable to HCC patients undergoing LDLT without PVT [120].

3. Recurrence and its management

Post-LT HCC recurrence is among the leading causes of death in patients transplanted for this indication. Posttransplant HCC surveillance is important in this regard. Early diagnosis and aggressive treatment have been proven to improve survival outcomes [149, 150]. Recurrence is highly variable. HCC recurrence is rare in the first year after LT (cause of death rate in the first year is 5.3%). Likewise, post-LT occurs less frequently after 5 years. However, it is known that HCC recurrence is highest within 2–3 years after transplant [151, 152]. Recurrence time also emerges as a prognostic factor. Early HCC recurrence is associated with worse prognosis [152, 153]. Using the Milan criteria, recurrence was observed in 5.7–16% of cases. This became the benchmark for all subsequently proposed criteria. The Organ Procurement and Transplantation Network/United Network for Organ Sharing (OPTN/UNOS) records confirmed that some pretransplant factors are strongly predictive of post-LT recurrence. These include poor tumor differentiation, micro or macrovascular invasion, lymph node or extrahepatic spread, explant TNM stage T3 or T4, tumor stage > T2 after downstaging, and high AFP level [154]. The overall recurrence rate observed in a systematic review was 16%. The median time from LT to recurrence is 13 months. The most common area of recurrence is extrahepatic. According to explant pathology, 51% of the patients had LT performed outside the Milan criteria [155]. The University of California, Los Angeles (UCLA) also reported that recurrence is mainly extrahepatic. Recurrence in the liver allograft was 37.8% and had a large multinodular pattern [153]. According to these results, only liver recurrence is observed in 15–40%. Extrahepatic disease is probably due to the growth of occult metastases. Several risk factors were also identified, including dynamic and static AFP levels, neutrophil/lymphocyte ratio (NLR), AFP L3, and des gamma carboxy prothrombin. AFP response to LRT is consistently superior for post-LT outcomes [90].

Geographical and ethnic factors significantly affect recurrence. While the recurrence rate is 10% in South American studies, this rate is reported to be 21% in Asian studies. Asian and African American men have a higher recurrence rate compared to Caucasians [156]. Tumor histology provides important information in predicting the risk of recurrence. Significantly increased expression of angiopoietin-2 was demonstrated in the tumor endothelium of patients with HCC recurrence. This was not evident in hepatocytes in explant pathology. In univariate analysis, BMI, AFP level at the time of transplant, endothelial angiopoietin-2 expression, Milan criterion, Metroticket AFP score, and AFP pattern were significantly associated with recurrence. However, multivariate analysis showed that only angiopoietin-2 expression was an independent factor associated with recurrence [157].

Among the histological views evaluated, multivariate analysis revealed that immune cells were the strongest prognostic predictor of recurrence. Patients undergoing LT for HCC have a distinct immune profile that differs from that of patients transplanted for other reasons. In addition, the balance of immunosuppression may affect the risk of graft rejection, on the one hand, and the risk of HCC recurrence, on the other [158, 159].

In the early posttransplant period, a considerable but temporary increase in myeloid and B cells and a decrease in T cells and natural killer cells are observed. However, within 3 weeks, the cells return to pretransplant levels. Plasmacytoid dendritic cells increase progressively after LT and peak at 2 weeks. Patients with and without HCC show very different immune cell composition, but in transplanted patients, there is little difference in these levels in the early post-LT period between the patients who would develop recurrence and those who would not. In fact, large differences in immune subgroup cell levels exist even before transplantation. In a deeper analysis, significant T14 and T21 expansion is observed starting from 3 weeks post-LT in patients who would develop HCC recurrence [160].

The probability of recurrence in the MC group (at diagnosis) is 11.3% in the 5 years after LT and 13.3% in the 10 years after LT. However, in the DS group (at the time of LT), it is 19.1% in the 5 years and 20.6% in the 10 years. On the other hand, in patients with failed HCC, DS is 38.9% in 5 years and 41.4% in 10 years. Predictors of pre-LT recurrence here are tumor burden at transplantation, LRT count of >2, and AFP level of >20 ng/dl [42]. Early recurrence within 2 years of undergoing surgery accounts for 70% of HCC patients with recurrence, is almost incurable, and has been related to very poor survival [161].

3.1 Post-LT HCC surveillance

Post LT surveillance is very important to prescribe potential curative treatments and to identify early recurrence [162]. Evidence-based recommendations for post-LT surveillance in patients with HCC are lacking. In fact, even though close surveillance is associated with better prognosis and more applicable curative treatments [150], there is a scarcity of strong studies on its effect on LT outcomes, duration, and frequency of monitoring. Therefore, the surveillance strategy includes cross-section imaging such as contrast enhanced computer tomography (CT) or magnetic resonance imaging (MRI) for the abdomen and non-contrast CT for the lungs. If there is high clinical suspicion in other areas of metastasis, such as bone and brain, these areas are included in the examination. AFP is an effective monitoring method during surveillance. For this purpose, it is measured every 6 months. Although there is no standardization in surveillance protocols, at least 3 years of surveillance is recommended. There is no difference in RFS in patients whose scans are followed up at 3 to 6 month intervals.

UCSF recommends a surveillance strategy that uses the RETREAT score, which includes the diameter of the largest tumor in addition to the microvascular invasion, serum AFP during LT, and total tumor number in liver explant pathology [162]. Accordingly, if the RETREAT score is 0 (5-year recurrence rate is 3%), surveillance is not recommended for patients. Patients with RETREAT score 1–3 require surveillance every 6 months for 2 years. While RETREAT recommends surveillance every 6 months for 5 years in patients with a score of 4, it recommends surveillance every 6 months for 2–5 years and every 3 months for the first 2 years for patients with a RETREAT score of 5 or more.

3.2 The influence of immunosuppression on HCC

The effect of immunosuppression on recurrence after LT is also a matter of debate. Immunosuppression can inhibit the immune system’s ability to detect and destroy cancer cells, which may promote tumor growth. As a result, the risk of HCC recurrence also increases. Calcineurin inhibitor (CNI) is associated with a dose-dependent increased risk of recurrence [163].

Mammalian target of rapamycin (mTOR) was found to upregulate some mutations in HCC. In such cases, mTOR inhibitors have antiproliferative and antiangiogenic effects [164]. Therefore, mTOR reduces recurrence and increases survival. The International Liver Transplantation Society, in its Transplant Oncology Consensus Conference, recommended levels of lower than 10 ng/mL for tacrolimus and lower than 300 ng/mL for cyclosporine.

Although there were concerns that DAA treatment increased the aggressiveness and recurrence of HCC, this has now been proven to be unfounded. There was even a significant reduction in the risk of death in LT in patients treated with DAA [13, 165].

Management of recurrence after LT is parallel to the management of primary LT. The clear benefit of treatment modalities with curative intent has also been demonstrated. Treatment options also vary with the location and extent of recurrence, which affects survival [43]. In one study, in cases where HCC recurrence was treated after LT, the highest 3-year survival was observed in the surgery alone group as 60%, followed by the surgical and nonsurgical combined treatment group as 37%. However, in the nonsurgical group, this rate was only 11% [153].

In a large study involving 661 patients who underwent resection for HCC, 16% of patients with recurrent HCC were listed for LT. While 63% of them eventually underwent transplantation, 23% patients were dropped out due to tumor progression [166]. Salvage LT (SLT) is another recommended curative treatment option in post-liver resection. However, long-term results are not clear. Intent to treat analysis showed that 5-year ITT OS was 69% and DFS was 60%. The SLT option is successful in 56% of patients. The results of salvage transplantation appear to be comparable to the those of primary LT for HCC, even when examined on the basis of ITT [167].

LRT may be considered as a potential treatment modality for patients with unresectable recurrent HCC after LT. Systemic therapy has limited use for post-LT recurrent HCC. Compared with supportive care, sorafenib showed a median survival of 10.6 months versus 2.2 months in this group of patients. Another multikinase inhibitor, regorafenib, is used as second-line therapy. It provides median survival rate of 13.1 versus 5.5 months [168, 169]. Capecitabine has the same safety profile as sorafenib. Compared with the best supportive care, median OS of 22 versus 7 months was observed [170].

4. Conclusions

HCC recurrence after LT is still a dreadful event, occurring in up to 20% of cases. It remains challenging to individualize risk assessments for the recurrence of HCC after liver transplantation. This is increasingly important due to the rising number of patients transplanted outside the Milan criteria with extended criteria. It might be prevented by stringent pretransplant selection criteria incorporating biological markers of aggressiveness (such as response to therapy, serum markers, and histological factors) in addition to size and number of tumors. Another strategy for improvement is the further expansion of the selection criteria. In this regard, effective perioperative adjuvant therapy based on recent developments in anticancer drugs and the refinement of posttransplant immunosuppression regimens would play a key role. Routine use of recurrence risk assessment is advised to provide tailored advice for patient selection or prioritization and an adequate individualized surveillance strategy as this predicts outcomes and prognosis. İn this context should be futher provide a framework through which to design clinical trials adjusted to the risk of recurrence and test immunosuppressive strategies or new adjuvant therapies to prevent HCC recurrence after transplantation. Several advances in this sense have been made in the last decade, allowing patients with HCC broader access to LT with more precise prediction of outcomes. In the post-LT period, surveillance should be driven by post-LT risk stratification. No adjuvant treatments after LT have been validated to prevent HCC recurrence. Within the last decade, breakthroughs in immunotherapy have greatly expanded the treatment armamentarium for HCC. However, there is still an unlit corner for HCC patients awaiting LT or after LT due to the deep concern about lethal rejection induced by immunotherapy. On the one hand, there will be an increasing number of HCC patients after immunotherapy who are bridged or downstaged to be candidates for LT, as immunotherapy is now gradually becoming a part of routine or even preferred regimens for HCC systemic therapy. There are also many patients with HCC recurrence after LT who fail to respond to other therapies, and immunotherapy may be their last option. Variation in the selection criteria for liver transplantation for HCC, from solely morphological to the incorporation of surrogate markers for the biological behaviors, the acceptable tumor burden has gradually been clarified and the patients who can be expected to benefit from this procedure can be appropriately selected. To improve the thresholds of liver transplantation for HCC, one goal is to increase the number of eligible patients. The number of deceased donors, as well as the allocation rule, would directly influence the expansion of HCC patients who are eligible for liver transplantation. In terms of the tumor burden, early tumor detection is indispensable. Further, pretransplant bridging therapy to prevent dropout due to tumor progression beyond selection criteria and downstaging to an acceptable tumor burden in those who would otherwise be excluded from candidacy would contribute to the evolution of the thresholds. The criteria of Liver transplantation for HCC continue to evolve. The transplant oncology community has increasingly understood the need to move away from size and number criteria and include surrogates of biological behavior into decision-making. It is likely that in the near future, most algorithms to decide which patients will be eligible for liver transplant will incorporate indices of tumor biology.

Conflict of interest

The author declares no conflict of interest.