Objectives: Milan criteria is the most commonly used criteria for patients with hepatocellular carcinoma awaiting liver transplant. The effects of locoregional therapy on downstaging or bridging before liver transplant on survival remain controversial. Considering that the tumor size may change with locoregional therapy and formalin fixation after explantation, we aimed to evaluate the effects of locoregional therapy on radiological and pathological Milan criteria and survival.
Materials and Methods: Demographic data, etiology, preoperative alpha-fetoprotein value, Child-Pugh and Model for End-Stage Liver Disease-Na scores, status of being inside or outside of radiological Milan criteria, status of being inside or outside of Milan criteria in explant (pathological Milan criteria), and the locoregional therapy types and combinations were evaluated for their effects on inclusion in Milan criteria and survival. Results: During the study period, 396 patients underwent liver transplant at our center, with 97 because of cirrhosis and hepatocellular carcinoma. When we viewed patients according to preoperative radiologic evaluations, 67.9% were within Milan criteria and 32.1% were outside. When we viewed according to explant (pathological) evaluations, 80.7% of patients were within Milan criteria. Among 97 patients, 71 (73.2%) had locoregional therapy (22 [30.9%] for downstaging, 49 [69.0%] for bridging to transplant), and 12 patients (12.3%) were within Milan criteria on explant examination while outside of Milan criteria before LT. One-year, 3-year, and 5-year survival rates were 80.7%, 76.1%, and 71.6%, respectively.
Conclusions: As a result of radiological evaluations, in patients who were outside of Milan criteria and underwent locoregional therapy, explant pathology within Milan criteria had a positive effect on survival; however, after locoregional therapy, there was no significant effect on survival in patients who were still outside of Milan criteria.
Key words : Bridge to transplant, Downstaging, Locoregional therapy
When patients are within Milan criteria (MC), liver transplant (LT) is a generally accepted modality for treatment of hepatocellular cancer (HCC).1 According to MC, LT is performed in patients with a single tumor size not exceeding 5 cm or a maximum of 3 tumors not exceeding 3 cm and without vascular invasion. The number and size of nodules in patients with HCC are based on imaging methods. However, with increased numbers of patients on wait lists because of a limited organ pool, patients with HCC and tumor progression are excluded from wait lists.
Results of LT performed outside of MC are controversial in terms of survival. Therefore, bridging treatments have been used in the pretransplant period. These bridging treatments, which are used to reduce the tumor burden and can allow patients outside of MC to meet MC or can prevent patients on wait lists from being dropped from the list, are called locoregional therapy (LRT).2,3Pretransplant LRTs include transarterial chemoembolization (TACE), radiofrequency ablation (RFA), microwave ablation, and transarterial radioembolization.4,5 Studies have shown that LRTs can cause pretransplant tumor necrosis, preventing progression of viable tumors and preventing patients from being dropped from wait lists.6,7 However, limited studies have shown the effects of bridging LRT on explant pathology, tumor recurrence, or survival.
In addition to LRTs, there is another factor that affects tumor size. The explant is evaluated in formalin-fixed solutions for pathology examination. Some studies have shown that this solution has a tumor-reducing effect on tumor size.8,9 Transplant criteria, including MC, were established as a result of radiological or laboratory evaluations. However, from pathological examinations, the number and size of the tumor may change both because of the LRT and because of the formalin fixation solution and LRT. This study was undertaken to evaluate patients who underwent LT radiologically (with or without LRT) and pathologically in terms of MC and to investigate whether changes in tumor size and number as a result of LRT and formalin fixation solutions can affect postoperative results and survival.
Materials and Methods
Study design and patient selection
Between 2007 and 2022, 396 LTs were performed at our center, with 97 because of HCC. Seven patients (7.2%) had living-donor LT procedures (kinship relations up to fourth degree), and 90 patients (92.7%) had deceased-donor LT procedures. Magnetic resonance imaging and computed tomography were used to confirm lesion sizes between 1 and 2 cm. No additional radiological examinations were used for lesions with a typical staining pattern (intense contrast uptake in the arterial phase followed by extracellular contrast washout in the venous and/or delayed phases) and in lesions of 2 cm and above. Biopsy was not routinely performed, except for lesion with diameter of <1 cm or for suspicious lesions.
Liver transplant was performed in HCC patients who were inside of MC with viable tumor size in control imaging and after LRT in patients initially outside of MC. Patients outside of MC underwent varying numbers and combinations of LRT until less than 30% of viable tumor cells were observed in control imaging; patients with HCC within MC also received LRT for bridging to LT. Patients on wait lists who progressed and were found to have distant metastases were excluded from the study. In the pretransplant period, patients were required to have an α-fetoprotein (AFP) value of <400 ng/dL. Patients who remained above this value despite LRT treatment were also excluded from the study. Patients with mortality in the first 90-day period were considered to have early mortality, and these patients were also excluded from the study. This was a single-center, retrospective observational study approved by the local clinical ethics committee.
Demographic data, etiology, preoperative AFP level, duration on the wait list, Child-Pugh and Model for End-Stage Liver Disease (MELD)-Na scores, status of inside or outside of radiological MC status, type of LT procedure (living donor/whole), status of being inside or outside of MC in explant liver (pathological MC), and the type of LRT were analyzed to determine effects on inclusion in the MC and on survival.
Downstaging protocol and radiological assessment
Patients had LRT performed by a single interven-tional radiology team based on the Barcelona-Clinic Liver-Cancer staging system. Radiofrequency ablation was applied to liver lesions selected on ultrasonography in number ≤3 and size ≤3 cm. Transarterial chemoembolization procedures were planned for lesions of ≥3 cm in size and lesions that were not detectable on ultrasonography and difficult to reach percutaneously in the liver. Combinations of RFA and TACE were applied simultaneously to lesions of <3 cm whose borders could not be clearly distinguished on ultrasonography. Postprocedure control imaging was performed with magnetic resonance imaging or computed tomography at month 1, month 3, month 6, month 9, and month 12. After the imaging evaluations versus the baseline evaluation, physicians decided the number of sessions of the RFA or TACE procedure to be applied.
Evaluation of response to downstaging treatment and progression of LT
Response to downstaging sessions was evaluated according to the United Network for Organ Sharing downstaging protocol. Downstaging was considered successful when patients reached MC according to residual tumor cell size and number of tumors in control cross-sectional imaging. Tumor necrosis area was excluded from these measurements. Downstaging was considered unsuccessful if there was tumor progression, vascular invasion, extrahepatic spread, and AFP level that did not fall to <500 ng/mL despite LRT, and patients were excluded from LT. After a successful downstaging observation of at least 3 months of stable disease, LT was performed.
Livers removed during LT were placed in formalin fixation solutions without waiting and without any surgical procedure on the organ. Afterward, the explanted liver was evaluated by a single pathologist for the largest tumor diameter, number of tumor nodules, presence of macro- or microvascular invasion, and pathologically ascertainment of being inside or outside of MC. Cases with no viable tumor cells in explant pathology and diffuse necrosis were evaluated as having a complete response.
We used SPSS version 22 (IBM Corp) for statistical analyses. Categorical variables are presented as percentages. Other variables are presented as mean ± SD or as median (range); depending on whether the variable followed a normal distribution or not, we the Shapiro-Wilk normality test. We used the log-rank test to determine the differences between Kaplan-Meier curves for both overall survival and disease-free survival time. Mean survival time was reported. To determine the prognostic factors that affect overall survival and disease-free survival time, Cox proportional hazard regression analysis with backward selection procedure was performed after Kaplan-Meier analysis. P < .05 was considered significant.
Patient demographic characteristics
Ninety-seven patients with HCC because of cirrhosis secondary to various etiologies were included in this study. Of these patients, 13 (13.4%) were female and 84 (86.5%) were male. Mean age of patients was 64.4 ± 0.86 years. Patient demographic characteristics, preoperative Child-Pugh and MELD-Na scores, and preoperative AFP values are listed in Table 1. Median follow-up was 28.6 months (interquartile range,6.9-88.7). Cirrhosis was mostly because of hepatitis B virus (69.7%), as well as hepatitis C virus (9.2%), nonalcohol-related steatohepatitis (8.3%), alcohol-related (1.8%), cryptogenic (9.2%), and Wilson disease (0.9%). The average number of days on the wait list was 68 ± 28 (range, 0-1799).
Evaluation of patients according to Milan criteria
At the time of admission, 72.1% of all patients were within MC and 27.9% were outside of MC. In explant pathology, 82.4% of patients were evaluated as within MC and 17.6% as outside of MC. The explant pathology of 16 patients (16.4%) was evaluated as having a complete response to LRT. The distribution of patients with radiological and pathological MC is listed in Table 2.
Other important parameters in pathological exa-mination were differentiation and vascular invasion. Among our patient group, 65 (67.0%) were evaluated as well differentiated, 25 (25.7%) as moderately differentiated, and 7 (7.2%) as poorly differentiated. Although 81 patients (83.5%) had no venous invasion, microvascular invasion was found in 9 patients (9.2%) and macrovascular invasion was found in 7 patients (7.2%).
Locoregional treatment and radiological assessment
The same interventional radiologists treated some patients with RFA and TACE at varying numbers and combinations. Twenty-six patients (26.8%) had LT without LRT. Because patients outside of MC in control imaging as a center preference do not receive LT, all patients who did not undergo LRT were patients who underwent LT, considering that they were radiologically shown within MC. Only 1 patient (3.8%) was considered to be pathologically outside of MC. Of 71 patients (73.1%) who received LRT, 22 (30.9%) were evaluated as beyond radiological MC and had downstaging therapy and 49 patients (69.0%) were evaluated as within MC and received LRT as bridging during the waiting period. The type, number, and distribution of LRTs administered in the pretransplant period are listed in Table 3.
Eight patients who died from early surgical causes within the first 90 days were excluded from survival analyses. Mean follow-up was 8.9 ± 0.6 years (95% CI, 7.6-10.2). Of 89 patients included in the survival analyses, overall survival rates at 1, 3, 5, and 10 years were 91.6%, 82.4%, 66.3%, and 49.2% respectively. The 1-, 3-, 5-, and 10-year disease-free survival rates were 90.4%, 78.6%, 66.7%, and 49.5%, respectively.
Overall survival between patients within radio-logical MC versus outside of radiological MC was not significantly different (P = .08) (Figure 1). Although the mean overall survival of patients within radiological MC was 58.9 ± 4.6 months, survival of patients outside of radiological MC was 47.3 ± 5.5 months. When we compared patients who were within versus outside of pathological MC, survival was better in the group within MC (P = .007) (Figure 2). The mean survival of patients within pathological MC was 57.7 ± 5.0 months, and the mean survival of patients beyond pathological MC was 37.2 ± 5.6 months.
When the LRT group was compared with the group who did not receive LRT, no significant difference in survival was shown (P = .0467) (Figure 3). In the survival analysis performed by comparing patients who received various sessions of LRT with those who did not, the mean survival of those who did not receive LRT was 60.7 ± 5.5 months, whereas mean overall survival of all patients was 55.2 ± 3.6 months.
Liver transplant is known as a life-saving treatment option for patients with HCC. Tumor size reduction and better survival opportunities are available with LRT for patients with advanced HCC. Locoregional therapy includes transcatheter or ablative treatments such as TACE, microwave ablation, RFA, cryoablation, and Y-90 transarterial embolization.10,11 These LRTs basically have 2 purposes: to avoid wait list drop-out and to create transplant bridging therapy. Ablative techniques can be used in early HCC cases. Among these, the most commonly used is RFA, with results similar to surgical resection in HCC patients with tumors smaller than 2 cm.10
Many criteria used in the classification of LT candidates in patients with HCC have been defined in the literature. With the study conducted by Mazzaferro and colleagues,1 LT with the so-called Milan criteria in the treatment of HCC became important in terms of making significant contri-butions to survival12 and MC became the most widely used criteria. Although survival of patients outside MC posttransplant is poor, better survival can be achieved with LRT when applied to patients during the wait list period. The main purpose of LRT is to destroy or reduce residual, viable tumor cells, thus allowing patients to be within MC.13
All criteria known in the literature, including MC, were established as a result of radiological or laboratory evaluations. However, pathological examination has shown that the number and size of the tumor may change both because of the LCT performed and because of formalin fixative solutions and LRT.14,15 The significance of tumor size, which changes with both the effect of formalin solution and the effect of LRT, in terms of survival has not been previously discussed in the literature, to our knowledge.
Locoregional therapy targets are known as curative in early stages, as bridging to LT (increasing surgical resectability and disease control) in intermediate-stage disease, and as palliative in advanced-stage diseases. Various numbers and combinations of LRT can be applied, according to tumor localization, size, and number. Özen and colleagues showed that TACE and RFA procedures can be performed simultaneously safely and can be effective in providing local control, especially in solitary lesions.16 In our study, a single interventional radiologist applied various combinations of LRT.
Although LRTs can be applied to patients on wait lists for bridging purposes, the effect on survival is controversial.17,18 In our study, we performed LRTs to 73.1% of our patients with the aim of bridging and downstaging, but we showed that LRT alone had no effect on survival.
In Renner and colleagues, patients who had no tumor progression with LRT had better survival.17 Although these results are consistent with our study, the view stated in the study that MC does not increase survival is not supported in our study.17 We observed an increase in survival in patients who could be pathologically included into MC, although overall no survival advantage was observed among patients with and without LRT.
In a retrospective multicenter study from Mehta and colleagues, downstaging treatments had a positive effect on survival in MC.18 Parikh and colleagues stated that the rate of successful downstaging was 54%, but its effect on survival was controversial.19 In our study, although most of the patients received LRT, its effect on survival could not be demonstrated. On the other hand, with regard to overall survival, we found that patients within pathological MC had higher survival rate than patients within radiological MC.
Milan criteria classification of patients after pathological evaluation may be more significant in terms of survival; however, the inability to make a clear interpretation about the effect of LRT and formalin solutions constitutes a limitation of this study. Kshithi and colleagues showed that tumor size and surgical margin may affect head and neck tumors.8 In a study on breast cancer specimens, formalin fixation solutions may have an effect on tumor size.20 When these results are considered, although it has been shown that formalin fixation fluids cause changes in some tissues and tumor sizes, further studies are needed on this subject, since there is no study showing their effects on HCC and liver tissue.
Milan criteria contains strict criteria for transplant candidates. Some studies have shown that good survival can be achieved if MC are expanded.21,22 A consideration is determining the minimum accep-table 5-year survival rate in treatment of HCC. Although a 5-year survival rate of 50% may be sufficient,23,24 this rate is quite low.5 In our study, the 5-year survival was determined to be 66.3%, which is acceptable compared with the literature. However, we found our 10-year survival rate to be 49.5%, which can be considered important for long-term survival in HCC patients. Although MC continues to be significant in terms of survival in patients undergoing LT because of HCC, radiological and pathological MC are different and may vary in terms of whether they are included in these criteria as a result of various factors. Although LRT is known to be an effective treatment option for patients on waiting lists, especially in countries with deceased-donor LT, its effect on survival is controversial. We found no effect of LRT on overall survival, although our 10-year survival rate showed that LRT has acceptable results in HCC patients.
Volume : 21
Issue : 10
Pages : 820 - 825
DOI : 10.6002/ect.2023.0221
From the Organ Transplantation Center, Bursa Uludag University, Bursa, Turkey
Acknowledgements: The authors have not received any funding or grants in support of the presented research or for the preparation of this work and have no declarations of potential conflicts of interest.
Corresponding author: Ekrem Kaya, Organ Transplantation Center, Medical Faculty, Bursa Uludag University, Bursa, Turkey
Phone: +90 224 2952018
Table 1. Demographic Characteristic of Study Patients
Table 2. Evaluation of Patients According to Milan Criteria
Figure 1. Overall Survival According to Radiological Milan Criteria
Figure 2. Overall Survival According to Pathological Milan Criteria
Figure 3. Survival Analysis of Patients With and Without Locoregional Therapy
Table 3. Type, Number, and Distribution of Locoregional Therapies Administered Pretransplant