If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
The role of orthotopic liver transplantation in the management of hepatocellular carcinoma (HCC) has evolved significantly over the past decades. Initial experiences with orthotopic liver transplantation were limited to patients with extensive unresectable tumors, and were marked by uniformly dismal outcomes due to high rates of tumor recurrence.1, 2 Those results evoked considerable interest in reexamining the staging guidelines to determine eligibility for orthotopic liver transplantation. This lead to a study by Mazzaferro et al.3 who reported that 48 patients with a single tumor ≤5 cm in diameter or with ≤3 tumors all ≤3 cm in diameter, as identified by preoperative imaging, displayed survival rates comparable to those of non-HCC liver transplant recipients. These Milan criteria (MC) are currently widely accepted as an effective method of selecting patients with early-stage HCC for curative orthotopic liver transplantation and have been incorporated into organ allocation systems.4
The prevalence of HCC is high in Japan, with around 30,000 deaths annually.5 While deceased donor liver transplantation has barely been available for patients with HCC, various treatment modalities, such as hepatic resection, percutaneous ethanol injection, radiofrequency ablation, and transarterial chemoembolization, have been developed, leading to improved outcomes.5–8 Against this background, living donor liver transplantation (LDLT) has recently emerged as a new therapeutic option. Contrasting sharply with deceased donor liver transplantation, which utilizes a scarce public resource subject to an equitable allocation system, living donor liver grafts are dedicated to related recipients. LDLT thus allows the extension of selection criteria, since the acceptable recurrence rate is not absolute, but rather depends on donor organ availability.9 However, LDLT involves significant disadvantages, including risks to the live donor and higher perioperative morbidity and mortality compared with other treatment modalities. Given these risks, the argument has been made that early-stage HCC should initially be managed using resection or conventional methods if comparable overall survival is available, with LDLT held in reserve as a second-line option.10, 11
In February 1999, we started an LDLT program for patients with HCC using our own selection criteria that included tumors exceeding MC. We have previously reported initial results for 56 patients at a median follow-up of 11 months.12 The present study reported extended results for 93 patients with HCC over a longer follow-up period. Risk factors for posttransplant mortality and recurrence were analyzed. The role of LDLT was investigated with regard to treatment strategies for cirrhotic patients with HCC.
LDLT, living donor liver transplantation; HCC, hepatocellular carcinoma; MC, Milan criteria; MELD, model for end-stage liver disease; TNM, tumor, node, metastasis; JIS, Japan Integrated Staging.
PATIENTS AND METHODS
Patients with HCC met both of the following selection criteria for LDLT, approved by the ethics committee at Kyoto University: 1) HCC not suitable for resection or local ablation therapies due to advanced tumor spread, repeated uncontrolled recurrence or poor liver function reserve, and 2) exclusion of extrahepatic metastasis or macroscopic venous invasion on preoperative imaging. No restrictions were placed on the number or size of tumors. All patients were evaluated for the extent of tumor involvement using abdominal ultrasonography, contrast-enhanced computed tomography, brain and chest CT and bone scintigraphy.
Between February 1999 and September 2004, a total of 93 patients (64 men, 29 women) with HCC underwent LDLT at Kyoto University. During the same period, 51 patients were referred for but unable to undergo LDLT to our hospital. Reasons preventing LDLT included macroscopic vascular invasion (n = 30) or extrahepatic metastasis (n = 15) detected on imaging immediately before operation, no available donor (n = 3), and severely deteriorated patient condition (n = 3). Median age for the 93 patients who underwent LDLT was 54 years (range, 22–69 years). A total of 55 patients (59%) were hepatitis C virus antibody-positive, and 30 (32%) were hepatitis B surface antigen-positive, including 2 patients with coinfection. Child-Turcotte-Pugh classification was C for 44 patients (47%), B for 34 patients (37%), and A for 15 patients (16%). Median model for end-stage liver disease (MELD) score was 14 (range, 4–36). Of the 93 patients, 73 patients (78%) displayed a history of previous treatment for HCC using various nontransplant methods including transarterial chemoembolization (n = 63), percutaneous ethanol injection or radiofrequency ablation (n = 47), or hepatic resection (n = 11). These treatments were performed in other hospitals not as a bridge to transplant, but with intent for curative ablation before referral to our institute. The remaining 20 patients (22%) had no history of HCC treatment before LDLT, in most cases due to advanced liver dysfunction.
All 93 patients were diagnosed with HCC immediately before LDLT. Tumor staging was determined by counting only viable and enhancing nodules on pretransplant CT. For patients who had undergone previous therapies, resected tumors or nodules that were judged as nonviable after percutaneous ethanol injection, radiofrequency ablation, or transarterial chemoembolization were not counted. According to the tumor, node, metastasis (TNM) staging criteria of the International Hepato-Pancreato-Biliary Association and the Liver Cancer Study Group of Japan,13 HCC was classified as stage I (n = 14), II (n = 35), III (n = 42) or IV-A (n = 2). A total of 49 patients (53%) met MC according to preoperative imaging, while 44 patients did not.
Of the 73 patients who had a history of previous nontransplant treatment, records of tumor stage at first diagnosis of HCC were available for 58 patients: stage I for 24 patients, stage II for 17 patients, stage III for 16 patients, and stage IV-A for 1 patient, according to TNM classification. While 42 patients met MC, 16 patients did not. Median period between first diagnosis of HCC and LDLT was 26 months (range, 2–168 months) in these 58 patients. At the time of LDLT, tumor stage was up-staged in 31 patients and 18 patients, down-staged in 5 and 6, and unchanged in 22 and 34, according to TNM classification and MC, respectively.
The Japan Integrated Staging (JIS) score combines the Child-Turcotte-Pugh classification and TNM staging by the Liver Cancer Study Group of Japan, and it has recently been proposed as a new prognostic staging system for patients with HCC.14 JIS score is obtained by adding tumor stage score (stage I, 0; stage II, 1; stage III, 2; stage IV, 3) and Child-Turcotte-Pugh score (Child-Turcotte-Pugh A, 0; B, 1; C, 2). JIS score in the 93 patients was 0 (n = 2), 1 (n = 9), 2 (n = 26), 3 (n = 40) or 4 (n = 16).
LDLT was performed using a right lobe graft for 92 patients and a left lobe graft for 1 patient. Operative procedures for donor and recipient surgery have been described elsewhere.15, 16 Donors comprised 59 men and 44 women, with a median age of 41 years (range, 19–64 years). ABO blood-type matching was incompatible in 13 cases. Median graft-to-recipient body weight ratio was 1.03 (range, 0.73–1.69). As of the end of June 2005, median follow-up period was 24 months (range, 1–76 months).
The standard immunosuppression protocol comprised tacrolimus and low-dose steroid.17 However, 19 patients received steroid-free tacrolimus monotherapy as an induction procedure in an attempt to reduce HCC recurrence. Patients who received ABO blood-type incompatible transplants were treated with preoperative plasma exchange or double filtration plasmapheresis to reduce anti-A or anti-B antibody titer. During the first 3 weeks postoperatively, prostaglandin E1 and additional steroids were administered via the portal vein or hepatic artery.18 Acute rejection episodes were documented based on liver histology and treated with steroid bolus if moderate or severe.
Differences in qualitative variables were assessed using the Fisher exact or chi-square test, while differences in quantitative variables were analyzed using the Mann-Whitney test. Cumulative probability curves of survival or HCC recurrence were calculated using Kaplan-Meier methods, and differences between these curves were compared using the log-rank test. Any variable identified as significant (P < 0.05) in univariate analysis using the log-rank test was considered a candidate for multivariate analysis with Cox's proportional hazard regression model. Values of P < 0.05 were considered statistically significant. All statistical analyses were performed using the StatView 5 statistical software package (Abacus Concepts, Berkeley, CA).
As of the end of June 2005, a total of 64 patients remained alive. Cause of death included recurrent HCC (n = 10), sepsis (n = 7), pneumonia (n = 5), peritonitis (n = 4), chronic rejection (n = 1), and other tumor-unrelated causes (n = 2). Overall patient survival rate at 4 years was 64%. Survival rate tended to be better for hepatitis B surface antigen-positive patients (75%) than for hepatitis C virus-positive patients (62%), although no significant difference was identified (P = 0.4505). None of the recipient and donor variables listed in Table 1, including MELD score, history of pretransplant treatment, or ABO blood-type compatibility, were significantly associated with long-term postoperative survival rate.
Table 1. Preoperative Variables and Patient Survival (Univariate Analysis)
Number of dead patients: HCC-related death/HCC-nonrelated death.
A or B
Graft-to-recipient body weight ratio.
Analysis of preoperative tumor characteristics and patient 4-year survival (Table 2) revealed significantly lower survival rates in patients with tumors >5 cm in diameter than in patients with smaller tumors (20% vs. 69%, respectively; P = 0.0259). Unexpectedly, survival rates were similar for patients who met MC and those who did not (68% vs. 59% at 4 years, respectively; P = 0.6548) (Fig. 1). Patients with TNM stage III also tended to display poorer survival rates than patients with less advanced tumors, although no significant difference was identified (Table 2). With respect to JIS score, patients with a JIS score of 4 showed the lowest survival rate (58%), but no significant association was observed between JIS score and survival rate.
Table 2. Univariate Analysis of Patient Survival and Recurrence Rate According to Preoperative Tumor Characteristics
4-Year Survival Rate
4-Year Recurrence Rate
Abbreviations: TNM, tumor, node, metastasis; NA, not applied.
TNM stage: T factors are (1) single, (2) <2 cm, and (3) no vascular involvement. T1, fulfilling 3 factors; T2, fulfilling 2 factors; T3, fulfilling 1 factor; T4, fulfilling 0 factors. Stage I, T1N0M0; Stage II, T2N0M0; Stage III, T3N0M0; Stage IV-A, T4N0M0 or any T N1M0; Stage IV-B, any T N0-1M1.
As of the time of writing, postoperative recurrence of HCC had occurred in 16 patients. First sites of recurrence comprised lung (n = 5), bone (n = 3), graft liver (n = 3), abdominal lymph nodes (n = 2), adrenal gland (n = 1), brain (n = 1), and right subphrenic space (n = 1). After excluding death without recurrence, overall cumulative recurrence rate was 25% at 4 years.
Univariate analysis of risk factors for recurrence showed that recurrence rates did not differ significantly between hepatitis B surface antigen- and hepatitis C virus-positive patients (32% vs. 19%, respectively; P = 0.3416) or patients with and without steroid-free induction therapy (6% vs. 31%, respectively; P = 0.1076). Among preoperative tumor variables determined by imaging studies (Table 2), tumor diameter >5 cm, exceeding the MC and serum α-fetoprotein levels ≥400 ng/mL all represented significant predictors of higher recurrence rate. Recurrence rate was significantly higher in patients who exceeded MC than in patients who met these criteria (35% vs. 15%, respectively; P = 0.0190) (Fig. 2). However, multivariate analysis revealed that none of these variables represented significant independent risk factors (data not shown).
MC at First Diagnosis of HCC and Rates of Patient Survival and Recurrence after LDLT
For the 58 patients who had a history of previous nontransplant treatment for HCC and for whom records of tumor stage prior to any treatment were available, patient survival and HCC recurrence rates after LDLT were determined according to MC based on imaging at first diagnosis of HCC. Survival rates were similar for patients who had met MC (n = 42) and those who had not (n = 16) (71% vs. 69% at 4 years after LDLT, respectively; P = 0.8259) (Fig. 3A). Conversely, recurrence rate tended to be higher in patients who had exceeded MC than in patients who had met these criteria, although no significant difference was identified (37% vs. 13%, respectively; P = 0.0699) (Fig. 3B).
Pathological Findings and Recurrence
Correspondence of preoperative MC with the results of pathological analysis was evaluated in postoperative examinations of explanted livers. Completely ablated foci (100% necrosis) following pretransplant treatments were not counted as tumor. On pathological examination, 39 patients met so-called “pathological MC” and 54 did not. Accordingly, preoperative imaging underestimated the diagnosis in 13 patients (14%) and overestimated in 3 patients (3%). Of the 13 underestimated patients, HCC recurred in 4 patients. Univariate analysis of pathological findings (Table 3) revealed that tumor diameter >5 cm, tumor number ≥4, tumor grade indicating poor differentiation, and positive microvascular invasion were all significantly associated with postoperative recurrence. Using Cox's multivariate analysis (Table 4), pathological tumor number and poor differentiation were identified as independent risk factors for recurrence.
Table 3. Univariate Analysis of Pathological Tumor Characteristics and Recurrence Rate
Table 4. Multivariate Analysis of Pathological Tumor Factors and Recurrence
95% Confidence Interval
Well or Moderate
Effects of Previous Treatment Before LDLT
Patients were divided into 3 groups based on history of conventional treatment for HCC before LDLT: patients who primarily received LDLT without previous treatment (Group 1, n = 20); patients with 1–2 treatments (Group 2, n = 30); and patients with ≥3 treatments (Group 3, n = 43). Median period between first diagnosis of HCC and LDLT was 3 months in Group 1 (range, 1–15 months), 14 months in Group 2 (range, 2–70 months), and 36 months in Group 3 (range, 4–168 months). Preoperative liver function was most deteriorated in Group 1. Mean (± SD) preoperative MELD score was significantly higher in Group 1 (21 ± 8) than in Groups 2 (16 ± 7, P = 0.0414) or 3 (12± 6, P = 0.002). Conversely, the proportion of patients who exceeded MC before LDLT was significantly higher in Group 3 (63%) than in Groups 1 (35%) or 2 (33%, P = 0.020).
Based on postoperative examinations of explanted livers, microscopic venous invasion was significantly more frequent in Group 3 (51%) than in Groups 1 (21%) or 2 (33%, P = 0.0382). Postoperatively, HCC recurrence occurred in 3 patients in Group 1, 3 patients in Group 2, and 10 patients in Group 3. Survival rates at 4 years tended to be better for Group 2 (80%) than for Groups 1 (52%) and 3 (58%), although these differences were not significant (P = 0.0651 and P = 0.1042, respectively; Fig. 4A). Recurrence rates at 4 years were 23% for Group 1, 9% for Group 2, and 37% for Group 3. Rates were significantly lower for Group 2 than for Group 3 (P = 0.0411; Fig. 4B).
The present study showed that advanced stage of tumors in terms of tumor size, number, and α-fetoprotein levels on preoperative evaluation is associated with increased postoperative recurrence. These results corroborate the findings of previous studies on deceased donor liver transplantation.19–22 In a study by Gondolesi et al.23 of 36 patients with HCC treated using LDLT, tumor size >5 cm showed no significant effect on recurrence. Similarly, in our preliminary report on 56 patients,12 recurrence rate did not differ between patients within and beyond MC. However, as patient numbers increased, the present study demonstrated that patients with tumors >5 cm in diameter experienced significantly higher rates of recurrence and mortality. In addition, patients who exceeded MC displayed significantly higher risk of recurrence.
Of note was the finding that patient survival rates were similar for patients who met MC and for patients who did not (Fig. 1). This discrepancy may be due to a short follow-up period, but it can also be explained by HCC-unrelated deaths due to postoperative complications occurring within a few months after LDLT. Infectious complications such as sepsis, pneumonia, and peritonitis were the most common causes of early mortality. As a rule, patients with early HCC were referred to our hospital for LDLT due to advanced liver cirrhosis. Mortality rates within 3 months were 18% for those within MC and 9% for those beyond MC, which may be associated with poorer preoperative condition for the former group as reflected by higher MELD score (18 ± 8 vs. 13 ± 6, respectively; P = 0.002). Likewise, most patients who primarily received LDLT without previous treatment (Group 1, Fig. 4) were characterized by higher MELD scores and less advanced tumors. Survival rates for these patients were compromised by HCC-unrelated death due to perioperative complications.
In the present study, 78 patients had received nontransplant treatments for HCC prior to LDLT. For these patients, tumor staging was determined by counting only viable and enhancing nodules on pretransplant computed tomography. Pretransplant tumor staging thus differed from that at first diagnosis of HCC. In the 58 patients for whom records of tumor stage prior to any treatment were available, 6 patients with tumors originally beyond MC had been down-staged with treatment and were assigned to the within-MC group before LDLT. Conversely, 18 patients were up-staged before LDLT. In an attempt to analyze outcomes according to primary tumor staging, survival and recurrence curves were determined for patients meeting and patients exceeding MC based on imaging at first diagnosis of HCC for the 58 patients (Fig 3). While patient survival rates did not differ significantly, recurrence rate tended to be higher in patients who had originally exceeded MC than in patients who had met MC. However, these differences were not significant, probably due to small number of patients involved.
Another controversy remains regarding indications for patients with early HCC. On the one hand, use of LDLT has been proposed for patients with early HCC accompanied by early-stage cirrhosis (Child-Turcotte-Pugh A), as a reasonable survival rate should be expected in weighing the risk of live donation.24 Conversely, risking the life of a live donor for a patient who has alternative options of hepatic resection or other curative treatments with comparable long-term survival is unlikely to be ethically acceptable.25 Patient selection in our hospital is based on the latter policy: Patients with HCCs considered unsuitable for resection or local ablation therapies have been included in our program. In preoperative evaluations, a system for precise prognostic staging is essential for comparing outcomes between groups undertaking different therapeutic trials. Kudo et al.14 recently reported that JIS score offers a better system of prognostic staging for HCC than previous systems, in terms of both stratification ability and prognostic predictive power. In an analysis of 4,525 patients with HCC who received various conventional therapies, including hepatic resection, percutaneous ablation therapies and transarterial chemoembolization, patient survival was clearly stratified according to JIS score. The 5-year survival rates for patients were 73% for JIS 0 (n = 552), 52% for JIS 1 (n = 1,399), 33% for JIS 2 (n = 1,471), 13% for JIS 3 (n = 757) and 2% for JIS 4 (n = 244).14 This score system has not yet been validated outside Japan,26 but it was applied to our series of LDLT recipients. As a result, both 4- and 5-year survival rates were 58%, even in JIS 4 patients. Although these figures should be carefully compared, LDLT may result in improved prognosis for patients with a JIS score ≥2.
Conversely, for patients with early HCC and preserved liver function (that is, JIS score of 0 or 1), hepatic resection or ablation therapies would represent the treatments of choice. However, high rates of recurrence occur even after curative treatment,27 and the role of secondary or salvage transplantation has recently been discussed.24, 25, 28, 29 Although prior hepatic resection may complicate the operative transplant procedure and increase the risk of postoperative complications,24 postoperative survival in the 11 patients who received hepatic resection before LDLT was comparable to survival in the other 82 patients in this series (Table 1). On the other hand, transplantability at the time of recurrence is supposedly limited due to advanced tumor extension.24 Even for patients who are considered eligible for salvage LDLT, the present study revealed significantly higher recurrence rates for patients with a history of ≥3 treatments for HCC before LDLT (Group 3) than for patients with only 1–2 treatments (Group 2, Fig. 4B). This result implies that repeated nontransplant treatments for recurrent HCC may increase the risk of microscopic vascular invasion and impair the survival advantages conferred by LDLT. For patients who develop HCC recurrence after conventional therapies, feasibility, optimal timing, and efficacy of LDLT as a second-line treatment should be determined in further studies.
In conclusion, although efforts to decrease early mortality due to surgical complications are essential to improve outcomes for LDLT, this technique may constitute an optional treatment with a chance of cure for patients displaying otherwise uncontrolled HCC. Patients who develop recurrent HCC should be referred for LDLT before repeated nontransplant therapies, if a living donor is available.