- Email: [email protected]
Oral Medications Improve the Compliance to Regular Surveillance for Hepatocellular Carcinoma in Chronic Hepatitis B Patients
POSTER PRESENTATIONS p = .95) or ALBI grade 3 (mOS 3.5 vs 3.2, HR = 0.99; 95% CI 0.49– 2.00, p = .99). In the group with ALBI grade 1 and an AFP ≥ 400 ng/mL (N = 91), the mOS was 9.4 months on the ramucirumab arm and 4.9 months for placebo (HR = 0.60; 95% CI 0.38–0.94, p = .03); and with ALBI grade 2 and AFP ≥ 400 ng/mL (N = 176), mOS was 5.5 months on the ramucirumab arm and 4.0 months for placebo (HR = 0.77; 95% CI 0.55–1.06, p = .11). The ALBI grade 3 and AFP ≥ 400 ng/mL group was small (N = 20), limiting the assessment of survival benefit. Regardless of ALBI grade, no trend for survival improvement was observed in patients with AFP <400 ng/mL. Conclusions: In REACH, grouping by ALBI grade defined three patient populations with different prognosis. Unselected REACH patients with ALBI grade 1 were most likely to derive survival benefit from ramucirumab treatment. In patient populations with ALBI grade 1 and 2, those with a baseline AFP ≥ 400 ng/mL appeared to experience improved survival with ramucirumab treatment.
Received treatment by BCLC and HKLC stage is shown in Table 2.
Staging
Clinical
system
trial
Supportive Cyberknife
care
ablation RFA
Systemic therapy Receiving
Surgical Resection
Transplant TACE
Referral
Recommended Y90
Subtotal
11
139
Treatments (%)
BCLC stage A
0
7
6
28
3
32
28
24
73.3
B
1
3
8
3
5
7
15
3
11
56
26.8
C
0
5
21
0
21
14
12
0
10
83
25.3
D
0
1
20
0
3
2
1
6
1
34
58.8
2
2
20
1
19
14
11
8
77
63.7
HKLC stage I
0
IIa
0
3
3
9
1
1
7
12
4
40
42.5
IIb
0
4
2
0
2
19
13
2
7
49
38.8
1
2
3
6
1
21
IIIb
0
2
8
1
10
8
11
0
9
50
22
VIa
IIIa
1
1
5
0
8
3
2
0
2
21
38.1
VIb
0
0
1
6
0
6
0
3
0
0
1
0
0
9
28.6
100
Va
0
0
4
0
3
2
2
6
0
17
35.3
Vb
0
3
19
0
2
0
1
1
2
28
67.9
SAT-077 VALIDATION AND PERFORMANCE OF THE HONG KONG LIVER CANCER STAGING SYSTEM IN A WESTERN COHORT J. Umbel1, A. AlAhmadi1,2, S. Gebran3, C. Siegel1,4, P.M. Gholam1,4. 1 Digestive Health Institute, University Hospitals Case Medical Center, Cleveland, United States; 2Medicine, King AbdulAziz University, Jeddah, Saudi Arabia; 3Medicine, American University of Beirut, Beirut, Lebanon; 4 Transplant Institute, University Hospitals Case Medical Center, Cleveland, United States E-mail: [email protected]
Kaplan Meier survival curves for both systems showed a statistically significant decrement in survival by stage (log-rank p < 0.001). When prognostic performance was compared, HKLC offered higher homogeneity (LR and C-index) and lower AIC value compared to BCLC.
Staging
LR
AIC
Background and Aims: The Hong Kong Liver Cancer (HKLC) staging classification is validated in Asia where HBV drives hepatocellular carcinoma. Its proposed added value to the Barcelona Clinic Liver Cancer (BCLC) system is in identifying subsets that may benefit from more aggressive options. We sought to validate HKLC and compare its performance with BCLC for the first time in a real world Western cohort. Methods: 318/380 patients with full data who were managed by our Multidisciplinary Team between 1/2009 and 5/2015 were included. They were censored at time of death, last known follow up or transplant and analyzed by baseline BCLC and HKLC stage. KaplanMeier and Cox regression models were run to establish survival rates over time and assess prognostic performance of both systems using Akaike information criterion (AIC), concordance-index (c-index) and likelihood ratio. Results: Demographic, liver and tumor characteristics are shown in Table 1.
1
HKLC
122.7
2,041.2
2
BCLC
77.35
2,076.6
Age (y) Gender M:F Total bilirubin (mg/ dL) Creatinine (mg/dL) INR MELD Cirrhosis Hepatitis B/C Child Pugh A/B/C (%) Esophageal Varices Variceal Bleeding Hepatic Encephalopathy Ascites Spontaneous bacterial peritonitis Vascular invasion Extrahepatic disease Received Transplant
S694
62.8 (11.04) 243 (77.9%):69 (22.1) 2.06 (3.88) 1.154(1.1) 1.27(0.396) 11.5(5.79) 281 (90.1%) 15 (4.8)/180(57.7) 179(57.4%)/97(31.1%)/36(11.5%) 69 (22.1%) 27 (8.7%) 31 (9.9%) 86 (27.6) 6 (1.9%) 93 (29.8%) 35 (11.2%) 31 (10%)
Rank
Cindex (SE) 0.729 (0.023) 0.696 (0.021)
Conclusions: Though both systems have good survival prediction by stage, HKLC may provide more homogeneity across stages. This may translate into more tailored treatment allocation in Western hepatocellular carcinoma patients. SAT-078 ORAL MEDICATIONS IMPROVE THE COMPLIANCE TO REGULAR SURVEILLANCE FOR HEPATOCELLULAR CARCINOMA IN CHRONIC HEPATITIS B PATIENTS J.Y. Nam1, J.-H. Lee1, D.H. Lee1, Y. Chang1, H. Ahn1, H. Cho1, J.-J. Yoo1, M. Lee1, Y.Y. Cho1, E. Cho1, S.J. Yu1, Y.J. Kim1, J.-H. Yoon1. 1Department of Internal Medicine and Liver Research Institute, Seoul National University Hospital, Seoul, South Korea E-mail: [email protected] Background and Aims: The regular surveillance for hepatocellular carcinoma (HCC) in chronic hepatitis B (CHB) patients is essential to detect HCC earlier and to improve their prognosis. This study aimed to determine whether the prescription of oral medications (i.e., hepatotonics and antiviral agents), which might increase the psychological perceived severity according to the Health Belief Model, contributes to early tumor detection and overall survival. Methods: A total of 401 CHB patients who were newly diagnosed as HCC were included: 134 patients received no medications (group 1), 151 received hepatotonics such as ursodeoxycholic acid and silymarin (group 2), and 116 received nucleos(t)ide analogues (NAs) (group 3) within two years before the diagnosis of HCC. All patients have been educated about regular follow-up every 6 month. Primary endpoint was overall survival and secondary endpoints were compliance to regular surveillance and initial HCC status. Results: The rates of good compliance to regular surveillance (defined as >80% of visit intervals were <6 months) was significantly higher in both group 2 (95.0%) and 3 (96.7%) than in group 1 (63.4%) (all p < 0.001). HCC diagnosed in very early stage (BCLC stage 0) was significantly higher in both groups 2 (32.5%) and 3
Journal of Hepatology 2016 vol. 64 | S631–S832
POSTER PRESENTATIONS (36.2%) than in group 1 (20.9%) (all p < 0.05). HCC diagnosed in advanced or end stage (BCLC stage C or D) was significantly lower in both groups 2 (37.1%) and 3 (32.8%) than in group 1 (42.5%) (all p < 0.05). Mean maximal tumor size was significantly smaller in both groups 2 (1.9 ± 1.1 cm) and 3 (1.8 ± 0.9 cm) than in group 1 (2.8 ± 2.4 cm) (all p < 0.001). As compared to group 1, group 2 (HR, 0.63; 95% CI, 0.41–0.97; p = 0.032) as well as group 3 (HR, 0.40; 95% CI, 0.22– 0.71; p < 0.001) showed significantly longer overall survival (Figure 1).
Conclusions: The prescription of hepatotonics, as well as NAs, improves the patients’ compliance to HCC surveillance and, consequently, enables the early detection of HCC which is associated with survival gain. SAT-079 PERFORMANCE OF THE HAP SCORE AMONG PATIENTS WITH HEPATOCELLULAR CARCINOMA TREATED BY RADIOEMBOLIZATION J. Buades Mateu1, D. Martinez-Urbistondo1, M. De la Torre Alaez1, M. Iñarrairaegui1,2, J.I. Bilbao3, B. Sangro1,2. 1Liver Unit, Clinica Universidad de Navarra; 2IDISNA and CIBEREHD; 3Interventional Radiology, Clinica Universidad de Navarra, Pamplona, Spain E-mail: [email protected] Background and Aims: Patients with unresectable, unablatable hepatocellular carcinoma that have liver-limited disease are usually treated with chemoembolization. The HAP score was designed to establish the prognosis of these patients using 4 variables: albumin <36 g/dL, bilirubin >17 μmol/L, AFP > 400 ng/mL, tumor size >7 cm. The HAP score was shown to differentiate 4 groups of patients treated with chemoembolization with progressively worse prognosis. Radioembolization using resin microspheres loaded yttrium-90 is increasingly used in these patients, particularly those who are not good candidates for chemoembolization. We have therefore investigated the performance of the HAP score in this population. Methods: All consecutive patients with hepatocellular carcinoma treated by radioembolization from 2003 to 2012 were retrospectively analyzed. Patients not followed entirely in our center, those with a follow-up <3 months, and those in which the HAP score could not be calculated were excluded. Although the analysis is retrospective, most variables were prospectively recorded. Survival was plotted from the day of treatment until death or last visit using Kaplan-Meier
method and compared by log-rank test. Cox-regression analysis was used for multivariate analysis. Results: 116 patients analyzed had a mean age of 64 years and were predominantly males (82.8%), and cirrhotics (79%) in ChildPugh class A (85.3%). 26.3% had portal vein thrombosis and 26.7% had AFP > 400 UI/mL. Patients were in BCLC stages A (16.4%), B (53.4%) or C (30.2%) and HAP groups A (14.7%), B (31.9%), C (37.1%), and D (16.4%). Survival plots stratified by HAP score showed a statistical difference in overall survival ( p = 0.017). However, there were no differences between groups HAP A (median: 19.2 months) and HAP B (median: 21.3 months) or between groups HAP C (median: 7.9 months) and HAP D (median: 10.6 months). When the frequency of each component of the HAP score was compared, groups A and B differed significantly in the frequency of large tumor size (A:0% vs. B:27%) and high bilirubin (A:0% vs. B:46%), while groups C and D differed significantly in the frequency of large tumor size (C:44% vs. D:89%) and high AFP (C:32% vs. D:63%). Conclusions: The HAP scoring system has not shown a good prognostic accuracy in a large series of patients treated with radioembolization. A benefit of radioembolization compared to chemoembolization in patients with large tumors or high AFP may partially explain this impaired prognostic ability. SAT-080 HIGHER SERUM BILIRUBIN LEVELS OF LIVING LIVER DONORS MAY BE ASSOCIATED WITH DECREASED RISK OF POST-TRANSPLANT HEPATOCELLULAR CARCINOMA RECURRENCE S. Han1, J.D. Yang2, D.H. Sinn3, J. Ko1, J.M. Kim4, J.C. Shin1, H.J. Son5, J.-W. Joh4, G. Kim1. 1Anesthesiology and Pain Medicine, Samsung Medical Center, Seoul, South Korea; 2Medicine, Mayo Clinic College of Medicine, Rochester, United States; 3Medicine; 4Surgery, Samsung Medical Center, Seoul; 5Anesthesiology and Pain Medicine, Gangwon University School of Medicine, Chuncheon, South Korea E-mail: [email protected] Background and Aims: Serum bilirubin level, which may reflect the host defense against increased oxidative stress, is inversely associated with the risk of cancer development. In liver transplantation, the intrinsic bilirubin metabolism of donor liver is subsequently translated into recipient. Thus, we hypothesized that liver transplantation conducted with living donors with higher serum bilirubin reduces hepatocellular carcinoma recurrence. Methods: Two hundred fifty recipients who underwent living donor liver transplantation for treating hepatocellular carcinoma within the Milan criteria were included in the study. The association between donor preoperative total bilirubin concentration and the recurrence risk was analyzed using the competing risks Cox regression with the death as the competing event while adjusting for tumor biology including alpha-fetoprotein, number and size of nodules, microvascular invasion, and differentiation. Results: Donor preoperative total bilirubin concentration was 0.7 mg/dL in median and ranged from 0.2 to 2.7 mg/dL. Donor preoperative total bilirubin concentration was positively correlated with recipient post-transplant total bilirubin ( p < 0.01 at 3 months, 1 year, and 3 years after transplantation). The Cox model demonstrated that donor preoperative total bilirubin concentration was negatively associated with the recurrence risk in univariable analysis (hazard ratio 0.40, 95% confidence interval 0.18–0.92, p = 0.030) and multivariable analysis (hazard ratio 0.14, 95% confidence interval 0.04–0.53, p = 0.004). Conclusions: Our data give an evidence for bilirubin as a potent anticancer substance against post-transplant hepatocellular carcinoma recurrence. Further validation of the present study is warranted to verify the feasibility of the use of interventions targeting on post-transplant serum bilirubin level to decrease the recurrence risk.
Journal of Hepatology 2016 vol. 64 | S631–S832
S695
Received treatment by BCLC and HKLC stage is shown in Table 2.
Staging
Clinical
system
trial
Supportive Cyberknife
care
ablation RFA
Systemic therapy Receiving
Surgical Resection
Transplant TACE
Referral
Recommended Y90
Subtotal
11
139
Treatments (%)
BCLC stage A
0
7
6
28
3
32
28
24
73.3
B
1
3
8
3
5
7
15
3
11
56
26.8
C
0
5
21
0
21
14
12
0
10
83
25.3
D
0
1
20
0
3
2
1
6
1
34
58.8
2
2
20
1
19
14
11
8
77
63.7
HKLC stage I
0
IIa
0
3
3
9
1
1
7
12
4
40
42.5
IIb
0
4
2
0
2
19
13
2
7
49
38.8
1
2
3
6
1
21
IIIb
0
2
8
1
10
8
11
0
9
50
22
VIa
IIIa
1
1
5
0
8
3
2
0
2
21
38.1
VIb
0
0
1
6
0
6
0
3
0
0
1
0
0
9
28.6
100
Va
0
0
4
0
3
2
2
6
0
17
35.3
Vb
0
3
19
0
2
0
1
1
2
28
67.9
SAT-077 VALIDATION AND PERFORMANCE OF THE HONG KONG LIVER CANCER STAGING SYSTEM IN A WESTERN COHORT J. Umbel1, A. AlAhmadi1,2, S. Gebran3, C. Siegel1,4, P.M. Gholam1,4. 1 Digestive Health Institute, University Hospitals Case Medical Center, Cleveland, United States; 2Medicine, King AbdulAziz University, Jeddah, Saudi Arabia; 3Medicine, American University of Beirut, Beirut, Lebanon; 4 Transplant Institute, University Hospitals Case Medical Center, Cleveland, United States E-mail: [email protected]
Kaplan Meier survival curves for both systems showed a statistically significant decrement in survival by stage (log-rank p < 0.001). When prognostic performance was compared, HKLC offered higher homogeneity (LR and C-index) and lower AIC value compared to BCLC.
Staging
LR
AIC
Background and Aims: The Hong Kong Liver Cancer (HKLC) staging classification is validated in Asia where HBV drives hepatocellular carcinoma. Its proposed added value to the Barcelona Clinic Liver Cancer (BCLC) system is in identifying subsets that may benefit from more aggressive options. We sought to validate HKLC and compare its performance with BCLC for the first time in a real world Western cohort. Methods: 318/380 patients with full data who were managed by our Multidisciplinary Team between 1/2009 and 5/2015 were included. They were censored at time of death, last known follow up or transplant and analyzed by baseline BCLC and HKLC stage. KaplanMeier and Cox regression models were run to establish survival rates over time and assess prognostic performance of both systems using Akaike information criterion (AIC), concordance-index (c-index) and likelihood ratio. Results: Demographic, liver and tumor characteristics are shown in Table 1.
1
HKLC
122.7
2,041.2
2
BCLC
77.35
2,076.6
Age (y) Gender M:F Total bilirubin (mg/ dL) Creatinine (mg/dL) INR MELD Cirrhosis Hepatitis B/C Child Pugh A/B/C (%) Esophageal Varices Variceal Bleeding Hepatic Encephalopathy Ascites Spontaneous bacterial peritonitis Vascular invasion Extrahepatic disease Received Transplant
S694
62.8 (11.04) 243 (77.9%):69 (22.1) 2.06 (3.88) 1.154(1.1) 1.27(0.396) 11.5(5.79) 281 (90.1%) 15 (4.8)/180(57.7) 179(57.4%)/97(31.1%)/36(11.5%) 69 (22.1%) 27 (8.7%) 31 (9.9%) 86 (27.6) 6 (1.9%) 93 (29.8%) 35 (11.2%) 31 (10%)
Rank
Cindex (SE) 0.729 (0.023) 0.696 (0.021)
Conclusions: Though both systems have good survival prediction by stage, HKLC may provide more homogeneity across stages. This may translate into more tailored treatment allocation in Western hepatocellular carcinoma patients. SAT-078 ORAL MEDICATIONS IMPROVE THE COMPLIANCE TO REGULAR SURVEILLANCE FOR HEPATOCELLULAR CARCINOMA IN CHRONIC HEPATITIS B PATIENTS J.Y. Nam1, J.-H. Lee1, D.H. Lee1, Y. Chang1, H. Ahn1, H. Cho1, J.-J. Yoo1, M. Lee1, Y.Y. Cho1, E. Cho1, S.J. Yu1, Y.J. Kim1, J.-H. Yoon1. 1Department of Internal Medicine and Liver Research Institute, Seoul National University Hospital, Seoul, South Korea E-mail: [email protected] Background and Aims: The regular surveillance for hepatocellular carcinoma (HCC) in chronic hepatitis B (CHB) patients is essential to detect HCC earlier and to improve their prognosis. This study aimed to determine whether the prescription of oral medications (i.e., hepatotonics and antiviral agents), which might increase the psychological perceived severity according to the Health Belief Model, contributes to early tumor detection and overall survival. Methods: A total of 401 CHB patients who were newly diagnosed as HCC were included: 134 patients received no medications (group 1), 151 received hepatotonics such as ursodeoxycholic acid and silymarin (group 2), and 116 received nucleos(t)ide analogues (NAs) (group 3) within two years before the diagnosis of HCC. All patients have been educated about regular follow-up every 6 month. Primary endpoint was overall survival and secondary endpoints were compliance to regular surveillance and initial HCC status. Results: The rates of good compliance to regular surveillance (defined as >80% of visit intervals were <6 months) was significantly higher in both group 2 (95.0%) and 3 (96.7%) than in group 1 (63.4%) (all p < 0.001). HCC diagnosed in very early stage (BCLC stage 0) was significantly higher in both groups 2 (32.5%) and 3
Journal of Hepatology 2016 vol. 64 | S631–S832
POSTER PRESENTATIONS (36.2%) than in group 1 (20.9%) (all p < 0.05). HCC diagnosed in advanced or end stage (BCLC stage C or D) was significantly lower in both groups 2 (37.1%) and 3 (32.8%) than in group 1 (42.5%) (all p < 0.05). Mean maximal tumor size was significantly smaller in both groups 2 (1.9 ± 1.1 cm) and 3 (1.8 ± 0.9 cm) than in group 1 (2.8 ± 2.4 cm) (all p < 0.001). As compared to group 1, group 2 (HR, 0.63; 95% CI, 0.41–0.97; p = 0.032) as well as group 3 (HR, 0.40; 95% CI, 0.22– 0.71; p < 0.001) showed significantly longer overall survival (Figure 1).
Conclusions: The prescription of hepatotonics, as well as NAs, improves the patients’ compliance to HCC surveillance and, consequently, enables the early detection of HCC which is associated with survival gain. SAT-079 PERFORMANCE OF THE HAP SCORE AMONG PATIENTS WITH HEPATOCELLULAR CARCINOMA TREATED BY RADIOEMBOLIZATION J. Buades Mateu1, D. Martinez-Urbistondo1, M. De la Torre Alaez1, M. Iñarrairaegui1,2, J.I. Bilbao3, B. Sangro1,2. 1Liver Unit, Clinica Universidad de Navarra; 2IDISNA and CIBEREHD; 3Interventional Radiology, Clinica Universidad de Navarra, Pamplona, Spain E-mail: [email protected] Background and Aims: Patients with unresectable, unablatable hepatocellular carcinoma that have liver-limited disease are usually treated with chemoembolization. The HAP score was designed to establish the prognosis of these patients using 4 variables: albumin <36 g/dL, bilirubin >17 μmol/L, AFP > 400 ng/mL, tumor size >7 cm. The HAP score was shown to differentiate 4 groups of patients treated with chemoembolization with progressively worse prognosis. Radioembolization using resin microspheres loaded yttrium-90 is increasingly used in these patients, particularly those who are not good candidates for chemoembolization. We have therefore investigated the performance of the HAP score in this population. Methods: All consecutive patients with hepatocellular carcinoma treated by radioembolization from 2003 to 2012 were retrospectively analyzed. Patients not followed entirely in our center, those with a follow-up <3 months, and those in which the HAP score could not be calculated were excluded. Although the analysis is retrospective, most variables were prospectively recorded. Survival was plotted from the day of treatment until death or last visit using Kaplan-Meier
method and compared by log-rank test. Cox-regression analysis was used for multivariate analysis. Results: 116 patients analyzed had a mean age of 64 years and were predominantly males (82.8%), and cirrhotics (79%) in ChildPugh class A (85.3%). 26.3% had portal vein thrombosis and 26.7% had AFP > 400 UI/mL. Patients were in BCLC stages A (16.4%), B (53.4%) or C (30.2%) and HAP groups A (14.7%), B (31.9%), C (37.1%), and D (16.4%). Survival plots stratified by HAP score showed a statistical difference in overall survival ( p = 0.017). However, there were no differences between groups HAP A (median: 19.2 months) and HAP B (median: 21.3 months) or between groups HAP C (median: 7.9 months) and HAP D (median: 10.6 months). When the frequency of each component of the HAP score was compared, groups A and B differed significantly in the frequency of large tumor size (A:0% vs. B:27%) and high bilirubin (A:0% vs. B:46%), while groups C and D differed significantly in the frequency of large tumor size (C:44% vs. D:89%) and high AFP (C:32% vs. D:63%). Conclusions: The HAP scoring system has not shown a good prognostic accuracy in a large series of patients treated with radioembolization. A benefit of radioembolization compared to chemoembolization in patients with large tumors or high AFP may partially explain this impaired prognostic ability. SAT-080 HIGHER SERUM BILIRUBIN LEVELS OF LIVING LIVER DONORS MAY BE ASSOCIATED WITH DECREASED RISK OF POST-TRANSPLANT HEPATOCELLULAR CARCINOMA RECURRENCE S. Han1, J.D. Yang2, D.H. Sinn3, J. Ko1, J.M. Kim4, J.C. Shin1, H.J. Son5, J.-W. Joh4, G. Kim1. 1Anesthesiology and Pain Medicine, Samsung Medical Center, Seoul, South Korea; 2Medicine, Mayo Clinic College of Medicine, Rochester, United States; 3Medicine; 4Surgery, Samsung Medical Center, Seoul; 5Anesthesiology and Pain Medicine, Gangwon University School of Medicine, Chuncheon, South Korea E-mail: [email protected] Background and Aims: Serum bilirubin level, which may reflect the host defense against increased oxidative stress, is inversely associated with the risk of cancer development. In liver transplantation, the intrinsic bilirubin metabolism of donor liver is subsequently translated into recipient. Thus, we hypothesized that liver transplantation conducted with living donors with higher serum bilirubin reduces hepatocellular carcinoma recurrence. Methods: Two hundred fifty recipients who underwent living donor liver transplantation for treating hepatocellular carcinoma within the Milan criteria were included in the study. The association between donor preoperative total bilirubin concentration and the recurrence risk was analyzed using the competing risks Cox regression with the death as the competing event while adjusting for tumor biology including alpha-fetoprotein, number and size of nodules, microvascular invasion, and differentiation. Results: Donor preoperative total bilirubin concentration was 0.7 mg/dL in median and ranged from 0.2 to 2.7 mg/dL. Donor preoperative total bilirubin concentration was positively correlated with recipient post-transplant total bilirubin ( p < 0.01 at 3 months, 1 year, and 3 years after transplantation). The Cox model demonstrated that donor preoperative total bilirubin concentration was negatively associated with the recurrence risk in univariable analysis (hazard ratio 0.40, 95% confidence interval 0.18–0.92, p = 0.030) and multivariable analysis (hazard ratio 0.14, 95% confidence interval 0.04–0.53, p = 0.004). Conclusions: Our data give an evidence for bilirubin as a potent anticancer substance against post-transplant hepatocellular carcinoma recurrence. Further validation of the present study is warranted to verify the feasibility of the use of interventions targeting on post-transplant serum bilirubin level to decrease the recurrence risk.
Journal of Hepatology 2016 vol. 64 | S631–S832
S695