Clinical Significance of Biliary Invasion at Diagnosis in Barcelona Clinic Liver Cancer Stage B-C Hepatocellular Carcinoma: A Nationwide Cohort Analysis in South Korea.

INTRODUCTION

Evidence of biliary invasion as a prognostic factor in patients with hepatocellular carcinoma (HCC) is unclear. We aimed to verify the significance of clinically diagnosed biliary involvement in patients with Barcelona Clinic Liver Cancer stage B-C (BCLC B-C) HCC.

METHODS

The Korean Liver Cancer Study Group randomly extracted data of patients with HCC enrolled in the Korean Central Cancer Registry between 2011 and 2016 from approximately 50 hospitals nationwide. After excluding records without information regarding serum bilirubin level, alpha-fetoprotein (AFP) level, and Child-Pugh class, a pre-propensity score matching cohort comprising 4,077 patients was included. Considering age, sex, body mass index, viral cause, serum bilirubin level, AFP, Child-Pugh class, tumor size, multiplicity, portal invasion, and extrahepatic metastasis, patients with and without bile duct invasion at initial imaging diagnosis were matched at a ratio of 1:2 from the pre-propensity score matching cohort to form a matched cohort (propensity score matching cohort).

RESULTS

The pre-propensity score matching cohort included 4,077 patients with BCLC B-C and 165 (4.0%) with biliary invasion at diagnosis. Regarding biliary invasion at diagnosis, 1- and 2-year overall survival (OS) rates were 41.2% and 29.1% (with invasion) and 54% and 40.9% (without invasion), respectively (p < 0.0001). Corresponding cancer-specific survival (CSS) rates at 1 and 2 years were 43.4% and 30.7% (with invasion) and 56.6% and 44% (without invasion), respectively (p < 0.0001). Although biliary invasion was a significant factor affecting overall and CSS rates in a univariate analysis, it was not statistically significant in multivariate analyses for overall (p = 0.153) and cancer-specific (p = 0.198) survival rates. The propensity score matching cohort included 165 patients with biliary invasion at diagnosis and 330 without biliary invasion. In the propensity score matching cohort, biliary invasion at diagnosis was not a significant factor affecting overall (p = 0.603) or cancer-specific (p = 0.960) survival rates in the univariate analyses. One- and 2-year OS were 41.2% and 29.1% (with invasion) and 36.1% and 28.2% (without invasion), respectively. The corresponding CSS at one and 2 years were 43.4% and 30.7% (with invasion) and 39.8% and 31.4% (without invasion), respectively. Multivariate analyses revealed that AFP levels, Child-Pugh class, tumor singularity, tumor size, portal invasion, lymph node metastases, and distant metastases significantly affected both overall and CSS rates.

CONCLUSION

Biliary invasion at diagnosis in patients with BCLC B-C does not affect overall or CSS rates; however, other prognostic factors associated with biliary invasion could have a greater impact.

INTRODUCTION

Evidence of biliary invasion as a prognostic factor in patients with hepatocellular carcinoma (HCC) is unclear. We aimed to verify the significance of clinically diagnosed biliary involvement in patients with Barcelona Clinic Liver Cancer stage B-C (BCLC B-C) HCC.

METHODS

The Korean Liver Cancer Study Group randomly extracted data of patients with HCC enrolled in the Korean Central Cancer Registry between 2011 and 2016 from approximately 50 hospitals nationwide. After excluding records without information regarding serum bilirubin level, alpha-fetoprotein (AFP) level, and Child-Pugh class, a pre-propensity score matching cohort comprising 4,077 patients was included. Considering age, sex, body mass index, viral cause, serum bilirubin level, AFP, Child-Pugh class, tumor size, multiplicity, portal invasion, and extrahepatic metastasis, patients with and without bile duct invasion at initial imaging diagnosis were matched at a ratio of 1:2 from the pre-propensity score matching cohort to form a matched cohort (propensity score matching cohort).

RESULTS

The pre-propensity score matching cohort included 4,077 patients with BCLC B-C and 165 (4.0%) with biliary invasion at diagnosis. Regarding biliary invasion at diagnosis, 1- and 2-year overall survival (OS) rates were 41.2% and 29.1% (with invasion) and 54% and 40.9% (without invasion), respectively (p < 0.0001). Corresponding cancer-specific survival (CSS) rates at 1 and 2 years were 43.4% and 30.7% (with invasion) and 56.6% and 44% (without invasion), respectively (p < 0.0001). Although biliary invasion was a significant factor affecting overall and CSS rates in a univariate analysis, it was not statistically significant in multivariate analyses for overall (p = 0.153) and cancer-specific (p = 0.198) survival rates. The propensity score matching cohort included 165 patients with biliary invasion at diagnosis and 330 without biliary invasion. In the propensity score matching cohort, biliary invasion at diagnosis was not a significant factor affecting overall (p = 0.603) or cancer-specific (p = 0.960) survival rates in the univariate analyses. One- and 2-year OS were 41.2% and 29.1% (with invasion) and 36.1% and 28.2% (without invasion), respectively. The corresponding CSS at one and 2 years were 43.4% and 30.7% (with invasion) and 39.8% and 31.4% (without invasion), respectively. Multivariate analyses revealed that AFP levels, Child-Pugh class, tumor singularity, tumor size, portal invasion, lymph node metastases, and distant metastases significantly affected both overall and CSS rates.

CONCLUSION

Biliary invasion at diagnosis in patients with BCLC B-C does not affect overall or CSS rates; however, other prognostic factors associated with biliary invasion could have a greater impact.