Ablation of Hepatocellular Carcinoma: a Nationwide Study

Background Hepatocellular carcinoma (HCC) is the most common primary malignancy of the liver and accounts for 85-90% of liver cancer. The most important risk factor for HCC is cirrhosis of any etiology. Among other major risk factors for HCC are chronic viral infections with hepatitis C virus (HCV) and hepatitis B virus (HBV), chronic alcohol consumption and non-alcoholic steatohepatitis.

Three nationwide databases and medical records will be used to address these important knowledge gaps. First, the association between survival and recurrence after ablation and HCC tumor size will be investigated. Second, the survival and recurrence after ablation in patients with HCC related to HCV will be investigated. due to HCV is different from that in patients with HCC due to other reasons than HCV. The nationwide data in these databases will reflect general clinical practice in an unselected population and provide novel information.

Purpose

With data from the Danish Liver and Bile Duct Cancer Database (DLGCD), the Danish Database for Hepatitis B and C (DANHEP), the National hepatitis laboratory database (DANVIR), and medical records, prognostic factors for survival in a large cohort of patients with HCC treated with ablation therapy will be investigated with the following aims:

1. Investigate association between HCC tumor size and survival and recurrence after ablation therapy.
2. Investigate survival and recurrence after ablation therapy in patients with HCV-related HCC compared with HCC due to other etiologies.

Hypotheses

1. Survival and recurrence after ablation is lower in patients with large HCC tumors than in patients with small tumors. However, the cut-off in current Danish guidelines is too conservative.
2. The survival after ablation is lower in patients with HCC due to HCV than in patient with HCC due to other reasons.

Methods This study is based on data from three large, nationwide databases in Denmark. DLGCD is an ongoing prospective nationwide database for patients treated for HCC, cholangiocarcinoma, colorectal liver metastases, or non-colorectal liver metastases. The database was established in 2012 and started collecting data in 2013. So far, data from the DLGCD have not been published. To date, the database includes data on 5056 surgical procedures in 2975 patients with HCC. To study the effect of etiology on survival and recurrence after treatment for HCC, data from the DANHEP and DANVIR databases will be incorporated. The three databases can be linked together as they all use CPR number to identify patients. DLGCD does not contain any information on HCV. However, DANHEP and DANVIR are nationwide databases that contain detailed information on individuals with chronic viral hepatitis in Denmark. DANHEP is a database with ongoing prospective registration of adult patients treated at a hospital (outpatient or inpatient) with chronic viral hepatitis B or C, established in 2002. DANVIR is a laboratory database containing information about HCV test (bot positive and negative) since 1991. The observation period of the study will be from 2013 to 2022. With data from the DANHEP and DANVIR databases, patients with HCC that have been infected with HCV can be identified to study the effect of ablation in HCV-infected patients versus patients without HCV infection.

Statistics The dataset will be described with frequencies and percentages for categorical variables and means or medians for continuous variables. For investigating differences, Student's t -test or Mann-Whitney U for continuous data and Pearson chi2 or Fisher's exact tests for categorical data will be used as appropriate. To determine risk factors affecting prognosis, recurrence and survival outcomes after ablation, Cox proportional hazard regression model will be used to calculate hazard ratios with 95% confidence intervals in both univariable and multivariable analyses. The multivariable analysis will be adjusted for age, sex, number of tumor, size of largest tumor, HCV-status, HBV-status, and Child-Pugh score. In addition, survival will be estimated by Kaplan-Meier methods with log-rank tests for HCV vs. non-HCV groups. It is possible that some patients with HCC never had a test for HCV. In a sensitivity analysis, only patients with HCC that had a HCV test (positive or negative) will be included. For recurrence we will use Fine-Gray proportional subhazards model and Aalen-Johansen estimator with death and transplantation as competing risks. The level of statistical significance will be set to p<0.05. All statistical analyses will be performed using R, version 4.0.3.

Perspectives With data from nationwide databases, the aim is to investigate survival and recurrence for HCC and the significance of tumor size and etiology of HCC in a large group of patients. Ablation has been widely used for decades, but the majority of published studies within this area are from Asia and based on small cohorts of highly selected patients. Furthermore, to our knowledge, no previous larger studies have determined the effect of ablation in patients with HCV-related HCC. Evidence will be improved with a large, Danish cohort reflecting outcomes in general clinical practice in unselected patients. The study has potential to contribute to the future guidelines regarding indication for ablation.