Development and validation of a risk score to predict the overall survival following surgical resection of hepatocellular carcinoma in non-cirrhotic liver

Bobby V M Dasari, Yuk Ting Ma

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)

Abstract

BACKGROUND:
The aim of this study was to develop and validate a risk score to predict overall survival (OS) in patients undergoing surgical resection for hepatocellular carcinoma in non-cirrhotic liver (NC-HCC).
METHODS:
Patients who underwent resection for NC-HCC between 2004 and 2013 were identified from the SEER database. A derivation set of 75% of this cohort was used to develop a risk score. This was then internally validated on the remaining patients, and externally validated using a cohort of patients from The HPB Unit, Birmingham, UK.
RESULTS:
A total of 3897 patients were included from the SEER database, with a median post-diagnosis survival of 59 months. In the derivation set, multivariable analyses identified male sex, increasing tumour size, the presence of multiple tumours, bilobar tumours and major vascular invasion as adverse prognostic factors. A risk score generated from these factors was significantly predictive of OS, and was used to classify patients into low, medium and high-risk groups. These groups had a five-year OS of 69%, 51% and 19% in the internal, and 73%, 50% and 45% in the external validation sets.
CONCLUSION:
The proposed risk score is useful in the selection, pre-operative consenting and counselling of patients for surgery and to allow patients to make an informed decision regarding treatment.
Original languageEnglish
JournalHPB : the official journal of the International Hepato Pancreato Biliary Association
Early online date12 Aug 2019
DOIs
Publication statusE-pub ahead of print - 12 Aug 2019

Fingerprint

Dive into the research topics of 'Development and validation of a risk score to predict the overall survival following surgical resection of hepatocellular carcinoma in non-cirrhotic liver'. Together they form a unique fingerprint.

Cite this