Sultan's Score: A Novel Predictive Score to Predict Complete Response Following Drug-Eluting Bead Chemoembolization.

Abstract

Background: Transcatheter arterial chemoembolization (TACE) is a well-known standard treatment for hepatocellular carcinoma (HCC); however, the effectiveness of this treatment can vary among patients.

Objectives: This study aimed to develop a novel imaging-based prediction score (Sultan's score) to predict complete response after treatment.

Methods: From January 2015 to 2021, 41 patients with solitary HCC, admitted at King Saud University Medical City, Riyadh, Saudi Arabia, were treated with drug-eluting TACE (DEBTACE). Clinical data, tumor details, treatment specifics, and outcomes were gathered retrospectively. Sultan's score incorporates five imaging-based elements, namely, well-defined tumor borders, presence of tumor capsule, tumor size, arterial hyper enhancement, and hypertrophic arterial feeder. The cut-off value of Sultan's score was determined by calculating the maximum Youden index using the receiver operating characteristic curve to accurately predict complete treatment response to DEBTACE.

Results: Following two DEBTACE sessions, 28 (68.3%) and 13 (31.7%) patients showed complete and partial responses, respectively. The mean ± standard deviation and median (interquartile range) of the Sultan's score in patients with a complete treatment response were 3.93 ± 0.72 and 4 (4-4), and the corresponding values in patients with a partial response were 2.77 (0.93) and 3 (2-3.5), respectively. A cut-off value of 3.5 for the Sultan's score had a sensitivity and specificity of 78.6% and 76.9%, respectively, in predicting a complete treatment response. The area under the curve was 0.827 (95% confidence interval: 0.688-0.966).

Conclusions: We developed a novel imaging-based scoring system (Sultan's score) for predicting complete response in patients with HCC following DEBTACE.