Page 80 - Read Online

Ma et al. Hepatoma Res. 2026;12:43 Page 11 of 22

Evaluate whether
The dynamic patients are
Univariate changes of liver suitable for
AST increase > analysis, function after the continuing TACE,
25%, Child-Pugh multivariate Cox first TACE were
ART score [49] 2013 focusing on the OS
grade increase, regression first introduced as dynamic changes of
tumor response (backward the basis for liver function
elimination) retreatment deterioration and
decisions
tumor response
Used for prognostic
Univariate
Baseline AFP level, Retreatment evaluation before
BCLC stage, Child- analysis, decision score the second TACE to
ABCR score [53] 2015 Pugh grade multivariate Cox combining baseline identify patients OS
increase, tumor regression characteristics and who are not
response (backward treatment response suitable for
elimination)
continuing TACE
Based on the ART To guide the
score, it was decision-making of
Univariate optimized for TACE retreatment
Child-Pugh grade analysis, Chinese patients in Chinese HCC
increase, BCLC multivariate Cox
mART score [51] 2016 with HBV patients and OS
stage, tumor regression background, improve the
response (forward deleted “elevated accuracy of
selection)
AST,” and added prognosis
“BCLC stage” prediction
Dynamic Baseline AFP level,
prediction Child-Pugh grade,
model diameter of main
lesion, number The time-series
and size of hepatic clinical data were
lesions, vascular Time series data Incorporating time transformed into a
invasion, distant slicing, Cox- series data and visual tree survival Dynamic
Cascade survival 2018 metastasis, based feature simplifying survival path to achieve OS and
map [54] selection trajectories to treatment
vascular dynamic prognosis
invasion/N1/M1, (backward guide subsequent tracking and nodes
change of lesions, elimination) treatment treatment node
performance identification
status (all the
variables were
dichotomized)
Evaluate whether
patients are
suitable for
Univariate continuing TACE,
analysis, Treatment focusing on the
tumor number, multivariate Cox response (assessed dynamic changes of
tumor size, AFP, by mRECIST after
Post-TACE-predict regression tumor response
model [48] 2020 Bilirubin, vascular (backward the first TACE) was (mRECIST after OS
invasion, mRECIST elimination) with introduced as a first TACE), while
response prognostic variable
cluster-robust also incorporating
standard errors for the first time baseline tumor
burden, AFP,
bilirubin, and
vascular invasion