| Background and ObjectiveTransarterial chemoembolization(TACE)is one of the important treatments for intermediate to advanced hepatocellular carcinoma(HCC),but in some patients,TACE treatment is ineffective,and tumors continue to progress.To avoid repeated ineffective TACE treatment,the concept of TACE refractoriness has been developed.Although Chinese scholars first promulgated the Chinese version of the criteria for determining TACE resistance in 2022,the criteria developed by the Japanese Society of HepatologyLiver Cancer Study Group of Japan(JSH-LCSGJ)are more widely used.However,because of the late detection and large tumor load of Chinese HCC patients,the JSHLCSGJ’s version of TACE resistance are often ineffective in guiding treatment in Chinese HCC patients.It is controversial to determine whether TACE resistance occurs by assessing the efficacy after two consecutive TACE treatments.Therefore,this study intends to retrospectively investigate the effect of early TACE resistance on survival in HCC patients.In addition,it also aims to investigate for the first time whether viable lesions≥50%after two consecutive TACE treatments affects the prognosis of patients with hepatocellular carcinoma.MethodsThe clinical data of patients with primary hepatocellular carcinoma who received initial treatment with TACE at the Department of Interventional Therapy,Cancer Hospital of Chinese Academy of Medical Sciences between January 2014 and August 2017 were analyzed to determine early TACE resistance based on the JSH-LCSGJ(2021 version).Propensity score matching was performed using a 1:1 ratio(early TACE resistance group vs.no TACE resistance group).Survival analysis was performed using the Kaplan-Meier method with the Log-rank test,and prognosis-related risk factors were clarified by univariate and multivariate Cox proportional risk models.ResultsA total of 323 HCC patients were included in the study.51.08%of HCC patients(n=165)showed early TACE resistance and 48.92%of patients(n=158)were not.After propensity score matching at a ratio of 1:1,a total of 120 patients were matched in each group.The median overall survival(OS)of patients in the early T ACE-resistant group was significantly shorter than that of patients in the non-T ACE-resistant group[21 months(95%CI:15.7-26.3)vs.34 months(95%CI:27.5-40.5),P=0.001].Among them,subgroup analysis suggested that the median OS was shorter in stage A patients who developed early TACE resistance than in stage A patients without TACE resistance[28 months(95%CI:22.0-34.0)vs.57 months(95%CI:51.3-62.7),P<0.001)].Similarly,the median OS was shorter in patients with early TACE resistance in stage B than in patients without TACE resistance[20 months(95%CI:16.1-23.9)vs.36 months(95%CI:29.2-42.8),P=0.002].However,there was no statistically significant difference in the effect of the presence or absence of early TACE resistance on survival among stage C patients[14 months(95%CI:6.1-21.9)vs.25 months(95%CI:19.3-30.7),P=0.398].Of the 165 patients who developed early TACE resistance,a total of 38 patients were considered to have early TACE resistance only because of viable lesions≥50%after two consecutive TACE treatments,and this group of patients was compared with patients without TACE resistance and matched for propensity score in a 1:1 ratio.The difference in median OS between the two groups of patients was not statistically significant[35 months(95%CI:21.6-48.5)vs.31 months(95%CI:25.4-36.6),P=0.611].Multivariate Cox regression analysis suggested that Barcelona Clinic Liver Cancer staging(BCLC),tumor size,tumor capsule,tumor bilobar distribution,α-fetoprotein level(AFP),and early TACE resistance were independent risk factors for prognosis in HCC patients.ConclusionsEarly TACE refractoriness may shorten the OS of HCC patients.However,viable lesions≥ 50%after two consecutive TACE treatments did not impair the survival of patients.It may be inappropriate to consider these patients as having developed TACE refractoriness.Background and ObjectiveThe concept of TACE resistance was developed to effectively avoid ineffective and repetitive transarterial chemoembolization(TACE)therapy leading to deterioration of liver function in patients with primary hepatocellular carcinoma(HCC),and timely switch to other treatments.However,the current concept of TACE resistance is still a matter of debate and may impact the objective assessment of TACE efficacy.Therefore,it is important for clinical decision making to effectively predict the prognosis of HCC patients who continue received TACE therapy after the development of TACE resistance.In this study,we collected information on the clinical characteristics and enhanced magnetic resonance imaging(MRI,magnetic resonance imaging)radiomics features of whole liver of HCC patients before treatment and build a model using machine learning algorithms to effectively predict the prognosis of HCC patients who continue to receive TACE treatment after developing TACE resistance.MethodsPre-T ACE clinical data and pre-TACE enhanced MRI images of patients with primary HCC who underwent initial TACE treatment and developed TACE refractoriness at the Department of Interventional Therapy of our hospital were retrospectively collected from January 2014 to March 2018.The patients were divided into a training cohort and a test cohort according to the ratio of 7:3.Tumor and whole liver regions were outlined on the enhanced MRI portal phase images using 3D slicer software(version 5.0.3),and 1688 radiomics features were extracted separately using pyradiomic software package(https://pyradiomics.readthedocs.io).The models were trained by random forest(RF),Cox regression and support vector machine(S VM)algorithms using tumor features and whole liver features before treatment,respectively,and the prediction performance of the models were compared using c-index,receiver operating characteristic(ROC)curves,and decision curve analysis(DCA).ResultsA total of 111 patients with HCC were included in the study,77 in the training cohort and 34 in the validation cohort.There were 68 males and 9 females in the training cohort,with a mean age of 58.27 years;31 males and 3 females in the validation cohort,with a mean age of 55.53 years.Univariate Cox regression analysis suggested that Barcelona clinic liver cancer stage(BCLC),tumor burden(ratio of tumor to liver volume),body mass index(BMI),alpha-fetoprotein(AFP),and vascular invasion were strongly associated with prognosis.After dimensionality reduction by the Least absolute shrinkage and selection operator(LASSO)algorithm,five tumor radiomics features and 12 whole liver radiomics features were selected,respectively,Among them,the model constructed based on tumor burden&whole liver radiomics features&clinical features had the highest predictive efficacy,with c-indexes of 0.85 and 0.80 and areas under the ROC curve(AUCs)of 0.96 and 0.83 in the training and test cohorts,respectively.Based on whole liver(tumor&normal liver)radiomics score(TLrad-score),the patients were divided into low and high scoring subgroups based on the cut-off value(-0.34),and the median overall survival(mOS)of the low scoring subgroup was significantly better than that of the high scoring subgroup(P=0.0002).-Patients were divided into low tumor burden group and high tumor burden group according to the tumor burden cut-off value(0.18),and the mOS of low tumor burden group was also significantly better than that of high tumor burden group(P=0.01).ConclusionsA model based on tumor burden&whole liver radiomics features&clinical features has important value for accurately predicting the prognosis risk of HCC patients who continue receive TACE treatment after developing TACE refractoriness. |
PDF Full Text Request