Dataset Information

Deep Learning Predicts Overall Survival of Patients With Unresectable Hepatocellular Carcinoma Treated by Transarterial Chemoembolization Plus Sorafenib.

ABSTRACT:

Objectives

To develop and validate a deep learning-based overall survival (OS) prediction model in patients with hepatocellular carcinoma (HCC) treated with transarterial chemoembolization (TACE) plus sorafenib.

Methods

This retrospective multicenter study consisted of 201 patients with treatment-naïve, unresectable HCC who were treated with TACE plus sorafenib. Data from 120 patients were used as the training set for model development. A deep learning signature was constructed using the deep image features from preoperative contrast-enhanced computed tomography images. An integrated nomogram was built using Cox regression by combining the deep learning signature and clinical features. The deep learning signature and nomograms were also externally validated in an independent validation set of 81 patients. C-index was used to evaluate the performance of OS prediction.

Results

The median OS of the entire set was 19.2 months and no significant difference was found between the training and validation cohort (18.6 months vs. 19.5 months, P = 0.45). The deep learning signature achieved good prediction performance with a C-index of 0.717 in the training set and 0.714 in the validation set. The integrated nomogram showed significantly better prediction performance than the clinical nomogram in the training set (0.739 vs. 0.664, P = 0.002) and validation set (0.730 vs. 0.679, P = 0.023).

Conclusion

The deep learning signature provided significant added value to clinical features in the development of an integrated nomogram which may act as a potential tool for individual prognosis prediction and identifying HCC patients who may benefit from the combination therapy of TACE plus sorafenib.

SUBMITTER: Zhang L

PROVIDER: S-EPMC7556271 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep Learning Predicts Overall Survival of Patients With Unresectable Hepatocellular Carcinoma Treated by Transarterial Chemoembolization Plus Sorafenib.

Zhang Lei L Xia Wei W Yan Zhi-Ping ZP Sun Jun-Hui JH Zhong Bin-Yan BY Hou Zhong-Heng ZH Yang Min-Jie MJ Zhou Guan-Hui GH Wang Wan-Sheng WS Zhao Xing-Yu XY Jian Jun-Ming JM Huang Peng P Zhang Rui R Zhang Shen S Zhang Jia-Yi JY Li Zhi Z Zhu Xiao-Li XL Gao Xin X Ni Cai-Fang CF

Frontiers in oncology 20200930

<h4>Objectives</h4>To develop and validate a deep learning-based overall survival (OS) prediction model in patients with hepatocellular carcinoma (HCC) treated with transarterial chemoembolization (TACE) plus sorafenib.<h4>Methods</h4>This retrospective multicenter study consisted of 201 patients with treatment-naïve, unresectable HCC who were treated with TACE plus sorafenib. Data from 120 patients were used as the training set for model development. A deep learning signature was constructed us ...[more]

PMID: 33102242

Similar Datasets

Project description:IntroductionSeveral clinical trials comparing the efficacy and safety of transarterial chemoembolization (TACE) plus molecular-targeted agents versus TACE alone revealed no clinical benefits in progression-free survival (PFS) or overall survival (OS). Here, we report the final OS analysis from the TACTICS trial, which previously demonstrated significant improvement in PFS with TACE plus sorafenib in patients with unresectable hepatocellular carcinoma (HCC) (NCT01217034).MethodsPatients with unresectable HCC were randomized to a TACE plus sorafenib group (N = 80) or a TACE alone group (N = 76). Patients in the combination treatment group received sorafenib 400 mg once daily for 2-3 weeks before TACE, followed by 800 mg once daily during on-demand conventional TACE sessions until time to untreatable progression. In this trial, TACE-specific PFS was used. TACE-specific PFS is defined as the time from randomization to progressive disease (PD) or death from any cause, and PD was defined as untreatable progression, caused by the inability of a patient to further receive or benefit from TACE for reasons that include intrahepatic tumor progression (25% increase vs. baseline) according to response evaluation criteria in cancer of the liver, the detection of extrahepatic spread, vascular invasion, or transient deterioration of liver function to Child-Pugh C after TACE.ResultsAt the cut-off date of July 31, 2020, 131 OS events were observed. The median OS was 36.2 months with TACE plus sorafenib and 30.8 months with TACE alone (hazard ratio [HR] = 0.861; 95% confidence interval [CI], 0.607-1.223; p = 0.40, ΔOS, 5.4 months). The updated PFS was 22.8 months with TACE plus sorafenib and 13.5 months with TACE alone (HR = 0.661; 95% CI, 0.466-0.938; p = 0.02). Post-trial treatments with active procedures/agents were received by 47 (58.8%) patients in the TACE plus sorafenib group and 58 (76.3%) in the TACE alone group (p = 0.01). In post hoc analysis, PFS and OS benefit were shown in HCC patients with tumor burden beyond up-to-7 criteria.ConclusionsIn TACTICS trial, TACE plus sorafenib did not show significant OS benefit over TACE alone; however, clinical meaningful OS prolongation and significantly improved PFS was observed. Thus, the TACE plus sorafenib can be considered a choice of treatment in intermediate-stage HCC, especially in patients with high tumor burden. Trial Registration: NCT01217034.

Project description:BackgroundTransarterial chemoembolization (TACE) is the recommended treatment for hepatocellular carcinoma (HCC) patients at Barcelona Clinic Liver Cancer (BCLC) B-stage, whereas sorafenib is an orally administered small molecule target drug for BCLC C-stage. This updated systemic review and meta-analysis focuses on identifying the efficacy of the combination of TACE with sorafenib, which remains controversial despite years of exploration.MethodsPubMed, EMBASE, Scopus and the Cochrane Library were systematically reviewed to search for studies published from January 1990 to May 2017. Studies focusing on the efficacy of combination therapy for unresectable HCC were eligible. The hazard ratio (HR) with 95% confidence intervals (95% CIs) for time to progression (TTP), overall survival (OS), disease control rate (DCR) and aetiology were collected. The data were then analysed through fixed/random effects meta-analysis models with STATA 13.0. The incidence and severity of treatment-related adverse events (AEs) were also evaluated.ResultsTwenty-seven studies were included. Thirteen non-comparative studies reported median OS (ranging from 18.5 to 20.4 months), median TTP (ranging from 7 to 13.9 months) and DCR (ranging from 18.4 to 95%). Fourteen comparative studies provided median OS (ranging from 7.0 to 29.7 months) and median TTP (ranging from 2.6 to 10.2 months). Five comparative studies provided DCR (ranging from 32 to 97.2%). Forest plots showed that combination therapy significantly improved TTP (HR = 0.66, 95% CI 0.50-0.81, P = 0.002) rather than OS (HR = 0.63, 95% CI 0.55-0.71, P = 0.058), compared to TACE alone. DCR increased significantly in the combination therapy group (OR = 2.93, 95% CI 1.59-5.41, P = 0.005). Additional forest plots were drawn and no significant differences were observed with regard to survival outcome among various aetiologies. Forest plots for separate analysis of regions showed the HR for TTP was 0.62 (95% CI 0.45-0.79, P = 0.002) in the Asian countries group, and 0.82 (95% CI 0.59-1.05, P = 0.504)) in western countries. The HR for OS was 0.61 (95% CI 0.48-0.75, P = 0.050) in the Asian countries group and was 0.88 (95% CI 0.56-1.20, P = 0.845) in western countries. These data may indicate positive TTP outcome in Asian patients but not in European patients while no positive findings regarding OS were observed in either region. The most common AEs included fatigue, hand-foot skin reaction, diarrhoea and hypertension.ConclusionsCombination therapy may benefit unresectable HCC patients in terms of prolonged TTP and DCR. More well-designed studies are needed to investigate its superiority for OS.

Project description:BackgroundThe combination of tyrosine kinase inhibitors (TKIs) and transarterial chemoembolization (TACE) fulfills an important role in the treatment of unresectable hepatocellular carcinoma (uHCC). Among the combination therapies, both lenvatinib and sorafenib combined with TACE are recommended as first-¬line treatments for uHCC. However, at present, limited data are available concerning the efficacy and safety of these two combination therapies in uHCC.MethodsA detailed systematic search for studies on lenvatinib plus TACE (LEN+TACE) and sorafenib plus TACE (SOR+TACE) was conducted in the online databases PubMed, Embase and The Cochrane Library. The outcome data including overall survival (OS), progression free survival (PFS), time to progression (TTP), tumor response and adverse events (AEs), were independently extracted by two authors in a standardized way.ResultsOne randomized controlled trial and five cohort studies with 598 patients (LEN+TACE: 261, SOR+TACE: 337) were included in the meta-analysis. A higher rate of odds ratio (OR) for the objective response rate (ORR) [OR: 3.63; 95% confidence intervals (95% CI): 1.89-6.95; I squared statistic (I2) = 57%, P < 0.001] and disease control rate (DCR) (OR: 3.78; 95% CI: 2.00-7.16; I2 = 52%, P = 0.0001) were observed in the LEN+SOR group compared with the SOR+TACE group. The LEN+TACE group also had significant longer OS [hazard ratio (HR): 0.67; 95% CI: 0.52-0.85; I2 = 1%, P = 0.001], PFS (HR: 0.49; 95% CI: 0.38-0.62; I2 = 0%, P? 0.001) and TTP (HR: 0.62; 95% CI: 0.45-0.84; I2 = 0%, P = 0.002) compared with the SOR+TACE group. The incidence of hypertension (OR: 3.05; 95% CI: 1.45-6.39; P = 0.003) and proteinuria (OR: 5.25; 95% CI: 1.73-15.89; P = 0.003) were significantly higher in the LEN+TACE group than SOR+TACE group, while LEN+TACE group exhibited a lower rate of hand-foot-skin reaction (HFSR) (OR: 0.51; 95% CI: 0.27-0.95; P = 0.03) compared with the SOR+TACE group.ConclusionThe combination therapy of LEN+TACE showed significant superiority compared with SOR+TACE in terms of its efficacy for patients with uHCC. SOR+TACE should be recommended as a replacement therapy when serious AEs occur during the administration of LEN+TACE as the combination therapy.

Project description:BackgroundSorafenib has been recommended as the first-line treatment and shown to prolong median overall survival (OS) of patients with advanced unresectable hepatocellular carcinoma (HCC). Recently, a growing amount of research has supported the application of transarterial chemoembolization (TACE) in patients with advanced-stage HCC. The aim of this study was to compare the outcomes of TACE and sorafenib and identify the prognostic factors related to OS for Barcelona Clinic Liver Cancer (BCLC) stage C patients with PS 1 but without vascular invasion or extrahepatic spread.MethodsA total of 323 consecutive patients in BCLC stage C with PS 1 but without vascular invasion or extrahepatic spread were enrolled in this retrospective study. Survival analyses were performed using the Kaplan-Meier analysis, and the statistical differences between the TACE and sorafenib groups were examined by the log-rank test. Univariate and multivariate Cox regression analyses were performed to investigate the prognostic factors for OS.ResultsBased on the Kaplan-Meier curves, patients treated with TACE showed a better OS than those undergoing sorafenib, with respective OS at 1, 3, and 5 years (67.7%, 41.5%, 23.2% vs. 55.6%, 29.6%, 4.8%; log-rank P=0.002). The univariate analysis indicated that tumor size, tumor number, and treatment method, along with platelet (PLT), white blood cell (WBC), and α-fetoprotein (AFP) count, were associated with OS. The multivariate analysis demonstrated that tumor size, tumor number, and treatment method were significant prognostic factors for OS. According to the subgroups analyses based on the tumor size and tumor number, there were significant differences in OS among overall subsets between TACE and sorafenib therapy.ConclusionsTACE provided better prognostic performance than sorafenib and should be suggested as an alternative treatment modality to sorafenib for BCLC stage C patients with PS 1 but without vascular invasion or extrahepatic spread.

Dataset Information

Deep Learning Predicts Overall Survival of Patients With Unresectable Hepatocellular Carcinoma Treated by Transarterial Chemoembolization Plus Sorafenib.

Objectives

Methods

Results

Conclusion

Publications

Deep Learning Predicts Overall Survival of Patients With Unresectable Hepatocellular Carcinoma Treated by Transarterial Chemoembolization Plus Sorafenib.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets