Abstract
Objective
To explore the potential of pre-therapy computed tomography (CT) parameters in predicting the treatment response to initial conventional TACE (cTACE) in intermediate-stage hepatocellular carcinoma (HCC) and develop an interpretable machine learning model.
Methods
This retrospective study included 367 patients with intermediate-stage HCC who received cTACE as first-line therapy from three centers. We measured the mean attenuation values of target lesions on multi-phase contrast-enhanced CT and further calculated three CT parameters, including arterial (AER), portal venous (PER), and arterial portal venous (APR) enhancement ratios. We used logistic regression analysis to select discriminative features and trained three machine learning models via 5-fold cross-validation. The performance in predicting treatment response was evaluated in terms of discrimination, calibration, and clinical utility. Afterward, a Shapley additive explanation (SHAP) algorithm was leveraged to interpret the outputs of the best-performing model.
Results
The mean diameter, ECOG performance status, and cirrhosis were the important clinical predictors of cTACE treatment response, by multiple logistic regression. Adding the CT parameters to clinical variables showed significant improvement in performance (net reclassification index, 0.318, P < 0.001). The Random Forest model (hereafter, RF-combined model) integrating CT parameters and clinical variables demonstrated the highest performance on external validation dataset (AUC of 0.800). The decision curve analysis illustrated the optimal clinical benefits of RF-combined model. This model could successfully stratify patients into responders and non-responders with distinct survival (P = 0.001).
Conclusion
The RF-combined model can serve as a robust and interpretable tool to identify the appropriate crowd for cTACE sessions, sparing patients from receiving ineffective and unnecessary treatments.
Similar content being viewed by others
Abbreviations
- AER:
-
Arterial enhancement ratio
- PER:
-
Portal venous enhancement ratio
- APR:
-
Arterial portal venous enhancement ratio
- CT:
-
Computed tomography
- cTACE:
-
Conventional transarterial chemoembolization
- ML:
-
Machine learning
- HCC:
-
Hepatocellular carcinoma
- mRECIST:
-
Modified Response Evaluation Criteria in Solid Tumors
- LR:
-
Logistic regression
- RF:
-
Random forest
- SVM:
-
Support vector machine
- SHAP:
-
Shapley additive explanation
References
Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F (2021) Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 71(3):209–249. https://doi.org/10.3322/caac.21660
Heimbach JK, Kulik LM, Finn RS, Sirlin CB, Abecassis MM, Roberts LR, Zhu AX, Murad MH, Marrero JA (2018) AASLD guidelines for the treatment of hepatocellular carcinoma. Hepatology 67(1):358–380. https://doi.org/10.1002/hep.29086
EASL (2018) EASL clinical practice guidelines: management of hepatocellular carcinoma. J Hepatol 69(1):182–236. https://doi.org/10.1016/j.jhep.2018.03.019
Bolondi L, Burroughs A, Dufour JF, Galle PR, Mazzaferro V, Piscaglia F, Raoul JL, Sangro B (2012) Heterogeneity of patients with intermediate (BCLC B) hepatocellular carcinoma: proposal for a subclassification to facilitate treatment decisions. Semin Liver Dis 32(4):348–359. https://doi.org/10.1055/s-0032-1329906
Han K, Kim JH (2015) Transarterial chemoembolization in hepatocellular carcinoma treatment: Barcelona clinic liver cancer staging system. World J Gastroenterol 21(36):10327–10335. https://doi.org/10.3748/wjg.v21.i36.10327
Ma W, Jia J, Wang S, Bai W, Yi J, Bai M, Quan Z, Yin Z, Fan D, Wang J, Han G (2014) The prognostic value of 18F-FDG PET/CT for hepatocellular carcinoma treated with transarterial chemoembolization (TACE). Theranostics 4(7):736–744. https://doi.org/10.7150/thno.8725
Pinato DJ, Sharma R, Allara E, Yen C, Arizumi T, Kubota K, Bettinger D, Jang JW, Smirne C, Kim YW, Kudo M, Howell J, Ramaswami R, Burlone ME, Guerra V, Thimme R, Ishizuka M, Stebbing J, Pirisi M, Carr BI (2017) The ALBI grade provides objective hepatic reserve estimation across each BCLC stage of hepatocellular carcinoma. J Hepatol 66(2):338–346. https://doi.org/10.1016/j.jhep.2016.09.008
Sieghart W, Hucke F, Pinter M, Graziadei I, Vogel W, Müller C, Heinzl H, Trauner M, Peck-Radosavljevic M (2013) The ART of decision making: retreatment with transarterial chemoembolization in patients with hepatocellular carcinoma. Hepatology 57(6):2261–2273. https://doi.org/10.1002/hep.26256
Adhoute X, Penaranda G, Naude S, Raoul JL, Perrier H, Bayle O, Monnet O, Beaurain P, Bazin C, Pol B, Folgoc GL, Castellani P, Bronowicki JP, Bourlière M (2015) Retreatment with TACE: the ABCR SCORE, an aid to the decision-making process. J Hepatol 62(4):855–862. https://doi.org/10.1016/j.jhep.2014.11.014
Boas FE, Brody LA, Erinjeri JP, Yarmohammadi H, Shady W, Kishore S, Sofocleous CT (2016) Quantitative measurements of enhancement on preprocedure triphasic CT Can predict response of colorectal liver metastases to radioembolization. AJR Am J Roentgenol 207(3):671–675. https://doi.org/10.2214/AJR.15.15767
Liu Y, Chen W, Cui W, Liu H, Zhou X, Chen L, Li J, Chen M, Chen J, Wang Y (2020) Quantitative pretreatment CT parameters as predictors of tumor response of neuroendocrine tumor liver metastasis to transcatheter arterial bland embolization. Neuroendocrinology 110(7–8):697–704. https://doi.org/10.1159/000504257
Marrache F, Vullierme MP, Roy C, El Assoued Y, Couvelard A, O’Toole D, Mitry E, Hentic O, Hammel P, Lévy P, Ravaud P, Rougier P, Ruszniewski P (2007) Arterial phase enhancement and body mass index are predictors of response to chemoembolisation for liver metastases of endocrine tumours. Br J Cancer 96(1):49–55. https://doi.org/10.1038/sj.bjc.6603526
Syha R, Grözinger G, Grosse U, Maurer M, Zender L, Horger M, Nikolaou K, Ketelsen D (2016) Parenchymal blood volume assessed by C-arm-based computed tomography in immediate posttreatment evaluation of drug-eluting bead transarterial chemoembolization in hepatocellular carcinoma. Invest Radiol 51(2):121–126. https://doi.org/10.1097/rli.0000000000000215
Liu D, Liu F, Xie X, Su L, Liu M, Xie X, Kuang M, Huang G, Wang Y, Zhou H, Wang K, Lin M, Tian J (2020) Accurate prediction of responses to transarterial chemoembolization for patients with hepatocellular carcinoma by using artificial intelligence in contrast-enhanced ultrasound. Eur Radiol 30(4):2365–2376. https://doi.org/10.1007/s00330-019-06553-6
Kong C, Zhao Z, Chen W, Lv X, Shu G, Ye M, Song J, Ying X, Weng Q, Weng W, Fang S, Chen M, Tu J, Ji J (2021) Prediction of tumor response via a pretreatment MRI radiomics-based nomogram in HCC treated with TACE. Eur Radiol 31(10):7500–7511. https://doi.org/10.1007/s00330-021-07910-0
Peng J, Kang S, Ning Z, Deng H, Shen J, Xu Y, Zhang J, Zhao W, Li X, Gong W, Huang J, Liu L (2020) Residual convolutional neural network for predicting response of transarterial chemoembolization in hepatocellular carcinoma from CT imaging. Eur Radiol 30(1):413–424. https://doi.org/10.1007/s00330-019-06318-1
Lundberg SM, Nair B, Vavilala MS, Horibe M, Eisses MJ, Adams T, Liston DE, Low DK, Newman SF, Kim J, Lee SI (2018) Explainable machine-learning predictions for the prevention of hypoxaemia during surgery. Nat Biomed Eng 2(10):749–760. https://doi.org/10.1038/s41551-018-0304-0
Caruana R, Lou Y, Gehrke J, Koch P, Sturm M, Elhadad N (2015) Intelligible models for healthcare: predicting pneumonia risk and hospital 30-day readmission. Paper presented at the proceedings of the 21th ACM SIGKDD international conference on knowledge discovery and data mining, Sydney, NSW, Australia
Lencioni R, Llovet JM (2010) Modified RECIST (mRECIST) assessment for hepatocellular carcinoma. Semin Liver Dis 30(1):52–60. https://doi.org/10.1055/s-0030-1247132
Weir JP (2005) Quantifying test-retest reliability using the intraclass correlation coefficient and the SEM. J Strength Cond Res 19(1):231–240. https://doi.org/10.1519/15184.1
Liu Y, Chen W, Cui W, Liu H, Zhou X, Chen L, Li J, Chen M, Chen J, Wang Y (2019) Quantitative pretreatment CT parameters as predictors of tumor response of NET liver metastasis to TAE. Neuroendocrinology. https://doi.org/10.1159/000504257
Lundberg SM, Erion G, Chen H, DeGrave A, Prutkin JM, Nair B, Katz R, Himmelfarb J, Bansal N, Lee S-I (2020) From local explanations to global understanding with explainable AI for trees. Nat Mach Intell 2(1):56–67. https://doi.org/10.1038/s42256-019-0138-9
Zhang L, Jiang Y, Jin Z, Jiang W, Zhang B, Wang C, Wu L, Chen L, Chen Q, Liu S, You J, Mo X, Liu J, Xiong Z, Huang T, Yang L, Wan X, Wen G, Han XG, Fan W, Zhang S (2022) Real-time automatic prediction of treatment response to transcatheter arterial chemoembolization in patients with hepatocellular carcinoma using deep learning based on digital subtraction angiography videos. Cancer Imaging. https://doi.org/10.1186/s40644-022-00457-3
Berenguer R, Pastor-Juan MDR, Canales-Vázquez J, Castro-García M, Villas MV, Mansilla Legorburo F, Sabater S (2018) Radiomics of CT features may be nonreproducible and redundant: influence of CT acquisition parameters. Radiology 288(2):407–415. https://doi.org/10.1148/radiol.2018172361
Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D Grad-cam: visual explanations from deep networks via gradient-based localization. In: Proceedings of the IEEE international conference on computer vision, 2017. pp 618–626
Fronda M, Doriguzzi Breatta A, Gatti M, Calandri M, Maglia C, Bergamasco L, Righi D, Faletti R, Fonio P (2021) Quantitative assessment of HCC wash-out on CT is a predictor of early complete response to TACE. Eur Radiol 31(9):6578–6588. https://doi.org/10.1007/s00330-021-07792-2
Rokach L, Maimon O (2005) Decision trees. IEEE Trans Syst Man Cybern Part C 35(4):476–487. https://doi.org/10.1007/0-387-25465-X_9
Golden CE, Rothrock MJ Jr, Mishra A (2019) Comparison between random forest and gradient boosting machine methods for predicting Listeria spp. prevalence in the environment of pastured poultry farms. Food Res Int 122:47–55. https://doi.org/10.1016/j.foodres.2019.03.062
Doupe P, Faghmous J, Basu S (2019) Machine learning for health services researchers. Value Health 22(7):808–815. https://doi.org/10.1016/j.jval.2019.02.012
Li K, Yao S, Zhang Z, Cao B, Wilson CM, Kalos D, Kuan PF, Zhu R, Wang X (2022) Efficient gradient boosting for prognostic biomarker discovery. Bioinformatics. https://doi.org/10.1093/bioinformatics/btab869
Thorsen-Meyer H-C, Nielsen AB, Nielsen AP, Kaas-Hansen BS, Toft P, Schierbeck J, Strøm T, Chmura PJ, Heimann M, Dybdahl L, Spangsege L, Hulsen P, Belling K, Brunak S, Perner A (2020) Dynamic and explainable machine learning prediction of mortality in patients in the intensive care unit: a retrospective study of high-frequency data in electronic patient records. Lancet Digit Health 2(4):e179–e191. https://doi.org/10.1016/S2589-7500(20)30018-2
Lundberg SM, Nair B, Vavilala MS, Horibe M, Eisses MJ, Adams T, Liston DE, Low DK-W, Newman S-F, Kim J, Lee S-I (2018) Explainable machine-learning predictions for the prevention of hypoxaemia during surgery. Nat Biomed Eng 2(10):749–760. https://doi.org/10.1038/s41551-018-0304-0
Lu J, Zhang S, Jiang H, Yang L, Hao D, Yang Y, Li X, Chen A, Shao J, Liu X (2021) Gestational diabetes mellitus prediction model: A risk factor analysis of pregnant women with gestational diabetes mellitus and a normal oral glucose tolerance test in the second trimester of pregnancy. Technol Health Care Off J Eur Soc Eng Med. https://doi.org/10.3233/THC-218006
Zhang S, Zhao B, Chen D, Qi Y, Ma Y, Ma J, Xie W, Guo H (2021) Anesthetic management of precise radiotherapy under apnea-like condition. J Int Med Res 49(3):300060521990260. https://doi.org/10.1177/0300060521990260
Funding
This research was supported by grants from the National Key Research and Development Program of China (2023YFF1204600); the National Natural Science Foundation of China (82227802, 82302306); the Clinical Frontier Technology Program of the First Affiliated Hospital of Jinan University, China (No. JNU1AF-CFTP-2022-a01201); the Science and Technology Projects in Guangzhou (202201020022, 2023A03J1036, 2023A03J1038); the Science and Technology Youth Talent Nurturing Program of Jinan University (21623209); and the Postdoctoral Research Foundation of China (2022M721349).
Author information
Authors and Affiliations
Contributions
LZ, LW, CM, ZH, and SZ conceived and designed the project; ZJ, CL, BZ, QC, ZH, JY, XM, and HS performed the research and collected the data; LW and CL analyzed the data; and LZ, ZJ, ZH, FW, and CM wrote the paper. All authors read and approved the final manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors have no relevant financial or non-financial interests to disclose.
Ethical approval
This retrospective study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of the institutional Ethics Committees of the First Affiliated Hospital of Jinan University, and the requirement for informed consent was waived.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Figure S1
. Performance of different models on the cross-validation cohort. The following order: A was LR-clinical model, B was LR-combined model, C was SVM-clinical model, D was SVM-combined model, E was RF-clinical model, F was RF-combined model. The performance of RF-combined model was the best among models in the cross-validation cohort. LR, Logistic regression; SVM, support vector machine; RF, random forest. (TIF 3099 kb)
Supplementary Figure S2.
The learning curve of the different models in the training and validation cohorts. The following order: A was LR-clinical model, B was LR-combined model, C was SVM-clinical model, D was SVM-combined model, E was RF-clinical model, F was RF-combined model. LR, Logistic regression; SVM, support vector machine; RF, random forest. (TIF 226 kb)
Supplementary Figure S3.
Stratified analysis of the predictive value of mean diameter for response to cTACE therapy based on tumor blood supply. ROC curve analysis for mean AER (A) and mean PER (B); C Hierarchical logistic regression analysis of mean diameter for response to cTACE; D Individual-level SHAP force plot of a patient from the external validation cohort with a highly vascularized but non-responsive HCC tumor. (TIFF 1871 kb)
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zhang, L., Jin, Z., Li, C. et al. An interpretable machine learning model based on contrast-enhanced CT parameters for predicting treatment response to conventional transarterial chemoembolization in patients with hepatocellular carcinoma. Radiol med 129, 353–367 (2024). https://doi.org/10.1007/s11547-024-01785-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11547-024-01785-z