Development and validation of a radiomic nomogram for preoperative Lauren classification in gastric cancer
#br# DING Yi1, LU Chao2, WANG Xiao-xiao2, CHEN Jian1, SHAN Xiu-hong2
(1. School of Medicine, Jiangsu University, Zhenjiang Jiangsu 212013; 2. Department of Radiology, the Affiliated People′s Hospital of Jiangsu University, Zhenjiang Jiangsu 212002, China)
Abstract:Objective: To explore the feasibility of CT-based radiomics nomogram to preoperatively differentiate Lauren diffuse type from intestinal type in gastric cancer. Methods: The clinical data of 539 patients with pathologically diagnosed gastric cancer were retrospectively analyzed. All patients randomly separated into two cohorts at a 7 ∶3 ratio for training and validation. Venous phase CT images were segmented by radiologists with ITK-SNAP software manually.Two sets of radiomic features were derived from tumor region and peritumor on venous phase CT images.With the least absolute shrinkage and selection operator(LASSO) logistic regression,the effective characteristics were selected. A tumor-based model, a peripheral ring-based model, a combined radiomic signature,clinical model 1 and clinical model 2 were proposed. Afterwards, a radiomic nomogram integrating the combined radiomic signature and clinical characteristics was developed. Receiver operating characteristic (ROC) curves as well as corresponding area under ROC curves (AUC) were estimated for both cohorts to assess the corresponding discrimination ability.We adopted Delong-test to compare the predictive performance between each two models. Calibration curves were conducted to verify the good fitness of model predictive outputs with actual values for the radiomic nomogram. Decision curves were conducted in the validation to quantify the usefulness in clinical trials. Results: The combined radiomic signature achieving an AUC of 0.715 (95% CI: 0.663-0.767) in the training cohort and 0.714 (95%CI: 0.636-0.792) in the validation cohort. The radiomic nomogram incorporating the combined radiomic signature,and clinical characteristics supassed all the other models with a training AUC of 0.745 (95%CI: 0.696-0.795) and a validation AUC of 0.758 (95%CI:0.685-0.831). Further, calibration curves and decision curves demonstrated its great model fitness and clinical usefulness. Conclusion: The radiomic nomogram based on the combined radiomic signature and clinical characteristics held potential in differentiating Lauren diffuse type from intestinal type, which is benefical for reasonable clinical treatment strategy.
[1]Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018:GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries\[J\].CA Cancer J Clin,2018,68(6):394-424.
[2]Chen YC, Fang WL, Wang RF, et al.Clinicopathological variation of Lauren classification in gastric cancer\[J\].Pathol Oncol Res,2016,22(1):197-202.
[3]Lee SE, Han K, Kwak JY, et al. Radiomics of US texture features in differential diagnosis between triplenegative breast cancer and fibroadenoma\[J\].Sci Rep,2018,8(1): 13546-13553.
[4]Huang YQ, Liang CH, He L, et al. Development and validation of a radiomics nomogram for preoperative prediction of lymph node metastasis in colorectal cancer\[J\]. J Clin Oncol,2016,34(18):2157-2164.
[5]Dong D, Tang L, Li ZY, et al. Development and validation of an individualized nomogram to identify occult peritoneal metastasis in patients with advanced gastric cance\[J\].Ann Oncol,2019,30(3): 431-438.
[6]Lafata KJ,Hong JC,Geng R,et al. Association of pretreatment radiomic features with lung cancer recurrence following stereotactic body radiation therapy\[J\]. Phys Med Biol,2019,64(2):025007.
[7]Nie K , Shi L , Chen Q , et al. Rectal cancer: assessment of neoadjuvant chemoradiation outcome based on radiomics of multiparametric MRI\[J\].Clin Cancer Res,2016, 22(21):5256-5264.
[8]Park H, Lim Y, Ko ES, et al. Radiomics signature on magnetic resonance imaging: association with diseasefree survival in patients with invasive breast cancer\[J\]. Clin Cancer Res,2018,24(19): 4705-4714.
[9]Jiang Y, Chen C, Xie J, et al. Radiomics signature of computed tomography imaging for prediction of survival and chemotherapeutic benefits in gastric cancer\[J\]. EBio Medicine, 2018,36:171.
[10]Liu S, Liu S, Ji C, et al. Application of CT texture analysis in predicting histopathological characteristics of gastric cancers\[J\].Eur Radiol, 2017,27(12):4951-4959.
[11]Ma L, Xu X, Zhang M, et al. Dynamic contrastenhanced MRI of gastric cancer: Correlations of the pharmacokinetic parameters with histological type, Lauren classification, and angiogenesis\[J\].Magn Reson Imaging,2017,37:27-32.
[14]Hundahl SA, Phillips JL, Menck HR. The National Cancer Data Base Report on poor survival of US gastric carcinoma patients treated with gastrectomy: Fifth edition American Joint Committee on cancer staging, proximal disease, and the “different disease” hypothesis\[J\]. Cancer,2000,88(4): 921-932.
[15]Satoh A , Shuto K , Okazumi S , et al. Role of perfusion CT in assessing tumor blood flow and malignancy level of gastric cancer\[J\]. Dig Surg,2010,27(4):253-260.
[16]Qiu MZ, Shi SM, Chen M,et al. Comparison of HER2 and Lauren classification between biopsy and surgical resection samples, primary and metastatic samples of gastric cancer\[J\]. J Cancer,2017,8(17):3531-3537.
[17]Lauren P. The two histological main types of gastric carcinoma: diffuse and socalled intestinaltype carcinoma: an attempt at a histoclinical classification\[J\]. Acta Pathol Microbiol Scand,1965,64(1):31-49.
[18]Qiu MZ, Cai MY, Zhang DS, et al. Clinicopathological characteristics and prognostic analysis of Lauren classification in gastric adenocarcinoma in China\[J\]. J Transl Med,2013,11:58-64.
[19]Liu S, Guan W, Wang H, et al. Apparent diffusion coefficient value of gastric cancer by diffusionweighted imaging: correlations with the histological differentiation and Lauren classification\[J\]. Eur J Radiol,2014,83(12): 2122-2128.