MR T2WI intratumoral and peritumoral CT radiomics for predicting von Hippel-Lindau(VHL)gene mutation in renal carcinoma
10.13929/j.issn.1672-8475.2025.02.008
- VernacularTitle:MR T2WI瘤内及瘤周影像组学预测肾癌希佩尔-林道(VHL)基因突变
- Author:
Liping LI
1
;
Ruiguang MA
1
;
Rui QIAN
1
;
Shuang BAO
1
;
Qingjuan MENG
1
Author Information
1. 长春中医药大学附属医院 长春市中医院放射科,吉林 长春 130022
- Publication Type:Journal Article
- Keywords:
kidney neoplasms;
magnetic resonance imaging;
von Hippel Lindau gene;
radiomics
- From:
Chinese Journal of Interventional Imaging and Therapy
2025;22(2):118-122
- CountryChina
- Language:Chinese
-
Abstract:
Objective To observe the value of MR T2WI intratumoral and peritumoral CT radiomics for predicting von Hippel-Lindau(VHL)gene mutation in renal carcinoma.Method Totally 150 patients with renal carcinoma were retrospectively enrolled and divided into training set(n=105)and validation set(n=45)at the ratio of 7∶3,and furtherly assigned into mutation subgroup and wild subgroup according to with VHL gene mutation or not.Multivariate logistic regression analysis was performed to screen the independent risk factors of VHL gene mutation in renal carcinoma,then a clinical model was constructed.The optimal radiomics features were extracted and screened based on intratumoral+peritumoral 2 mm regions shown on MR T2WI.Logistic regression algorithm was used to construct radiomics model,and finally a combined model was established based on radiomics model and clinical model.Receiver operating characteristic curves were drawn,and the area under the curves(AUC)were calculated to evaluate the efficacy of each model for predicting VHL gene mutation in renal carcinoma.Result Patients'age,smoking history,hypertension history,other family histories and β2-microglobulin were all independent clinical risk factors for VHL gene mutation in renal carcinoma(all P＜0.05).The efficacy of clinical model,radiomics model and combined model for predicting VHL gene mutation in renal carcinoma increased successively(all P＜0.05),with AUC of 0.758,0.831 and 0.952 in training set,0.729,0.803 and 0.896 in validation set,respectively.Conclusion MR T2WI intratumoral and peritumoral CT radiomics had good efficacy for predicting VHL gene mutation in renal carcinoma.Combining with clinical features could further improve predicting efficacy of model.
