浙江医学2025，Vol.47Issue(20)：2164-2169,2175,7.DOI:10.12056/j.issn.1006-2785.2025.47.20.2025-808
CT影像组学模型预测安罗替尼治疗晚期非小细胞肺癌疗效的研究
Prediction of CT-based radiomics in the efficacy of anlotinib on advanced non-small cell lung cancer
摘要
Abstract
Objective To build a predictive model based on CT radiomics to evaluate the application prospects of anlotinib in patients with advanced non-small cell lung cancer(NSCLC).Methods The study retrospectively analyzed 122 chest CT images and clinical characteristics from 78 patients with pathologically confirmed advanced NSCLC who received anlotinib treatment in Hangzhou First People's Hospital Affiliated to Westlake University School of Medicine and Hangzhou Cancer Hospital from January 2018 and December 2020.Each lesion was evaluated per RECIST 1.1 criteria with complete response,partial response and stable disease defined as effective outcomes while progressive disease as ineffective outcome.The lesions were partitioned into a training set(92 lesions)and a validation set(30 lesions)at a 3:1 ratio.The region of interest delineation and radiomics feature extraction were performed for all tumor lesions.The least absolute shrinkage and selection operator(LASSO)regression was used to select meaningful radiomic features.Subsequently,logistic regression was used to establish the clinical-radiomic model and draw the nomogram.The ROC curve,calibration curve and decision curve analysis were used to analyze the efficacy,calibration degree and clinical benefits of the model,respectively.Results Among 78 patients,anlotinib treatment was efficacious in 64 cases and inefficacious in 14 cases.Regarding 122 lesions,98 were responsive while 24 were non-responsive.Three clinical features and two radiomic features were screened out by LASSO regression,namely smoking history,pathological examination,concomitant medications,and F3.GrayLevelCooccurenceMatrix 250.7Entropy,and F5.IntensityDirectLocalRangeMin.According to ROC curve analysis,AUC was 0.796(95％CI:0.698-0.894)in the training set and 0.757(95％CI:0.581-0.933)in the validation set for the clinical-radiomic model.Regarding the calibration curve,the clinical-radiomic model fitted fine.In the decision curve analysis,the threshold probability of the clinical-radiomic model was greater than 0.6,indicating capability of predicting clinical benefits.Conclusion CT-based radiomics can predict the efficacy of the anlotinib treatment in advanced NSCLC.关键词
安罗替尼/CT影像组学/肺癌/预测模型Key words
Anlotinib/CT-based radiomics/Lung cancer/Prediction model引用本文复制引用
李嘉威,马胜林,陈雪琴,朱鲁程,李鑫,杨邵瑜,黄杰,莫威行,陶志刚,夏冰..CT影像组学模型预测安罗替尼治疗晚期非小细胞肺癌疗效的研究[J].浙江医学,2025,47(20):2164-2169,2175,7.基金项目
杭州市医药卫生科技计划项目(A20251402) （A20251402）
杭州市医学重点学科建设项目(2025HZGF06) （2025HZGF06）
