目的 探讨MRI影像组学模型在预测无强化弥漫性胶质瘤术后5年内复发中的价值。方法 回顾性分析2016-06至2020-06在首都医科大学附属北京天坛医院经病理证实的无强化弥漫性胶质瘤116例的临床资料,收集临床病理信息、MRI影像及随访信息。通过Python pyradiomics及Atlasquery工具提取影像组学特征和肿瘤位置特征,使用LASSO算法对特征进行降维和筛选,分别建立性别年龄预测模型(Model-1)、肿瘤位置预测模型(Model-2)、影像组学预测模型(Model-3)、肿瘤位置+影像组学预测模型(Model-4)、性别年龄+肿瘤位置+影像组学预测模型(Model-5),绘制受试者工作特征曲线及决策曲线,评估模型预测效能。结果 不同模型对预测无强化弥漫性胶质瘤术后复发的AUC分别为0.529(Model-1)、0.680(Model-2)、0.912(Model-3)、0.911(Model-4)及0.917(Model-5),其中Model-5预测术后复发效能最佳,在测试集中敏感度、特异度及准确率分别为0.837、0.866及0.863。结论 基于MRI影像组学的预测模型对预测无强化弥漫性胶质瘤术后5年内复发情况,具有较高的应用价值。
Abstract
Objective To investigate the value of the MRI-based radiomics model in predicting the recurrence of non-enhancing diffuse gliomas within 5 years after surgery. Methods A retrospective analysis was conducted on 116 patients with pathologically confirmed non-enhanced diffuse gliomas treated at Beijing Tiantan Hospital Affiliated to Capital Medical University from June 2016 to June 2020. Clinical pathological data, MRI images, and follow-up information were collected. Radiomics features and tumor location features were extracted using Python pyradiomics and Atlasquery tools. The LASSO algorithm was applied for feature dimensionality reduction and selection. Five predictive models were constructed as follows: Model-1 (Sex and Age), Model-2 (Tumor Location), Model-3 (Radiomics), Model-4 (Tumor Location + Radiomics), and Model-5 (Sex+Age+Tumor Location+Radiomics). Receiver operating characteristic (ROC) curves and decision curve analysis (DCA) were plotted to evaluate the predictive performance of the models. Results The areas under the ROC curve (AUC) for predicting postoperative recurrence were 0.529 (Model-1), 0.680 (Model-2), 0.912 (Model-3), 0.911 (Model -4), and 0.917 (Model-5), respectively. Model-5 demonstrated the best performance, achieving a sensitivity of 0.837, specificity of 0.866, and accuracy of 0.863, respectively in the test set. Conclusions The MRI radiomics-based model has high clinical value in predicting recurrence of non-enhancing diffuse gliomas within 5 years after surgery.
关键词
胶质瘤 /
复发 /
磁共振成像 /
影像组学 /
预测价值
Key words
glioma /
recurrence /
magnetic resonance imaging /
radiomics /
predictive value
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Louis D N, Perry A, Wesseling P, et al. The 2021 WHO classification of tumors of the central nervous system: a summary[J]. Neuro Oncol,2021,23(8):1231-1251.
[2] Wen P Y, Kesari S. Malignant gliomas in adults[J]. N Engl J Med,2008,359(5):492-507.
[3] Horbinski C, Nabors L B, Portnow J, et al. NCCN guidelines insights: central nervous system cancers, version 2.2022[J]. J Natl Compr Canc Netw, 2023, 21(1): 12-20.
[4] Karschnia P, Dietrich J, Bruno F, et al. Surgical management and outcome of newly diagnosed glioblastoma without contrast enhancement (low-grade appearance): a report of the RANO resect group[J].Neuro Oncol,2024, 26(1):166-177.
[5] Nabors L B, Portnow J, Ahluwalia M, et al. Central nervous system cancers, version 3.2020, NCCN clinical practice guidelines in oncology[J]. J Natl Compr Canc Netw,2020,18(11):1537-1570.
[6] Tesileanu M S, Dirven L, Wijnenga M J, et al. Survival of diffuse astrocytic glioma, IDH1/2 wildtype, with molecular features of glioblastoma, WHO grade IV: a confirmation of the cIMPACT-NOW criteria[J]. Neuro Oncol, 2020, 22(4): 515-523.
[7] Foltyn-Dumitru M, Banan R, Schell M, et al. Histopathological and molecular characteristics of IDH-wildtype glioblastoma without contrast enhancement: Implications for clinical outcomes[J]. Neuro Oncol,2025,27(7):1878-1887.
[8] Kang K M,Song J,Choi Y,et al.MRI scoring systems for predicting isocitrate dehydrogenase mutation and chromosome 1p/19q codeletion in adult-type diffuse glioma lacking contrast enhancement[J].Radiology,2024, 311(2):e233120.
[9] Cancer N, Brat D J,Verhaak R G,et al.Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas[J].N Engl J Med,2015,372(26):2481-2498.
[10] Wijnenga M J, Mattni T, French P J, et al. Does early resection of presumed low-grade glioma improve survival? A clinical perspective[J]. J Neurooncol,2017,133(1):137-146.
[11] Mellinghoff I K, Lu M, Wen P Y, et al. Vorasidenib and ivosidenib in IDH1-mutant low-grade glioma: a randomized, perioperative phase 1 trial[J].Nat Med,2023,29(3):615-622.
[12] Mellinghoff I K, Penas P M, Peters K B, et al.Vorasidenib, a dual inhibitor of mutant IDH1/2,in recurrent or progressive glioma; results of a first-in-human phase I trial[J].Clin Cancer Res,2021,27(16): 4491-4499.
[13] Ding X, Wang Z, Chen D, et al. The prognostic value of maximal surgical resection is attenuated in oligodendroglioma subgroups of adult diffuse glioma: a multicenter retrospective study[J]. J Neurooncol, 2018, 140(3): 591-603.
[14] Park Y W, Kim S, Han K, et al. Rethinking extent of resection of contrast-enhancing and non-enhancing tumor:different survival impacts on adult-type diffuse gliomas in 2021 World Health Organization classification[J]. Eur Radiol,2024, 34(2):1376-1387.
[15] Molinaro A M, Hervey-Jumper S, Morshed R A, et al. Association of maximal extent of resection of contrast-enhanced and non-contrast-enhanced tumor with survival within molecular subgroups of patients with newly diagnosed glioblastoma[J]. JAMA Oncol,2020, 6(4):495-503.
[16] Chai R C, Zhang K N, Chang Y Z, et al. Systematically characterize the clinical and biological significances of 1p19q genes in 1p/19q non-codeletion glioma[J]. Carcinogenesis,2019,40(10): 1229-1239.
[17] Li J, Niu X, Gan Y, et al. Clinical and pathologic features and prognostic factors for recurrent gliomas[J]. World Neurosurg,2019,128(8):e21-e30.
[18] Khorasani A, Dadashi Serej N, Jalilian M, et al. Performance comparison of different medical image fusion algorithms for clinical glioma grade classification with advanced magnetic resonance imaging (MRI) [J].Sci Rep, 2023,13(1):17646.
[19] Tang C, Chen L, Xu Y, et al. Prediction of TERT mutation status in gliomas using conventional MRI radiogenomic features[J].Front Neurol,2024,15(7):1439598.
[20] Kartal A S, Kartal M O, Gülleroglu N B, et al. Texture analysis and metabolic parameters of 18F-FDG PET/CT to predict primary tumour response and prognosis of paediatric soft tissue sarcomas[J]. Eur J Nucl Med Mol Imaging, 2025,52(13):4913-4923.
[21] Shamshiripour P, Hajiahmadi F, Lotfi S, et al. Next-generation anti-angiogenic therapies as a future prospect for glioma immunotherapy; from bench to bedside[J]. Front Immunol,2022,13(6):859633.
PDF(1497 KB)
