分享:
分享到微信朋友圈
X
综述
较低级别脑胶质瘤分子分型的磁共振影像组学研究进展
陈瑞宏 谭艳

Cite this article as: CHEN R H, TAN Y. Research advances of radiomics in prognosis prediction of lower-grade gliomas[J]. Chin J Magn Reson Imaging, 2023, 14(3): 159-164.本文引用格式:陈瑞宏, 谭艳. 较低级别脑胶质瘤分子分型的磁共振影像组学研究进展[J]. 磁共振成像, 2023, 14(3): 159-164. DOI:10.12015/issn.1674-8034.2023.03.029.


[摘要] 脑胶质瘤是脑实质最常见的原发恶性肿瘤。脑胶质瘤分为1~4级,其中2级和3级被称为较低级别脑胶质瘤(lower-grade gliomas, LGGs)。2021年世界卫生组织中枢神经系统分类(World Health Organization central nervous system, WHO CNS)深化了分子分型对LGGs诊疗的重要性,以LGGs分子分型进行病理分级升级诊断。影像组学作为新兴领域,可术前无创预测LGGs分子分型,为LGGs的治疗评估及预后预测提供依据。目前已有较多研究通过分析MRI常规序列、功能序列,结合临床信息,运用机器学习和深度学习建立影像组学模型,术前无创预测LGGs分子分型。虽然其存在局限性,但仍具有一定的科研与临床意义。本文就LGGs分子分型在MRI影像组学的研究进展予以综述,以期通过MRI影像组学术前预测LGGs分子分型,便于指导临床个体化诊疗方案制订及预后预测。
[Abstract] Glioma is the most common primary malignant tumor of brain parenchyma. Gliomas are divided into grades 1-4, of which grades 2 and 3 are called lower-grade gliomas (LGGs). The 2021 World Health Organization Central Nervous System (WHO CNS) deepens the importance of molecular typing for the diagnosis and treatment of LGGs, using LGGs molecular typing for pathological grading upgrade diagnosis. As an emerging field, radiomics can noninvasively predict the molecular subtypes of LGGs before surgery, providing a basis for treatment evaluation and prognosis prediction of LGGs. At present, many studies have established radiomics models by analyzing MRI routine sequences, functional sequences, combined with clinical information, using machine learning and deep learning to predict LGGs molecular typing noninvasively before surgery. Although it has limitations, it still has certain scientific research and clinical significance. This article reviews the research progress of LGGs molecular typing in MRI radiomics, in order to predict LGGs molecular typing by MRI radiomics, and to facilitate the formulation of clinical individualized diagnosis and treatment plans and prognosis prediction.
[关键词] 较低级别脑胶质瘤;脑胶质瘤;磁共振成像;影像组学;分子分型
[Keywords] lower-grade gliomas;glioma;magnetic resonance imaging;radiomics;molecular typing

陈瑞宏 1   谭艳 2*  

1 山西医科大学医学影像学院,太原 030001

2 山西医科大学第一医院影像科,太原 030001

通信作者:谭艳,E-mail:tanyan123456@sina.com

作者贡献声明:谭艳设计本综述的方案,对稿件重要的智力内容进行了修改,获得了国家自然科学基金的资助;陈瑞宏起草和撰写稿件,获取、分析或解释本综述的数据/文献;全体作者都同意最后的修改稿发表,都同意对本综述的所有方面负责,确保本研究的准确性和诚信。


基金项目: 国家自然科学基金 82071893
收稿日期:2022-09-17
接受日期:2023-03-03
中图分类号:R445.2  R730.264 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2023.03.029
本文引用格式:陈瑞宏, 谭艳. 较低级别脑胶质瘤分子分型的磁共振影像组学研究进展[J]. 磁共振成像, 2023, 14(3): 159-164. DOI:10.12015/issn.1674-8034.2023.03.029.

0 前言

       脑胶质瘤是脑实质最常见的原发恶性肿瘤。2021年世界卫生组织中枢神经系统分类(World Health Organization central nervous system, WHO CNS)采用阿拉伯数字取代罗马数字将脑胶质瘤分为1~4级[1]。目前认为2级和3级脑胶质瘤为较低级别脑胶质瘤(lower-grade gliomas, LGGs)[2]。随着对脑胶质瘤研究的不断深入,证实不同分子分型的脑胶质瘤治疗疗效和预后差异明显,这表明将脑胶质瘤分子分型用于指导个体化治疗是有价值的[3]。2016 年WHO CNS首次将分子学特征用于脑胶质瘤病理分级诊断,如异柠檬酸脱氢酶(isocitrate dehydrogenase, IDH)、1号染色体短臂和19号染色体长臂(1p/19q)等。这种“整合”诊断的模式提升了脑胶质瘤诊断的准确性,有助于制订个性化治疗方案,以便于改善患者预后[4]。2021年WHO CNS在“整合”诊断模式的基础上新增分子分型指标,如IDH野生型LGGs伴有端粒酶逆转录酶(telomerase reverse tranase, TERT)启动子突变、表皮生长因子受体(epidermal growth factor receptor, EGFR)基因扩增、7号染色体扩增/10号染色体缺失三者之一,则升级诊断为胶质母细胞瘤,IDH野生型;存在CDKN2A/B纯合子缺失的IDH突变型星形细胞瘤升级为WHO 4级。2021年WHO CNS更客观、精确地定义了肿瘤类型,并认为分子分型在病理分级证据级别上优于组织学形态[1]

       MRI包括常规序列和功能序列,这些序列反映不同的肿瘤组织特征并互相补充,是用于诊断脑胶质瘤和划定手术区域的重要影像工具。但是其作用有限,难以明确分子分型[5]。目前LGGs分子分型主要依靠术后病理活检,但其操作有创、价格昂贵、诊断滞后。另外,由于LGGs高度异质性导致的取样偏倚影响了分子分型诊断的准确性。基于MRI的影像组学更大程度地发挥了MRI影像图像的内在价值,较术后病理活检具有术前无创、价格合理和诊断及时的优势。其运用计算机图像处理及大数据挖掘技术,从庞大的人眼无法识别的影像数据中提取图像特征并进行分析,从而反映LGGs的内部信息及异质性,可以术前无创且有效地预测LGGs分子分型,以便于及时地为患者制订最佳的个性化诊疗方案[6]。因此,笔者将就MRI影像组学在预测LGGs分子分型中的研究进展予以综述。

1 影像组学预测IDH突变

       研究表明超过80%的LGGs存在IDH突变。对于IDH突变型LGGs,最大安全范围切除肿瘤可有效提高总生存率,并且术后对适当剂量放疗同步替莫唑胺化疗的放化疗方案更为敏感。而对于IDH野生型LGGs,残留部分不易切除的肿瘤对总生存率影响较小,并且从化疗中受益较少。术前预测IDH突变可更好地划定手术切除范围及制订术后早期放化疗方案,尽可能改善患者预后[7, 8, 9]

       众多研究表明,基于MRI常规序列,运用机器学习建立的MRI影像组学模型可术前无创且有效地预测IDH突变。如ARITA等[10]基于299名LGGs患者的T1WI、T2WI、T2加权流体衰减反转恢复(T2-weighted fluid attenuation inversion recovery, T2-FLAIR)序列、对比增强T1加权成像(contrast enhanced T1-weighted imaging, CE-T1WI)序列图像,利用最小绝对收缩和选择算子(least absolute shrinkage and selection operator, LASSO)回归建立IDH突变预测模型,预测准确度为83%,当加入病灶位置信息后准确度提升到87%。ZHANG等[11]基于73名LGGs患者的T2WI、T2-FLAIR、CE-T1WI图像,运用支持向量机(support vector machine, SVM)建立IDH突变预测模型,受试者工作特征(receiver operating characteristic, ROC)曲线下面积(area under the curve, AUC)为0.830,并且发现T2WI图像特征最为重要。上述研究表明仅基于MRI常规序列建立的MRI影像组学模型对IDH突变具有较高的预测效能,其中T2WI图像特征表现更好,并且加入病灶位置信息可提高预测效能。

       有研究表明基于MRI常规序列并加入功能序列建立的MRI影像组学模型可进一步提高预测效能。如PARK等[12]从常规序列与扩散张量成像(diffusion tensor imaging, DTI)图像中提取影像特征,运用随机森林建立IDH突变预测模型,加入DTI后的影像组学模型预测效能高于仅基于常规序列的影像组学模型(AUC 0.900 vs. 0.869)。REN等[13]从57名LGGs患者的T2-FLAIR、表观扩散系数(apparent diffusion coefficient, ADC)、指数化ADC(exponential ADC, eADC)和脑血流量(cerebral blood flow, CBF)中提取影像特征,建立最优SVM预测模型,AUC为0.931。但是KIM等[14]认为ADC参数特征未能提高IDH预测效能,却可以提高肿瘤分级的预测能力。

       研究发现,结合影像语义特征建立的MRI影像组学模型同样能够提升预测效能。如CAO等[15]从102名LGGs患者的T1WI、T2WI、CE-T1WI、T2-FLAIR及ADC图像中提取影像特征,运用随机森林建立影像组学模型预测IDH突变,AUC=0.849。结合常规序列提取的伦勃朗视觉感受图像(visually accessible Rembrandt images, VASARI)特征建立的融合组学模型预测效能得到提升,AUC=0.879。SUN等[16]发现除VASARI特征外,将T2-FLAIR错配标志(即肿瘤在T2WI上为完整或接近完整且几乎同质的高信号,而肿瘤在T2-FLAIR上为主体低信号,但有高信号的薄边缘[17])添加到多参数MRI影像组学模型中同样提高了IDH状态的预测效能。除影像语义外,ZHOU等[18]认为结合年龄信息也可提高模型的预测能力,并且相较于影像特征,年龄信息的预测价值更高。尽管基于MRI图像,运用机器学习建立的影像组学模型对LGGs中IDH突变的预测效能优异,但是图像运动伪影程度的增加可能降低预测效能。NALAWADE等[19]发现应用运动校正技术可恢复运动损坏图像,提高IDH分类精度,这种技术可能助力MRI影像组学预测IDH突变。

       上述研究证明基于MRI常规序列建立的影像组学模型在预测IDH突变方面具有重要价值,并且可以通过结合功能序列、影像语义特征或者年龄信息提高预测效能。除此之外,利用运动校正技术可降低图像运动伪影对IDH预测模型构建的影响。

2 影像组学预测1p/19q共缺失

       1p/19q共缺失的LGGs对放化疗更加敏感,是对总体生存率有利的预后因素[20]。对于具有1p/19q共缺失的3级少突胶质细胞瘤,推荐术后早期足量放疗加PCV化疗方案,对于无1p/19q共缺失者推荐放疗加辅助替莫唑胺化疗[21]。研究认为,无论IDH状态如何,基于MRI常规序列并运用机器学习算法建立的MRI影像组学模型可术前无创且高效地预测1p/19q 共缺失。如KONG等[22]从96名LGGs患者的CE-T1WI和T2WI图像中提取影像特征,运用随机森林建立1p/19q共缺失预测模型,AUC为0.889。CASALE等[23]从209名LGGs患者的CE-T1WI及T2WI图像中提取特征,同样运用随机森林分类器建立1p/19q共缺失预测模型,准确度为81%。KHA等[24]从159名LGG患者的T1WI或T2WI图像中提取7个影像特征并运用极端梯度提升算法(eXtreme Gradient Boosting, XGBoost)构建1p/19q共缺失预测模型,其准确度为82.8%。虽然上述机器学习算法对1p/19q共缺失效能不同,但KOCAK等[25]认为其性能差异不具有统计学意义。

       尽管运用机器学习算法建立的预测模型表现良好,但其需要纳入大量临床数据以增加病例数。不同来源的图像数据存在扫描机器和扫描参数的差异,这会对预测结果产生负面影响,而利用深度学习可减少上述影响从而提升预测性能[26]。YAN等[27]从330名LGGs患者的CE-T1WI、T1WI、T2WI、T2-FLAIR图像中提取影像特征,运用卷积神经网络(convolutional neural networks, CNN)建立预测模型,AUC为0.986。AKKUS等[28]基于CE-T1WI和T2WI图像,同样运用CNN建立1p/19q共缺失预测模型,其预测敏感度、特异度、准确度分别为93.30%、82.22%、87.70%。上述研究表明,基于MRI常规序列运用深度学习算法预测1p/19q共缺失效能更佳。除MRI常规序列外,有学者发现功能序列也对1p/19q共缺失具有良好预测能力。如LEWIS等[29]基于74名LGGs患者未经过滤的ADC影像特征预测1p/19q共缺失的敏感度、特异度、AUC分别为80.6%、89.3%、0.811。

       上述研究表明,仅基于MRI图像并运用机器学习及深度学习算法预测1p/19q状态是可靠的,这能为LGGs患者的术前个性化放化疗方案制订提供依据。但不同于年龄信息在IDH突变预测模型中的效能提升作用,年龄和性别信息的加入未能提高1p/19q共缺失预测模型的效能[30]

3 影像组学预测MGMT启动子甲基化

       氧6-甲基化鸟嘌呤DNA甲基转移酶(O6-methylguanine-DNA methyltransferase promoter methylation, MGMT)启动子甲基化在LGGs中的发生率约40%~80%[31]。众所周知,烷基化药物替莫唑胺是LGGs治疗中最重要的药物之一。研究表明MGMT启动子甲基化可以通过降低LGGs对替莫唑胺的抵抗性来延长患者生存期[32]。术前预测MGMT启动子甲基化可以提早制订化疗方案以便于及时进行术后干预,改善患者预后。

       研究发现,无论IDH状态如何,基于MRI常规序列建立的MRI影像组学模型预测MGMT启动子甲基化都是可行的。如YOGANANDA等[33]基于163名LGGs患者的单一T2WI图像,运用深度学习建立MGMT-net模型预测MGMT启动子甲基化,准确度、敏感度和特异度分别为94.73%、96.31%和91.66%,AUC为0.93。SHBOUL等[34]从108名LGGs患者的T1WI、CE-T1WI、T2WI、T2-FLAIR中提取影像特征,运用XGBoost建立MGMT启动子甲基化预测模型,AUC、敏感度和特异度分别为0.83、93%和73%。HUANG等[35]基于CE-T1WI、T2WI及T2-FLAIR图像并运用多因素逻辑回归建立MGMT启动子甲基化预测模型,AUC为0.833,敏感度为70.2%,特异度为90.6%。沙永建等[36]则以IDH突变型LGGs为研究对象,从158名IDH突变型LGGs患者的CE-T1WI及T2-FLAIR图像中提取特征,同样运用多因素逻辑回归建立MGMT启动子甲基化预测模型,AUC为0.935。这可能提示当研究队列为IDH突变型LGGs时,预测效能更高。JIANG等[37]发现加入临床因素未能提升预测效能,WEI等[38]的研究同样表明常规影像组学特征预测MGMT甲基化的表现优于临床因素。这可能是由于临床因素的加入使预测模型更为复杂,从而导致预测效能难以提升。

       上述研究表明,无论IDH状态如何,MRI影像组学模型预测MGMT启动子甲基化的预测效能较高。而临床因素预测能力次于影像组学特征,难以提升预测效能。

4 影像组学预测TERT突变

       几乎所有的IDH突变合并1p/19q共缺失的LGGs都存在TERT突变,这证明TERT突变广泛存在于LGGs中的少突胶质细胞瘤中[39, 40]。有研究发现,IDH突变合并TERT突变时患者预后更好,而IDH野生合并TERT突变时预后较差[41]

       基于MRI常规序列建立的MRI影像组学模型可以术前有效预测LGGs中TERT突变。FANG等[42]从164名LGGs患者的CE-T1WI、T1WI、T2WI中提取影像特征,并利用线性SVM建立TERT突变预测模型,AUC为0.8446,敏感度为93.55%,特异度为61.97%,总体准确度为79.88%。JIANG等[43]从83名LGGs患者的3D-CE-T1WI和T2WI图像中提取特征并构建影像组学模型预测TERT突变,AUC=0.82,并且发现肿瘤区域特征较瘤周区域更为重要。FUKUMA等[44]认为当研究队列仅限于IDH突变型LGGs时预测效能更高。他们基于MRI常规序列建立SVM模型预测TERT突变,准确度为59%,当队列仅限于IDH突变时准确度提升至84%。上述研究表明,无论IDH状态如何,基于MRI常规序列建立的影像组学模型预测TERT突变效能良好,并且当研究队列仅限于IDH突变型LGGs时可提高TERT突变的预测效能。除此之外,有研究表明将临床信息与MRI常规序列结合可提高预测效能,并可对LGGs患者进行预后风险分组[45]

5 影像组学预测其他基因状态

       研究发现约50%的LGGs患者中存在TP53突变[46],但对预后影响仍不清楚。某些TP53突变型LGGs通过Yes相关蛋白1(Yes-associated protein 1, YAP1)的参与而具有化学敏感性,TP53中密码子273突变和YAP1可能为显著的预后标记物[47]。ZHANG等[11]建立的SVM模型预测TP53突变,最佳准确度、敏感度、特异度和AUC分别为92.0%、96.6%、85.7%、0.949。这提示了MRI影像组学预测TP53突变具有不错的预测效能。

       LGGs中α-地中海贫血伴智力低下综合征基因(alpha-thalassemia/mental retardation syndrome, nondeletion type, X-linked, ATRX)突变概率大约为55%~80%[48, 49],并且ATRX突变与1p/19q共缺失几乎不会同时存在。研究发现ATRX突变是新的治疗靶点并且具有较好的预后[50]。LI等[51]基于186名LGGs患者的T2WI图像提取特征,运用SVM分类器建立ATRX突变预测模型,AUC为0.940,最佳敏感度、特异度和准确度分别为92.0%、89.5%和95.2%。由于90%以上的ATRX突变都伴随着IDH突变,所以部分研究以IDH突变型LGGs患者作为研究对象。如WU等[52]以78名IDH突变型LGGs患者为研究对象,从T2-FLAIR、CE-T1WI、CBF、ADC、eADC图中提取影像特征并与性别年龄结合建立预测模型,一致性指数为0.84。REN等[13]从36名IDH突变型LGGs患者的T2-FLAIR、ADC、eADC和CBF图像中提取特征并建立最优SVM预测模型,其预测ATRX未突变的精确度为91.67%。上述研究表明,不论IDH状态如何,基于MRI常规序列或功能序列,结合临床特征建立的MRI影像组学模型对ATRX突变具有良好预测效能。

       在对IDH突变型LGGs的研究中发现,存在CDKN2A/B纯合缺失的肿瘤患者生存率显著降低[53]。因此,2021年WHO CNS将CDKN2A/B纳入分子分型,并将CKDN2A/B纯合缺失的IDH突变型星形细胞瘤升级为WHO 4级,即便它不满足WHO 4级脑胶质瘤的组织学标准(即不存在坏死和微血管增生[8])。但目前尚无对于该基因的影像组学研究,这可能是LGGs影像组学的后续研究方向。

6 影像组学对多基因联合预测

       由于LGGs分子表型的多样性,目前部分研究基于MRI常规序列建立影像组学模型对多种分子表型进行联合预测,进一步细化LGGs分子亚型。如LAM等[54]基于MRI常规序列,运用XGBoost结合遗传算法建立MRI影像组学模型联合预测IDH和1p/19q状态,对LGGs三种亚型(IDH突变1p/19q共缺失、IDH突变1p/19q非共缺失、IDH野生型1p/19q非共缺失)进行分类,验证集总体精度为69.05%。ARITA等[10]基于MRI常规序列建立影像组学模型对3种分子亚型(IDH1/2突变型、TERT启动子突变合并IDH1/2突变和IDH野生型)进行分类,准确度为74%。ZHANG等[11]基于MRI常规序列构建影像组学模型联合预测IDH与TP53状态,对IDH突变TP53未突变、IDH突变TP53突变和IDH野生三种亚型预测的准确度分别为72.8%、71.9%、70%。上述研究表明,基于MRI常规序列建立的MRI影像组学模型在预测IDH联合1p/19q共缺失、TERT突变、TP53突变方面效能良好,可为LGGs患者个体化诊疗提供依据。

7 局限性与前景展望

       虽然MRI影像组学在预测LGGs分子分型方面效能良好,但是其还处于相对不成熟的阶段,仍然存在一些局限性。(1)多数研究为单机构的小样本研究,其研究结果单一且缺乏广泛验证。因此,建设更大的数据共享平台以推动多中心、大样本研究,便于检验和完善研究成果。(2)虽然影像组学可以挖掘图像隐藏的海量信息,但是不同的扫描机器、扫描方式以及算法往往难以达到高重复性,会影响模型的适用性。因此,需要进一步规范图像采集和信息处理技术。(3)大多数研究均为回顾性研究,存在选择偏倚和回忆偏倚的问题,且结论有限。应注重开展前瞻性研究,以便于临床应用。(4)尽管有部分功能序列被列入研究,但主体仍是常规序列。因此需要更多功能序列的加入以进一步探索影像组学预测能力。(5)虽然影像组学研究纳入了LGGs的部分关键基因,但是2021年WHO CNS 5新增的7号染色体扩增/10号染色体缺失以及CKDN2A/B突变未纳入。这可能是LGGs影像组学的后续研究方向。

8 小结

       影像组学作为新兴领域,更大程度地发挥了影像图像的内在价值。基于MRI常规序列及功能序列,运用机器学习和深度学习分析构建MRI影像组学模型,可术前无创、有效地预测LGGs分子分型,并且实现多基因联合预测。结合影像语义特征及临床信息,并选择最优算法,可进一步提升预测效能。这为LGGs的治疗评估及预后预测提供影像依据,可助力LGGs患者的临床精准诊疗。

[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. DOI: 10.1093/neuonc/noab106.
[2]
TOM M C, CAHILL D P, BUCKNER J C, et al. Management for Different Glioma Subtypes: Are All Low-Grade Gliomas Created Equal?[J]. Am Soc Clin Oncol Educ Book, 2019, 39: 133-145. DOI: 10.1200/EDBK_238353.
[3]
BRAT D J, VERHAAK R G, ALDAPE K D, et al. Comprehensive, Integrative Genomic Analysis of Diffuse Lower-Grade Gliomas[J]. N Engl J Med, 2015, 372(26): 2481-2498. DOI: 10.1056/NEJMoa1402121.
[4]
LOUIS D N, PERRY A, REIFENBERGER G, et al. The 2016 World Health Organization Classification of Tumors of the Central Nervous System: a summary[J]. Acta Neuropathol, 2016, 131(6): 803-820. DOI: 10.1007/s00401-016-1545-1.
[5]
BUCHLAK Q D, ESMAILI N, LEVEQUE J C, et al. Machine learning applications to neuroimaging for glioma detection and classification: An artificial intelligence augmented systematic review[J]. J Clin Neurosci, 2021, 89: 177-198. DOI: 10.1016/j.jocn.2021.04.043.
[6]
GILLIES R J, KINAHAN P E, HRICAK H. Radiomics: Images Are More than Pictures, They Are Data[J/OL]. Radiology, 2016, 278(2): 563-577 [2022-07-09]. https://doi.org/10.1148/radiol.2015151169. DOI: 10.1148/radiol.2015151169.
[7]
PATEL S H, BANSAL A G, YOUNG E B, et al. Extent of Surgical Resection in Lower-Grade Gliomas: Differential Impact Based on Molecular Subtype[J]. AJNR Am J Neuroradiol, 2019, 40(7): 1149-1155. DOI: 10.3174/ajnr.A6102.
[8]
DONO A, BALLESTER L Y, PRIMDAHL D, et al. IDH-Mutant Low-grade Glioma: Advances in Molecular Diagnosis, Management, and Future Directions[J/OL]. Curr Oncol Rep, 2021, 23(2): 20 [2022-07-09]. https://pubmed.ncbi.nlm.nih.gov/33492489/. DOI: 10.1007/s11912-020-01006-6.
[9]
HAN S, LIU Y, CAI S J, et al. IDH mutation in glioma: molecular mechanisms and potential therapeutic targets[J]. Br J Cancer, 2020, 122(11): 1580-1589. DOI: 10.1038/s41416-020-0814-x.
[10]
ARITA H, KINOSHITA M, KAWAGUCHI A, et al. Lesion location implemented magnetic resonance imaging radiomics for predicting IDH and TERT promoter mutations in grade Ⅱ/Ⅲ gliomas[J/OL]. Sci Rep, 2018, 8(1): 11773 [2022-07-09]. https://pubmed.ncbi.nlm.nih.gov/30082856/. DOI: 10.1038/s41598-018-30273-4.
[11]
ZHANG X, TIAN Q, WANG L, et al. Radiomics Strategy for Molecular Subtype Stratification of Lower-Grade Glioma: Detecting IDH and TP53 Mutations Based on Multimodal MRI[J]. J Magn Reson Imaging, 2018, 48(4): 916-926. DOI: 10.1002/jmri.25960.
[12]
PARK C J, CIHOL Y S, PARK Y W, et al. Diffusion tensor imaging radiomics in lower-grade glioma: improving subtyping of isocitrate dehydrogenase mutation status[J]. Neuroradiology, 2020, 62(3): 319-326. DOI: 10.1007/s00234-019-02312-y.
[13]
REN Y, ZHANG X, RUI W, et al. Noninvasive Prediction of IDH1 Mutation and ATRX Expression Loss in Low-Grade Gliomas Using Multiparametric MR Radiomic Features[J]. J Magn Reson Imaging, 2019, 49(3): 808-817. DOI: 10.1002/jmri.26240.
[14]
KIM M, JUNG S Y, PARK J E, et al. Diffusion- and perfusion-weighted MRI radiomics model may predict isocitrate dehydrogenase (IDH) mutation and tumor aggressiveness in diffuse lower grade glioma[J]. Eur Radiol, 2020, 30(4): 2142-2151. DOI: 10.1007/s00330-019-06548-3.
[15]
CAO M, SUO S, ZHANG X, et al. Qualitative and Quantitative MRI Analysis in IDH1 Genotype Prediction of Lower-Grade Gliomas: A Machine Learning Approach[J/OL]. Biomed Res Int, 2021, 2021: 1235314 [2022-07-09]. https://doi.org/10.1155/2021/1235314. DOI: 10.1155/2021/1235314.
[16]
SUN C, FAN L, WANG W, et al. Radiomics and Qualitative Features From Multiparametric MRI Predict Molecular Subtypes in Patients With Lower-Grade Glioma[J/OL]. Front Oncol, 2021, 11: 756828 [2022-07-09]. https://doi.org/10.3389/fonc.2021.756828. DOI: 10.3389/fonc.2021.756828.
[17]
JAIN R, JOHNSON D R, PATEL S H, et al. "Real world" use of a highly reliable imaging sign: "T2-FLAIR mismatch" for identification of IDH mutant astrocytomas[J]. Neuro Oncol, 2020, 22(7): 936-943. DOI: 10.1093/neuonc/noaa041.
[18]
ZHOU H, CHANG K, BAI H X, et al. Machine learning reveals multimodal MRI patterns predictive of isocitrate dehydrogenase and 1p/19q status in diffuse low- and high-grade gliomas[J]. J Neuro-Oncol, 2019, 142(2): 299-307. DOI: 10.1007/s11060-019-03096-0.
[19]
NALAWADW S S, YU F F, BANGALORE Y C G, et al. Brain tumor IDH, 1p/19q, and MGMT molecular classification using MRI-based deep learning: an initial study on the effect of motion and motion correction[J/OL]. J Med Imaging (Bellingham), 2022, 9(1): 016001 [2022-09-09]. https://doi.org/10.1117/1.Jmi.9.1.016001. DOI: 10.1117/1.Jmi.9.1.016001.
[20]
LU L, ZHANG Y, ZHAO Y, et al. Effects of 1p/19q Codeletion on Immune Phenotype in Low Grade Glioma[J/OL]. Front Cell Neurosci, 2021, 15: 704344 [2022-07-09]. https://doi.org/10.3389/fncel.2021.704344. DOI: 10.3389/fncel.2021.704344.
[21]
国家卫生健康委员会医政医管局, 中国抗癌协会脑胶质瘤专业委员会, 中国医师协会脑胶质瘤专业委员会. 脑胶质瘤诊疗指南(2022版)[J]. 中华神经外科杂志, 2022, 38(8): 757-777. DOI: 10.3760/cma.j.cn112050-20220510-00239.
National Health Commission Medical Administration Bureau, Glioma Professional Committee of Chinese Anti-Cancer Association, Glioma Professional Committee of Chinese Medical Doctor Association. Guidelines for diagnosis and treatment of glioma (2022 edition)[J]. Chin J Neurosurg, 2022, 38(8): 757-777. DOI: 10.3760/cma.j.cn112050-20220510-00239.
[22]
KONG Z, JIANG C, ZHANG Y, et al. Thin-Slice Magnetic Resonance Imaging-Based Radiomics Signature Predicts Chromosomal 1p/19q Co-deletion Status in Grade Ⅱ and Ⅲ Gliomas[J/OL]. Front Neurol, 2020, 11: 551771 [2022-07-09]. https://doi.org/10.3389/fneur.2020.551771. DOI: 10.3389/fneur.2020.551771.
[23]
CASALE R, LAVROVA E, SANDULEEANU S, et al. Development and external validation of a non-invasive molecular status predictor of chromosome 1p/19q co-deletion based on MRI radiomics analysis of Low Grade Glioma patients[J/OL]. Eur J Radiol, 2021, 139: 109678 [2022-07-09]. https://doi.org/10.1016/j.ejrad.2021.109678. DOI: 10.1016/j.ejrad.2021.109678.
[24]
KHA Q H, LE V H, HUNG T N K, et al. Development and Validation of an Efficient MRI Radiomics Signature for Improving the Predictive Performance of 1p/19q Co-Deletion in Lower-Grade Gliomas[J/OL]. Cancers, 2021, 13(21): 5398 [2022-07-09]. https://doi.org/10.3390/cancers13215398. DOI: 10.3390/cancers13215398.
[25]
KOCAK B, DURMAZ E S, ATES E, et al. Radiogenomics of lower-grade gliomas: machine learning-based MRI texture analysis for predicting 1p/19q codeletion status[J]. Eur Radiol, 2020, 30(2): 877-886. DOI: 10.1007/s00330-019-06492-2.
[26]
ALI M B, GU I Y, BERGER M S, et al. Domain Mapping and Deep Learning from Multiple MRI Clinical Datasets for Prediction of Molecular Subtypes in Low Grade Gliomas[J/OL]. Brain Sci, 2020, 10(7): 463 [2022-09-09]. https://doi.org/10.3390/brainsci10070463. DOI: 10.3390/brainsci10070463.
[27]
YAN J, ZHANG S, SUN Q, et al. Predicting 1p/19q co-deletion status from magnetic resonance imaging using deep learning in adult-type diffuse lower-grade gliomas: a discovery and validation study[J]. Lab Invest, 2022, 102(2): 154-159. DOI: 10.1038/s41374-021-00692-5.
[28]
AKKUS Z, ALI I, SEDLAR J, et al. Predicting Deletion of Chromosomal Arms 1p/19q in Low-Grade Gliomas from MR Images Using Machine Intelligence[J]. J Digit Imaging, 2017, 30(4): 469-476. DOI: 10.1007/s10278-017-9984-3.
[29]
LEWIS M A, GANESHAN B, BARNES A, et al. Filtration-histogram based magnetic resonance texture analysis (MRTA) for glioma IDH and 1p/19q genotyping[J]. Eur J Radiol, 2019, 113: 116-123. DOI: 10.1016/j.ejrad.2019.02.014.
[30]
LI Y, AMMARI S, LAWRANCE L, et al. Radiomics-Based Method for Predicting the Glioma Subtype as Defined by Tumor Grade, IDH Mutation, and 1p/19q Codeletion[J/OL]. Cancers, 2022, 14(7): 1778 [2022-09-09]. https://doi.org/10.3390/cancers14071778. DOI: 10.3390/cancers14071778.
[31]
中国脑胶质瘤协作组, 中国脑胶质瘤基因组图谱计划. 中国脑胶质瘤分子诊疗指南[J]. 中华神经外科杂志, 2014, 30(5): 435-444. DOI: 10.3760/cma.j.issn.1001-2346.2014.05.002.
China Glioma Collaborative Group, Chinese Glioma Genome Atlas Project Chinese Glioma Molecular Diagnosis and Treatment Guidelines[J]. Chin J Neurosurg, 2014, 30(5): 435-444. DOI: 10.3760/cma.j.issn.1001-2346.2014.05.002.
[32]
张宁, 马辉辉, 王凡, 等. 脑胶质瘤患者MGMT启动子甲基化和IDH1突变与临床预后相关[J].基础医学与临床, 2021, 41(10): 1476-1480. DOI: 10.3969/j.issn.1001-6325.2021.10.010.
ZHANG N, MA H H, WANG F, et al. Correlations of MGMT promoter methylation and IDH1 mutations with the clinical prognosis of malignant glioma patients[J]. Basic Clin Med, 2021, 41(10): 1476-1480. DOI: 10.3969/j.issn.1001-6325.2021.10.010.
[33]
YOGANANDA C G B, SHAN B R, NALAWADE S S, et al. MRI-Based Deep-Learning Method for Determining Glioma MGMT Promoter Methylation Status[J]. Am J Neuroradiol, 2021, 42(5): 845-852. DOI: 10.3174/ajnr.A7029.
[34]
SHBOUL Z A, CHEN J, KHAN M I. Prediction of Molecular Mutations in Diffuse Low-Grade Gliomas using MR Imaging Features[J/OL]. Sci Rep, 2020, 10(1): 3711 [2022-09-09]. https://pubmed.ncbi.nlm.nih.gov/32111869/. DOI: 10.1038/s41598-020-60550-0.
[35]
HUANG W Y, WEN L H, WU G, et al. Radiological model based on the standard magnetic resonance sequences for detecting methylguanine methyltransferase methylation in glioma using texture analysis[J]. Cancer Sci, 2021, 112(7): 2835-2844. DOI: 10.1111/cas.14918.
[36]
沙永建, 王效春, 谭艳, 等. MRI影像组学术前预测较低级别胶质瘤IDH突变合并MGMT启动子甲基化亚型的价值[J].磁共振成像, 2022, 13(7): 6-11. DOI: 10.12015/issn.1674-8034.2022.07.002.
SHA Y J, WANG X C, TAN Y, et al. The value of predicting the subtype of IDH mutation combining with MGMT promoter methylation in lower grade gliomas by radiomics based on preoperative MRI[J]. Chin J Magn Reson Imaging, 2022, 13(7): 6-11. DOI: 10.12015/issn.1674-8034.2202.07.002.
[37]
JIANG C, KONG Z, LIU S, et al. Fusion Radiomics Features from Conventional MRI Predict MGMT Promoter Methylation Status in Lower Grade Gliomas[J/OL]. Eur J Radiol, 2019, 121: 108714 [2022-07-09]. https://doi.org/10.1016/j.ejrad.2019.108714. DOI: 10.1016/j.ejrad.2019.108714.
[38]
WEI J, YANG G, HAO X, et al. A multi-sequence and habitat-based MRI radiomics signature for preoperative prediction of MGMT promoter methylation in astrocytomas with prognostic implication[J]. Eur Radiol, 2019, 29(2): 877-888. DOI: 10.1007/s00330-018-5575-z.
[39]
TERZI N K, YILMAZ I, OZ A B. The Place and Prognostic Value of TERT Promoter Mutation in Molecular Classification in Grade Ⅱ-Ⅱ Glial Tumors and Primary Glioblastomas[J]. Turk Patoloji Derg, 2022, 38(2): 90-98. DOI: 10.5146/tjpath.2021.01555.
[40]
ŚLEDZIŃSKA P, BEBYN M G, FURTAK J, et al. Prognostic and Predictive Biomarkers in Gliomas[J/OL]. Int J Mol Sci, 2021, 22(19): 10373 [2022-07-09]. https://pubmed.ncbi.nlm.nih.gov/34638714/. DOI: 10.3390/ijms221910373.
[41]
ARITA H, ICHIMURA K. Prognostic significance of TERT promoter mutations in adult-type diffuse gliomas[J]. Brain Tumor Pathol, 2022, 39(3): 121-129. DOI: 10.1007/s10014-021-00424-z.
[42]
FANG S, FAN Z, SUN Z, et al. Radiomics Features Predict Telomerase Reverse Transcriptase Promoter Mutations in World Health Organization Grade Ⅱ Gliomas via a Machine-Learning Approach[J/OL]. Front Oncol, 2020, 10: 606741 [2022-07-09]. https://doi.org/10.3389/fonc.2020.606741. DOI: 10.3389/fonc.2020.606741.
[43]
JIANG C, KONG Z, ZHANG Y, et al. Conventional magnetic resonance imaging-based radiomic signature predicts telomerase reverse transcriptase promoter mutation status in grade Ⅱ and Ⅲ gliomas[J]. Neuroradiology, 2020, 62(7): 803-813. DOI: 10.1007/s00234-020-02392-1.
[44]
FUKUMA R, YANAGISAWA T, KINOSHITA M, et al. Prediction of IDH and TERT promoter mutations in low-grade glioma from magnetic resonance images using a convolutional neural network[J/OL]. Sci Rep, 2019, 9(1): 201311 [2022-07-09]. https://pubmed.ncbi.nlm.nih.gov/31889117/. DOI: 10.1038/s41598-019-56767-3.
[45]
LU J, LI X, LI H. A radiomics feature-based nomogram to predict telomerase reverse transcriptase promoter mutation status and the prognosis of lower-grade gliomas[J/OL]. Clin Radiol, 2022, 77(8): e560-e567 [2022-09-09]. https://doi.org/10.1016/j.crad.2022.04.005. DOI: 10.1016/j.crad.2022.04.005.
[46]
GILLET E, ALENTORN A, DOUKOURE B, et al. TP53 and p53 statuses and their clinical impact in diffuse low grade gliomas[J]. Neuro Oncol, 2014, 118(1): 131-139. DOI: 10.1007/s11060-014-1407-4.
[47]
NOOR H, BRIGGS N E, MCDONALD K L, et al. TP53 Mutation Is a Prognostic Factor in Lower Grade Glioma and May Influence Chemotherapy Efficacy[J/OL]. Cancers, 2021, 13(21): 5362 [2022-07-09]. https://doi.org/10.3390/cancers13215362. DOI: 10.3390/cancers13215362.
[48]
IKEMURA M, SHIBAHRAR J, MUKASA A, et al. Utility of ATRX immunohistochemistry in diagnosis of adult diffuse gliomas[J]. Histopathology, 2016, 69(2): 260-267. DOI: 10.1111/his.12927.
[49]
NADNAKUMAP P, MANSOURI A, DAS S. The Role of ATRX in Glioma Biology[J/OL]. Front Oncol, 2017, 7: 236 [2022-07-09]. https://doi.org/10.3389/fonc.2017.00236. DOI: 10.3389/fonc.2017.00236.
[50]
XIE Y, TAN Y, YANG C, et al. Omics-based integrated analysis identified ATRX as a biomarker associated with glioma diagnosis and prognosis[J]. Cancer Biol Med, 2019, 16(4): 784-796. DOI: 10.20892/j.issn.2095-3941.2019.0143.
[51]
LI Y, LIU X, QIAN Z, et al. Genotype prediction of ATRX mutation in lower-grade gliomas using an MRI radiomics signature[J]. Eur Radiol, 2018, 28(7): 2960-2968. DOI: 10.1007/s00330-017-5267-0.
[52]
WU S, ZHANG X, RUI W, et al. A nomogram strategy for identifying the subclassification of IDH mutation and ATRX expression loss in lower-grade gliomas[J]. Eur Radiol, 2022, 32(5): 3187-3198. DOI: 10.1007/s00330-021-08444-1.
[53]
HUANG L E. Impact of CDKN2A/B Homozygous Deletion on the Prognosis and Biology of IDH-Mutant Glioma[J/OL]. Biomedicines, 2022, 10(2): 246 [2022-09-09]. https://doi.org/10.3390/biomedicines10020246. DOI: 10.3390/biomedicines10020246.
[54]
LAM L H T, DO D T, DIEP D T N, et al. Molecular subtypes classification of low-grade gliomas patients using MRI-based radiomics and machine learning[J/OL]. NMR Biomed, 2022, 35(11): e4792 [2022-09-09]. https://doi.org/10.1002/nbm.4792. DOI: 10.1002/nbm.4792.

上一篇 卒中后失语症的静息态功能磁共振成像相关研究进展
下一篇 MRI在显示腮腺内面神经中的应用进展
  
诚聘英才 | 广告合作 | 免责声明 | 版权声明
联系电话:010-67113815
京ICP备19028836号-2