分享:
分享到微信朋友圈
X
临床研究
动态对比增强及平均表观传播子磁共振成像在预测高级别胶质瘤MGMT启动子甲基化状态中的应用价值
袁鹏翾 高阳 吴琼 张华鹏 王少彧

Cite this article as: YUAN P X, GAO Y, WU Q, et al. The value of DCE and MAP-MRI in predicting the methylation status of MGMT promoter in high-grade glioma[J]. Chin J Magn Reson Imaging, 2023, 14(5): 85-91.本文引用格式:袁鹏翾, 高阳, 吴琼, 等. 动态对比增强及平均表观传播子磁共振成像在预测高级别胶质瘤MGMT启动子甲基化状态中的应用价值[J]. 磁共振成像, 2023, 14(5): 85-91. DOI:10.12015/issn.1674-8034.2023.05.016.


[摘要] 目的 探讨应用平均表观传播子MRI(mean apparent propagator-magnetic resonance imaging, MAP-MRI)及动态对比增强MRI(dynamic contrast enhanced magnetic resonance imaging, DCE-MRI)预测3、4级胶质瘤患者O6-甲基鸟嘌呤-DNA-甲基转移酶(O6-methylguanine-DNA methyhransferase, MGMT)启动子甲基化状态的可行性。材料与方法 前瞻性纳入本院自2018年6月至2022年1月经病理证实的14例MGMT启动子甲基化和17例MGMT启动子非甲基化胶质瘤患者,进行了术前常规MRI、DCE-MRI及MAP-MRI扫描。通过手动勾画肿瘤实质区域为感兴趣区(region of interest, ROI)并提取ROI定量参数特征,测量DCE-MRI参数及MAP-MRI参数。参数与MGMT甲基化间的相关性采用Pearson相关分析。所有参数均采用两独立样本t检验,比较DCE-MRI和MAP-MRI对预测3、4级胶质瘤MGMT启动子甲基化状态的诊断效能。进一步建立单因素和多因素logistic回归模型,分析构建受试者工作特征(receiver operating characteristic, ROC)曲线,以DeLong检验比较DCE-MRI参数、MAP-MRI参数与多参数联合模型预测MGMT甲基化的诊断效果。结果 DCE-MRI参数容积转运常数(volume transfer constant, Ktrans)、血管外细胞外容积分数(fractional volume of the extravascular-extracellular space, Ve)以及MAP-MRI参数非高斯(non-Gaussianity, NG)、非高斯轴向(non-Gaussianity axial, NGAx)、Q空间逆方差(Q-space inverse variance, QIV)、返回原点概率(return to the origin probability, RTOP)、返回轴线概率(return to the axis probability, RTAP)与MGMT启动子甲基化间呈中等相关性,对预测3、4级胶质瘤MGMT启动子甲基化与非甲基化组间差异具有统计学意义(P<0.05);ROC曲线下面积(area under the curve, AUC)分别为0.803、0.815、0.807、0.803、0.765、0.790、0.739。多因素logistic分析显示Ve是预测MGMT启动子甲基化的最佳预测因子,其准确性最高,AUC为0.815(95% CI:0.659~0.971),比值比(odds ratio, OR)为0.891(95% CI:0.815~0.975)。DeLong检验结果表明DCE-MRI和MAP-MRI多参数联合后预测胶质瘤MGMT启动子甲基化状态的诊断效能最高,AUC为0.992。结论 DCE-MRI和MAP-MRI对于预测高级别胶质瘤MGMT启动子甲基化状态具有一定的应用价值,同时应用两者联合诊断将有助于进一步提高诊断的效能。
[Abstract] Objective To investigate the feasibility of predicting the methylation status of O6-methylguanine-DNA-methyltransferase (MGMT) promoter in patients with grade 3 and 4 gliomas by mean apparent propagator-MRI (MAP-MRI) and dynamic contrast enhanced MRI (DCE-MRI).Materials and Methods From June 2018 to January 2022, 14 patients with MGMT promoter methylation and 17 patients with MGMT promoter non-methylated gliomas confirmed by pathology in our hospital were prospectively enrolled. Preoperative routine magnetic resonance imaging, DCE-MRI and MAP-MRI scans were performed.The tumor parenchymal region was manually delineated as region of interest (ROI) and the quantitative parameters of ROI were extracted to measure the parameters of DCE-MRI and MAP-MRI. Pearson correlation analysis was used to analyze the correlation between parameters. Two independent sample t-test was used to compare the diagnostic efficacy of DCE-MRI and MAP-MRI in predicting MGMT promoter methylation status of grade 3 and 4 gliomas.Univariate and multivariate logistic regression models were further established, and receiver operating characteristic (ROC) curves were analyzed and constructed. DeLong test was used to compare the diagnostic effects of DCE-MRI parameters, MAP-MRI parameters and multi-parameter combined model in predicting MGMT methylation.Results The parameters volume transfer constant (Ktrans) and fractional volume of the extravascular-extracellular space (Ve) of DCE-MRI and non-Gaussianity (NG), non-Gaussianityaxial (NGAx), Q-spaceinversevariance (QIV), return to the origin probability (RTOP), return to the axis probability (RTAP) of MAP-MRI were moderately correlated with MGMT promoter methylation, and the differences between the two groups were statistically significant (P<0.05). The area under the ROC curve (AUC) was 0.803, 0.815, 0.807, 0.803, 0.765, 0.790, 0.739, respectively. Multivariate logistic analysis showed that Ve was the best predictor of MGMT promoter methylation, with the highest accuracy and AUC of 0.815 (95% CI: 0.659-0.971), odds ratio (OR) of 0.891 (95% CI: 0.815-0.975). The results of DeLong test showed that the combined multi-parameter model of DCE-MRI and MAP-MRI had the highest diagnostic efficiency in predicting the methylation status of the MGMT promoter in glioma, with an AUC of 0.992.Conclusions DCE-MRI and MAP-MRI are valuable for predicting the methylation status of the MGMT promoter in high-grade gliomas, and the simultaneous application of the two combined diagnoses will help to further improve the diagnostic efficacy.
[关键词] 高级别胶质瘤;磁共振成像;动态对比增强磁共振成像;平均表观传播子磁共振成像;O6-甲基鸟嘌呤-DNA-甲基转移酶;分子分型
[Keywords] high-grade glioma;magnetic resonance imaging;dynamic contrast-enhanced-magnetic resonance imaging;mean apparent propagator-magnetic resonance imaging;O6-methylguanine-DNA methyltransferase;molecular subtype

袁鹏翾 1   高阳 1*   吴琼 1   张华鹏 2   王少彧 2  

1 内蒙古医科大学附属医院影像诊断科,呼和浩特 010050

2 西门子医疗系统有限公司,上海 201318

通信作者:高阳,E-mail:1390903990@qq.com

作者贡献声明:高阳设计本研究的方案,对稿件重要内容进行了修改;袁鹏翾起草和撰写稿件,获取、分析或解释本研究的数据;吴琼、张华鹏、王少彧获取、分析或解释本研究的数据,对稿件重要内容进行了修改;袁鹏翾获得了基金项目支持;全体作者都同意发表最后的修改稿,同意对本研究的所有方面负责,确保本研究的准确性和诚信。


基金项目: 内蒙古自治区科技计划项目 2019GG047
收稿日期:2022-09-16
接受日期:2023-05-05
中图分类号:R445.2  R730.264 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2023.05.016
本文引用格式:袁鹏翾, 高阳, 吴琼, 等. 动态对比增强及平均表观传播子磁共振成像在预测高级别胶质瘤MGMT启动子甲基化状态中的应用价值[J]. 磁共振成像, 2023, 14(5): 85-91. DOI:10.12015/issn.1674-8034.2023.05.016.

0 前言

       高级别胶质瘤术后需要进行联合放射治疗和替莫唑胺化疗或靶向治疗。然而,即使积极采用上述的多种方法对患者进行治疗,患者的预后通常依旧无法令人满意[1, 2]。高级别胶质瘤患者的预后受到多重因素的影响,其中对放化疗的敏感程度尤为重要,目前已有文献指出患者在治疗过程中对常用化疗药替莫唑胺产生耐药是治疗效果不佳的原因之一[3]。而O6-甲基鸟嘌呤-DNA-甲基转移酶(O6-methylguanine-DNA methyhransferase, MGMT)甲基化的患者更倾向于在替莫唑胺治疗中获益,因此,检测胶质瘤患者MGMT启动子甲基化的状态对于临床治疗具有重要的指导价值[4, 5]。评价MGMT启动子甲基化状态的金标准目前仍是通过手术或病理活检对组织标本进行相应的分子检测,但是,这种有创性检测手段对于无法进行手术或病理取材的部分患者将无法进行有效的MGMT的启动子甲基化状态检测。近年来,随着磁共振功能成像技术的快速发展,其能够从不同的角度评价神经胶质瘤的生物学特性,而不只是局限于单纯的解剖特征。越来越多的相关性研究在胶质瘤的术前诊断分子基因型、预测预后等方面显示出了较高的可靠性和准确性,功能MRI技术因此被认为对于胶质瘤患者的诊疗计划具有重大价值。

       平均表观传播子MRI(mean apparent propagator-MRI, MAP-MRI)是近几年发展出的一种新空间数据获取分析模型,是一种多b值多方向的q空间成像,可以有效地反映组织复杂微结构中质子的状态,从而反映出复杂微观结构(如扩散限制、多腔室)中的扩散情况,捕获到神经组织中微观结构的内在特征[6, 7]。而动态对比增强MRI(dynamic contrast enhanced MRI, DCE-MRI)技术对血脑屏障(blood brain barrier, BBB)破坏敏感,与微血管密度密切相关,可有效评价肿瘤血管生成,能够获得反映组织微循环功能的各项参数。这两种不同的技术模型或许对预测肿瘤的分级及分子分型方面存在互补作用。目前,已有部分研究表明DCE-MRI及MAP-MRI具有评估胶质瘤分级以及分子基因型的能力[8, 9, 10, 11],但是目前为止关于应用MAP-MRI及DCE-MRI来预测胶质瘤MGMT启动子甲基化状态的报道却少有涉及,本研究通过DCE-MRI联合MAP-MRI对胶质瘤MGMT启动子甲基化状态进行定量分析,旨在利用MAP-MRI及DCE-MRI的定量参数来评估其在预测胶质瘤患者MGMT甲基化状态的准确性。

1 材料与方法

1.1 病例资料

       本研究遵守《赫尔辛基宣言》,经内蒙古医科大学附属医院伦理委员会批准,全体受试者均签署了知情同意书,批准文号:WZ(2022037)。本项研究的纳入标准:(1)按照2021年世界卫生组织(World Health Organization, WHO)中枢神经系统肿瘤分类诊断为成人型弥漫性胶质瘤;(2)所有患者均在术前行MRI检查,包括常规MRI、DCE及DSI,且图像质量理想。排除标准:(1)检查前或术前进行过治疗(包括放疗、化疗、放化疗或激素治疗);(2)难以勾画感兴趣区(region of interest, ROI)的胶质瘤;(3)低级别胶质瘤(WHO 1级和2级)。

1.2 图像采集

       所有受试者均在3.0 T 磁共振扫描仪(Skyra,Siemens Healthcare, Erlangen, Germany)和32通道正交头线圈上进行了术前MRI扫描,扫描序列包括常规扫描(T1WI、T2WI、扩散加权成像及对比增强T1WI)、扩散谱成像(diffusion spectrum imaging, DSI)和DCE-MRI。DSI序列的扫描参数如下:TR/TE 7000 ms/107 ms,FOV 260 mm×260 mm,一般性自动校准部分并行采集2次,层厚3.0 mm,体素大小2.2 mm×2.2 mm×3.0 mm,50层,Q空间扩散模式,最大b值3000 s/mm2,扫描时间为15 min 40 s。轴位DCE-MRI采用体积内插梯度回波序列,每位患者均扫描35个时相,成像时间为3 min 48 s。注药时机选择为第3个时相,采用双筒高压注射器,经肘静脉套管针(20 G)注射对比剂(钆喷酸葡胺,北京北陆药业股份有限公司,中国),剂量为0.1 mmol/kg,注射速率为2.5 mL/s,对比剂注射完毕后,立刻注射20 mL生理盐水以相同速率冲洗连接管。DCE-MRI的扫描参数如下:TR/TE 5.08 ms/1.02 ms,FOV 220 mm×220 mm,层厚3.5 mm,体素大小1.1 mm×1.1 mm×3.5 mm,矩阵192 ×192。

1.3 图像后处理及分析

       MAP-MRI图像后处理使用西门子公司内部开发的名为NeuDiLab的工具进行计算,该工具是基于开放资源工具DIPY(Diffusion Imaging In Python,http://nipy.org/dipy)的基础上用Python内部开发的。获得的MAP-MRI参数包括非高斯(non-Gaussianity, NG)、非高斯轴向(non-Gaussianity axial, NGAx)、非高斯纵向(non-Gaussianity vertical, NGRad)、Q空间逆方差(Q-space inverse variance, QIV)、返回原点概率(return to the origin probability, RTOP)、返回轴线概率(return to the axis probability, RTAP)、返回平面概率(return to the plane probability, RTPP)和均方位移(mean square displacement, MSD)。DCE-MRI图像是在Siemens Syngo.via工作站上通过标准软件包进行后处理,应用Toft extended linear血流动力学模型计算出各个病灶的定量参数,包括容积转运常数(volume transfer constant, Ktrans)、回流速率常数(the rate constant, Kep)和血管外细胞外容积分数(fractional volume of the extravascular-extracellular space, Ve)。使用ITK-SNAP软件(www.itkSnap.org)手动勾画肿瘤实质区域为ROI,肿瘤实体部分分别测量3次,测量过程中尽量避开坏死、出血、囊变及钙化的部分,在肿瘤最大直径的层面上圈画出肿瘤实质所在的区域,每个ROI取其测量平均值。如果存在肿瘤存在强化,则选择肿瘤强化区域。如果肿瘤无强化,则选择表观扩散系数(apparent diffusion coefficient, ADC)图信号最低的区域作为ROI,参考T2液体衰减反转恢复图像排除瘤旁的正常组织和瘤周水肿区。

1.4 统计学分析

       采用统计学软件SPSS 24.0(SPSS, Inc., Chicago, IL, USA)及MedCalc(version 12.1.0. Inc., Mariakerke, Belgium)进行数据分析,采用两独立样本t检验鉴别MGMT甲基化组与非甲基化组的年龄之间的差异及DCE-MRI、MAP-MRI参数在MGMT甲基化别组与非甲基化别组间的差异,应用卡方检验鉴别MGMT甲基化组与非甲基化组的性别和3级、4级胶质瘤级别之间的差异。DCE-MRI、MAP-MRI参数与MGMT甲基化间的相关性采用Pearson相关分析检验。对差异具有统计学意义(P<0.05)的DCE-MRI参数和MAP-MRI参数通过采用向前逐步回归构建单变量和多变量logistic回归模型,并进一步绘制受试者工作特征(receiver operating characteristic, ROC)曲线,计算曲线下面积(area under the curve, AUC)、敏感度、特异度以DeLong检验比较DCE-MRI、MAP-MRI参数与多参数联合模型对MGMT甲基化的诊断效果。

2 结果

2.1 患者的基本人口统计学资料结果

       本研究共纳入31名胶质瘤患者(男15例,女16例,年龄范围为23~70岁)。根据WHO分类,8例为3级胶质瘤,23例为4级胶质母细胞瘤。根据MGMT启动子状态,将患者分为两组,即MGMT甲基化组(n=14)及MGMT非甲基化组(n=17)。在此项研究中MGMT的状态与患者的年龄、性别和胶质瘤级别间的差异均无统计学意义(P值均>0.05)。MGMT不同表达状态的组间人口统计学资料结果如表1所示。

表1  两组患者的一般资料比较
Tab. 1  Comparison of general data between two groups of patients

2.2 DCE-MRI参数对高级别胶质瘤MGMT启动子状态的诊断效能

       统计结果显示,MGMT非甲基化组的Ktrans值及Ve值明显高于MGMT甲基化组,且差异有统计学意义(P值均<0.05),而两组之间的Kep值差异无统计学意义(P值>0.05;表2图1、2)。

图1  女,67岁,WHO 4级胶质母细胞瘤,MGMT甲基化。1A~1C分别为Kep、Ktrans、Ve图。
图2  男,55岁,WHO 4级胶质母细胞瘤,MGMT非甲基化。1A~1C分别为Kep、Ktrans、Ve图。MGMT:O6-甲基鸟嘌呤-DNA-甲基转移酶;Kep:回流速率常数;Ktrans:容积转运常数;Ve:血管外细胞外容积分数。
Fig. 1  Female, 67 years old, WHO grade 4 glioblastoma, MGMT methylation. DCE-MRI with Kep (1A), Ktrans (1B), Ve (1C) maps demonstrating the lesion.
Fig. 2  Male, 55 years old, glioblastoma WHO 4 grade, MGMT non-methylation. DCE-MRI with Kep (2A), Ktrans (2B), Ve (2C) maps demonstrating the lesion. MGMT: O6-methylguanine-DNA methyhransferase; DCE: dynamic contrast enhanced; Kep: the rate constant;Ktrans: volume transfer constant; Ve: fractional volume of the extravascular-extracellular space.
表2  两组DCE-MRI参数统计学分析结果
Tab. 2  Statistical analysis of DCE-MRI parameters in two groups

2.3 MAP-MRI参数对胶质瘤MGMT启动子状态的诊断效能

       对于MAP-MRI参数,甲基化组的NG、NGAx、RTOP及RTAP值明显低于非甲基化组(P值均<0.05),而甲基化组的QIV的值明显高于非甲基化组(P=0.033;表3图3、4)。

图3  女,67岁,WHO 4级胶质母细胞瘤,MGMT甲基化。3A~3H分别为MSD、NG、NGAx、NGRad、QIV、RTAP、RTOP、RTPP图。
图4  男,55岁,WHO 4级胶质母细胞瘤,MGMT非甲基化。4A~4H分别为MSD、NG、NGAx、NGRad、QIV、RTAP、RTOP、RTPP图。MGMT:O6-甲基鸟嘌呤-DNA-甲基转移酶;MSD:均方位移;NG:非高斯;NGAx:非高斯轴向;NGRad:非高斯纵向;QIV:Q空间逆方差;RTAP:返回轴线概率;RTOP:返回原点概率;RTPP:返回平面概率。
Fig. 3  Female, 67 years old, WHO grade 4 glioblastoma, MGMT methylation. MAP-MRI with MSD (3A), NG (3B), NGAx (3C), NGRad (3D), QIV (3E), RTAP (3F), RTOP (3G) and RTPP (3H) maps demonstrating the lesion.
Fig. 4  Male, 55 years old, WHO grade 4 glioblastoma, MGMT non-methylation. MAP-MRI with MSD (4A), NG (4B), NGAx (4C), NGRad (4D), QIV (4E), RTAP (4F), RTOP (4G) and RTPP (4H) maps demonstrating the lesion. MGMT: O6-methylguanine-DNA methyhransferase; MSD: mean square displacement; NG: non-Gaussianity; NGAx: non-Gaussianity axial; NGRad: non-Gaussianity vertical; QIV: Q-space inverse variance; RTAP: return to the axis probability; RTOP: return to the origin probability; RTPP: return to the plane probability.
表3  两组MAP-MRI参数统计学分析结果
Tab. 3  Statistical analysis of MAP-MRI parameters in two groups

2.4 DCE-MRI及MAP-MRI参数与MGMT甲基化的相关性

       在不同的参数中,MGMT甲基化与NG(r=-0.548,P=0.001)、NGAx(r=-0.509,P=0.003)、RTOP(r=-0.499,P=0.004)、RTAP(r=-0.484,P=0.006)、Ktransr=-0.492,P=0.005)、Ve(r=-0.537,P=0.002)、QIV(r=0.383,P=0.033)存在中等相关性,与MSD、NGRad、Kep无显著相关性(P>0.05)。

2.5 Logistic回归分析、建立联合模型与ROC曲线分析

       单因素分析显示,NG、NGAx、QIV、RTOP、RTAP、Ktrans及Ve是MGMT启动子甲基化的预测因子,MAP-MRI的参数值NG、NGAx、RTOP、RTAP和QIV的AUC分别为0.807、0.803、0.790、0.739、0.765。DCE-MRI的参数值Ktrans和Ve的AUC分别为0.803和0.815。多因素logistic回归分析显示,Ve是预测MGMT启动子甲基化的最佳预测因子,其准确性最高,AUC为0.815(95% CI:0.659~0.971),比值比(odds ratio, OR)为0.891(95% CI:0.815~0.975),P=0.012。多参数联合模型对区分MGMT启动子甲基化具有最高的诊断效能,AUC为0.992(95% CI:0.970~1.000),R2为0.923,敏感度为92.9%,特异度为100%(表4图5)。

图5  DCE-MRI、MAP-MRI参数及DCE-MRI和MAP-MRI多参数联合鉴别高级别胶质瘤MGMT甲基化与非甲基化的ROC曲线。DCE:动态对比增强;MAP:平均表观传播子;ROC:受试者工作特征;Ktrans:容积转运常数;NG:非高斯;NGAx:非高斯轴向;RTAP:返回轴线概率;RTOP:返回原点概率;Ve:血管外细胞外容积分数;QIV:Q空间逆方差;联合:多参数联合模型。
Fig. 5  ROC curve of DCE-MRI, MAP-MRI parameters and DCE-MRI and MAP-MRI multi-parameters to distinguish MGMT methylation from non-methylation in high-grade gliomas. ROC: receiver operating characteristics; DCE: dynamic contrast-enhanced; MAP: mean apparent propagator; Ktrans: volume transfer constant; NG: non-Gaussianity; NGAx: non-Gaussianity axial; RTAP: return to the axis probability; RTOP: return to the origin probability; Ve: fractional volume of the extravascular-extracellular space; QIV: Q-space inverse variance.
表4  两组DCE-MRI及MAP-MRI参数及多参数联合模型的ROC分析
Tab. 4  ROC analysis of DCE-MRI and MAP-MRI parameters and multi-parameter combined model of the two groups

3 讨论

       本研究的主要目的是评估DCE-MRI以及MAP-MRI技术区分胶质瘤患者MGMT启动子甲基化状态的能力。研究结果显示,MAP-MRI的参数NG、NGAx、RTOP、RTAP和QIV以及DCE-MRI的参数Ktrans、Ve在预测3、4级胶质瘤MGMT启动子甲基化状态中具有一定的应用价值,其中Ve是独立预测MGMT甲基化的最佳预测因子,其可能反映了肿瘤微血管的细胞密度。在对DCE-MRI和MAP-MRI参数进行了相关性分析后,NG、NGAx、RTOP、RTAP、QIV、Ktrans、Ve与MGMT甲基化呈中等相关性。将以上7种参数联合后的多参数联合模型与单一参数相比,多参数预测模型具有最高的诊断效能。另外,构建出的多参数模型的在拟合度方面具有良好的性能,这表明联合应用DCE-MRI与MAP-MRI评估MGMT甲基化状态优于单一应用以上两种磁共振技术评估MGMT甲基化状态,这将有助于对个别患者作出更好的治疗选择,或可成为一种新的预测MGMT甲基化的无创性检查方法。据我们所知,本研究是国内外首次同时将DCE-MRI与MAP-MRI两种技术应用于预测MGMT启动子甲基化状态的研究。基于2021版WHO指南,MGMT启动子甲基化被认为与胶质瘤患者的预后具有密切联系[12, 13]。MGMT是一种DNA修复蛋白,其与烷化剂作用的主要位点——O6-烷基鸟嘌呤O6位的烷基相关联,具有修复烷化剂对DNA造成的损伤的功能,并最终导致对烷化剂耐药,是相同级别的脑胶质瘤患者对化疗药物敏感性不同的重要原因。MGMT启动子甲基化直接决定了MGMT的表达水平,大量研究结果表明MGMT启动子甲基化的胶质瘤患者对于替莫唑胺化疗具有更高的敏感性[14, 15, 16],与MGMT启动子甲基化的患者相比较,MGMT非甲基化的胶质瘤患者的化疗效果则更差。另一方面,最近的一部分研究还发现,MGMT蛋白的表达与患者生存期显著相关,MGMT启动子甲基化往往预示着患者有更长的无进展生存期和总生存期[17]。因此,术前无创且精确地预测MGMT启动子甲基化状态对于患者的临床治疗有着至关重要的作用。

3.1 MAP-MRI各参数评估MGMT启动子甲基化状态的效能

       MAP-MRI作为一种新的空间数据获取分析模型,其对于脑组织的复杂微观结构改变更具有敏感性。在本次研究中,与MGMT非甲基化组相比,MGMT甲基化组具有更高的QIV值以及更低的NG、NGAx、RTOP、RTAP值。QIV被定义为q空间逆方差,其对于组织变化以及扩散受限较为敏感[18, 19, 20, 21]。此前的研究表明RTOP 值可以反映组织异质性和局限性,还能很好地反映白质和灰质的微观结构变化,其中白质中的RTOP高于灰质,另外,RTOP可以在扩散张量方向确定的局部解剖参考系统中分解为RTAP及RTPP[6]。NG也可以反映组织结构的复杂性,NG的轴向和径向分裂分别为NGAx和NGRad,其也可以间接地反映出组织结构的变化程度[22]。另外,NG是其他扩散异质性来源的高度敏感指标,能够更好地评估白质纤维束跨区域的各向异性,并且在不同组织类型中具有更大的动态范围,这很有可能是本研究中NG诊断效能更高的原因。胶质瘤微观结构的复杂性限制了水分子的自由扩散,水分子周围的细胞结构越复杂,扩散速度越慢,本研究结果显示MGMT非甲基化组的QIV显著低于MGMT甲基化组,证实了MGMT非甲基化的胶质瘤水分子在扩散中遇到了更多的障碍,并且肿瘤具有较高的细胞密度或细胞结构复杂性,这与以往研究的结果相符[23]。有研究指出,MGMT甲基化的肿瘤细胞生长更为缓慢,对周围组织结构的浸润较少,同时具有较少的组织结构复杂性,肿瘤细胞的构成也更为均匀,内皮细胞增殖异质性较好,因此更不容易形成囊变、坏死等[24]。这充分解释了本研究中MGMT甲基化组中的NG、NGAx、RTOP、RTAP值相较MGMT非甲基化组更低的原因。本研究中MSD、NGRad、RTPP在MGMT甲基化和非甲基化组之间不具显著差异,其中NGRad和RTPP的P值接近0.05,我们推测有可能是本研究的样本数量较少所导致。目前为止,尚未有研究将MAP-MRI应用于预测胶质瘤MGMT甲基化中,而通过本研究,MAP-MRI对于评估胶质瘤MGMT甲基化状态展现出巨大潜力,能够很好地预测胶质瘤MGMT甲基化状态,我们的研究结果为更准确地对预测MGMT甲基化状态提供了依据。

3.2 DCE-MRI各参数评估MGMT启动子甲基化状态的效能

       胶质瘤新生血管的活跃程度是评价肿瘤恶性程度的一个重要指标,胶质瘤肿瘤区域的BBB功能及结构通常破坏严重,因此造成胶质瘤微血管的通透性增加,形成了胶质瘤具有高度侵袭性的特点[25]。DCE-MRI对于反映肿瘤和瘤周微循环的情况具有巨大优势,可以得到量化BBB破坏程度以及反映组织微循环功能和结构的多种参数,而得出的参数在评价血脑屏障损伤程度方面具有巨大的潜力[26]。有研究表明,MGMT启动子甲基化会使血管内皮生长因子受体(vascular endothelial growth factor receptor, VEGFA)水平发生一定程度上的变化,进而影响肿瘤新生血管的生成以及血管通透性的改变。这一机制解释了本研究中MGMT甲基化组的Ktrans和Ve值低于MGMT非甲基化组的结果,证实了MGMT启动子非甲基化组的胶质瘤具有更高的血管通透性、更快的血管生成速度以及更严重的血脑屏障破坏程度。本研究结果中,Ktrans与Ve对于预测MGMT甲基化展现出优势,另外Ve是独立预测MGMT甲基化的最佳预测因子,这与ZHANG等[27]的研究结果相类似,而与HILARIO等[9]及AHN等[28]的结果相反,这可能与上述研究未充分考虑样本的异柠檬酸脱氢酶(isocitrate dehydrogenase, IDH)突变状态有关,已有学者证实IDH突变型胶质瘤更容易同时发生MGMT启动子甲基化[29, 30],另外,还有研究认为,IDH野生型合并MGMT甲基化的高级别胶质瘤患者对烷化剂更敏感[31, 32],而本研究中的样本是IDH野生型合并MGMT甲基化的高级别胶质瘤。此外,本研究中MGMT甲基化组与MGMT非甲基化组之间的Kep值不具有显著差异。这与上述研究结果相一致。在另一种灌注成像技术——动态磁敏感对比增强成像中,有研究者[33]发现非甲基化组的相对脑血容量高于甲基化组,他们推测是由于MGMT非甲基化组的血管密度较高所导致,这同样与本研究的研究结果相一致。

3.3 局限性

       本研究同样存在一定的局限性及不足:首先,本研究的样本数量相对较少且为前瞻性研究,这可能造成结果的可解释性不足。其次,勾画肿瘤区域时在一定程度上会受到勾画者的主观影响且由于肿瘤具有异质性和肿瘤的采样问题,勾画的ROI与病理检测区域或许并不完全一致,这可能会造成一定程度上的误差。最后,本研究的样本组成主要是IDH野生型的胶质母细胞瘤,这有可能因此产生选择上的偏倚。未来随着研究样本量的进一步扩大以及影像组学及纹理分析的进展,上述问题或将得到解决。

4 结论

       综上所述,我们的研究证明了DCE-MRI及MAP-MRI对于评估MGMT启动子甲基化状态具有重要意义,其中Ve是以上两种技术独立预测MGMT甲基化的最佳预测因子,而结合以上两种技术的多种参数则会进一步提高MGMT甲基化的诊断效能,能够为患者对于化疗药物替莫唑胺的使用决策提供重要的依据,将更好地促进患者的正确治疗决策。

[1]
RENI M, MAZZA E, ZANON S, et al. Central nervous system gliomas[J]. Crit Rev Oncol Hematol, 2017, 113: 213-234. DOI: 10.1016/j.critrevonc.2017.03.021">10.1016/j.critrevonc.2017.03.021">10.1016/j.critrevonc.2017.03.021.
[2]
KAN L K, DRUMMOND K, HUNN M, et al. Potential biomarkers and challenges in glioma diagnosis, therapy and prognosis[J/OL]. BMJ Neurol Open, 2020, 2(2): e000069 [2022-08-10]. https://pubmed.ncbi.nlm.nih.gov/33681797/. DOI: 10.1136/bmjno-2020-000069">10.1136/bmjno-2020-000069">10.1136/bmjno-2020-000069.
[3]
MCALEENAN A, KELLY C, SPIGA F, et al. Prognostic value of test(s) for O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation for predicting overall survival in people with glioblastoma treated with temozolomide[J/OL]. Cochrane Database Syst Rev, 2021, 3(3): CD013316 [2022-08-10]. https://pubmed.ncbi.nlm.nih.gov/33710615/. DOI: 10.1002/14651858.CD013316.pub2">10.1002/14651858.CD013316.pub2">10.1002/14651858.CD013316.pub2.
[4]
YU Z, CHEN Y, WANG S, et al. Inhibition of NF-kappa B results in anti-glioma activity and reduces temozolomide-induced chemoresistance by down-regulating MGMT gene expression[J]. Cancer Lett, 2018, 428: 77-89. DOI: 10.1016/j.canlet.2018.04.033">10.1016/j.canlet.2018.04.033">10.1016/j.canlet.2018.04.033.
[5]
WELLER M, VAN DEN BENT M, PREUSSER M, et al. EANO guidelines on the diagnosis and treatment of diffuse gliomas of adulthood[J]. Nat Rev Clin Oncol, 2020, 18(3): 170-186. DOI: 10.1038/s41571-020-00447-z">10.1038/s41571-020-00447-z">10.1038/s41571-020-00447-z.
[6]
AVRAM A V, SARLLS J E, BARNETT A S, et al. Clinical feasibility of using mean apparent propagator (MAP) MRI to characterize brain tissue microstructure[J]. Neuroimage, 2016, 127: 422-434. DOI: 10.1016/j.neuroimage.2015.11.027">10.1016/j.neuroimage.2015.11.027">10.1016/j.neuroimage.2015.11.027.
[7]
FICK R H J, WASSERMANN D, CARUYER E, et al. MAPL: Tissue microstructure estimation using Laplacian-regularized MAP-MRI and its application to HCP data[J]. Neuroimage, 2016: 365-385. DOI: 10.1016/j.neuroimage.2016.03.046">10.1016/j.neuroimage.2016.03.046">10.1016/j.neuroimage.2016.03.046.
[8]
SU C Q, LU S S, HAN Q Y, et al. Intergrating conventional MRI, texture analysis of dynamic contrast-enhanced MRI, and susceptibility weighted imaging for glioma grading[J]. Acta Radiol, 2019, 60(6): 777-787. DOI: 10.1177/0284185118801127">10.1177/0284185118801127">10.1177/0284185118801127.
[9]
HILARIO A, HERNANDEZ-LAIN A, SEPULVEDA J M, et al. Perfusion MRI grading diffuse gliomas: Impact of permeability parameters on molecular biomarkers and survival[J]. Neurocirugía, 2019, 30(1): 11-18. DOI: 10.1016/j.neucir.2018.06.004">10.1016/j.neucir.2018.06.004">10.1016/j.neucir.2018.06.004.
[10]
SUN Y, SU C, DENG K, et al. Mean apparent propagator-MRI in evaluation of glioma grade, cellular proliferation, and IDH-1 gene mutation status[J/OL]. Eur Radiol, 2022, 32(6): 3744-3754 [2022-08-10]. https://pubmed.ncbi.nlm.nih.gov/35076759/. DOI: 10.1007/s00330-021-08522-4">10.1007/s00330-021-08522-4">10.1007/s00330-021-08522-4.
[11]
WANG P, WENG L, XIE S, et al. Primary application of mean apparent propagator-MRI diffusion model in the grading of diffuse glioma[J/OL]. Eur J Radiol, 2021, 138: 109622 [2022-08-10]. https://pubmed.ncbi.nlm.nih.gov/33721768/. DOI: 10.1016/j.ejrad.2021.109622">10.1016/j.ejrad.2021.109622">10.1016/j.ejrad.2021.109622.
[12]
ERIKSSON M, KAHARI J, VESTMAN A, et al. Improved treatment of glioblastoma-changes in survival over two decades at a single regional Centre[J]. Acta Oncol, 2019: 1-8. DOI: 10.1080/0284186X.2019.1571278">10.1080/0284186X.2019.1571278">10.1080/0284186X.2019.1571278.
[13]
SZYLBERG M, SOKAL P, ŚLEDZIŃSKA P, et al. MGMT Promoter Methylation as a Prognostic Factor in Primary Glioblastoma: A Single-Institution Observational Study[J]. Biomedicines, 2022, 10(8): 2030. DOI: 10.3390/biomedicines10082030">10.3390/biomedicines10082030">10.3390/biomedicines10082030.
[14]
CHANG S, QIONG W, XIAOLING Y, et al. The TERT promoter mutation status and MGMT promoter methylation status, combined with dichotomized MRI-derived and clinical features, predict adult primary glioblastoma survival[J]. Cancer Med, 2018, 7(8): 3704-3712. DOI: 10.1002/cam4.1666">10.1002/cam4.1666">10.1002/cam4.1666.
[15]
DELLA MONICA R, CUOMO M, BUONAIUTO M, et al. MGMT and Whole-Genome DNA Methylation Impacts on Diagnosis, Prognosis and Therapy of Glioblastoma Multiforme[J]. Int J Mol Sci, 2022, 23(13): 7148. DOI: 10.3390/ijms23137148">10.3390/ijms23137148">10.3390/ijms23137148.
[16]
HAQUE W, TEH C, BUTLER E B, et al. Prognostic and predictive impact of MGMT promoter methylation status in high risk grade Ⅱ glioma[J]. J Neurooncol, 2022, 157(1):137-146. DOI: 10.1007/s11060-022-03955-3">10.1007/s11060-022-03955-3">10.1007/s11060-022-03955-3.
[17]
POHCHOO S, DING W J H, AZLINA A A, et al. Quantitative magnetic resonance imaging and radiogenomic biomarkers for gliomas characterisation: a systematic review[J]. Br J Radiol, 2018, 91: 20170930. DOI: 10.1259/bjr.20170930">10.1259/bjr.20170930">10.1259/bjr.20170930.
[18]
EL-SEROUGY L, ABDEL RAZEK A A, EZZAT A, et al. Assessment of diffusion tensor imaging metrics in differentiating low-grade from high-grade gliomas[J]. Neuroradiol J, 2016, 29(5): 400-407. DOI: 10.1177/1971400916665382">10.1177/1971400916665382">10.1177/1971400916665382.
[19]
RAJA R, SINHA N, SAINI J, et al. Assessment of tissue heterogeneity using diffusion tensor and diffusion kurtosis imaging for grading gliomas[J]. Neuroradiology, 2016, 58(12): 1-15. DOI: 10.1007/s00234-016-1758-y">10.1007/s00234-016-1758-y">10.1007/s00234-016-1758-y.
[20]
ZHANG J, CHEN X, CHEN D, et al. Grading and proliferation assessment of diffuse astrocytic tumors with monoexponential, biexponential, and stretched-exponential diffusion-weighted imaging and diffusion kurtosis imaging[J]. Eur J Radiol, 2018, 109: 188-195. DOI: 10.1016/j.ejrad.2018.11.003">10.1016/j.ejrad.2018.11.003">10.1016/j.ejrad.2018.11.003.
[21]
ZENG Q, DONG F, SHI F, et al. Apparent diffusion coefficient maps obtained from high b value diffusion-weighted imaging in the preoperative evaluation of gliomas at 3T: comparison with standard b value diffusion-weighted imaging[J]. Eur Radiol, 2017, 27(12): 5309-5315. DOI: 10.1007/s00330-017-4910-0">10.1007/s00330-017-4910-0">10.1007/s00330-017-4910-0.
[22]
ÖZARSLAN E, KOAY C G , SHEPHERD T M, et al. Mean apparent propagator (MAP) MRI: A novel diffusion imaging method for mapping tissue microstructure[J]. NeuroImage, 2013, 78: 16-32. DOI: 10.1016/j.neuroimage.2013.04.016">10.1016/j.neuroimage.2013.04.016">10.1016/j.neuroimage.2013.04.016.
[23]
MOON W J, CHOI J W, ROH H G, et al. Imaging parameters of high grade gliomas in relation to the MGMT promoter methylation status: the CT, diffusion tensor imaging, and perfusion MR imaging[J]. Neuroradiol, 2012, 54: 555-563. DOI: 10.1007/s00234-011-0947-y">10.1007/s00234-011-0947-y">10.1007/s00234-011-0947-y.
[24]
ILIADIS G, KOTOULA V, CHATZISOTIRIOU A, et al. Volumetric and MGMT parameters in glioblastoma patients: Survival analysis[J]. BMC Cancer, 2012, 12: 3. DOI: 10.1186/1471-2407-12-3">10.1186/1471-2407-12-3">10.1186/1471-2407-12-3.
[25]
OSTROM Q T, CIOFFI G, GITTLEMAN H, et al. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2012-2016[J]. Neuro Oncol, 2019, 21(Suppl 5): v1-v100. DOI: 10.1093/neuonc/noz150">10.1093/neuonc/noz150">10.1093/neuonc/noz150.
[26]
YOU S H, CHOI S H, KIM T M, et al. Differentiation of High-Grade from Low-Grade Astrocytoma: Improvement in Diagnostic Accuracy and Reliability of Pharmacokinetic Parameters from DCE MR Imaging by Using Arterial Input Functions Obtained from DSC MR Imaging[J]. Radiology, 2018, 286(3): 981-991. DOI: 10.1148/radiol.2017170764">10.1148/radiol.2017170764">10.1148/radiol.2017170764.
[27]
ZHANG H W, LYU G W, HE W J, et al. DSC and DCE Histogram Analyses of Glioma Biomarkers, Including IDH, MGMT, and TERT, on Differentiation and Survival[J/OL]. Acad Radiol, 2020, 27(12): e263-e271 [2022-08-10]. https://pubmed.ncbi.nlm.nih.gov/31983532/. DOI: 10.1016/j.acra.2019.12.010">10.1016/j.acra.2019.12.010">10.1016/j.acra.2019.12.010.
[28]
AHN S S, SHIN N Y, CHANG J H, et al. Prediction of methylguanine methyltransferase promoter methylation in glioblastoma using dynamic contrast-enhanced magnetic resonance and diffusion tensor imaging[J]. J Neurosurg, 2014, 121(2): 367-373. DOI: 10.3171/2014.5.JNS132279">10.3171/2014.5.JNS132279">10.3171/2014.5.JNS132279.
[29]
WOJCIECH S, BURLEY T A, GABRIELA K M, et al. Diagnostic and Therapeutic Biomarkers in Glioblastoma: Current Status and Future Perspectives[J]. Biomed Res Int, 2017, 2017: 8013575. DOI: 10.1155/2017/8013575">10.1155/2017/8013575">10.1155/2017/8013575.
[30]
KIM M, YOO J, CHANG J H, et al. Association of MGMT Gene Promoter Methylation With Clinicopathological Parameters in Patients With Wild-type IDH Glioblastoma[J]. Anticancer Res, 2022, 42(1): 335-341. DOI: 10.21873/anticanres.15490">10.21873/anticanres.15490">10.21873/anticanres.15490.
[31]
CHAI R, LI G, LIU Y, et al. Predictive value of MGMT promoter methylation on the survival of TMZ treated IDH-mutant glioblastoma[J]. Cancer Biol Med, 2021, 18(1): 272-282. DOI: 10.20892/j.issn.2095-3941.2020.0179">10.20892/j.issn.2095-3941.2020.0179">10.20892/j.issn.2095-3941.2020.0179.
[32]
LAM K, ELDRED B S C, KEVAN B, et al. Prognostic value of O6-methylguanine-DNA methyltransferase methylation in isocitrate dehydrogenase mutant gliomas[J/OL]. Neurooncol Adv, 2022, 4(1): vdac030 [2022-08-10]. https://pubmed.ncbi.nlm.nih.gov/35386566/. DOI: 10.1093/noajnl/vdac030">10.1093/noajnl/vdac030">10.1093/noajnl/vdac030.
[33]
LU J, LI X, LI H. Perfusion parameters derived from MRI for preoperative prediction of IDH mutation and MGMT promoter methylation status in glioblastomas[J]. Magn Reson Imaging, 2021, 83: 189-195. DOI: 10.1016/j.mri.2021.09.005">10.1016/j.mri.2021.09.005">10.1016/j.mri.2021.09.005.

上一篇 对比增强高分辨率血管壁成像评估不同钆基对比剂对颅内斑块强化效果的差异
下一篇 动脉自旋标记成像诊断干燥综合征早期腮腺损伤的临床价值
  
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