Radiogenomic modeling predicts survival-associated prognostic groups in glioblastoma.
Radiogenomic modeling predicts survival-associated prognostic groups in glioblastoma.
复制标题
DOI：
10.1093/noajnl/vdab004
复制
发表时间：
2021-01
期刊：
影响因子：
--
通讯作者：
Cimino PJ
中科院分区：
文献类型：
--
作者：
Nuechterlein N;Li B;Feroze A;Holland EC;Shapiro L;Haynor D;Fink J;Cimino PJ
结合全外显子组测序（WES）和体细胞拷贝数改变（SCNA）信息可以将异柠檬酸脱氢酶（IDH）1/2-野生型胶质母细胞瘤分为两种预后分子亚型，这两种亚型不能通过表观遗传或临床特征来区分。放射学特征区分这些分子亚型的潜力尚未建立。放射学特征（n = 35 340）从46个多序列，术前磁共振成像（MRI）扫描IDH 1/2-野生型胶质母细胞瘤患者从癌症成像档案（TCIA），所有的人都有相应的WES/SCNA数据。我们开发了一种新的特征选择方法，该方法利用提取的MRI特征的结构来减轻大量特征与我们队列中有限患者之间的差异所带来的维度挑战。使用我们的特征选择方法训练六个传统的机器学习分类器来区分分子亚型，并将其与最小绝对收缩和选择算子（LASSO）特征选择，递归特征消除和方差阈值进行比较。我们能够将胶质母细胞瘤分为两个预后亚组，交叉验证曲线下面积评分为0.80（±0.03），使用岭逻辑回归对我们的新特征选择方法选择的特征进行15维主成分分析（PCA）嵌入。对所选特征的询问表明，描述T2加权液体衰减反转恢复（FLAIR）MRI序列T2信号异常区域轮廓的特征可能最能区分这两组。我们成功地训练了一个机器学习模型，该模型允许从标准MRI中提取相关的靶向特征，以准确预测分子定义的风险分层IDH 1/2野生型胶质母细胞瘤患者组。
Combined whole-exome sequencing (WES) and somatic copy number alteration (SCNA) information can separate isocitrate dehydrogenase (IDH)1/2-wildtype glioblastoma into two prognostic molecular subtypes, which cannot be distinguished by epigenetic or clinical features. The potential for radiographic features to discriminate between these molecular subtypes has yet to be established. Radiologic features (n = 35 340) were extracted from 46 multisequence, pre-operative magnetic resonance imaging (MRI) scans of IDH1/2-wildtype glioblastoma patients from The Cancer Imaging Archive (TCIA), all of whom have corresponding WES/SCNA data. We developed a novel feature selection method that leverages the structure of extracted MRI features to mitigate the dimensionality challenge posed by the disparity between a large number of features and the limited patients in our cohort. Six traditional machine learning classifiers were trained to distinguish molecular subtypes using our feature selection method, which was compared to least absolute shrinkage and selection operator (LASSO) feature selection, recursive feature elimination, and variance thresholding. We were able to classify glioblastomas into two prognostic subgroups with a cross-validated area under the curve score of 0.80 (±0.03) using ridge logistic regression on the 15-dimensional principle component analysis (PCA) embedding of the features selected by our novel feature selection method. An interrogation of the selected features suggested that features describing contours in the T2 signal abnormality region on the T2-weighted fluid-attenuated inversion recovery (FLAIR) MRI sequence may best distinguish these two groups from one another. We successfully trained a machine learning model that allows for relevant targeted feature extraction from standard MRI to accurately predict molecularly-defined risk-stratifying IDH1/2-wildtype glioblastoma patient groups.
登录
查看更多内容
影响因子：
4.6
作者：
Arita H;Kinoshita M;Kawaguchi A;Takahashi M;Narita Y;Terakawa Y;Tsuyuguchi N;Okita Y;Nonaka M;Moriuchi S;Takagaki M;Fujimoto Y;Fukai J;Izumoto S;Ishibashi K;Nakajima Y;Shofuda T;Kanematsu D;Yoshioka E;Kodama Y;Mano M;Mori K;Ichimura K;Kanemura Y
通讯作者：
Kanemura Y
影响因子：
5.7
作者：
Jenkinson, M;Bannister, P;Smith, S
通讯作者：
Smith, S
影响因子：
4.4
作者：
Clark, Kenneth;Vendt, Bruce;Prior, Fred
通讯作者：
Prior, Fred
影响因子：
15.9
作者：
Cimino, Patrick J.;McFerrin, Lisa;Holland, Eric C.
通讯作者：
Holland, Eric C.
影响因子：
7.1
作者：
Cimino PJ;Zager M;McFerrin L;Wirsching HG;Bolouri H;Hentschel B;von Deimling A;Jones D;Reifenberger G;Weller M;Holland EC
通讯作者：
Holland EC