Koc University Hospital, Department of Radiology, Zeytinburnu, Istanbul, Türkiye.
Turk Neurosurg. 2024;34(6):1056-1065. doi: 10.5137/1019-5149.JTN.46251-24.3.
To assess the performance metrics of perfusion and permeability magnetic resonance imaging (MRI) parameters with optimal cut-offs in differentiating isocitrate dehydrogenase (IDH) genotype and tumor grade in patients with grade 2-4 gliomas.
This retrospective study included 36 patients surgically diagnosed with grade 2-4 glioma (six grade 2, seven grade 3, and 23 grade 4) with known IDH genotypes (23 IDH wild-type, 13 IDH mutant) between November 2021 and August 2023. All patients underwent preoperative perfusion and permeability MRI examinations with a 3.0 Tesla scanner. Parameters were calculated on colored map images. Using the intraclass correlation coefficient, intra- and inter-observer agreement was assessed. Following multiple testing correction, the perfusion parameters with statistically significant differences were subjected to receiver operating characteristic (ROC) analysis.
Five MRI parameters (rCBV and rCBF from perfusion; Ktrans, Ve, and Vp from permeability) showed a significant difference between groups in terms of IDH genotype (p < 0.001). In ROC analysis, the best parameters in differentiating IDH genotype included rCBV and Ktrans; rCBV with a cut-off of 5.58 achieved an area under the ROC curve (AUC), sensitivity, specificity, and accuracy of 0.883, 95.7%, 76.9%, and 88.8%, respectively. For Ktrans, with a cut-off of 0.0727 min-1, these values were 0.893, 100%, 69.2%, and 88.8%, respectively. In ROC analysis, these two parameters with rCBF and Ve also showed good performance in differentiating low- and high-grade gliomas with an AUC, sensitivity, and accuracy exceeding 0.940, 86%, and 88%, respectively.
Perfusion and permeability MRI may provide useful parameters in differentiating the IDH genotype and grade of gliomas.
评估灌注和渗透性磁共振成像（MRI）参数的性能指标，以区分 IDH 基因型和 2-4 级胶质瘤患者的肿瘤分级的最佳截断值。
本回顾性研究纳入了 2021 年 11 月至 2023 年 8 月期间经手术诊断为 2-4 级胶质瘤（6 级 2 例，3 级 7 例，4 级 23 例）且 IDH 基因型已知（23 例 IDH 野生型，13 例 IDH 突变型）的 36 例患者。所有患者均在 3.0T 扫描仪上进行术前灌注和渗透性 MRI 检查。参数在彩色图图像上计算。使用组内相关系数评估观察者内和观察者间的一致性。经过多重检验校正后，对具有统计学差异的灌注参数进行受试者工作特征（ROC）分析。
在 IDH 基因型方面，5 个 MRI 参数（灌注的 rCBV 和 rCBF；渗透性的 Ktrans、Ve 和 Vp）在组间存在显著差异（p<0.001）。在 ROC 分析中，区分 IDH 基因型的最佳参数包括 rCBV 和 Ktrans；rCBV 截断值为 5.58 时，ROC 曲线下面积（AUC）、敏感度、特异度和准确度分别为 0.883、95.7%、76.9%和 88.8%。对于 Ktrans，截断值为 0.0727 min-1 时，这些值分别为 0.893、100%、69.2%和 88.8%。在 ROC 分析中，rCBF 和 Ve 这两个参数在区分低级别和高级别胶质瘤方面也表现出良好的性能，AUC、敏感度和准确度均超过 0.940、86%和 88%。
灌注和渗透性 MRI 可为区分 IDH 基因型和胶质瘤分级提供有价值的参数。
