Department of Neuroradiology, National Hospital for Neurology and Neurosurgery, University College London Hospitals NHS Foundation Trust, London, UK.
Department of Imaging, Imperial College Healthcare NHS Trust, London, UK.
Clin Radiol. 2021 Oct;76(10):785.e9-785.e16. doi: 10.1016/j.crad.2021.06.015. Epub 2021 Jul 18.
AIM: To identify imaging predictors of molecular subtype and tumour grade in patients with isocitrate dehydrogenase (IDH) gene mutant (IDH) World Health Organization (WHO) grade 2 or 3 gliomas. MATERIALS AND METHODS: Patients with histologically confirmed WHO grade 2 or 3 IDH gliomas between 2016 and 2019 were included in the study. Magnetic resonance imaging (MRI) images were evaluated for the presence or absence of potential imaging predictors of tumour subtype, such as T2/fluid attenuated inversion recovery (FLAIR) signal match, and these factors were examined using regression analysis. On perfusion imaging, the maximum relative cerebral blood volume (rCBV) was evaluated as a potential predictor of tumour grade. The performance of two experienced neuroradiologists in correctly predicting tumour type on MRI was evaluated. RESULTS: Eighty-five patients were included in the study. The presence of T2/FLAIR signal match >50% of tumour volume (p<0.01) and intratumoural susceptibility (p=0.02) were independent predictors of 1p/19q co-deletion. Mean rCBV max was significantly higher in WHO grade 3 astrocytomas (p=0.04) than WHO grade 2 astrocytomas. The consensus prediction of 1p/19q co-deletion status by two neuroradiologists of tumour was 95% sensitive and 86% specific. CONCLUSION: The presence of matched T2/FLAIR signal could be used to identify tumour subtype when biopsy is inconclusive or genetic analysis is unavailable. rCBV predicted astrocytoma grade. Experienced neuroradiologists predict tumour subtype with good sensitivity and specificity.
目的:确定 IDH 基因突变(IDH)世界卫生组织(WHO)2 级或 3 级胶质瘤患者中分子亚型和肿瘤分级的影像学预测因子。 材料和方法:本研究纳入了 2016 年至 2019 年间经组织学证实的 WHO 2 级或 3 级 IDH 胶质瘤患者。评估磁共振成像(MRI)图像是否存在肿瘤亚型的潜在影像学预测因子,如 T2/液体衰减反转恢复(FLAIR)信号匹配,并用回归分析检查这些因素。在灌注成像上,评估最大相对脑血容量(rCBV)作为肿瘤分级的潜在预测因子。评估两位有经验的神经放射科医生在 MRI 上正确预测肿瘤类型的表现。 结果:本研究纳入了 85 例患者。肿瘤体积中 T2/FLAIR 信号匹配>50%(p<0.01)和瘤内顺磁性(p=0.02)是 1p/19q 共缺失的独立预测因子。WHO 3 级星形细胞瘤的平均 rCBV max 显著高于 WHO 2 级星形细胞瘤(p=0.04)。两位神经放射科医生对肿瘤 1p/19q 共缺失状态的共识预测具有 95%的敏感性和 86%的特异性。 结论:当活检结果不确定或基因分析不可用时,匹配的 T2/FLAIR 信号的存在可用于识别肿瘤亚型。rCBV 预测星形细胞瘤分级。有经验的神经放射科医生对肿瘤亚型具有良好的敏感性和特异性预测。
AJNR Am J Neuroradiol. 2019-1-31
J Magn Reson Imaging. 2017-11-15