Department of Neurosurgery, University Medicine of Rostock, Schillingallee 35, 18057, Rostock, Germany.
Institute for Biostatistics and Informatics in Medicine, University Medicine of Rostock, Ernst-Heydemann-Str. 8, 18057, Rostock, Germany.
J Neurooncol. 2017 Nov;135(2):391-402. doi: 10.1007/s11060-017-2587-5. Epub 2017 Jul 28.
The potential impact of different radiological features of glioblastoma multiforme (GBM) on overall survival (OS) like tumor volume, peritumoral edema (PTE), necrosis volume, necrosis-tumor ratio (NTR) and edema-tumor ratio (ETR) is still very controversial. To determine the influence of volumetric data on OS und to compare different measuring techniques described in literature. We prospectively evaluated preoperative MR images from 30 patients harboring a primary supratentorial GBM. All patients received gross-total tumor resection followed by standard radiation and chemotherapy (temozolomide). By 3D semi-automated segmentation, we measured tumor volume, necrosis volume, PTE, postoperative residual tumor volume and calculated ETR, NTR and the extent of resection. After critical review of the existing literature we compared alternative measuring techniques with the gold standard of 3D segmentation. Statistical analysis showed a significant impact of the preoperative tumor and necrosis volumes on OS (p = 0.041, respectively p = 0.039). Furthermore, NTR also showed a significant association with OS (p = 0.005). Comparison of previously described measuring techniques and scorings with our results showed that no other technique is reliable and accurate enough as a predictive tool. The critical review of previously published studies revealed mainly inaccurate measurement techniques and patient selection as potential reasons for inconsistent results. Preoperatively measured necrosis volume and NTR are the most important radiological features of GBM with a strong influence on OS. No other measuring techniques are specific enough and comparable with 3D segmentation.
多形性胶质母细胞瘤（GBM）的不同放射学特征（如肿瘤体积、瘤周水肿（PTE）、坏死体积、坏死-肿瘤比（NTR）和水肿-肿瘤比（ETR））对总生存期（OS）的潜在影响仍存在很大争议。为了确定容积数据对 OS 的影响，并比较文献中描述的不同测量技术。我们前瞻性地评估了 30 名患有原发性幕上 GBM 的患者的术前磁共振图像。所有患者均接受了大体全切除肿瘤手术，随后接受标准的放疗和化疗（替莫唑胺）。通过 3D 半自动分割，我们测量了肿瘤体积、坏死体积、PTE、术后残留肿瘤体积，并计算了 ETR、NTR 和切除程度。在对现有文献进行批判性回顾后，我们将替代测量技术与 3D 分割的金标准进行了比较。统计分析显示术前肿瘤和坏死体积对 OS 有显著影响（p=0.041，分别为 p=0.039）。此外，NTR 与 OS 也有显著相关性（p=0.005）。与我们的结果相比，对以前描述的测量技术和评分的比较表明，没有其他技术作为预测工具具有足够的可靠性和准确性。对以前发表的研究的批判性回顾显示，主要是不准确的测量技术和患者选择是导致结果不一致的潜在原因。术前测量的坏死体积和 NTR 是 GBM 最重要的放射学特征，对 OS 有很强的影响。没有其他测量技术具有足够的特异性和可比性，无法与 3D 分割相比。
