qCEST: Quantitative Chemical Exchange Saturation Transfer MR imaging of brain tumors at 7 Tesla

CEST MRI被认为是生物医学成像领域的一个令人兴奋的话题。它的主要优点是成像的生理和分子信息与空间分辨率可媲美传统的MRI。CEST信号与蛋白质含量和蛋白质结构、代谢物浓度和细胞内pH值的相关性使CEST成为一种有趣的癌症研究成像方式：蛋白质含量是肿瘤浸润过程中细胞数量增加的指标；此外，pH梯度与肿瘤迁移之间存在相关性。然而，最近的研究表明，需要改进技术来正确分离通常称为CEST对比的不同贡献者，例如交换饱和酰胺质子或蛋白质的脂肪族质子。由于高频分辨率高，可以在超高场强（B0 = 7 T）下分离这些单独的CEST效应。在先前的模型溶液和动物研究中，我们可以证明CEST数据也必须对水质子弛豫特性的变化进行校正。这在胶质母细胞瘤的成像中特别重要，因为这些肿瘤的水松弛参数发生了改变。因此，需要定量CEST方法（qCEST）来开发CEST成像的全部潜力，以提供蛋白质含量和pH图。CEST MRI的第二个挑战是扫描时间较长，这阻碍了它的临床应用。为了在人类患者中推广常规的CEST成像，我们打算在两个方向上推进该技术：(i)开发定量CEST序列，从而深入了解CEST的基本参数并提供参考方法；（ii）开发快速临床CEST序列并将其引入临床应用。首先，继之前的动物实验之后，将为人类的CEST MRI建立重要的校正步骤。基于先前工作的序列，将开发能够定量获取蛋白质含量和pH值的方法。在用模型溶液进行实验验证后，将在健康受试者中进行体内评估。同时，开发一套优化扫描时间的临床CEST序列。本文将针对CEST MRI数据的具体特征，对k空间采集、饱和度和读出分割、视图共享和压缩感知技术进行优化。利用定量CEST序列的结果，该临床CEST序列将在幻影和志愿者的研究中进行验证和进一步优化。通过与弥散加权成像和31P磁共振光谱的比较，可以分析qCEST与细胞结构和ph值的相关性。最后，对胶质母细胞瘤患者和低级别脑肿瘤患者在7 T时进行检查。与标准MR对比的比较将允许评估定量CEST的临床益处。

CEST MRI is regarded as an exciting topic in the field of biomedical imaging. Its major advantage is imaging of physiological and molecular information with spatial resolution comparable to that of conventional MRI. Correlations of CEST signals with protein content and protein structure, metabolite concentration, and intracellular pH have made CEST an interesting imaging modality for studies of cancer: Protein content is a measure of increased cellularity during tumor infiltration; furthermore there is a correlation between gradients in pH and tumor migration. However, recent studies have revealed the need for improved techniques to properly isolate different contributors to what is commonly called a CEST contrast, such as exchange of saturated amide protons or aliphatic protons of proteins. Separation of these individual CEST effects is possible at ultra-high field strength (B0 = 7 T) due to the high frequency resolution. In previous studies with model solutions and animals we could show that CEST data must also be corrected for changes in the relaxation properties of water protons. This is of particular importance in imaging of glioblastoma, since water relaxation parameters are altered in these tumors. Hence, a quantitative CEST approach (qCEST) is required to develop the full potential of CEST imaging to provide protein content and pH maps.A second challenge of CEST MRI is a comparatively long scanning time which impedes its clinical application. To promote routine qCEST imaging in human patients, we intend to advance the technique in two directions: (i) Development of a quantitative CEST sequence that yields insights into the fundamental CEST parameters and provides a reference method, and (ii) development of a fast clinical CEST sequence and its introduction into clinical use. First, important correction steps, following previous animal work, will be established for CEST MRI in humans. Based on the sequence available from prior work, methods that enable quantitative access to protein content and pH will be developed. After their validation in experiments with model solutions, in vivo evaluation in healthy subjects will be performed. Simultaneously, a scanning-time-optimized clinical CEST sequence will be developed. Here, k-space acquisition, saturation and readout segmentation, view-sharing, and compressed sensing techniques will be optimized with respect to the specific features of CEST MRI data. Using the results of the quantitative CEST sequence, this clinical CEST sequence will be validated and further optimized in studies with phantoms and volunteers. Comparison with diffusion-weighted imaging and 31P MR spectroscopy will allow analysis of correlations of qCEST with cellularity and pH. Finally, examinations will be performed at 7 T in patients with glioblastoma and patients with lower grade brain tumors. Comparison with standard MR contrasts will permit the assessment of the clinical benefit of quantitative CEST.