Hayes Carmel, Padhani Anwar R, Leach Martin O
CRC Clinical Magnetic Resonance Research Group, The Institute of Cancer Research and The Royal Marsden NHS Trust, Downs Road, Sutton SM2 5PT, UK.
NMR Biomed. 2002 Apr;15(2):154-63. doi: 10.1002/nbm.756.
Dynamic contrast-enhanced MRI (DCE-MRI) is widely used in the diagnosis and staging of cancer and is emerging as a promising method for monitoring tumour response to treatment. However, DCE-MR imaging techniques are still evolving and methods of image analysis remain variable and non-standard, and range from relative changes in the pattern of enhancement to pharmacokinetic modelling of contrast agent uptake. The combination of results from different institutions is therefore difficult and the sensitivities of different methods have not been compared. The purpose of this study is to investigate correlations between qualitative and quantitative methods of analysis for DCE-MR images from breast cancer patients undergoing neo-adjuvant chemotherapy. Fifteen patients underwent DCE-MRI examinations before and after one course of chemotherapy. Changes in the temporal pattern of signal enhancement, the rate and amplitude of enhancement and the volume transfer constant of contrast agent between the blood plasma and the extravascular extracellular space (EES), K(trans), and the EES fractional volume, nu(e), were determined. In addition, whole tumour region-of-interest analysis was compared with histogram analysis to investigate the extent of tumour heterogeneity. It was found that changes in the rate of enhancement correlated strongly with changes in K(trans) values (Kendall's tau = 0.68, P < 0.001). Furthermore, it was found that the shape of the signal enhancement curve only changed when the K(trans) values changed by 50% or more. Median K(trans) values determined following histogram analysis of pixel maps of K(trans) were approximately equal to those determined by whole tumour region-of-interest analysis. The absolute change in the K(trans) values correlated negatively with the pre-treatment values, particularly for responding patients. Thus, for higher pre-treatment K(trans) values, a greater decrease was observed. Greater changes were observed in the upper extremes of the K(trans) histogram than in the median values after one course of treatment.
动态对比增强磁共振成像(DCE-MRI)广泛应用于癌症的诊断和分期,正逐渐成为监测肿瘤治疗反应的一种有前景的方法。然而,DCE-MR成像技术仍在不断发展,图像分析方法仍然多变且不标准,范围从增强模式的相对变化到造影剂摄取的药代动力学建模。因此,不同机构结果的合并很困难,且不同方法的敏感性尚未进行比较。本研究的目的是调查接受新辅助化疗的乳腺癌患者DCE-MR图像定性和定量分析方法之间的相关性。15例患者在一个化疗疗程前后接受了DCE-MRI检查。测定了信号增强的时间模式、增强速率和幅度以及血浆与血管外细胞外间隙(EES)之间造影剂的容积转移常数K(trans)和EES分数容积nu(e)的变化。此外,将整个肿瘤感兴趣区分析与直方图分析进行比较,以研究肿瘤异质性的程度。结果发现,增强速率的变化与K(trans)值的变化密切相关(肯德尔tau系数=0.68,P<0.001)。此外,还发现只有当K(trans)值变化50%或更多时,信号增强曲线的形状才会改变。对K(trans)像素图进行直方图分析后确定的K(trans)中位数与通过整个肿瘤感兴趣区分析确定的中位数大致相等。K(trans)值的绝对变化与治疗前值呈负相关,尤其是对于有反应的患者。因此,对于较高的治疗前K(trans)值,观察到更大的下降。在一个化疗疗程后,K(trans)直方图的上限极端值比中位数观察到更大的变化。