Dynamic contrast-enhanced (DCE)-MRI

Dynamic contrast-enhanced (DCE)-MRI is a technique whereby a standard low-molecular weight Gadolinium agent is administered intravenously at a standard rate, followed by sequential imaging of a region of interest (ROI) within the target lesion for up to 30 minutes. Signal intensity changes are measured within the ROI relative to normal tissue and plotted against time, allowing analysis of contrast "wash-in" and "wash out" components of the enhancement curve. Simple semi-quantitative analysis is performed by assessing the nature of such curves, an example being the evaluation of breast lesions. The slope of the wash-in and wash-out components of the curve, time to maximal enhancement or time to peak (TTP)] and area under the curve (AUC) provide semi-quantitative data that can be utilized to derive information on tumor blood flow, concentration, and tissue permeability.From these data, additional metrics such as mean transit time (MTT), a measure of the time taken for blood to perfuse a tissue, can be derived. Drawbacks of this method include the influence of protocol parameters such as contrast agent concentration, rate of injection, and variation in imaging hardware settings.
Alternatively, post-processing techniques based on kinetic modeling can be applied wherein a computer analyzes the contrast enhancement patterns on a pixel-by-pixel basis to derive quantitative measures of permeability . Parameters such as Ktrans (a measure of blood flow and permeability) and kep (the reverse flow constant) are obtained, thereby serving as markers of angiogenesis. DCE-MRI benefits from the widespread availability of MR, a lack of ionizing radiation, the transferability of protocols to existing scanners, and the use of standard gadolinium-based agents. Indeed, DCE-MRI has been studied in Phase II and III chemotherapeutic trials of anti-angiogenic agents and remains the most widely adopted imaging method for quantifying angiogenesis. However, DCE-MRI is not without its problems. The required post-processing and additional interpretation add to the reporting time; realistically an MR physicist needs to be on site, which is not possible in every center. In addition, the relationship of gadolinium concentration to signal intensity is not linear and depends on the T1 relaxivity of the tissue imaged - this can be partially overcome by the acquisition of a T1 map. Furthermore, there is disagreement on the optimal kinetic model to use and although the selection of an arterial input function should aid standardization, it often results in an additional variable. These problems mean that standardization is still an issue, making comparison between centers problematic.