Establishing the validity of brain tumor perfusion imaging

点击翻译按钮获取中文摘要

ABSTRACT The principal goal of this competitive renewal is to establish clinical brain tumor perfusion imaging protocols strategically designed for high accuracy, physiologic sensitivity and clinical applicability. Dynamic susceptibility contrast (DSC) MRI is one of the most widely used advanced imaging techniques in neuro-oncology, with a reported use of 85% in all routine brain tumor scans at sites across the US and Europe. During the previous funding cycle, the singular aim of the project was to identify the most accurate DSC-MRI acquisition protocol(s) by developing, validating and applying a population-based digital reference object (DRO) and then verifying these protocols in patients with brain tumors. The most significant finding of this effort was the identification and clinical validation of an accurate single contrast agent dose and single-echo DSC-MRI protocol, which has now been adopted as the consensus recommendation by the brain tumor imaging community. Another key result of this project was the validation, using the DRO and a prospective patient study, of a single-dose, dual-echo DSC- MRI sequence, an approach that enables simultaneous assessment of DSC and dynamic contrast enhanced (DCE)-MRI data, as the protocol with highest accuracy, even across variable pulse sequence parameters and tissue properties. Building on the success of this prior work, we now aim to overcome two obstacles that still limit DSC-MRI’s clinical utility, accuracy, and multi-site consistency: i) a reliance on echo planar imaging (EPI) based pulse sequences that undermine the geometric fidelity, reliable colocalization of perfusion and anatomic images and accuracy of derived DCE-MRI data, and ii) lack of validation of, and a benchmark for, brain tumor DSC/DCE- MRI post-processing algorithms and software, for both single- and dual-echo acquisitions. These limitations represent critical and clinically relevant challenges that urgently need to be addressed. To address these issues, we propose to: 1) establish an anthropomorphic benchmark for validating brain tumor DSC DCE-MRI analysis tools and 2) develop a three-dimensional, dual-echo pulse sequence for simultaneous DSC/DCE MRI. In this project we will provide the neuro-oncology community with validated image acquisition and analysis methods for accurate, physiologic sensitive and clinically applicable DSC/DCE-MRI mapping methods in brain tumor patients. We will provide the first DSC/DCE-MRI anthropomorphic benchmark that can be used to validate existing and future algorithms and software, thereby improving multi-site and clinical trial consistency. Ultimately, validated DSC-MRI techniques will improve its reliability and relevancy across a range of clinical scenarios, including tumor localization, therapy response assessment, surgical and biopsy guidance, and multi-site clinical trials of conventional and targeted brain tumor therapies.