University of Wisconsin–Madison

Quantification of Perfusion in Liver Tumors with MRI – Sequence Optimization

February 2011 to December 2011

In the last several years, there has been a remarkable transformation in the field of cancer therapeutics, with a shift from traditional “cytotoxic” agents to targeted therapies. This trend has necessitated the development of innovative biomarkers that allow rapid assessment of tumor response to a specific agent. Such biomarkers are urgently needed to personalize the treatment of tumors with effective agents. The overall purpose of this work is to develop new imaging strategies that measure early response to anti-angiogenic chemotherapeutic agents in liver tumors, using new magnetic resonance imaging (MRI) methods that measure blood flow to tumors (perfusion). Currently, the assessment of tumor response relies on the measurement of tumor size based on standard size criteria guidelines over the course of months. We hypothesize that early changes of tumor blood flow are predictive of long-term tumor regression or stabilization in patients treated with anti-angiogenic chemotherapy. Our proposed studies could lead to more efficient evaluation of treatment options for tumors such as hepatocellular carcinoma (HCC) and metastatic liver tumors and may provide new measurements of tumor response in an effort to deliver more individualized cancer care in the future. As described below, we propose to evaluate the clinical and cost effectiveness of advanced perfusion MRI methods to characterize blood flow to liver tumors before and immediately after initiation of systemic therapy. We aim to determine the comparative effectiveness of advanced perfusion MRI with conventional cross-sectional imaging methods (CT) that measure tumor response based on tumor size criteria. By identifying tumors that will not respond to these drugs earlier, their toxic side effects are avoided, and tremendous cost savings can be realized. Unfortunately, perfusion imaging in the liver, the largest organ in the body, is challenging, and to date no methods have adequately provided the necessary combination of: 1) volumetric coverage, 2) high spatial resolution and 3) high temporal resolution, while maintaining 4) good signal to noise ratio (SNR), and 5) fidelity of Gadolinium contrast uptake curves in tissue. In this proposal we will use a novel perfusion MRI method based on a time-resolved 3D “radial” sampling methods developed at UW Madison. Using a state-of-the-art high-field (3.0T) clinical MRI scanner, we will combine this approach with a “constrained reconstruction” algorithm to maximize the SNR performance and temporal resolution. Using the combination of 3D radial imaging and constrained reconstruction, we will acquire a time series of volumetric 3D images every 3 seconds, with very high spatial resolution (2.0mmx2.0mmx2.0mm). To achieve these goals, we will 1) implement and test the new method on a state-of-the-art 3.0T clinical scanner in volunteers, and in patients with hepatocellular carcinoma, and 2) determine the impact of the optimized method on clinical outcome through a prospective longitudinal pilot study in HCC patients undergoing antiangiogenic therapy. Successful validation of the proposed MRI measures of tumor response will provide a new image-based approach for clinical decision-making for chemotherapeutic regimens by transforming how MRI is used as a measure of tumor response.

This project led by: Scott B Reeder, MD, PhD