Advancing MRI technology for early diagnosis of liver metastases

Abstract
Liver is commonly involved in metastatic disease in colorectal cancer (CRC) and knowledge about
the presence and location of these tumors affects treatment decisions. In patients with CRC,
surgical or ablative treatment of liver metastases improves overall survival. Early diagnosis of
colorectal metastases (i.e. while lesions are small) is expected to improve treatment outcomes by
increasing the number of subjects that can undergo surgical resection or by identifying subjects
early on, when non-surgical options are an alternative treatment. Magnetic Resonance Imaging
(MRI) is regarded as the most effective imaging modality for the detection and characterization of
liver neoplasms; T2-weighted (T2w) and T1-weighted (T1w) images - combined with
administration of a gadolinium chelate agent and multi-phase dynamic contrast enhancement
(DCE) - are the foundational acquisitions used for the detection and characterization of liver
tumors. However, challenges remain for the detection and characterization of small lesions due
to factors including inadequate spatial resolution, partial volume effects, physiological motion, and
variations in timing of contrast arrival in DCE imaging. In this academic-industrial partnership the
scientific and engineering teams at the University of Arizona and Siemens Medical Solutions are
coming together to develop robust radial MRI techniques for T2w/T2 mapping and DCE imaging
of the liver to improve detection and characterization of small tumors with the goal of bringing
these techniques to routine clinical practice. The proposed work is based on a radial turbo spin-
echo technique pioneered by the team at the University of Arizona for abdominal imaging and a
radial stack-of-stars technique with continuous acquisition for DCE imaging. The specific aims of
the partnership are: Aim 1: To develop radial T2w acquisition and reconstruction techniques with
efficient full coverage of the liver for small tumor detection and accurate T2 quantification for tumor
characterization. Aim 2: To implement a self-navigated 3D radial stack-of-stars technique for
continuous acquisition of DCE data and retrospective reconstruction of the dynamic phases. Aim
3: To conduct a clinical evaluation of the techniques from Aims 1 and 2 against conventional T2w
and DCE techniques. Aim 4: To streamline translation of the new radial methods to the clinic by
developing a computationally efficient reconstruction pipeline. The endpoints of our study include
technical advances in MRI acquisitions that markedly overcome limitations of current liver MRI for
the diagnosis of early metastases. We expect our proposal to yield technology improvements
that will increase precision of care and outcomes in patients with metastatic malignancies, in
particular those with colorectal cancer.

Grant Number: 5R01CA245920-05
NIH Institute/Center: NIH
Principal Investigator: Maria Altbach