Independent Component Analysis of Dynamic Susceptibility Contrast MRI in Brain Tumor: A New Biomarker for Measuring Tumor Perfusion Patterns

Peter Sherman LaViolette¹, Alex D. Cohen¹, Scott D. Rand², Wade Mueller³, Kathleen M. Schmainda^1,2

It is well known that tumor growth beyond a size of about 2 mm requires the development of its own blood vessels, a process termed angiogenesis. While physiologic angiogenesis, such as that occurs with wound healing, results in the formation of well-ordered mature vessels, pathologic angiogenesis such as that observed with tumors, results in the formation of chaotic and immature vessels. It is therefore not surprising that the resulting tumor perfusion patterns are likewise altered. We hypothesized that application of ICA (independent component analysis) to DSC-MRI signals would provide a new approach for distinguishing tumor from normal tissue, thus demonstrating the potential to serve as a novel biomarker to predict response to anti-angiogenic drugs thought to normalize tumor vasculature.