ABSTRACT

Tumors are spatially heterogeneous, stemming from selection pressures on physiologically distinct sub-regions, also called tumor habitats in an analogy to evolutionary ecology. Different habitats may respond differently to therapy, which has a great impact in the clinical outcome. Non-invasive imaging analyses capable of assessing distinct tumor sub-regions rather than the entire tumor, are promising to longitudinally and more realistically monitor the tumor progression and response to therapy. Habitat imaging is a technique that aspires to interrogate these distinct local environments, in which Magnetic Resonance Imaging (MRI) has been explored due to the biophysical principles that govern MRI signal generation. Different functional imaging sequences can be used to segment the tumor into distinct sub-regions. However, biological assumptions from imaging should be made with caution, and a ground truth must be provided by corresponding histological data. This chapter presents studies that have proposed different analyses in specific MRI sequences, as well as multiparametric MRI, to define tumor sub-regions and to assess intratumoral heterogeneity.