Multiparametric MRI and Coregistered Histology Identify Tumor Habitats in Breast Cancer Mouse Models

Area of Science:

• Oncology
• Radiology
• Biophysics

Background:

• Solid tumors exhibit significant genomic, anatomical, and physiological heterogeneity.
• Tumor heterogeneity is linked to poorer patient outcomes and therapy resistance.
• Understanding the relationship between physiological and genomic heterogeneity is crucial.

Purpose of the Study:

• To investigate the relationship between physiological and genomic tumor heterogeneity.
• To develop methods for defining microscopic cytotypes within physiologically distinct tumor subregions.
• To utilize multiparametric MRI (mpMRI) to interrogate physiological tumor parameters.

Main Methods:

• Multiparametric MRI (mpMRI) was used to assess perfusion, cell density, and edema.
• MRI signals from six acquisition schemas were combined and clustered using a Gaussian mixture model.
• Coregistration with histologic and IHC data was achieved using MRI-guided 3D printed tumor molds.

Main Results:

• Specific MRI parameters were identified to classify four tissue types: viable-normoxic, viable-hypoxic, nonviable-hypoxic, and nonviable-normoxic.
• These classifications were validated in orthotopic 4T1 and MDA-MB-231 breast tumor models.
• The study demonstrated the ability to define physiologically distinct tumor habitats using mpMRI.

Conclusions:

• This is the first coregistered study to define physiologically distinct tumor habitats in breast tumor models using mpMRI.
• Noninvasive imaging metrics can effectively distinguish subregions within heterogeneous tumors.
• mpMRI provides a valuable tool for correlating imaging findings with histopathology in heterogeneous tumors.