Abstract Introduction: Heterogeneity in the tumor microenvironment affects therapeutic delivery, providing a major challenge for cancer treatment. We investigate the ability of multi-parametric, voxel-based characterizations of tumor heterogeneity from quantitative magnetic resonance imaging (MRI) to spatially resolve physiological tumor subregions of response to anti-HER2 targeted therapy in a murine model of HER2+ breast cancer. Methods: BT474 cells were implanted in nude athymic mice (n=20) and tumors grown to ~225 mm3. Mice were randomly assigned to saline control or trastuzumab groups, with treatments given on days 0 and 3. Dynamic contrast enhanced (DCE) MRI and diffusion weighted (DW) MRI data were collected on days 0 (pre-treatment), 1 and 4. Tumors, excised on day 4, were processed for histology with anti-CD31 staining. Apparent diffusion coefficients (ADC, a measure of cell density) were extracted from DW-MRI data. DCE-MRI data was modeled to extract the extravascular, extracellular volume fraction, ve, and the volume transfer coefficients, Ktrans and kep, which correspond to the rate of wash-in and wash-out of contrast agent, respectively. Hierarchical clustering of tumor voxel data (ADC, ve, Ktrans, and kep) was used to identify physiological tumor subregions. The contribution of each subregion to tumor volume was quantified as percent tumor volume for each mouse and time point. Image analysis of histology data was used to generate vessel area and nuclei maps for whole-slice histology data. CD31 stained slices were subsequently divided into physiological subregions in terms of vessel and nuclei density, and the contribution of each subregion to the whole-slice area was quantified as percent tumor area. Results: Hierarchical clustering of the MRI data yielded four clusters: low vascularity - low cellularity (LV-LC), low vascularity - high cellularity (LV-HC), high vascularity - low cellularity (HV-LC), high vascularity - high cellularity (HV-HC). At day 0, no significant differences in cluster percent tumor volume were observed between control and treated tumors (p>0.05). At day 4, a significant decrease in LV-HC percent tumor volume was observed in treated tumors compared to control (mean 13.7% vs. 35.0%, p=0.04). Histology subregion analysis corroborated in vivo imaging findings, with treated tumors having significantly lower LV-HC percent tumor area compared to control (mean 26.8% vs 53.6%, p<0.01). Conclusion: High-dimensional analysis of quantitative MRI parameter maps can be utilized to identify physiological subregions of tumor response, and can be biologically validated using histology data. The results suggest trastuzumab therapy decreases the hypoxic (LV-HC) tumor volume. Quantifying tumor microenvironment alterations in response to therapy can potentially be used to predict response for patients with HER2+ breast cancer. Citation Format: Anum Syed, Jennifer Whisenant, Anna Sorace, Thomas Yankeelov. Quantifying trastuzumab-induced alterations of intratumoral heterogeneity using MRI-derived tumor subregions in the preclinical setting [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 1946.