Introduction and Motivation. Neoadjuvant therapy (NAT) is the standard-of-care for patients with locally advanced breast cancer. The development of novel NAT regimens provides an opportunity to tailor treatment for individual patients by predicting response to therapy early during NAT. Population-based approaches are limited in capturing tumor heterogeneity and longitudinal changes for individual patients. Thus, there is a need for approaches employing patient-specific data to make patient-specific predictions in the clinical setting. Previously, we developed a biology-based mathematical model (Jarrett et al. Nat Protoc. 2021 Nov;16(11):5309-5338.) that achieved a concordance correlation coefficient (CCC) of 0.97 between the measured and predicted change in total tumor cellularity for triple-negative (TN) breast cancer patients (n = 56; Wu et al. Cancer Res. 2022 Sep 16;82(18):3394-3404.). In this work, we have extended this model to the I-SPY 2 data set. Methods. I-SPY 2 is an adaptive clinical trial for locally advanced breast cancer that acquired dynamic contrast enhanced (DCE) and diffusion-weighted (DW) MRI scans before and during NAT. Our initial subset of 17 patients consists of 7 with TN and 10 with estrogen- or progesterone-positive breast cancer. Imaging data was acquired before (V1), after three weeks (V2), and after completion (V3) of paclitaxel treatment. The first image processing step is to correct for motion within and between the DCE-MRI and DW-MRI series across time (i.e., V1 to V3) using a combination of rigid and tumor-constrained non-rigid registration. A map of the apparent diffusion coefficient (ADC) was generated by fitting DW-MRI scans with b-values of 0, 100, 600, and 800 s/mm2. We segmented the tumor ROI by applying fuzzy c-means clustering to a radiologist-guided bounding box drawn on the DCE-MRI scans. Adipose and fibroglandular tissues were segmented using k-means clustering. The voxel-wise number of tumor cells (NTC) was calculated from the ADC map and drug distribution from DCE-MRI data. Our biologically-based model consists of a reaction-diffusion partial differential equation that describes the voxel-wise rate of change in NTC as the balance of cell diffusion, logistic proliferation, and death due to treatment. Diffusion was mechanically coupled to the surrounding tissue. We calibrated model parameters (proliferation rate, diffusion constant, and drug efficacy rate) using data from V1 and V2. The calibrated model was applied to make patient-specific predictions for tumor status at V3. Results and Conclusion. For this initial subset of 17 patients, the CCC between the measured and calibrated change from V1 to V2 was 0.95 for total tumor cellularity change and 0.95 for total tumor volume change. The CCC between the measured and predicted change from V1 to V3 was 0.85 for total tumor cellularity change and 0.83 for total tumor volume change. These preliminary results suggest our tumor forecasting pipeline can make accurate predictions using MRI data obtained in the clinical setting. Future directions include modeling patients treated with experimental therapies and quantifying predictions as function of breast cancer subtype and drug class.