PROJECT SUMMARY This R21 proposal seeks to develop a cell-size-based MRI technique, Selective Size MR Imaging using Filters via diffusion Times (SSIFT), as a novel approach for imaging brain metastases (BM) in patients with high specificity, and to evaluate its potential to differentiate recurrent BM from radiation necrosis (RN) induced by stereotactic radiosurgery (SRS). Approximately 10–30% of all cancer patients will eventually develop BM in their lifetime, and SRS is a common treatment for patients with BM to improve the local control. However, about 10-20% of treated BM will eventually develop new contrast-enhancing lesions that are either RN or recurrent BM. These two types of lesions usually occur within a similar time frame but need to be treated differently: RN can be managed conservatively to alleviate symptoms; by contrast, tumor recurrence is often managed with surgery or further radiation to control tumor progression. Unfortunately, RN is indistinguishable from tumor recurrence on standard-of-care MRI with gadolinium (Gd)-based contrast agents due to the breakdown of the blood-brain barrier (BBB) in both types of lesions. Therefore, there is a need for clinical radiation oncology to develop a reliable non-invasive imaging method to differentiate tumor recurrence from RN. Despite numerous attempts to develop a variety of advanced imaging methods to tackle this clinical challenge, there is still no single imaging method that has been widely accepted as a reliable imaging biomarker. We hereby propose a new idea, i.e., to use endogenous information on cell size to differentiate BM from RN. Different from our previous quantitative cell size MRI with a long scan time, we propose here a fast SSIFT method that serves as a cell-size-based filter to selectively emphasize signals arising from large cancer cells with simultaneous suppression of signals from other brain abnormalities, such as RN. We have validated SSIFT in preclinical animal models to demonstrate its ability to differentiate BM from RN, but it remains unclear if SSIFT will work in patients with BM. We hereby hypothesize that SSIFT is a practical, reliable, and specific approach for imaging BM in patients with the capability to differentiate recurrent BM from RN. To test this hypothesis, we have assembled a multidisciplinary team and propose [Aim 1] to refine a reliable and efficient SSIFT protocol in imaging BM patients and [Aim 2] to evaluate the clinical ability of SSIFT in differentiating recurrent BM from RN induced by SRS. Upon completion, the proposed study will break new ground to establish a new type of cell-size-based MRI technique geared towards addressing an important, common dilemma in the management of BM patients with SRS treatments. In addition, the ability to differentiate BM from peri-tumor edema makes this new method another possible alternative imaging approach for BM patients with kidney dysfunction who are not candidates for gadolinium-agent-based MRI.