191 Physiological MRI Biomarkers & Cognition after SRS for Brain Metastases We observed post-radiotherapy changes within edematous regions. Notably, regions that recovered from edema three months after SRS showed increased OEF, CBF, and CMRO2. Contrarily, CVR remained unchanged within these regions, with values lower than those observed in healthy-appearing tissue. These findings suggest that while these resolved edema regions show metabolic recuperation, the vascular reserve continues to be impacted. While there was a trend towards decreased OEF, CMRO2, and CVR in regions with new edema post-radiotherapy, statistical significance for CVR was almost reached (p = .078) after removing outliers. This lack of significance may be attributed to the limited number of patients available for analysis, emphasizing the need for further investigation. Altogether, our findings suggest that CVR might be more susceptible to, and serve as a sensitive marker of edema-induced effects, warranting continued exploration in future studies. We observed a global increase in CBF throughout the brain after SRS, while other physiological measures remained constant. Studies have shown that exposure to radiation leads to an inflammatory response in the tissue.59,60 The observed increase in CBF may thereby be either reflective of a restorative mechanism of the brain in response to radiation-induced inflammation or ongoing inflammation. However, as the majority of patients did not use dexamethasone three months after SRS, it is most likely indicative of a restorative mechanism. Previous studies have demonstrated that tissue exposed to a radiation dose exceeding 15 Gy suffers irreversible damage to its vasculature.17 Surprisingly, our regional analysis did not reveal differences in post-radiotherapy changes among regions exposed to low (<10 Gy), medium (10-15 Gy), or high (>15 Gy) doses. However, it is important to consider all patients in the current study received SRS, which effectively limits the radiation dose to the healthy brain tissue surrounding the tumor. This may have influenced our analysis, particularly in the high dose regions, where the number of voxels available for examination was considerably lower. In order to gain further insights, we therefore conducted regional analysis using equivalently sized ROIs (bins) that corresponded to increasing radiotherapy dose. Our analysis revealed a negative correlation between the dose and CBF, CMRO2, and CVR, but not OEF. This suggests that as the radiotherapy dose increases, the impact on both the metabolic and vascular capacities become more pronounced. While voxel-wise analyses did not support the associations observed in this ROI-based analyses, regional analysis is often more precise in evaluating these association as demonstrated in previous research14,61, due the inherent measurement noise associated with these physiological MRI measures. As our current analyses focused solely on linear relationships, exploring non-linear relationships, especially those seen at low radiotherapy dose, would be valuable for future research. Further investigation 7
RkJQdWJsaXNoZXIy MTk4NDMw