565 Background: While radiation portals are tailored to a patient’s unique anatomy and the selection of systemic agents routinely employs biomarker data, the selection of radiotherapy based on a patient’s tumor biology is not routinely utilized in breast cancer. The purpose of this study was to identify which genetic markers are possible predictors for local recurrence as a surrogate for radiation response. Methods: We identified 200 patients who received radiotherapy for breast cancer. Selected tumor markers included: Androgen receptor (AR), Hypoxia Inducible Factor 1-α (HIF-1), Phosphotidylinosotol-4,5-bisphosphate 3-kinase (PI3K), and Interleukin 13 (IL-13). Biomarkers were analyzed in terms of “extent” and “intensity” on a scale of 0-3 and scored by 2 separate pathologists. The primary endpoint of local recurrence (LR) & secondary endpoint of overall survival were analyzed using Kaplan-Meier survival curves, log-rank test for differences, and Cox regression models. Results: Median follow up was 7.98 years. At 5 years, the rate of LR was 92.6% and overall survival was 89.4%. On multivariate Cox regression analysis, a one unit increase in IL-13 extent increased the hazard of LR by 73%. A one unit decrease in AR extent increased the hazard of LR by 134%. The hazard of death increased 3.2 times for each unit increase in HIF1 extent. The hazard of death increased 1.5 times for each unit increase in PI3K extent. PI3K extent and intensity was increased, and AR extent and intensity was decreased in triple negative breast cancer (TNBC) (n = 68) vs non-TNBC (p < 0.0001). African Americans had a 4.2 times hazard of LR vs Caucasians. Conclusions: Expression of IL-13 was associated with a higher risk of LR; expression of AR was associated with decreased LR. These two markers may be instrumental in predicting radiation response. If this study is validated, cancers that express more IL-13 may require higher doses or targeted therapy. In contrast, those cancers expressing AR may not require as aggressive therapy. Lastly, PI3K and HIF1 α expression were significant predictors of worse overall survival. The clinical implications of these biologic markers are significant as they may help to guide biologically-driven, personalized breast cancer radiotherapy.