Abstract Objective: Change in functional tumor volume (FTV) measured by dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), has been shown to be associated with response to neoadjuvant chemotherapy (NACT) for patients with stage II/III breast cancer. FTV reflects the vascularized volume of tumor and is calculated by applying minimum thresholds PEt for the initial percent enhancement and SERt for the early-to-late signal enhancement ratio following gadolinium contrast injection. In this retrospective study, we investigate the association of FTV influenced by PEt and SERt values with recurrence-free survival (RFS) in breast cancer subtypes defined by hormone receptor (HR) and HER2 status. Materials and Methods: 64 patients with locally advanced breast cancer were imaged by DCE-MRI before treatment (MRI1), after one cycle of adriamycin-cytoxan (AC) therapy (MRI2), inter-regimen (MRI3, taxane receivers only) and at the completion of chemotherapy prior to surgery (MRI4). Because treatment length and regimen varied among patients, the MRI exam performed after all NACT and before surgery was designated as final MRI (MRIf). FTV was calculated with varying PEt (30–200% in steps of 10%) and SERt (0–2 in steps of 0.2) values. A Cox proportional hazard model was used to analyze association with RFS, defined as the time between surgery and disease recurrence or last disease-free follow-up, for the following imaging predictors: early percent change in FTV (ΔFTV2: MRI1 vs. MRI2), final percent change (ΔFTVf: MRI1 vs. MRIf), and final FTV in MRIf (FTVf). RFS association was evaluated for the full cohort and within subsets defined by tumor subtype (HR+/HER2-, HER2+, triple negative). Estimated hazard ratios per unit change in predictors, associated 95% confidence intervals (CI) and p-values were evaluated. The cutoff of p<0.05 was used to differentiate PEt/SERt values with higher predictive performance. Results: For the full cohort, FTVf showed the most robust association with RFS among the three predictors, with p<0.05 across all measured combinations of PEt/SERt. The PEt/SERt combination with smallest p-value was found for PE=30% and SER=0.2, with hazard ratio 1.09 (95% CI 1.06−1.13, p<0.001). P-values less than 0.05 for the ΔFTVf predictor was found for the full range of SERt values tested, but was mostly confined to PEt in 30−110% range; ΔFTV2 resulted in p<0.05 for very few PEt/SERt combinations. When analyzed by tumor subtypes, different prediction profiles were observed. FTVf showed p<0.05 for the lower PEt range only in the HER2+ (n=15) and triple-negative (n=11) groups. A limited number of PEt/SERt pairs showed p<0.05 for ΔFTVf and ΔFTV2, with the exception of the ΔFTVf in the triple-negative group, for which p <0.05 for many PEt/SERt combinations in the low PEt range. Conclusions: This study was undertaken to explore the significance of FTV for predicting breast cancer recurrence following NACT. While this study is retrospective in nature and had small sample size, the findings nevertheless suggest that performance of imaging predictors based on FTV may be improved if threshold optimization is performed separately for the clinically-relevant subtypes defined by HR and HER2 receptor expression. Citation Format: Wen Li, Wei-Ching Lo, Ella F Jones, David C Newitt, John Kornak, Lisa J Wilmes, Nola M Hylton. Optimization of magnetic resonance imaging predictive performance by breast cancer subtypes for predicting response to neoadjuvant chemotherapy [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P1-01-09.