The purpose of the present study is to explore the relation between glycolytic metabolism assessed by (18)F-fluoro-2-deoxy-D-glucose positron emission tomography/computed tomography ((18)F-FDG PET/CT) and final neoadjuvant chemotherapy (NC) response in locally advanced breast tumors. Of women with breast cancer, 126 were prospectively evaluated. All patients underwent (18)F-FDG PET/CT previous to NC. Standard uptake value (SUV) max was calculated in the primary tumor. After NC, residual primary tumor specimen was histopathologically classified according to Miller and Payne tumor regression grades (TRG), from G1 to G5 and in response groups as good responders (G4 or G5), partial responders (G2 or G3), and non-responders (G1). Furthermore, residual lesions were classified following a binary assessment as responders (G4 or G5) and non-responders (the rest of cases). The relationship between SUV max with TRG and response groups was evaluated. Of tumors, 127 were assessed (a patient had bilateral breast lesions). TRG were as follows: G1 (27), G2 (27), G3 (32), G4 (11), and G5 (30). Forty-one were classified as good responders, 59 as partial responders, and 27 as non-responders. For the binary assessment, 41 lesions were classified as responders and 86 as non-responders. We found statistical differences (p=0.02) between the mean SUV max and TRG with greater SUV values for G5 compared to the other TRG. Good responders showed greater mean SUV max ± SD compared to partial responders and non-responders (10.51 ± 6.64 for good responders, 6.94 ± 5.81 for partial responders, and 5.23 ± 2.76 for non-responders; p=0.001). Baseline tumor metabolism assessing by FDG PET/CT was associated with the final histopathologic status after neoadjuvant chemotherapy, with greater SUV max values for good responders compared to the less responder cancers.