Abstract Purpose: We previously demonstrated that mitochondrial inhibitors' efficacy was restricted to a metabolic context in which mitochondrial respiration was the predominant energy source, a situation achievable by inducing vascular normalization/hypoxia correction with antiangiogenics. Vascular normalization can be tracked with 2[18F]fluoro-2-deoxy-d-glucose (FDG)-PET. We tested the efficacy of the mitochondrial inhibitor ME-344 or placebo added to bevacizumab in early breast cancer. Patients and Methods: Treatment-naïve HER2-negative patients with T > 1 cm (any N) underwent a breast-centered 18F-fluorodeoxyglucose (FDG)-PET (day 1) and received a single dose of bevacizumab (15 mg/kg), followed by a second FDG-PET (day 8). Patients were then randomized (1:1) to Arm A (ME-344 10 mg/kg intravenous on days 8, 15, and 21) or Arm B (placebo). Tumors were biopsied on days 0 and 29. Succinate dehydrogenase enzyme histochemistry (SDH-EHC), confocal microscopy of vessel architecture, and HIF1α staining were performed in pre- and posttreatment biopsies to assess the pharmacodynamics, vessel normalization, and tissue re-oxygenation by bevacizumab, respectively. Results: ME-344 displayed significant biological activity versus placebo: compared with a 186% increase in Arm B, Ki67 decreased by 23.4% from days 0 to 28 in Arm A (P < 0.001) (N = 42 patients). FDG-PET predicted vascular normalization in about one-third of the patients in each arm, which was confirmed using confocal microscopy and HIF1α staining. In the subgroup with vascular normalization, ME-344 induced a Ki67 decrease of 33.4% (placebo: 11.8 increase). SDH-EHC suggested on-target effects of ME-344. Conclusions: ME-344 has significant biological antitumor activity in HER2-negative breast cancer, particularly after induction of vascular normalization and tissue reoxygenation with bevacizumab.