Vascular endothelial growth factor (VEGF)-A is overexpressed in most malignant and premalignant breast lesions. VEGF-A can be visualized noninvasively with PET imaging and using the tracer 89 Zr-labeled bevacizumab. In this clinical feasibility study, we assessed whether VEGF-A in primary breast cancer can be visualized by 89 Zr-bevacizumab PET. Methods: Before surgery, breast cancer patients underwent a PET/CT scan of the breasts and axillary regions 4 d after intravenous administration of 37 MBq of 89 Zr-bevacizumab per 5 mg. PET images were compared with standard imaging modalities. 89 Zr-bevacizumab uptake was quantified as the maximum standardized uptake value (SUV max ). VEGF-A levels in tumor and normal breast tissues were assessed with enzyme-linked immunosorbent assay. Data are presented as mean 6 SD. Results: Twenty-five of 26 breast tumors (mean size 6 SD, 25.1 6 19.8 mm; range, 4-80 mm) in 23 patients were visualized. SUV max was higher in tumors (1.85 6 1.22; range, 0.52-5.64) than in normal breasts (0.59 6 0.37; range, 0.27-1.69; P , 0.001). The only tumor not detected on PET was 10 mm in diameter. Lymph node metastases were present in 10 axillary regions; 4 could be detected with PET (SUV max , 2.66 6 2.03; range, 1.32-5.68). VEGF-A levels in the 17 assessable tumors were higher than in normal breast tissue in all cases (VEGF-A/mg protein, 184 6 169 pg vs. 10 6 21 pg; P 5 0.001), whereas 89 Zr-bevacizumab tumor uptake correlated with VEGF-A tumor levels (r 5 0.49). Conclusion: VEGF-A in primary breast cancer can be visualized by means of 89 Zr-bevacizumab PET.