The challenges and opportunities in functional imaging HER2 positive breast cancers

摘要： Breast cancer is the most common malignancy in women and 20‐30% of these cancers overexpress the HER2 protein making them candidates for HER2‐directed therapies. Trastuzumab (Herceptin?), the first HER2‐directed therapy, is a humanized monoclonal antibody that binds to the extracellular domain of HER2 preventing downstream signaling and cell proliferation. Trastuzumab is also an immunologic agent stimulating antibody‐dependent cytotoxicity. Investigators have developed radiolabeled trastuzumab as a PET imaging agent for use in HER2‐positive breast cancer patients. However, its clinical role has yet to be established. In the Journal, Dr. Woo and his colleagues report the use of NOTA as a chelator for 64Cu labeling of trastuzumab and claim more favorable pharmacokinetics than 64Cu DOTA trastuzumab [1]. But is NOTA a better chelator for 64Cu than DOTA? They base their comparison on a report by Paudyal et al. (Cancer Sci 2010; 101: 1045‐1050) in which the uptake of 64Cu‐DOTA‐trastuzumab in the liver was 26.9 ±7.4% ID/g at 24 hours[12], while in contrast, the uptake of 64Cu‐NOTA‐trastuzumab was 5.44 ±1.84%D/g in the liver at 24 hours (Figure 4)[1]. They conclude the difference in liver uptake between the two studies was due to release of 64Cu from the DOTA but not from the NOTA chelate. However, the two studies cannot be compared since they were not performed using the same tumor models. Paudyal et al. (Cancer Sci 2010; 101: 1045‐1050) performed PET imaging on Her2+ non‐small cell xenografts, while Woo et al. performed PET imaging on Her2+ breast cancer xenografts. The difference in liver uptake may simply reflect differential shedding of Her2 antigen to the liver between the two types of tumors. To make their point, Woo et al. should have compared DOTA vs NOTA conjugated trastuzumab in the same tumor model. This is a clear example of confusing the chemical stability of metal chelates with their metabolic clearance in different tumor models.