Sarah Verhoeff

143 Summary, general discussion and future perspectives is currently no place for [89Zr]Zr-DFO-girentuximab PET/CT as a tool to discriminate between mccRCC patients with indolent or aggressive disease. Our results cannot be extrapolated to the role of CAIX as visualized by [89Zr]Zr-DFO-girentuximab PET/CT in other tumor types where CAIX is upregulated in hypoxic conditions. Position of antibody-based PD-L1 PET/CT The available data on antibody-based PD-(L)1 PET/CT imaging have shown variable results regarding the correlating between tracer-uptake and PD-L1 expression as well as ICI response (Table 1). The complexity of underlying mechanisms of action, the use of various monoclonal antibodies and the differences in trial designs, leaves room for the interpretation and explanation of these results. Beyond the reported variable correlation between tracer-uptake and treatment response, there are several other limitations associated with PET imaging using 89Zr-labeled antibodies. First the long circulating half-lives of antibodies results in high background signal that requires imaging several days after injection of the radiotracer. This can be impractical to implement in the clinical workflow. Second, this circulation time also results in the need for long-lived radioisotopes such as 89Zr and a consequently higher radiation exposure to patients. This is a problem because it limits the use of 89Zr-antibody PET/CT in younger patients, patients with a curative treatment option, or in case of repeated radiotracer injections. Third, the size of antibodies may impede their diffusion into solid tissues15. But not only their size, also local features affect the antibody uptake such as the blood vessel density, vascular permeability, interstitial fluid pressures and tumor growth kinetics16. Variations could hamper antibody tumor penetration and might also lead to non-specific uptake, which can challenge the interpretation of the PET signal in the tumor17. Last but not least, the Fc-tail modification affects the circulation time and effector functions of the antibody, explaining differences between different antibodies used to visualize the same target18. Altogether, antibody-based PD-L1 PET imaging is an excellent tool to non-invasively visualize antibody drug distribution. However, the impractical design of antibody-based PD-L1 PET-imaging, long acquisition time, radiation exposure and costs, limit potential future implementation as a biomarker in daily clinical practice. Therefore, we do not expect that antibody-based PD-L1 PET/ CT will replace the current golden standard of PD-L1 immunohistochemistry to select patients for ICI treatment (chapter 5,6,7). 8