Page 208 - 89Zr-Immuno-PET:Towards a Clinical Tool to Guide Antibody-based Therapy in Cancer

Chapter 10
now this type of study design has been limited by the relative high radiation exposure of 89Zr-immuno-PET.
In addition, reduction of radiation exposure will allow application of 89Zr-immuno-PET for indications beyond oncology, for example in auto-immune diseases where mAb-based therapy is also used (20). Finally, increase in sensitivity is expected to allow molecular imaging of other slow-kinetic drugs (e.g. nanoparticles) and cell imaging.
2. Further biological and clinical validation (including cost effectiveness) Before 89Zr-immuno-PET can be applied in large scale clinical trials or in daily clinical practice, the link with underlying biology and clinical outcome should be confirmed. We observed a potential false-positive finding for 89Zr-rituximab in relapsed/refractory non-Hodgkin lymphoma (Chapter 7) and two false-positive findings for 89Zr-cmAb U36 in head and neck cancer (Chapter 3). Visual assessment of 89Zr- immuno-PET is used to localize non-physiological accumulation of the radiotracer, indicating target-mediated uptake. Visual assessment provides a qualitative biomarker, assuming that target-mediated, specific uptake increases over time, and dominates the total PET signal at late time points (e.g. 3-6 days p.i.) when tumor uptake is visible.
Future studies are required to develop a method for quantification of target-engagement in tumors. However, tumor characteristics are more complex and more variable (within and between patients) compared to normal tissues. Therefore, tumors with and without target expression can be studied to investigate non-specific uptake in tumors.
Although tumor biopsies are often used as reference, immunohistochemistry may not be a true gold standard for target engagement of mAbs. Biopsies have several limitations, for example sampling error due to heterogeneity in target expression. In addition, detection of target expression by IHC does not provide information on accessibility of the target in-vivo (whether the therapeutic mAb has reached the target). Future work to relate quantification of 89Zr-labeled mAbs to PBPK modeling is expected to provide more insight in how the PET signal is linked to the underlying biology. This information is required to define which measure (visual, SUV, Ki) is suitable for assessment of target engagement. The type of measure (e.g. single time point, or multiple time point) has impact on study design for further clinical validation studies and cost-effectiveness.
206