Challenges of quantification of TSPO in the human brain.

The first positron emission tomography (PET) imaging studies in humans of the translocator protein 18 kDa (TSPO) were conducted in the 1980s with a primary interest in quantifying the binding in peripheral organs such as the heart, spleen and kidneys to what was then known as the peripheral benzodiazepine receptor. However, the number of studies rapidly increased when the focus of the research shifted to the brain, and [11C](R)-PK11195 became de facto the reference radiotracer for all in vivo TSPO binding assays. For the quantitative analysis of the data which initially was performed with compartmental models and plasma input functions, this led to the adoption of the reference tissue kinetic models which were developed at the same time in the mid 1990s. In contrast to many neuro-receptor studies of the dopaminergic or serotonergic system, it was not possible to anatomically define a brain region devoid of TSPO that could serve as a reference region. Instead, data-driven techniques were adopted that extracted at the voxel level reference tissue kinetics without incorporating anatomical information. In this review, an overview of the development, use and challenges of the various quantitative analysis methods for TSPO brain PET data is given. The different approaches to (automatically) extract reference tissue input curves from the dynamic images are discussed. Descriptions of key PET imaging studies exploring TSPO binding quantitatively in disease populations are included.