Trends in PET quantification: opportunities and challenges

Abstract

Since its inception, positron emission tomography (PET) has emerged as a non-invasive imaging modality that allows, in different fields (neurology, cardiology and oncology), in vivo quantitative assessment of molecular and physiological biomarkers in healthy and disease states [1–4]. Quantitative analysis makes it possible to establish a direct relationship between the time-varying activity concentration in organs/tissues of interest and the functional parameters representing the underlying biological processes at the cellular level [5–8]. It should, however, be emphasized that the term quantification has often been used inappropriately in the medical imaging literature to indicate different measurement approaches such as [5]: (1) semi-quantification (a contradiction in terms) or relative quantification (e.g., measurement of SUV), (2) absolute quantification of activity concentration, usually incorporating careful corrections for physical degrading factors (e.g., measurement of tracer uptake in MBq), and (3) proper physiological quantification, where the absolute activity concentration [obtained in (2)] is converted into molecular parameters of interest [e.g., glucose metabolic rate (rGMCglc) expressed as mol/100 g/min]. The concentration of tracers in organs/tissues of interest depends on their specific kinetic properties, i.e., various factors including, but not limited to, the rate of delivery through circulation, the biochemical reactions involved in the specific biological process under examination, biological clearance, and so on. Furthermore, the measurement of radioactivity in volumes of interest must take into account the physical half-life of the radionuclide employed for the pharmaceutical labeling. These physiological and physical factors must be fully taken into consideration if quantitative PET is to realize its full potential, and thus allow assessment of the physiological and molecular characteristics of the cells and organs/tissues under examination. Using these approaches, it is possible to quantify a number of processes, including the rate of glucose utilization, receptor binding, receptor occupancy, and so on. The resulting estimates can then be linked to clinical outcomes (e.g., disease evolution, response to treatment, survival) so that disease activity can be assessed and related to the underlying pathological states. Moreover, these quantitative measures can provide surrogate endpoints in therapy trials. The major challenges to quantitative preclinical PET imaging, when the aim is to quantify biological or pharmacokinetic processes, can be categorized in five classes [9, 10]: