TY - JOUR T1 - F18-florbetapir PET/CT imaging of cardiac amyloidosis: metabolite correction in compartmental modeling JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 224 LP - 224 VL - 59 IS - supplement 1 AU - Marie Kijewski AU - Hendrik Harms AU - Shipra Dubey AU - Anthony Belanger AU - M. Samir El-Sady AU - Sophia Jacob AU - Ariana Nodoushani AU - Shuyan Wang AU - William Sticka AU - Mi-Ae Park AU - Paco Bravo AU - Marcelo Di Carli AU - Rodney Falk AU - Sharmila Dorbala Y1 - 2018/05/01 UR - http://jnm.snmjournals.org/content/59/supplement_1/224.abstract N2 - 224Objectives: F18-florbetapir, an amyloid tracer, was developed for imaging Alzheimer’s disease (AD). It is now being used in cardiac amyloidosis (CA), a disease in which amyloid infiltrates the myocardium. In healthy controls and in AD, tracer metabolism is similar enough across individuals that a population metabolite curve (1) can be used for compartmental analysis; the advantage of this approach is that it eliminates the need for blood sampling and time-consuming blood processing and analysis. Our goal was to determine whether individual metabolite analysis is necessary for compartmental modeling of F18-florbetapir PET images of CA, as we have previously found for F18-FLT PET/CT in glioblastoma patients (2). Methods: Thirty-four CA patients were imaged with F18-florbetapir PET/CT, as part of the Myocardial Imaging of Cardiac Amyloidosis (MICA) project at BWH, at baseline and 6 and 12 months later. Images were analyzed using a two-compartment, reversible model; the major parameter of interest was the volume of distribution (Vt). Four to six blood samples were acquired during each 60-min dynamic acquisition for metabolite analysis using high performance liquid chromatography (HPLC). For a subset of 12 patients, individual metabolite curves were derived by weighted fitting of parent fractions at each sampling point to a Hill function. In order to determine the errors in Vt that would result from using an incorrect metabolite function, we simulated a typical CA time-activity curve assuming the population metabolite curve, and performed compartmental analysis using metabolite curves representing 1%-10% slower metabolism. Results: Metabolite curves varied widely among the 12 patients; furthermore, curves for individual patients were not constant over time. In almost all cases, individual metabolite curves suggested slower tracer metabolism compared to the published population curve. The simulation experiments yielded errors in Vt ranging from 3.4% for 1% slower metabolism to 24.6% for 10% slower metabolism. For many of the cases we studied, patient metabolite curves represent more than 10% slower metabolism compared to the population curve. We plan to analyze the distribution of metabolite curves for the entire group in order to determine whether subgroups with similar tracer metabolism can be identified based on clinical characteristics. Conclusions: Individual metabolite analysis may be required for compartmental modeling of F18-florbetapir PET/CT images of cardiac amyloidosis. ER -