TY - JOUR T1 - Measuring in vivo receptor affinity: Comparison of methods using the H3 ligand [11C]GSK189254 in baboon JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 213 LP - 213 VL - 51 IS - supplement 2 AU - Cristian Salinas AU - David Weinzimmer AU - Graham Searle AU - David Labaree AU - Yu-Shin Ding AU - Yiyun Huang AU - Eugenii Rabiner AU - Richard Carson AU - Roger Gunn Y1 - 2010/05/01 UR - http://jnm.snmjournals.org/content/51/supplement_2/213.abstract N2 - 213 Objectives PET can measure the in vivo receptor affinity of a compound using a number of different approaches that require either radiolabelling of the compound itself or the use of an established radioligand for the target of interest. We present a comparison of these techniques in the baboon using the high affinity H3 ligand [11C]GSK189254. Methods One baboon underwent a total of 9 [11C]GSK189254 scans over 5 separate days. Each scan included a 120min dynamic emission and arterial plasma activity, parent fraction, and plasma free fraction measures. Days 1, 2, and 3 consisted of a high specific activity (SA) scan followed by a lower SA scan. On day 4 and 5 unlabelled GSK189254 was pre-administered before high SA scans at 1, 6 hrs (on day 4), and 15 hrs (on day 5) allowing the generation of a time occupancy curve (TOC). In vivo GSK189254 affinity was estimated using; (a) a full non linear kinetic model fitted simultaneously in tracer and non-tracer conditions (with Bmax and Kd as the main parameters), (b) a direct occupancy model applied to the total volumes of distribution, Vt = Vnd+BPp(Kd/(Kd+Cp)) where Vt, BPp, and Kd are estimated and Cp is the plasma drug concentration, and (c) a PK-RO model that relates the plasma drug concentration time course and the TOC, dRO/dt=kon Cp (1-RO) - koff RO. All estimates were corrected by the appropriate free fraction. Method (a) requires a labelled version of the drug whilst methods (b&c) can be applied with an established radioligand distinct from the drug. Results All approaches generated similar estimates of affinity: (a) Kd=42±17pM, (b) Kd=28±9pM, (c) Kd=72±8.7pM. The consistency of method (c) with (a&b) suggests that a direct relation between plasma concentration and TOC is reasonable. Conclusions The three approaches yielded consistent estimates for the in vivo affinity of GSK189254. However, for cases where a drug has an increased target residence time, leading to an indirect relationship, method (c) may provide more accurate estimates of in vivo affinity ER -