Evaluation of [¹¹C]TMI as a COX-2 selective PET tracer in baboon

Objectives: Cyclooxygenase-2 (COX-2) is an inducible enzyme under normal physiological conditions. Overexpression of COX-2 is associated with pathogenesis of inflammation, cancers, stroke, arthritis and several neurological disorders. Positron-emitting labeling of COX-2 selective inhibitors may permit in vivo imaging of COX-2 localization, quantification of its activity in diseases, as well as relating nonsteroidal anti-inflammatory drug (NSAID) blockade of COX-2 to therapeutic effect. At present there is no target-specific or validated PET tracer available for the in vivo monitoring of COX-2. Here we validate [¹¹C] 3-(4-methylsulfonylphenyl)-4-phenyl-5-trifluoromethyl isoxazole ([¹¹C]TMI), a selective COX-2 inhibitor (Ki = <1 nM) in nonhuman primate, using PET.

Methods: Synthesis of reference standard TMI, corresponding thiobuturic ester precursor were achieved as previously described.¹ [¹¹C]TMI radiosynthesis is accomplished by reacting the corresponding butyrate ester with [¹¹C]CH₃I, followed by oxidation with oxone in a one pot procedure.¹ PET data were acquired in male baboons (N=2) (papio anubis) with an ECAT ACCEL (Siemens Medical Solutions, Inc. Knoxville) in two-dimensional mode for 120 minutes following an intravenous injection of 3 + 0.5 mCi of [¹¹C]TMI. Blocking study was performed by administering 1 mg/kg meloxicam 30 minute prior to [¹¹C]TMI injection (N=1). [¹¹C]TMI volume of distribution (V_T) was estimated using a metabolite-corrected arterial input functions in a one-tissue compartment model (1TC) and graphical methods such as likelihood estimation in graphical analysis (LEGA) and Logan. A revised Lassen plot² was used to calculate occupancy and nondisplaceable binding (V_ND) using baseline and blocking data.

Results: Radiosynthesis of [¹¹C]TMI was achieved in 37 + 3% radiochemical yield (EOS) with >99% radiochemical purity and a specific activity of 2.5 + 0.5 Ci/μmol. PET studies in baboon showed that [¹¹C]TMI penetrates the blood brain barrier (BBB) and accumulates in brain in a somewhat homogeneous pattern. Metabolite analyses in baboon indicated that [¹¹C]TMI undergoes no significant metabolism with >90% of parent tracer retained in the baboon plasma after 90 minute injection. All the tested models show comparable V_T measurement in the range of 4-7-5.7 (mL/cm³). We found about 24% lower V_T values in blocking studies compared to baseline binding of [¹¹C]TMI. Lassen plot estimates a modest occupancy of ~43%, and V_ND of ~2. Conclusion: Our PET imaging studies in baboon show that [¹¹C]TMI penetrates the BBB and distributes to COX-2 enriched brain regions. Details of the in vivo evaluation of [¹¹C]TMI in baboon brain will be presented. References 1. Majo VJ et al., Bioorg Med Chem Let., 2005, 15, 4268. 2. Cunningham VJ et al., J Cereb Blood Flow Metab., 2010; 30(1): 46-50. Grant Support: Diane Goldberg Foundation (NYSPI/CUMC) and Pfizer Inc.