Radiochemistry challenges and progression for incorporation of 18F into a complex substituted 6-18F-fluoroquinoline BMS-986205 for IDO imaging

605 Objectives: Indoleamine-2,3-dioxygenase (IDO) is a monomeric, heme-containing dioxygenase which catalyzes the initial, rate-limiting step in tryptophan metabolism. IDO induces immunosuppression in the tumor microenvironment and some patients treated with PD1 checkpoint inhibitors have shown upregulation of the IDO enzyme. Existing clinical methods for measuring IDO include ex vivo immunohistochemistry (IHC) of a tumor biopsy and in vivo serum kynurenine measurements. Many patients will have only one pre-treatment biopsy making it difficult to monitor progression and treatment response. Additionally, kynurenine is a downstream metabolite that does not give a complete picture of intra-tumor IDO levels. PET imaging may provide a way to directly measure tumor levels of IDO and serve as a method to monitor disease progression and response to treatment. BMS-986205 is a selective, IDO1 inhibitor that contains a 6-fluoroquinoline moiety possibly suited to 18F labeling to afford an IDO1 PET radioligand. The complexity of this molecule, which required incorporation of the 18F label into the unactivated 6-position of a quinoline presented several synthetic challenges. Our goal was to develop a translatable method for late-stage nucleophilic fluorination of BMS-986205 suitable for clinical use. Methods: Several fluorination methods were investigated to synthesize this ligand. Multiple precursors were synthesized including mesityl iodonium salts (1), sulfonium salts (2), boronic acids (3) and pinacol boronate esters (4). Each were generated from a common intermediate of substituted 6-iodoquinoline. For each precursor the corresponding literature radiolabeling conditions were tested first. The conditions were then modified to increase yield and find a promising synthetic route for optimization. Modifications included, but were not limited to, changing the QMA size, fluoride trapping and eluting conditions, synthetic reagents, and reaction solvents. Results: The copper-mediated fluorination of mesityl iodonium salts was unsuccessful with -OTf counter ion resulting in fluorination of the leaving group or fluoride exchange with -BF4 counter ion. Copper-mediated fluorination of a boronic acid was also unsuccessful, but the pinacol boronate ester did generate product in very low yield of less than 37 MBq (1 mCi) following the corresponding literature procedure. Radiolabeling of the sulfonium salt was successful in low yield around 185 MBq (5 mCi, 1.5% decay corrected yield from 775 mCi of 18F) with >98% radiochemical purity and >99% ee. Using the pinacol boronate ester precursor, the reaction conditions were modified and optimized as shown in Scheme 1 to yield 1.048±0.5169 GBq (28.32±13.97 mCi, n=24, 3.5% decay corrected yield from 1 Ci of 18F) with >98% radiochemical purity and >99% ee. Specific activity was between 74-925 MBq/nmol (2-25 mCi/nmol) with an average 231 MBq/nmol (6.24 mCi/nmol). This method was translated to a clinical PET synthesis facility and validated for use in clinical studies. Conclusions: We have evaluated several methods for 18F labeling of [18F]BMS-986205 with varying degrees of success. Copper-mediated nucleophilic fluorination of a pinacol boronate ester precursor was optimized for semi-automated synthesis with good radiochemical yield, high radiochemical purity and high enantiomeric excess suitable for clinical imaging studies. References: (1) Ichiishi N, Brooks AF, Topczewski JJ, Rodnick ME, Sanford, MS, Scott, PJ. Org Lett. 2014;16:3224-3227. (2) Sander K, Gendron T, Yiannaki E, Cybulska K, Kalber TL, Lythgoe MF, Arstad E. Sci Rep. 2015;5: 9941. (3) Mossine AV, Brooks AF, Makaravage KJ, Miller JM, Ichiishi N, Sanford MS, Scott PJ. Org Lett. 2015;17:5780-5783. (4) Tredwell M, Preshlock SM, Taylor NJ, Gruber S, Huiban M, Passchier J, Mercier J, Genicot C, Gouverneur V. Angew Chem Int Ed Engl. 2014;53:7751-7755.