Universität zu Köln

Cologni, Roberta (2023). Synthesis and evaluation of 18F-labeled inhibitors for targeting mutant isocitrate dehydrogenase 1 (mIDH1). PhD thesis, Universität zu Köln.

PDF (Dissertation Cologni Roberta) Diss_Cologni_Roberta.pdf - Published Version Download (27MB)

Abstract

PET is a fundamental imaging method used for in vivo visualization of physiological and pathophysiological processes on the molecular level. This technique relies on radiotracers, which consist of a pharmacophore addressing the biological target of interest labeled with a radionuclide to enable in vivo detection. For the design of radiotracers to target the mutant isocitrate dehydrogenase 1 enzyme (mIDH1), known scaffolds based on structures with inhibitory activity for this mutated enzyme were employed. Such mutations are prevalent in diffuse gliomas (WHO grade 2-4) setting them apart from glioblastoma (WHO grade 4). Consequently, these mutations have become crucial for differential diagnosis and prediction of patient prognosis. Unlike existing methods that require surgical sampling for detection of mIDH1, positron emission tomography (PET) with mIDH- targeting probes could offer a non-invasive alternative for determining and longitudinally assessing the mIDH status. This could lead to an early and accurate diagnosis of low-grade gliomas and facilitate to differentiate glioma recurrence from post radiation treatment effects. Furthermore, tumor heterogeneity can be determined. Thus, the application of these probes opens up new opportunities for monitoring of tumor growth and thus patient care since neither MRI nor the PET tracer [18F]fluoroethyltyrosine have these capabilities. In the present study, specific inhibitors of mIDH1 were selected as lead structures to prepare 18F- labeled mIDH1 selective PET tracers using copper mediated radiofluorination. To this end, olutasidenib, presented in 2020, was selected as a starting scaffold. However, a first attempt to label it in position-7 failed. Therefore, a sub potent version of olutasidenib was selected as a mIDH1 selective candidate to use this compound as a proof of concept. In this case a 6-fluoro substituent was introduced into the quinolone ring system instead of the 6-chloro substituent present on the left-hand side of the original structure. The reference compound 4-(((6-fluoro-2-oxo-1,2-dihydroquinolin-3- yl)methyl)amino)-2-methoxybenzonitrile was obtained in an overall yield of 13% over 5 steps starting from 4-fluoroacetanilide and tested in enzymatic assays to show an inhibitory potency of 616 nM. Two radiofluorination precursors were obtained via multi-step syntheses which consisted in the conversion of 4-bromoaniline into 6-bromo-3-formylquinolone, followed by pivaloyloxymethyl (POM) protection of the amide moiety prior to conjugation to 4-amino-2-methoxybenzonitrile. For 18F-labeling by copper-mediated radiofluorination, the brominated lead structures had to be converted into boronic acid pinacol esters, to furnish the precursors 3-(((4-cyano-3- methoxyphenyl)amino)methyl)-2-oxo-6-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)quinolin- 1(2H)-yl)methyl pivalate and ((3-(((4-cyano-3-methoxyphenyl)amino)methyl)-6-(4,4,5,5- tetramethyl-1,3,2-dioxaborolan-2-yl)quinolin-2-yl)oxy)methyl pivalate in a total yield of 10% over 10 synthetic steps. For radiofluorination, [18F]fluoride was fixed on an anion exchange cartridge (AEC), eluted with 1 mg of tetraethylammonium bicarbonate (Et4NHCO3) in 1 mL methanol and dried. An equimolar solution of precursor and Cu(4-PhePy)4(ClO4)2 (10 μmol) in 500 μL dimethylacetamide (DMA) was then added to the reactor and the mixture was stirred at 110 °C for 10 min. Subsequent deprotection with 200 μL 0.25 M NaOH in H2O/MeOH (1:1) at 80°C for 3 min yielded the desired product with a radiochemical conversion of 26 ± 12% for the N-protected and 41 ± 20% for the O-protected precursor. After HPLC-purification, 4-(((6-[18F]fluoro-2-oxo-1,2- dihydroquinolin-3-yl)methyl)amino)-2-methoxybenzonitrile ([18F]mIDH-138) was obtained in radiochemical yields of 22% for the N-protected and 29% for the O-protected precursor, respectively. The molar activity amounted to 13-540 GBq/μmol (start activities in the range of 1-4 GBq). The cellular uptake of [18F]mIDH-138 was evaluated in mIDH1 and wtIDH1 cells and revealed a statistically significant higher uptake in mIDH1 cells. The in vitro results were confirmed by in vivo experiments in a chorio allantoic membrane (CAM) model and showed negligible defluorination and high stability of the tracer over 1 hour. After the successful proof of concept, a radiotracer based on olutasidenib (6-F-(S)-olutasidenib) was pursued. Moreover, the R- and S-enantiomers (6-F-olutasidenib) were prepared to confirm the importance of the spatial orientation of the methyl group on the benzylic linker. The two reference compounds were obtained starting from 6-fluoro-2-chloro-3-formylquinoline and conjugated with (R)-Ellman’s sulfinamide to direct the methylation step in order to obtain the desired S- diastereoisomer as the major product. The different diastereoisomers were isolated and conjugated to 5-fluoro-1-methyl-6-oxo-1,6-dihydropyridine-2-carbonitrile, which itself was obtained via a 4 step synthesis from 5-fluoropicolinonitrile in a total yield of 13%. Enzymatic assay of 6-F-(S)-olutasidenib and 6-F-(R)-olutasidenib confirmed that the former is a more potent inhibitor with respect to the latter. The boronic acid ester precursors for the radiosynthesis of 6-[18F]fluoro-(S)-olutasidenib and 6- [18F]fluoro-(R)-olutasidenib were obtained in a similar way starting from the brominated quinolone which was POM protected after the conjugation with the N-methylpyridone and converted into boronic acid pinacol esters in a total yield of 2% and 1.5% respectively over 14 steps. Cu-mediated radiosynthesis was carried out following a similar procedure to that used for the first radiotracer and optimized (DMI, 100°C, 15 min), followed by basic cleavage of the POM group. The radiofluorination proved successful only for the O-protected precursor. No explanation could be found for the incompatibility of the N-protected precursor with radiofluorination. However, 6- [18F]fluoro-(S)-olutasidenib and 6-[18F]fluoro-(R)-olutasidenib were obtained with RCCs of 61 ± 13%, isolated radiochemical yields of 50 ± 10% and molar activities of 102-275 GBq/μmol (for starting activity of 1 GBq) and were further evaluated in cell uptake experiments. However, unexpectedly both tracers showed negligible tracer uptake. Finally, encouraged by the results obtained with [18F]mIDH-138, a fourth radiotracer, 4-(((6-[18F]fluoro-2-oxo-1,2-dihydroquinolin-3- yl)methyl)amino)-2-(2-fluoroethoxy)benzonitrile ([18F]FE-mIDH), was designed. Here the left-hand side of the molecule consisted of a POM protected 6-fluoroquinolone, identical to that of the first radiotracer. The arene on the right-hand side already present in the first tracer was modified by replacing the methoxy group with a 2-fluoroethoxy substituent, obtained in 3 synthetic steps starting from 2-hydroxy-4-nitrobenzonitrile with a total yield of 70%. The reference compound was tested for its potency towards mIDH1 and revealed an IC50 value in the micromolar range, despite the minor structural difference compared to the first compound. Again, due to the impossibility to debenzylate the N-protected scaffold without inducing its degradation, only the O-protected precursor could be obtained with a total yield of 10% over 8 steps. For the radiosynthesis of [18F]FE-mIDH, a mesylate leaving group was installed and exchanged against fluoride-18 by an SN2 reaction by aliphatic radiofluorination. Fluoride-18 was trapped on an AEC cartridge and eluted with 0.5 mg of tetramethylammonium triflate (Me4NOTf) in 0.5 mL of methanol and dried. A solution of the precursor (2 mg) in 500 μL acetonitrile was then added to the reactor and stirred at 85°C for 15 minutes followed by deprotection with 100 μL 0.1 M NaOH in ethanol at 35°C for 10 minutes. The radiotracer [18F]FE-mIDH was obtained with RCCs of 30 ± 8%, RCYs of 26 ± 8% and Am of 14-47 GBq/μmol (start activities of 1 GBq) and further incubated in wild type and mutant cell lines. An unexpected higher uptake in wild type cells in comparison to mutant cells was observed. Recent literature demonstrated that the mechanism of selective inhibition of mIDH1 inhibitors disclosed so far is not based on a selective binding but is actually correlated to the different binding strength of the natural ligands into the active binding pocket and the dynamic nature of the dimeric interface. Conclusively, this project revealed that the radiolabeled inhibitors are not suitable for selective mIDH1 tumor imaging. However, selective targeting of mIDH to determine its status remains a desirable goal in nuclear medicine. To achieve this aim a paradigm shift in tracer development is required, and at the current moment, it is still not foreseeable what the next steps in this direction will look like.

Item Type:	Thesis (PhD thesis)
Creators:	Creators Email ORCID ORCID Put Code Cologni, Roberta cologniroberta@gmail.com UNSPECIFIED UNSPECIFIED
URN:	urn:nbn:de:hbz:38-717914
Date:	4 December 2023
Language:	English
Faculty:	Faculty of Mathematics and Natural Sciences
Divisions:	Ehemalige Fakultäten, Institute, Seminare > Faculty of Mathematics and Natural Sciences > Institute of Nuclear Chemistry
Subjects:	Chemistry and allied sciences
Uncontrolled Keywords:	Keywords Language mIDH English radiotracer English F-18 English
Date of oral exam:	20 October 2023
Referee:	Name Academic Title Neumaier, Bernd professor, PhD
Refereed:	Yes
URI:	http://kups.ub.uni-koeln.de/id/eprint/71791