PET imaging of multidrug resistance in tumors using 18F-fluoropaclitaxel.

The failure of solid tumors to respond to chemotherapy is a complicated and clinically frustrating issue. The ability to predict which tumors will respond to treatment could reduce the human and monetary costs of cancer therapy by allowing pro-active selection of a chemotherapeutic to which the tumor does not express resistance. PET/CT imaging with a radiolabeled form of paclitaxel, F-18 fluoropaclitaxel (FPAC), may be able to predict the uptake of paclitaxel in solid tumors, and as a substrate of P-glycoprotein, it may also predict which tumors exhibit multidrug resistance (MDR), a phenotype in which tumors fail to respond to a wide variety of chemically unrelated chemotherapeutic agents. This article reviews the synthetic, preclinical and early human data obtained during the development phase of this promising new radiopharmaceutical.

Animal model of metastatic pheochromocytoma: evaluation by MRI and PET.

OBJECTIVE: The development of metastatic pheochromocytoma animal model provides a unique opportunity to study the physiology of these rare tumors and to evaluate experimental treatments. Here, we describe the use of small animal imaging techniques to detect, localize and characterize metastatic lesions in nude mice. METHODS: Small animal positron emission tomography (PET) imaging and magnetic resonance imaging (MRI) were used to detect metastatic lesions in nude mice following intravenous injection of mouse pheochromocytoma cells. [18F]-6-fluoro-dopamine ([18F]-DA) and [18F]-L-6-fluoro-3,4-dihydroxyphenylalanine, which are commonly used for localization of pheochromocytoma lesions in clinical practice, were selected as radiotracers to monitor metastatic lesions by PET. RESULTS: MRI was able to detect liver lesions as small as 0.5mm in diameter. Small animal PET imaging using [18F]-DA and [18F]-DOPA detected liver, adrenal gland, and ovarian lesions. CONCLUSION: We conclude that MRI is a valuable technique for tumor growth monitoring from very early to late stages of tumor progression and that animal PET confirmed localization of metastatic pheochromocytoma in liver with both radiotracers.

Why does the agonist [(18)F]FP-TZTP bind preferentially to the M(2) muscarinic receptor?

PURPOSE: Preferential binding of FP-TZTP at the M(2) receptor in vivo led to investigation of [(18)F]FP-TZTP as a potential PET tracer for Alzheimer's disease, in which a substantial reduction of M(2) receptors has been observed in autopsy studies. We hereby investigated in vitro the FP-TZTP behavior to further elucidate the properties of FP-TZTP that lead to its M(2) selectivity. METHODS: Chinese hamster ovarian cells expressing the five subtypes of human muscarinic receptor as well as the wild type were harvested in culture to assess equilibrium binding. Specific binding was calculated by subtraction of non-specific binding from total binding. Internal specific binding was calculated by subtraction of external specific binding from the total specific binding. Saturation assays were also performed to calculate B(max), K(i), and IC(50). In addition, equilibrium binding and dissociation kinetic studies were performed on rat brain tissue. Selected regions of interest were drawn on the digital autoradiograms and [(18)F]FP-TZTP off-rates were determined by measurement of the rate of release into a buffer solution of [(18)F]FP-TZTP from slide-bound cells that had been preincubated with [(18)F]FP-TZTP. RESULTS: At equilibrium in vitro, M(2) subtype selectivity of [(18)F]FP-TZTP was not evident. We demonstrated that ATP-dependent mechanisms are not responsible for FP-TZTP M(2) selectivity. In vitro off-rate studies from rat brain tissue showed that the off-rate of FP-TZTP varied with the percentage of M(2) subtype in the tissue region. CONCLUSION: The slower dissociation kinetics of FP-TZTP from M(2) receptors compared with the four other muscarinic receptor subtypes may be a factor in its M(2) selectivity.

Regional brain uptake of the muscarinic ligand, [18F]FP-TZTP, is greatly decreased in M2 receptor knockout mice but not in M1, M3 and M4 receptor knockout mice.

A muscarinic receptor radioligand, 3-(3-(3-fluoropropyl)thio) -1,2,5,thiadiazol-4-yl)-1,2,5,6-tetrahydro-1-methylpyridine (fP-TZTP) radiolabeled with the positron emitting radionuclide (18)F ([(18)F]FP-TZTP) displayed regional brain distribution consistent with M2 receptor densities in rat brain. The purpose of the present study is to further elucidate the subtype selectivity of [(18)F]FP-TZTP using genetically engineered mice which lacked functional M1, M2, M3, or M4 muscarinic receptors. Using ex vivo autoradiography, the regional brain localization of [(18)F]FP-TZTP in M2 knockout (M2 KO) was significantly decreased (51.3 to 61.4%; P<0.01) when compared to the wild-type (WT) mice in amygdala, brain stem, caudate putamen, cerebellum, cortex, hippocampus, hypothalamus, superior colliculus, and thalamus. In similar studies with M1KO, M3KO and M4KO compared to their WT mice, [(18)F]FP-TZTP uptakes in the same brain regions were not significantly decreased at P<0.01. However, in amygdala and hippocampus small decreases of 19.5% and 22.7%, respectively, were observed for M1KO vs WT mice at P<0.05. Given the fact that large decreases in [(18)F]FP-TZTP brain uptakes were seen only in M2 KO vs. WT mice, we conclude that [(18)F]FP-TZTP preferentially labels M2 receptors in vivo.

Liquid chromatography-tandem mass spectrometry identification of metabolites of two 5-HT1A antagonists, N-[2-[4-(2-methoxylphenyl)piperazino]ethyl]-N-(2-pyridyl) trans- and cis-4-fluorocyclohexanecarboxamide, produced by human and rat hepatocytes.

Two 5-HT1A antagonists, t-FCWAY and c-FCWAY, were developed as imaging agents for positron emission tomography (PET). In order to evaluate these compounds, hepatocytes from both human and rat were utilized to produce metabolites and LC-MS-MS was used to identify metabolites. These in vitro metabolism studies indicate that hydrolysis of the amide linkage is the major metabolism pathway for humans, whereas aromatic ring-oxidation is the major metabolism pathway for rat. The rat hepatocyte results correlate well with in vivo rat metabolism studies. Based on the structures of the metabolites, we have developed an extraction procedure to determine the concentration of the parent compound in plasma.

Biologically stable [(18)F]-labeled benzylfluoride derivatives.

Use of the [(18)F]-fluoromethyl phenyl group is an attractive alternative to direct fluorination of phenyl groups because the fluorination of the methyl group takes place under milder reaction conditions. However, we have found that 4-FMeBWAY showed femur uptake equal to that of fluoride up to 30 min in rat whereas 4-FMeQNB had a significantly lower percent injected dose per gram in femur up to 120 min. For these and other benzylfluoride derivatives, there was no clear in vivo structure-defluorination relationship. Because benzylchlorides (BzCls) are known alkylating agents, benzylfluorides may be alkylating agents as well, which may be the mechanism of defluorination. On this basis, the effects of substitution on chemical stability were evaluated by the 4-(4-nitro-benzyl)-pyridine (NBP) test, which is used to estimate alkylating activity with NBP. The effect of substitution on the alkylating activity was evaluated for nine BzCl derivatives: BzCl; 3- or 4-methoxy (electron donation) substituted BzCl; 2-, 3-, or 4-nitro (electron withdrawing) substituted BzCl; and 2-, 3-, or 4-chloro (electron withdrawing) substituted BzCl. Taken together, the alkylating reactivity of 3-chloro-BzCl was the weakest. This result was then applied to [(18)F]-benzylfluoride derivatives and in vivo and in vitro stability were evaluated. Consequently, 3-chloro-[(18)F]-benzylfluoride showed a 70-80% decrease of defluorination in both experiments in comparison with [(18)F]-benzylfluoride, as expected. Moreover, a good linear relationship between in vivo femur uptake and in vitro hepatocyte metabolism was observed with seven (18)F-labeled radiopharmaceuticals, which were benzylfluorides, alkylfluorides, and arylfluorides. Apparently, the [(18)F]-fluoride ion is released by metabolism in the liver in vivo. In conclusion, 3-chloro substituted BzCls are the most stable, which suggests that 3-chloro benzylfluorides will be the most chemically stable compound. This result should be important in future design of radioligands labeled with a benzylfluoride moiety.

Using single photon emission tomography (SPECT) and positron emission tomography (PET) to trace the distribution of muscarinic acetylcholine receptor (MACHR) binding radioligands.

Two [18F] labeled ligands for the mAChR were prepared and evaluated in rodents and nonhuman primates. The properties of both compounds, one an agonist and the other an antagonist, were consistent with M2 subtype specificity.

In vivo muscarinic binding of 3-(alkylthio)-3-thiadiazolyl tetrahydropyridines.

Based on encouraging in vitro data indicating M2 subtype selectivity, we synthesized, radiolabeled with 18F, and evaluated 3-(3-(2-fluoroethylthio)-1,2,5-thiadiazol-4-yl)-1,2,5,6-tetr ahydro-1-methylpyridine [FE-TZTP], and 3-(3-(3-fluoropropylthio)-1,2,5-thiadiazol-4-yl)-1,2,5,6-tet rahydro-1-methylpyridine [FP-TZTP] for muscarinic subtype selectivity in vivo. [18F]FE-TZTP displays high uptake in vivo but is inhibited only weakly by coinjecting unlabeled P-TZTP. Contrarily, [18F]FP-TZTP shows significant inhibition of uptake by coinjecting unlabeled P-TZTP or the muscarinic agonist L-687,306 (3-(3-cyclopropyl-1,2,4-oxadiazol-5-yl)-1-azabicyclo[2.2.1]heptane ). Using in vivo autoradiography, [18F]FP-TZTP displays regional distribution consistent with M2 subtype distribution. In addition, [18F]FP-TZTP shows specific uptake in the heart at 5 min. Analysis of metabolites in the awake rat brain revealed that the parent compound represents >95% of the extractable activity at 30 min. In vivo studies in rhesus monkeys revealed rapid brain uptake of [18F]FP-TZTP, with clearance sustained over 2 h. Administration of P-TZTP or FP-TZTP (80 nmol/kg) at 60 min after injection of [18F]FP-TZTP results in a significant displacement of brain activity in all regions. Metabolite analysis in monkey plasma shows that parent compound represents 20% of the extractable radioactivity at 40 min postinjection. One metabolite, which increases with time, has similar lipophilicity to the parent. However, based on metabolism in rat we believe metabolites are not in the brain to any significant extent in monkeys during the time of imaging studies. Regional uptake, autoradiographic distribution, and clearance rates in the brain are consistent with the hypothesis that [18F]FP-TZTP is M2 selective in vivo.

Muscarinic cholinergic receptor measurements with [18F]FP-TZTP: control and competition studies.

[18F]Fluoropropyl-TZTP (FP-TZTP) is a subtype-selective muscarinic cholinergic ligand with potential suitability for studying Alzheimer's disease. Positron emission tomography studies in isofluorane-anesthetized rhesus monkeys were performed to assess the in vivo behavior of this radiotracer. First, control studies (n = 11) were performed to characterize the tracer kinetics and to choose an appropriate model using a metabolite-corrected arterial input function. Second, preblocking studies (n = 4) with unlabeled FP-TZTP were used to measure nonspecific binding. Third, the sensitivity of [18F]FP-TZTP binding to changes in brain acetylcholine (ACh) was assessed by administering physostigmine, an acetylcholinesterase (AChE) inhibitor, by intravenous infusion (100 to 200 microg x kg(-1) x h(-1)) beginning 30 minutes before tracer injection (n = 7). Tracer uptake in the brain was rapid with K1 values of 0.4 to 0.6 mL x min(-1) x mL(-1) in gray matter. A model with one tissue compartment was chosen because reliable parameter estimates could not be obtained with a more complex model. Volume of distribution (V) values, determined from functional images created by pixel-by-pixel fitting, were very similar in cortical regions, basal ganglia, and thalamus, but significantly lower (P < 0.01) in the cerebellum, consistent with the distribution of M2 cholinergic receptors. Preblocking studies with unlabeled FP-TZTP reduced V by 60% to 70% in cortical and subcortical regions. Physostigmine produced a 35% reduction in cortical specific binding (P < 0.05), consistent with increased ACh competition. The reduction in basal ganglia (12%) was significantly smaller (P < 0.05), consistent with its markedly higher AChE activity. These studies indicate that [18F]FP-TZTP should be useful for the in vivo measurement of muscarinic receptors with positron emission tomography.

In vivo muscarinic binding selectivity of (R,S)- and (R,R)-[18F]-fluoromethyl QNB.

We have developed a multistep radiochemical synthesis of two diastereomers of quinuclidinyl-4-[18F]-fluoromethyl-benzilate ([18F]-FMeQNB), a high-affinity ligand for muscarinic acetylcholine receptors. Previously, we have shown that the nonradioactive (R,R)-diastereomer displays an eightfold selectivity for M1 over M2 while the nonradioactive (R,S)-diastereomer displays a sevenfold selectivity for M2 over M1 in vitro. This paper reports the results of in vivo comparison studies. In the rat, uptake of (R,S)-[18F]-FMeQNB was nearly uniform in all brain regions following the concentration of M2 subtype. The uptake was reduced by 36-54% in all brain regions on coinjection with 50 nmol of unlabeled ligand. An injection of (R,S)-[18F]-FMeQNB followed at 60 min by injection of unlabeled ligand and subsequent sacrifice at 120 min displaced 30-50% of radioactivity in the pons, medulla, and cerebellum, which contain a high proportion of M2 subtype. The most dramatic displacement and inhibition of uptake on coinjection of (R,S)-[18F]-FMeQNB was observed in the heart. In rhesus monkey, the compound showed prolonged uptake and retention in the brain. In the blood, the parent compound degraded rapidly to a single radiolabeled polar metabolite believed to be fluoride. Within 30 min the parent compound represented less than 5% of the plasma activity. Displacement with (R)-QNB was generally slow, but was more rapid from those tissues which contain a higher proportion of M2 subtype. The results are consistent with the hypothesis that (R,S)-[18F]-FMeQNB is M2 selective in vivo. On the other hand, (R,R)-[18F]-FMeQNB showed higher uptake in those brain regions containing a higher concentration of M1 subtype. Uptake in the heart at 60 min was much lower than that observed with the (R,S)-diastereomer. Inhibition of uptake on coinjection with unlabeled (R,S)-FMeQNB is only significant in the heart, thalamus, and pons. Inhibition of uptake on coinjection with unlabeled (R,R)-FMeQNB is quite uniform in all brain regions. Displacement with (R)-QNB shows a more varying amount displaced. These results are consistent with (R,R)-[18F]-FMeQNB being M1 selective in vivo.

A review of new oncotropic tracers for PET imaging.

We have developed three biochemical probes to determine if they are sensitive probes of early biochemical change in a tumor. All three probes appear to have the appropriate properties for in vivo imaging, but must now be evaluated as probes for the sensitive detection of changes in early malignant disease.

Evaluation of stereoisomers of 4-fluoroalkyl analogues of 3-quinuclidinyl benzilate in in vivo competition studies for the M1, M2, and M3 muscarinic receptor subtypes in brain.

To develop a subtype selective muscarinic acetylcholine receptor (mAChR) antagonist for PET, fluorine-19 labeled alkyl analogues of quinuclidinyl benzilate (QNB) were synthesized by stereoselective reactions. To investigate these analogues for tissue subtype specificity, in vivo competitive binding studies were performed in rat brain using (R)-3-quinuclidinyl (R)-4-[125I]iodobenzilate (IQNB). Five, fifty, or five-hundred nmol of the non-radioactive ligands were coinjected intravenously with 8 pmol of the radioligand, Cold (R,R)-IQNB blocked (R,R)-[125I]IQNB in a dose-dependent manner, without showing regional specificity. For the (R,S)-fluoromethyl, -fluoroethyl and -fluoropropyl derivatives, a higher percent blockade was seen at 5 and 50 mmol levels in M2 predominant tissues (medulla, pons, and cerebellum) than in M1 predominant tissues (cortex, striatum and hippocampus). The blockade pattern of the radioligand also correlated qualitatively with the percentage of M2 receptors in the region. The S-quinuclidinyl analogues showed M2 selectivity but less efficient blockade of the radioligand, indicating lower affinities. Radioligand bound to the medulla was inversely correlated to the M2 relative binding affinity of the fluoroalkyl analogues. These results indicate that the nonradioactive ligand blocks the radioligand based on the affinity of the nonradioactive ligand for a particular receptor subtype compared to the affinity of the radioligand for the same receptor subtype. Of the seven compounds evaluated, (R,S)-fluoromethyl-QNB appears to show the most selectivity for the M2 subtypes in competition studies in vivo.

Syntheses and biological properties of chiral fluoroalkyl quinuclidinyl benzilates.

Previously, (R)-quinuclidinyl (R)-4-iodobenzilate ((R,R)-IQNB), a muscarinic receptor antagonist, has been labeled with 123I and 125I for use in in vitro and in vivo studies in animals and humans. We have prepared fluoroalkyl analogs of QNB, which are amenable to labeling with 18F, for potential imaging applications with positron emission tomography. The enantiomers of (fluoroalkyl)benzilic acids were prepared via an enantioselective Grignard addition reaction. Subsequent coupling of the enantiomeric (fluoroalkyl)benzilic acid with a selected enantiomer of quinuclidinol provides fluorinated analogs of QNB with known stereochemistry at each of the stereogenic centers. These compounds exhibit different affinities for the muscarinic receptor tissue subtypes in vitro. (R,R)-4-(Fluoromethyl)-QNB, and (R,R)-IQNB, and (R,R)-4-(fluoroethyl)-QNB exhibit selectivity for the M1 subtype, and (R,S)-4-(fluoromethyl)-QNB exhibits selectivity for the M2 subtype.

Synthesis of polymer-bound 6-thiolatomercury and 6-mercuric sulfonate DOPA precursors and their halodemercuration reactivity.

Fluorodemercuration has the greatest utility for the preparation of 6-[18F]DOPA, but requires separation from unreacted mercury precursor and other mercury-containing compounds. One approach is the development of a polymer-bound mercury precursor. In this study, polymer-bound 6-thiolatomercury and 6-mercuric sulfonate DOPA derivatives, and its monomeric analogs were synthesized. Fluorodemercuration of monomeric analog of mercuric sulfonate gave half the yield (14-15%) while iododemercuration gave the same yield (38%) compared with a 6-mercuric trifluoroacetate protected DOPA. The mercuric sulfonate undergoes halodemercuration, so polymer-bound halodemercuration precursors may be useful as precursors of 6-[18F]DOPA.

A single column, rapid quality control procedure for 6-[18F]fluoro-L-dopa and 6-[18F]fluorodopamine PET imaging agents.

6-[18F]Fluoro-L-dopa and 6-[18F]fluorodopamine are promising PET imaging agents for visualizing cerebral dopaminergic centers and cardiac sympathetic innervation and function. Administration to humans requires a means to determine the purity before injection. We describe such a method using HPLC with u.v. and radioactivity detection and a single high-speed C-18 column with gradient elution. The procedure can resolve within 10 min these fluorinated catechols, their isomers, and dihydroxyphenylalanine. The chemical and radiochemical purity, and specific activity, can be determined before injection.

11 beta-methoxy-, 11 beta-ethyl- and 17 alpha-ethynyl-substituted 16 alpha-fluoroestradiols: receptor-based imaging agents with enhanced uptake efficiency and selectivity.

We have prepared three analogues of 16 alpha-fluoroestradiol (FES) substituted either with an 11 beta-methoxy group (1, 11 beta-MeO-FES), an 11 beta-ethyl group (2, 11 beta-Et-FES), or a 17 alpha-ethynyl group (3, 17 alpha-ethynyl-FES). These substituents all lower the binding of FES to the serum proteins alphafetoprotein and sex steroid binding protein, but their effect on estrogen receptor binding varies: Receptor binding is increased by the 11 beta-ethyl and 17 alpha-ethynyl groups, but decreased by the 11 beta-methoxy group. These substituents also have a parallel effect on the lipophilicity, and hence the nonspecific binding estimated for these compounds. All three compounds were prepared in fluorine-18 labeled form, at effective specific activities of 90-1600 Ci/mmol, by fluoride ion displacement reactions as done previously with FES. Tissue distribution studies in immature rats show high uptake selectivity by target tissue (uterus) and effective competition by an excess of unlabeled estradiol. Percent injected dose per gram values (% ID/g) at 1 h are 6% for 11 beta-MeO-FES and 11-13% for 11 beta-Et-FES and 17 alpha-ethynyl-FES (FES itself has a % ID/g of 9%). Uptake selectivity in terms of uterus to blood or muscle ratios at 1 h is highest for 11 beta-MeO-FES and 17 alpha-ethynyl-FES (43-149). Metabolic consumption studies show that most activity in uterus is unmetabolized and in blood is rapidly and nearly completely metabolized. In muscle, FES and the substituted estrogens show intermediate levels of metabolic consumption; in some cases activity in muscle extracts is nearly unmetabolized. Thus, the substituents on FES cause major alterations in receptor and nonreceptor binding affinity, uptake efficiency and selectivity, and extent of metabolism. It is not readily clear, however, whether the alterations in uptake efficiency and selectivity are the result of differences in receptor or nonreceptor binding or lipophilicity, or altered patterns of metabolism. Nevertheless, these compounds should be useful in providing a spectrum of uptake properties that could be used for imaging different estrogen-receptor-containing structures.

Preparation and biological evaluation of 18F-labeled benzamide analogs as potential dopamine D2 receptor ligands.

Three 18F-labeled benzamide derivatives were prepared and evaluated as potential ligands to study the dopamine D2 receptor phenomenon. The compounds are analogs of iodobenzamide, eticlopride and raclopride and are labeled with an N-2-[18F]fluoroethyl functionality on the pyrrolidine ring. The compounds were tested in vitro for binding affinity and found to exhibit somewhat lower affinity than the non-fluorinated analog. In vivo distribution studies revealed that all compounds were more highly bound to plasma proteins than was raclopride. In addition, compartmentation of radioactivity demonstrated nonspecific binding to be the predominate retention in the brain as reflected by the low caudate to cerebellum ratios for these compounds. These three 18F-labeled benzamide derivatives are inferior to raclopride and iodobenzamide for studies of the D2 receptor system using positron emission tomography.

Synthesis of a fluorinated fatty acid, dl-erythro-9,10-[18F]difluoropalmitic acid, and biodistribution studies in rats.

9,10-Difluoropalmitic acid (DFPA) labeled with the cyclotron produced, positron emitting radionuclide 18F has been synthesized as a potential analogue of 9,10-[3H]palmitic acid, a fatty acid which has been used to study lipid metabolism in rat brain and pituitary. [18F]DFPA was prepared by the direct and stereoselective addition of [18F]F2 to the double bond of cis-9,10-palmitoleic acid. The fluorination was carried out in FCCl3 at -70 degrees C using a low concentration of F2 (0.5%) in neon. [18F]DFPA has been obtained in radiochemical yields of 12-16% from end-of-bombardment (EOB) in approx. 2.5 h. Chemical and radiochemical purity exceeded 95%, and specific activities calculated to EOB ranged from 500 to 1000 mCi/mmol. [18F]DFPA crosses the blood-brain barrier and is incorporated into rat brain at about twice the level of that of 9,10-[3H]palmitic acid. The synthesis of [18F]DFPA permits us to study the biological disposition and metabolism of a vicinal-difluoro fatty acid.

Synthesis of 11C-labeled (+-)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imin e [(+-)-[11C]MK801].

The synthesis of C5 labeled (+-)-5-[11C]methyl-10,11-dihydro-5H-dibenzo[a,d]cyclo-hepten- 5,10-imine [(+-)-[11C]MK801] has been accomplished via alkylation of (+-)-10,11-dihydro-5H-dibenzo[a,d]cyclo-hepten-5,10-imine-N-t- butylformamidine [(+-)-5-des-methyl MK801 formamidine). The 11C labeling is accomplished by reaction of the anion of (+-)-5-des-methyl MK801 formamidine, generated with s-butyllithium, and [11C]methyl iodide.