Evidence for the importance of 5'-deoxy-5-fluorouridine catabolism in humans from 19F nuclear magnetic resonance spectrometry.

The use of a new methodology, 19F nuclear magnetic resonance, has allowed detection of all the fluorinated metabolites in the biofluids of patients treated with 5'-deoxy-5-fluorouridine (5'-dFUrd) injected i.v. at a dose of 10 g/m2 over 6 h. This technique, which requires no labeled drug, allows a direct study of the biological sample with no need for extraction or derivatization and a simultaneous identification and quantitation of all the different fluorinated metabolites. As well as the already known metabolites, unmetabolized 5'-dFUrd, 5-fluorouracil, and 5,6-dihydro-5-fluorouracil, the presence of alpha-fluoro-beta-ureidopropionic acid, alpha-fluoro-beta-alanine (FBAL), N-carboxy-alpha-fluoro-beta-alanine, and the fluoride anion F- is reported. The catabolic pathway proposed for 5'-dFUrd is analogous to that of 5-fluorouracil, completed with FBAL----F- step, and the plasmatic equilibrium of FBAL with N-carboxy-alpha-fluoro-beta-alanine, its N-carboxy derivative. The quantitative analysis of the different metabolites found in plasma and urine emphasizes the significance of the catabolic pathway. High concentrations of alpha-fluoro-beta ureidopropionic acid and FBAL are recovered in plasma from 3 h after the beginning of the perfusion to 1 h after its end. The global urinary excretion results show that there is a high excretion of 5'-dFUrd and metabolites. Unchanged 5'-dFUrd and FBAL are by far the major excretory products and are at nearly equal rates. The protocol followed in this study produces relatively low but persistent plasmatic concentrations of 5-fluorouracil throughout the perfusion.

Cardiotoxicity of high-dose continuous infusion fluorouracil: a prospective clinical study.

PURPOSE: A prospective clinical study was performed to determine the incidence of high-dose continuous intravenous infusion fluorouracil (5FU-CIV) cardiotoxicity. PATIENTS AND METHODS: Three hundred sixty-seven patients who were given first-cycle high-dose 5FU-CIV were monitored for cardiac function by clinical examination, ECG, and laboratory tests. 5FU-CIV was administered during a 96- or 120-hour period at doses that ranged from 600 to 1,000 mg/m2/d. Associated drugs included cisplatin (56%), mitomycin (12.5%), folinic acid (leucovorin) (7%), and others (14%). Thirty-nine patients (10.5%) received 5FU as a single agent. RESULTS: 5FU-induced cardiac events occurred in 28 patients (7.6%; 95% confidence interval, 4.9% to 10.3%). Nine of them had a history of cardiac disease. Primary tumors included head and neck (n = 13), gastrointestinal (n = 6), breast (n = 3), and others (n = 6). The mean onset time of cardiac symptoms was 3 days (range, 2 to 5). Inaugural symptoms included angina pectoris (n = 18), hypotension (n = 6), hypertension (n = 5), malaise (n = 4), dyspnea (n = 2), arrhythmia (n = 1), or sudden death (n = 1). At 5FU discontinuation, six patients' cardiac symptoms returned to baseline, but 21 patients experienced unstable angina (n = 8), hypotension/cardiovascular collapse (n = 11), pulmonary edema (n = 1), or sudden death (n = 4). The lethality rate was 2.2% (five sudden deaths plus three irreversible collapses). ECG showed repolarization changes (ST segment deviation; T-wave inversion) in 65% and/or diffuse microvoltage in 22% of the patients who presented with cardiac events. Echocardiography showed partial or global hypokinesia in nine of the 16 patients who were examined, and one case of prolonged akinesia. Cardiac enzymes rarely showed an increase (n = 2). In severe but reversible cases, clinical, ECG, and echographic parameters returned to baseline status within 48 hours after the drug discontinuation. A fluorine 19 nuclear magnetic resonance (19F NMR) analysis of urine was performed on 14 patients; six had cardiac symptoms and eight did not. Fluoroacetate (FAC), a known cardiotoxic compound, was detected in all cases. CONCLUSION: In our study, the incidence of high-dose 5FU-CVI cardiotoxicity was 7.6%. The hypothesis of a toxic cardiomyopathic process requires further confirmation.

Magnetic resonance spectroscopy: a powerful tool for drug metabolism studies.

Studies on the metabolism and disposition of drugs using nuclear magnetic resonance spectroscopy (MRS) as the analytical technique are reviewed. An overview of the main studies classed in terms of the observed magnetic nucleus (1H, 2H, 7Li, 13C, 19F, 31P, 77Se) is followed by some typical examples of the way in which 19F and 31P MRS can be profitably employed to gain more understanding about the metabolism and disposition of the anticancer fluoropyrimidines (5-fluorouracil (FU) and its prodrugs) and ifosfamide (IF). The results of three recent studies carried out in our laboratory are developed. They concern the direct quantitative monitoring of the hepatic metabolism of FU in the isolated perfused mouse liver, the elucidation of the origin of the cardiotoxicity of FU and the metabolism of IF from an analysis of biofluids of patients. Finally, the advantages and limitations of MRS for investigations on drug metabolism are discussed.

Chemical and biological evaluation of hydrolysis products of cyclophosphamide.

31P NMR spectroscopy was used to study the products of the decomposition of cyclophosphamide (1) in buffered solutions at pH's ranging between 1.2 and 8.6 at 20 degrees C and at pH 7.4 at 37 degrees C. At pH 1.2, 1 undergoes a rapid breakdown (t1/2 = 1.4 days) of the two P-N bonds, giving compounds 2 [HN(CH2CH2Cl)2] and 3 [H2N(CH2)3OP(O)(OH)2] as hydrochlorides. No intermediates were detected. At pH's between 5.4 and 8.6, hydrolysis of 1 during 17 days leads to the sole and previously unknown nine-membered ring compound 13. 13 results from the intramolecular alkylation of 1 giving the bicyclic compound 7 followed by the exothermal hydrolytic breakdown of the P-N bond of its six-membered ring. At pH 2.2 and 3.4, the two hydrolytic pathways coexist since, beside compounds 2 and 3, the hydrochloride of compound 9 [Cl(CH2)2NH(CH2)2NH(CH2)3OP(O)(OH)2] is formed, resulting from the acid-catalyzed breakdown of the P-N bond in the nine-membered ring compound 13. At pH 2.2, the presence of chloride ion affected neither the stability of 1 nor the contribution of the two competing hydrolytic pathways. At pH's ranging from 3.4 to 8.6, there is little degradation of 1 since more than 95% of initial 1 was still present after 7 days at 20 degrees C. Under physiological conditions (pH 7.4, 37 degrees C) after 6 days, 45% of 1 is hydrolyzed (t1/2 = 6.6 days), leading essentially (30% of initial 1) to the nine-membered ring compound 13. The rate of hydrolysis of 13 and the nature of its hydrolysis products were found to depend on pH over the range 0-8.6. After a single ip injection to mice, compounds 3, 9, and 13 were less toxic than 1. They did not exhibit any direct cytotoxic efficacy on the colony-forming capacity of L1210 cells in vitro, and they had no antitumor activity in vivo against P388 leukemia.

Noninvasive fluorine-19 NMR study of fluoropyrimidine metabolism in cell cultures of human pancreatic and colon adenocarcinoma.

Fluorine-19 NMR spectrometry was used to monitor the metabolism of two antineoplastic fluoropyrimidines, 5-fluorouracil (5FU) and 5'-deoxy-5-fluorouridine (5'dFUrd), in cell cultures of human pancreatic (Capan-1) and colon (HT-29) adenocarcinoma. The preliminary results showed, for the two tumor cell lines treated with 5FU, the presence in nonperfused cells of three signals corresponding to intracellular metabolites: 5FU, F-nucleotides and F-nucleosides. When the cells were perfused only the signals of F-nucleotides and 5FU were present. The F-nucleosides observed during the analysis of the nonperfused cells came from the conversion of F-nucleotides. During the NMR recording of Capan-1 cells at 37 degrees C the first metabolite of the catabolic pathway of 5FU, 5,6-dihydro-5-fluorouracil, occurred. At the beginning of the NMR recording of Capan-1 cells treated with 5'dFUrd, two signals corresponding to F-nucleotides and F-nucleosides (consistent with 5'dFUrd) were observed; during the analysis, a supplementary signal corresponding to 5FU appeared. Even after pretreatment with methotrexate the signal of 5FU incorporated into RNA was not detected. Our experiments, performed in attempts to observe the signal of the ternary complex between thymidylate synthetase (TS), 5-fluoro-2'-deoxyuridine-5'-monophosphate (FdUMP) and 5,10-methylene-tetrahydrofolate (5,10-CH2FH4), allowed detection in some cases of a broad signal, whose chemical shift was similar to that reported in the literature following incubation of TS with FdUMP and 5,10-CH2FH4, but our results were not always reproducible.

Determination of the urinary excretion of ifosfamide and its phosphorated metabolites by phosphorus-31 nuclear magnetic resonance spectroscopy.

Phosphorus-31 nuclear magnetic resonance spectroscopy was used to analyze urine samples obtained from patients treated with ifosfamide (IF). This technique allows the individual assay of all phosphorated metabolites of IF in a single analysis without the need for prior extraction. In addition to the classic IF metabolites 2-dechloroethylifosfamide (2DEC1IF), 3-dechloroethylifosfamide (3DEC1IF), carboxyifosfamide (CARBOXYIF), and isophosphoramide mustard (IPM), several signals corresponding to unknown phosphorated compounds were observed. Four of them were identified: one is alcoifosfamide (ALCOIF), two come from the degradation of 2,3-didechloroethylifosfamide (2,3-DEC1IF), and one results from the decomposition of 2DEC1IF. The total cumulative drug excretion as measured over 24 h in nine patients was 51% of the injected IF dose; 18% of the dose was recovered as unchanged IF. The major urinary metabolites were the dechloroethylated compounds, with 3DEC1IF excretion (11% of the injected dose) always being superior to 2DEC1IF elimination (4% of the injected dose). Degradation compounds of 2DEC1IF and 2,3DEC1IF represented 0.4% of the injected dose. The metabolites of the dechloroethylation pathway always predominated over those of the activation pathway (CARBOXYIF, ALCOIF, and IPM, representing 3%, 0.8%, and 0.2% of the injected dose, respectively). In all, 14% of the injected dose was excreted as unknown phosphorated compounds. The interpatient variation in levels of IF metabolites was obvious and involved all of the metabolites. Renal excretion was not complete at 24 h, since 11% of the injected dose was recovered in the 24- to 48-h urine samples.

New approach to metabolism of 5'-deoxy-5-fluorouridine in humans with fluorine-19 NMR.

The metabolism of 5'-deoxy-5-fluorouridine (5'dFUrd), an antitumor fluoropyrimidine, has been investigated in human biofluids (blood, plasma, urine) using a new method: fluorine-19 NMR spectrometry. This method allows direct study of the biological sample and simultaneous identification of all the fluorinated metabolites. In the blood of a patient treated with 5'dFUrd during a 6-h continuous perfusion, we observed unmetabolized 5'dFUrd, 5-fluorouracil, 5,6-dihydrofluorouracil, and another metabolite which has not previously been reported alpha-fluoro-beta-alanine. The two major metabolites in urine are unmetabolized 5'dFUrd and alpha-fluoro-beta-alanine.

Excretion of doxifluridine catabolites in human bile assessed by 19F NMR spectrometry.

The biliary excretion of doxifluridine (5'dFUR) catabolites was studied in a patient with external bile derivation using 19F NMR spectrometry, alpha-fluoro-beta-alanine (FBAL) and fluoride ion were detected in patient's bile samples but represented only about 10% of the excreted fluorinated metabolites. The major biliary metabolite (congruent to 90%), whose structure is still unknown, is a conjugate of FBAL. The cumulative biliary excretion of 5'dFUR fluorinated metabolites was low and represented 0.8% of the injected dose.

Application of fluorine-19 nuclear magnetic resonance to the determination of plasma-protein binding of 5'-deoxy-5-fluorouridine, a new antineoplastic fluoropyrimidine.

Two distinct fluorine-19 nuclear magnetic resonance ((19)F NMR) signals have been observed in human serum for free and plasma-protein bound 5'-deoxy-5-fluorouridine (5'dFUrd). The binding of this drug was studied directly in serum using (19)F NMR. To evaluate the validity of this method, a parallel study was conducted with equilibrium dialysis as the reference method. Two assay methods were applied after equilibrium dialysis, UV spectrophotometry and (19)F NMR spectrometry, the UV assay being used to validate the (19)F NMR assay. A study of the binding of 5'dFUrd to human serum albumin was also reported. The reliability of (19)F NMR as a technique to measure directly the binding of the drug and as an assay after equilibrium dialysis was demonstrated. The percentage of 5'dFUrd bound to plasma proteins is low and concentration-dependent in the 0.04-3.5 mmol l(-1) range.

Fluosol 43 intravascular persistence in mice measured by 19F nmr.

Quantitative determination of the intravascular persistence of F-tri-n-butylamine (FC 43 as Fluosol 43) in mice was carried out using 19F nmr spectroscopy. The method allows the direct study of whole blood, neither separation nor extraction of the sample being required. Accuracy (reproducibility) is better than +/- 3%, and is comparable to that of the gas chromatographic (gc) method. The sensitivity of detection is less than that of the gc method but is sufficient for this biological study. It was observed that the intravascular elimination of F-tri-n-butylamine follows a non-linear kinetic and becomes faster about 40 h after the injection. This phenomenon may be explained by the size-increase of the FC 43 droplets in the emulsion. Indeed, at about 40 h after injection, the level of Pluronic F-68 in the bloodstream was no longer sufficient to maintain the stability of the FC-43 droplets. They therefore tended to coalesce forming larger droplets that were phagocytosed more rapidly by the histiocytes of the reticuloendothelial system.

[Fluorine-19 nuclear magnetic resonance spectroscopy: a privileged tool for the study of the metabolism and pharmacokinetics of fluoropyrimidines]. / La spectroscopie de résonance magnétique nucléaire du fluor-19: un outil privilégié pour l'étude du métabolisme et de la pharmacocinétique des fluoropyrimidines.

This review reports on the various studies using fluorine-19 nuclear magnetic resonance spectroscopy (19F NMR) to study the metabolism of antineoplastic or antifungal fluoropyrimidines. It is divided into 2 parts: the first examines ex vivo studies, ie, of biofluids or excised tissue samples from patients. In vivo studies, ie where the biotransformation of the drug is followed by non-invasively both in animals and in humans, are described in the second part. For ex vivo studies, 19F NMR can already be considered as complementary to the classical analytical methodologies used for drug metabolism studies. In vivo 19F NMR spectroscopic studies, especially in humans, are still at an early stage of development. Several improvements, both methodological (development of volume-selective localization techniques and quantification methods) and clinical (more rigorous definition of pathologies under study and administered treatments), are a prerequisite for useful clinical application.

[Cardiotoxicity of 5-fluorouracil: a question of formulation]. / Cardiotoxicité du 5-fluorouracile: une question de formulation.

The cardiotoxicity of 5-fluorouracil (FU) was attributed to degradation compounds present in the injected vials, fluoroacetaldehyde (Facet) and fluoromalonaldehydic acid (FMald). These compounds are formed with time in the basic medium necessary to solubilize FU. FU-NaOH vials were much less cardiotoxic than FU-Tris vials on the isolated perfused rabbit heart model since, in FU-Tris vials, Facet and FMald are stored in stable "depot" forms, which are adducts with Tris, whereas, in FU-NaOH vials, they are extensively chemically transformed. Cardiotoxic fluoroacetate (FAC), arising from Facet metabolization, was found in urine of patients, with a ratio FAC/FU catabolites 10-30 fold lower in patients treated with FU-NaOH than in those treated with FU-Tris.

The application of nuclear magnetic resonance spectroscopy to drug metabolism studies.

1. The applications of n.m.r. spectroscopy studies to drug metabolism, both in vitro (biofluids, isolated cells, excised tissue samples, isolated organs) and in vivo (animals, humans) are reviewed. 2. N.m.r. is a relatively insensitive technique, but it has the great advantage of being non-invasive and non-destructive, i.e. it allows a direct study of intact biological samples. 3. The majority of studies examined deal with 19F-n.m.r. spectroscopy mainly because of the favourable n.m.r. characteristics of this nucleus, and the low level of endogenous fluorine which gives no detectable 19F signal. However, the potential utility of 1H-31P- and 13C-n.m.r. is also emphasized.

F Nuclear Magnetic Resonance Analysis of 5-Fluorouracil Metabolism in Four Differently Pigmented Strains of Nectria haematococca.

F nuclear magnetic resonance spectroscopy was used to study the metabolism of 5-fluorouracil in four strains of Nectria haematococca which displayed similar sensitivities to growth inhibition by this compound but differed in their pigmentation. The major metabolites, 5-fluorouridine and alpha-fluoro-beta-alanine, were excreted into the medium by all four strains. The classical ribofluoronucleotides (5-fluorouridine-5'-monophosphate, -diphosphate, and -triphosphate) and alpha-fluoro-beta-alanine were identified in the acid-soluble fraction of perchloric acid extracts of mycelia. Two hydrolysis products of 5-fluorouracil incorporated into RNA were found in the acid-insoluble pool. They were unambiguously assigned to 5-fluorouridine-2'-monophosphate and 3'-monophosphate with specific hydrolysis reactions on isolated RNA. The lack of fluorodeoxyribonucleotides and the fact that the four strains incorporated similar amounts of fluororibonucleotides into their RNAs strongly suggest an RNA-directed mechanism of cytotoxicity for 5-fluorouracil. The heavily pigmented wild type differed from the three low-pigmented strains in its low uptake of 5-fluorouracil and, consequently, in its reduced biosynthesis of 5-fluorouridine and alpha-fluoro-beta-alanine. At present, it is not clear whether this change in 5-fluorouracil metabolism is a side effect of pigment production or results from another event.

Direct detection of new flucytosine metabolites in human biofluids by 19F nuclear magnetic resonance.

19F nuclear magnetic resonance was used for the analysis of flucytosine (FC; 5-fluorocytosine) metabolites in biological fluids of a patient with cryptococcal meningitis who was intravenously injected with this drug at a daily dose of 7.5 g (2.5 g at 8-hr intervals). This method allows a direct, simultaneous, and quantitative determination of all the fluorinated metabolites of FC, in the range of sensitivity allowed by the spectrometer (sensitivity threshold, 0.01 mM). In urine, in addition to the already reported metabolites [unmetabolized FC and alpha-fluoro-beta-alanine (FBAL)], three new metabolites were identified: a glucuronide of FC (GLFC), 6-hydroxy-5-fluorocytosine (60HFC), and fluoride ion F-. The same metabolites (except F-) were found in plasma. In cerebrospinal fluid, only unchanged FC and GLFC were observed. The total urinary excretion during an 8-hr period between two injections of FC was 100.4% of the injected dose. Unchanged FC was the major excretory product accounting for 96.1% of the total. GLFC and 6OHFC made up, respectively, 2.7% and 1.2% of the excreted metabolites. The proportions of F- and FBAL were very low, respectively, 0.3% and 0.1% of the excreted metabolites. The global urinary excretion over a 24-hr period was 102% of the injected dose. The proportions of metabolites were very close to those obtained for the 8-hr period. In plasma, the proportions of metabolites were analogous to those determined in urine. In cerebrospinal fluid, GLFC represents 1% of the fluorinated metabolites.(ABSTRACT TRUNCATED AT 250 WORDS)

19F nuclear magnetic resonance analysis of the carbamate reaction of alpha-fluoro-beta-alanine (FBAL), the major catabolite of fluoropyrimidines. Application to FBAL carbamate determination in body fluids of patients treated with 5'-deoxy-5-fluorouridine.

alpha-Fluoro-beta-alanine (FBAL), the major catabolite of the antineoplastic fluoropyrimidines, is an amino acid which is in equilibrium with its carbamate derivative in weakly alkaline aqueous solutions containing carbonate. In both water and control biological fluids (urine, plasma) spiked with FBAL (and sodium bicarbonate, in some cases), 19F NMR was used: (i) to determine the pH range over which FBAL carbamate is present (pH greater than or equal to 7), the maximum concentration formed occurring around pH 9, (ii) to show that the amino group of FBAL interacts very slowly with a non-protein plasma component to form a compound X, unstable in acid medium. The presumed structure of X is RCONHCH2CHFCOOH, with R different from an alkyl group but still unidentified. The behavior of FBAL in urine and plasma of rats treated with FBAL or 5'-deoxy-5-fluorouridine (5'-dFUrd), a prodrug of 5-fluorouracil, and from patients treated with 5'-dFUrd was investigated. FBAL carbamate was not present in acid medium and was therefore absent in acidic human urine. However, it was found in alkaline rat urine. FBAL carbamate was found in plasma along with the compound X. The 19F NMR spectra of FBAL and derivatives are complex since alpha-fluoro-beta-ureido-propionic acid, the precursor of FBAL in the catabolic pathway of antineoplastic fluoropyrimidines, produces a signal overlapping that of FBAL carbamate, and very close to that of compound X.(ABSTRACT TRUNCATED AT 250 WORDS)