Ojima I. (ed.) Fluorine in Medicinal Chemistry and Chemical Biology

Подождите немного. Документ загружается.

512 Fluorine in Medicinal Chemistry and Chemical Biology

metabolism and tumor half lives by a range of compounds including metholtrexate,

α - interferon and Leucovorin [3, 7 – 10] .

19.5 Fluorine MR Spectroscopy in Neuropharmacology

MR - visible ﬂ uorinated compounds that can be measured in the human brain comprise a

large number of psychiatric medications including most of the serotonin - speciﬁ c reuptake

inhibitors (SSRIs) such as ﬂ uoxetine (Prozac ® ) [1, 28, 41] , as well as pharmaceuticals

with mechanisms of action outside the central nervous system such as dexfenﬂ uramine

(fen - phen, a serotoninergic anorectic drug) [1, 31] . A recent review [1] covers the impact

of MR spectroscopy in psychiatry.

In some compounds with long elimination half - lives, such as ﬂ uoxetine, the plasma

concentration reaches steady - state levels after 4 – 5 weeks of dosing. As a result of 30 days

of dosing at 40 mg/day, plasma concentrations of ﬂ uoxetine and its active metabolite,

norﬂ uoxetine, are in the range 91 – 302 ng/mL and 72 – 258 ng/mL, respectively [51] . The

total ﬂ uorine serum concentration is thus 1.5 – 5.4 µ M. However, tissue accumulation

occurs with many ﬂ uorinated compounds, in part due to their lipophilicity and pH trapping

in acidic vesicles in the brain [1] , and allows the ﬂ uorine concentration to be in a range

detectable using MR spectroscopy [1, 33] . Even with tissue accumulation, the vast major-

ity of brain

F MR spectroscopy assessments are unlocalized (i.e., the entire brain is

contained within a transmit/receive volume coil) [27 – 33, 41, 42, 45] due to low in vivo

ﬂ uorine concentrations.

Since MR spectroscopy detects signals from unbound forms of ﬂ uorinated drugs,

tissue accumulation in which the drug is bound may pose compound visibility problems.

For example, ﬂ uoxetine is known to undergo extensive protein binding (94.5%) [1, 51] ,

whereas dexfenﬂ uramine/dexnorfenﬂ uramine have more moderate protein binding (36%)

[52] . Despite moderate protein binding in solution, in vivo binding was not found for

dexfenﬂ uramine/dexnorfenﬂ uramine. The in vivo

F MR concentration in the brain was

compared with that determined post - mortem by LC - MS [31, 44] . In these primate experi-

ments, dexfenﬂ uramine/dexnorfenﬂ uramine was found to be fully detectable in vivo . In

contrast, signiﬁ cant ﬂ uoxetine binding and MR invisibility in the brain were found from

an autopsy of a patient who received 40 mg/day [32] . However, in the latter study [32] the

tissue was not preserved cryogenically but was ﬁ xed in formalin for 2 weeks prior to per-

forming the

F MR measurement. Potential changes in the cellular environment caused

by use of formalin [32] raise questions about the accuracy of “ bound fraction ” of ﬂ uoxetine

measured post - mortem.

Strauss and Dager [42] also measured the relative contribution of unbound versus

bound ﬂ uoxetine and metabolites to the

F MR visible signal in vivo in the human brain

by applying magnetization transfer methods to acquire

F MR spectra. Signals from the

bound form of ﬂ uoxetine/norﬂ uoxetine were found to be approximately 14.2% of those

detected from the unbound form [42] . A similar binding or interaction between ﬂ eroxacin

in plasma in vitro and in tissue in vivo was found [23] . Understanding the nature and

characteristics of the in vivo spectrum was a prerequisite for quantitation of the resultant

ﬂ eroxacin

F spectra.

Study of Metabolism of Fluorine-containing Drugs 513

While Christensen et al. [31, 44] clearly demonstrated that in vivo

F MR spectros-

copy can provide reliable drug concentration estimates in the brain, the brain to serum

drug ratios may vary widely between species [1, 31] . For irrefutable determination of MR

visibility of a compound, an in vivo

F MR spectroscopy study must be performed either

with organ biopsy or with sacriﬁ ce and necropsy, for drug concentration correlation. Such

validation studies need to be planned and performed carefully since the method of tissue

handling can dramatically inﬂ uence results. Karson et al. [41] reported that levels of drug

detected in formalin - ﬁ xed, postmortem brain tissue from a single subject were much lower

than the level of drug detected immediately following extraction. For neuropharmaceuti-

cals,

F MR spectroscopy may provide one of the only direct in vivo measurement of

drugs and their metabolites in the target organ.

Sassa et al. [2] used in vivo

F MR to evaluate the haloperidol decanoate intramus-

cular injection site in four well - controlled schizophrenia patients. An initial MR assess-

ment of the deltoid muscle with a 15 cm diameter surface coil was performed immediately

after injection with subsequent measurements at 1 and 2 weeks later. For comparison,

haloperidol plasma concentrations were analyzed with high - performance liquid chroma-

tography at the same time (half - life 3 weeks). A broad ﬂ uorine signal was detectable up

to 16 days post injection. The ﬂ uorine signal decreased during the ﬁ rst week and

then appeared to stabilize; this decline was much faster than expected based on the increase

in plasma concentration. The local distribution of haloperidol was assessed with 2D

CSI on the day post injection, but the signal was too low for CSI after 8 days. On day 1,

ﬂ uorine CSI could detect the drug ’ s oil carrier, but the oil was not visible on proton

imaging.

19.6 Fluorine MR Spectroscopy for Measuring Biodistribution

Long - term sequestration of ﬂ uorinated psychotropic compounds has been found by

MRS at 3 T in peripheral tissue several months after cessation of treatment [39] . The high -

ﬁ eld - strength (3 T) MR system signiﬁ cantly enhanced the ability to detect the compound -

speciﬁ c resonance in the lower limbs of patients with major depressive disorder treated

with ﬂ uoxetine or ﬂ uvoxamine.

F MR CSI several months after complete disappearance

of compound signal from the plasma and brain [27] suggested that the post - drug with-

drawal signal originated from bone marrow (see Figure 19.8 ). Mean in vivo concentrations

of the triﬂ uorinated compounds in the lower - extremity bone marrow, estimated by an

external reference standard, were 38 ± 17 µ M for ﬂ uoxetine and 40 ± 17 µ M for ﬂ uvox-

amine [39] .

Conversely, long - term accumulation of tecastemizole in the liver was not found. After

8 days of tecastemizole treatment at a daily dose of 270mg,

F MR signal was found in

the livers, but not the hearts, of healthy volunteers with a concentration range of 2 – 30 µ M

[11] . In contrast to the psychotropic compounds, no ﬂ uorine signal was found in the liver

after 28 days of washout, suggesting that tecastemizole and its ﬂ uorinated metabolites

were eliminated from liver.

F MR spectroscopy has also been used for evaluation of the short - term accumulation

and elimination of niﬂ umic acid, a nonsteroidal anti - inﬂ ammatory drug (NSAID), in the

514 Fluorine in Medicinal Chemistry and Chemical Biology

liver [12] . At about 55 min following oral administration, the parent compound was visible

in the liver. After about 127 min, at least one MR - visible metabolite compound was also

visible in the liver. The triﬂ uorinated parent compound and its MR - visible metabolite both

had linewidths of 30 Hz (T

* ∼ 33 ms). An external standard was used to estimate in vivo

concentrations of the two compounds, with the minimum detectable concentration for

niﬂ umic acid of 1.4 µ M (4.2 µ M in ﬂ uorine). The standard deviation for the concentration

measurements was estimated to be ± 40 – 50% across the detected range (4.2 – 376 µ M in

ﬂ uorine). The rapid compound visibility, as well as the narrow in vivo ﬂ uorine linewidth

suggest that the niﬂ umic acid and metabolites may have been located in plasma rather

than in tissue.

Payne et al. [13] used

F MR to measure the hepatic concentration of sitaﬂ oxacin,

a ﬂ uoroquinolone antimicrobial agent. Elevated liver enzymes have been reported after

use of sitaﬂ oxacin, even though it is eliminated predominantly by renal excretion ( > 99%).

This evaluation was to determine whether drug accumulation in the liver occurs. Sitaﬂ oxa-

cin has two nonequivalent ﬂ uorines ( cis - oriented (1 R ,2 S ) - 2 - ﬂ uorocyclopropylamine

moiety) with resonances separated by 92 ppm. The experiments were focused only on one

Figure 19.8 Distribution of ﬂ uvoxamine at steady - state concentration in the brain of a vol-

unteer measured by

F MR CSI at 3 T. (a) Spatially localized

F MR CSI spectra overlaid on

the axial proton image of the brain and including the external reference sample.

F spectra

of a voxel containing (b) the background (e.g. , no structure and no signal), (c) posterior brain

tissue, and (d) external reference compound (triﬂ uoroethanol). The density of the compounds

is reﬂ ected by peak amplitude and area under the curve (AUC).

( Source: Reprinted by permission from Macmillan Publishers Ltd from Bolo, N. R., Hode, Y.,

Nedelec, J. F., et al. Brain pharmacokinetics and tissue distribution in vivo of ﬂ uvoxamine and

ﬂ uoxetine by ﬂ uorine magnetic resonance spectroscopy. Neuropsychopharmacology (2000)

23 , 428 – 438.)

Study of Metabolism of Fluorine-containing Drugs 515

resonance with smaller

H –

F splitting and longer T

* and T

values. The calculated liver

half - life from the patients with evaluable ﬂ uorine spectroscopic data was 5.85 ± 5.3 h

compared with 5.9 ± 0.7 h in plasma. Because of the large liver standard deviation, it is

possible that there was a difference between liver and plasma half - lives; however, these

study subjects were not found to have any signiﬁ cant changes in liver enzymes. It was

concluded that there was no evidence of hepatic accumulation of sitiﬂ oxacin [13] .

Seddon et al. [25] examined the pharmacodynamics of a tissue hypoxia probe (SR -

4554) in a phase 1 dose escalation study. Three patients, with tumors of at least 3 cm

diameter within 4 cm depth from the skin surface, underwent unlocalized in vivo

F MR

spectroscopy at 1.5 T to determine whether SR - 4554 could be detected. Fluorine signal

was detected in all patients immediately after the intravenous infusion, in the highest

two doses within 8 h ( ∼ 2.5 times the mean plasma half - life), and no signal was detected

even at the highest dose at 27.5 h ( ∼ 9.5 plasma half - lives). The tumor concentration of

SR - 4554 in the tumor was of the same order as the plasma concentration; hence it is

unclear whether the ﬂ uorine signal represented tumor blood ﬂ ow or uptake by the tumor

[25] .

Long term in vivo biodistribution and metabolism cannot be measured using radio-

pharmaceuticals. MR spectroscopy and tissue biopsy remain the only choices for investi-

gating the long - term disposition of pharmaceuticals. While challenging, deep tissue in vivo

biodistribution and metabolism can be measured over time using

F MR after multiple

doses, after the initial washout phase, and noninvasively.

19.7 Analysis of In Vivo

F MR Spectra

Considerable processing must occur following collection of the MR signal decay (free

induction decay or FID) and careful attention must be paid throughout the steps. Each step

has the potential to alter the ability to quantify the signal based on the assumptions made

and the rigor with which the parameters are chosen and applied.

First, either a Lorentzian or Gaussian ﬁ lter is applied to the FID to reduce the amount

of noise. The choice of lineshape will depend on the shape of the frequency domain spec-

trum, the lineshape is related to how the ﬂ uorine spins interact with their environment.

The ﬁ lter linewidth is generally similar to or slightly less than the T

value (T

can be

estimated from the spectral linewidth). After application of the time domain ﬁ lter, a fast

Fourier transform (FFT) is performed. The resultant frequency domain spectrum will then

need to undergo phase adjustment to obtain a pure absorption spectrum. The amount of

receiver dead time (time lost between the end of the excitation pulse and the ﬁ rst useful

detection time point) will determine the presence and extent of baseline artifact present as

well as how difﬁ cult phase adjustment will be to accomplish.

Each resonance in the in vivo frequency domain spectrum is then ﬁ tted to either a

Gaussian or Lorentzian lineshape. The ﬁ ts of the individual resonances are separately

integrated to achieve an area under the curve (AUC). By using a reference for calibration

and accounting for different relaxation rates on the integrated AUC, it is possible to obtain

either relative or absolute concentration from in vivo

F MR spectroscopy [1, 5 – 7, 11, 12,

23, 25, 27, 28, 39, 68 – 70] .

516 Fluorine in Medicinal Chemistry and Chemical Biology

Accounting for different T

* and T

relaxation times of the different resonances cor-

rects for signal loss due to incomplete T

relaxation of the ﬂ uorine spins. This step is

required if the T

of the compound and reference differ and if the TR is less than 5 multiples

of the longest T

value. In addition, signal from short - T

* species require correction for

signal loss owing to the time between the midpoint of the excitation pulse and the start of

the ﬁ rst useful detection time point.

To obtain normalized MR signal, the compound/metabolite resonance AUC is divided

by the AUC of a reference compound. A linear response of

F MR integrated signal

intensity as a function of concentration (see Figure 19.5 d) has been shown [5, 11, 43, 46] .

Alternatively, it is also possible to normalize the

F MR spectral resonances to the proton

water resonance because the water content of liver is estimated to be ∼ 75% at 55 M [13,

25, 68 – 70] .

To derive absolute organ concentrations from the raw spectral data, it is necessary to

segment the organ of interest and to account for the effect of the

F coil sensitivity proﬁ le

[5, 11, 23, 28, 43, 56, 69, 70] . Several additional experiments are required to perform this

calculation. If the coil is tuned to both proton and ﬂ uorine resonance frequencies [13, 25,

68 – 70] , direct measurement of coil sensitivity for both

H and

F is needed. This method

is preferable at high magnetic ﬁ eld strength due to tissue - speciﬁ c dielectric resonance [35]

and RF power absorption [36] . The relaxation time corrected AUCs of the proton CSI

spectra are then used to normalize the

F spectra. With either reference method, for quality

assurance and for accuracy estimates, the transmit – receive sensitivity of the

F RF coil

is mapped using a low - ﬂ ip - angle proton 3D acquisition on a uniform phantom.

For external reference sample methods, a further step is required whereby the

AUC of the

F MR signal using the same coil on a series of different concentration

phantoms (see Figure 4 B) is used to help determine the absolute ﬂ uorine calibration

concentration. Then, for each subject these maps are aligned with the segmented images

to calculate the effective volumes of the target organ. The normalized signal multiplied

by the slope of the calibration curve and divided by the effective organ volume will

provide the absolute organ concentration. The key to successful phantom calibration

is to ensure that coil tuning and matching on the patients is identical to that on the

phantoms.

Within the previously described limitations of co - resonant metabolites and parent

compounds [11, 41] , few studies have measured the accuracy of in vivo

F MR concentra-

tion determination because of the need for a tissue sample. However accuracy measure-

ments have been made for other spin - ½ nuclei in other species (for example,

P [69] ) and

found no signiﬁ cant difference. Even with preclinical preparation and clinical quality

control, SSRIs have been documented to be underestimated by in vivo

F MR spectros-

copy due to their protein interactions. Strauss and Drager [42] estimated the MR - invisible

component of ﬂ uoxetine/norﬂ uoxetine to be 14.2%. In primates, with the external refer-

ence sample method, Christensen et al. [44] measured the accuracy of brain

F MR con-

centration determination of dexfenﬂ uoramine/norfenﬂ uoramine concentrations. In contrast

to SSRIs, they [44] determined that the MR concentrations of dexfenﬂ uoramine/norfen-

ﬂ uoramine were greater than or equivalent to, but in all cases within one standard deviation

of, those measured using gas chromatography. In addition to carefully designed and exe-

cuted

F MR examinations, quality assurance throughout the experiment, preliminary in

vitro and preclinical animal experiments are needed to help interpret in vivo spectroscopic

Study of Metabolism of Fluorine-containing Drugs 517

signal changes in multiple investigations [23, 31, 32, 41, 42, 44] . The accuracy of quanti-

ﬁ cation will always be best at the highest SNR levels.

19.8 Conclusions and Perspectives

In vivo

F MR spectroscopy has been shown to be a highly speciﬁ c, noninvasive tool for

identifying and monitoring ﬂ uorinated compounds and metabolites. This technology is

complementary to other structural and functional imaging tools, and often a combination

of approaches is required to address speciﬁ c questions. Although almost all

F MR exami-

nations are limited by the relatively low SNR of the spectra, ﬂ uorine MR spectroscopy

has a unique role in clinical research applications and can provide a direct measurement

of drug uptake in tissue.

While in vivo

F MR has provided unique insight into ﬂ uoropyrimidine metabolism,

most spectroscopic studies have limited numbers of patients and thus are subject to poten-

tial bias due to the variability in individual biometabolism. The small number of subjects

can further limit generalization of the results to a larger population. In addition, studies of

deep tissues or small lesions are challenging using

F MR spectroscopy because they

cannot be adequately evaluated using surface coil techniques and the low signal levels

preclude use of a volume coil.

The number of facilities capable of performing in vivo

F MR spectroscopy is

limited, in part due to the substantial hardware and software requirements associated with

data acquisition and analysis. Because there is no reimbursement for MR spectroscopic

examinations in the United States [54, 55] , the number of clinical applications will con-

tinue to lag behind development of structural MR imaging.

Instrumentation improvements for

F and other multinuclear MR spectroscopy

examinations have focused on increasing the SNR using similar technologies that proved

valuable in clinical MR imaging. SNR gains are sought both to obtain higher spatial reso-

lution [16] and to detect lower in vivo concentrations with MR spectroscopy [1, 5, 11, 38] .

Increases in SNR have been achieved by performing spectroscopic examinations at higher

magnetic ﬁ eld strengths, usually 3 T or 4 T [11, 38, 39] , and by optimizing the detection

hardware [24] . Detection systems for multinuclear MR spectroscopy now use digital

receivers and frequency - speciﬁ c components including tuned preampliﬁ ers and switches

[11, 24] .

On the basis of trends over the past two decades, the availability and utilization of

higher strength magnets should continue to increase, and thus increases in SNR will con-

tinue to be realized in the future. As lower - noise components become available and are

incorporated into the receiver chain, incremental SNR gains will also be achieved. Use of

cooled receivers and high - temperature superconducting coils [64] may also be of interest

to gain SNR for low - gamma nuclei that resonate at low frequencies. In addition, use of

phased - array detection coils [65] or in situ detection coils such as endovascular coils [66]

may further improve SNR, depending upon the anatomical target. Interestingly, the

MR spectroscopic signature has been recently used in the laboratory setting as a valuable

tool for identifying potentially counterfeit or otherwise illegally manufactured pharmaceu-

ticals such as those with lower than advertised concentrations of active ingredient [67] .

518 Fluorine in Medicinal Chemistry and Chemical Biology

In spite of the challenges, the in vivo biodistribution as well as absolute concentrations

of ﬂ uorinated compounds can be measured noninvasively using

F MR. Human studies

require careful planning and knowledge of the interactions of the molecules with the

environment in vivo . All compounds should be assessed for the extent of protein binding,

compound sequestration, or compartmentalization prior to going into the clinic [1, 11, 20,

31, 41, 42] . With a modest amount of preparation and planning, assessment of individual

drug and metabolite levels is feasible using

F MR, as is measurement of the pharmaco-

kinetics and pharmacodynamics. Since long - term in vivo biodistribution cannot be mea-

sured using radiopharmaceuticals, MR spectroscopy and tissue biopsy remain the only

choices for investigating the disposition of pharmaceuticals. Due to the widespread use of

ﬂ uorinated compounds in medicine, ﬂ uorine MR spectroscopy is a useful noninvasive

method for assessing absorption, biodistribution, metabolism, and excretion during the

drug development process because

F MR allows repeat measurement and accurate in

vivo evaluation of metabolism of ﬂ uorine - containing compounds.

Acknowledgments

We thank Scott Wilkinson, MS, Steve Jones, PhD, Tom Rollins, and other members of the

Sepracor Inc. Clinical and Pre - Clinical Development teams, and Blaise Frederick, PhD,

Nick Bolo, PhD and Perry Renshaw, MD, PhD at McLean Hospital Brain Imaging Center

in Boston, MA for their scientiﬁ c contributions to the tecastemizole work. In addition, we

are grateful to Sepracor, Inc. for permission to publish the tecastemizole spectra and data.

In addition, we thank Dennis Klomp, PhD (University Medical Center Utrecht,

Utrecht, the Netherlands) and Nick Bolo, PhD (McLean Hospital, Boston, MA, U.S.A.)

for their generous contribution of the materials for Figures 19.7 and 19.8 , respectively.

Abbreviations

1D one - dimensional

two - dimensional

three - dimensional

5 - FU

5 - ﬂ uorouracil

AUC

area under the curve

CSA

chemical shielding anisotropy

CSI

chemical shift imaging

DMSO

dimethyl sulfoxide

fen - phen

dexfenﬂ uramine

FBAL

2 - ﬂ uoro - β - alanine

FDG

ﬂ uorodeoxyglucose

FID

free induction decay

FFT

fast Fourier transform

LC - MS

liquid chromatography - mass spectrometry

Study of Metabolism of Fluorine-containing Drugs 519

References

Lyoo , I. K. and Renshaw , P. F. ( 2002 ) Magnetic resonance spectroscopy: current and future

applications in psychiatric research . Biological Psychiatry , 51 , 195 – 207 .

Sassa , T. , Suhara , T. , Ikehira , H. , et al. ( 2002 )

F - magnetic resonance spectroscopy and chemi-

cal shift imaging for schizophrenic patients using haloperidol decanoate . Psychiatry and Clinical

Nuerosciences , 56 , 637 – 642 .

Port , R. E. and Wolf , W. ( 2003 ) Noninvasive methods to study drug distribution . Investigational

New Drugs , 21 , 157 – 168 .

Martino , R. , Gilard , V. , Desmoulin , F. and Malet - Martino , M. ( 2006 ) Interest of ﬂ uorine - 19

nuclear magnetic resonance spectroscopy in the detection, identiﬁ cation and quantiﬁ cation of

metabolites of anticancer and antifungal ﬂ uoropyrimidine drugs in human bioﬂ uids . Chemo-

therapy , 52 ( 5 ), 215 – 219 .

Bachert , P. ( 1998 ). Pharmacokinetics using ﬂ uorine NMR in vivo . Progress in NMR Spectros-

copy , 33 , 1 – 56 .

Wolf , W. , Albright , M. J. , Silver , M. S. , et al. ( 1987 ) Fluorine - 19 NMR spectroscopic studies

of the metabolism of 5 - ﬂ uorouracil in the liver of patients undergoing chemotherapy . Magnetic

Resonance Imaging , 5 , 165 – 169 .

Wolf , W. , Presant , C. A. and Waluch , V. ( 2000 ) 19F - MRS studies of ﬂ uorinated drugs in humans .

Advanced Drug Delivery Reviews , 41 , 55 – 74 .

Malet - Martino , M. , Gilard , V. , Desmoulin , F. and Martino , R. ( 2006 ) Fluorine nuclear magnetic

resonance spectroscopy of human bioﬂ uids in the ﬁ eld of metabolic studies of anticancer and

antifungal ﬂ uoropyrimidine drugs . Clinica Chimica Acta , 366 , 61 – 73 .

van Laarhoven , H. W. M. , Punt , C. J. A. , Kamm , Y. J. L. and Heerschap , A. ( 2005 ) Monitoring

ﬂ uoropyrimidine metabolism in solid tumors with in vivo

F magnetic resonance spectroscopy .

Critical Reviews in Oncology/Hematology , 56 , 321 – 343 .

Martino , R. G. , Filard , V. , Desmoulin , F. and Malet - Martino , M. ( 2005 ) Fluorine - 19 or phos-

phorus - 31 NMR spectroscopy: a suitable analytical technique for quantitative in vitro metabolic

studies of ﬂ uorinated or phosphorylated drugs . Journal of Pharmaceutical and Biomedical

Analysis , 38 ( 5 ), 871 – 891 .

10.

magnetic resonance

NEX

number of acquisitions

NOE

nuclear Overhauser effect

NSAID

nonsteroidal anti - inﬂ ammatory drug

PET

positron emission tomography

radiofrequency

SAR

speciﬁ c absorption rate

SNR

signal - to - noise ratio

SPECT

single - photon emission computerized tomography

SR - 4554

SR - 4554 (compound number; sensitive to tissue hypoxia)

SSRI

serotonin - speciﬁ c reuptake inhibitor

spin – lattice relaxation time

spin – spin relaxation time

* effective T

relaxation time

pulse repetition time

520 Fluorine in Medicinal Chemistry and Chemical Biology

Schneider , E. , Bolo , N. R. , Frederick , B. , et al. ( 2006 ) Magnetic resonance spectroscopy for

measuring the biodistribution and in situ in vivo pharmacokinetics of ﬂ uorinated compounds:

validation using an investigation of liver and heart disposition of tecastemizole . Journal of

Clinical Pharmacy and Therapeutics , 31 , 261 – 273 .

Bilecen , D. , Sculte , A. C. , Kaspar , A. , et al. ( 2003 ) Detection of the non - steroidal anti -

inﬂ ammatory drug niﬂ umic acid in humans: a combined

F MRS in vivo and in vitro study .

NMR in Biomedicine , 16 , 144 – 151 .

Pellecchia , M. , Sem , D. S. and W ü thrich , K. ( 2002 ) NMR in drug discovery . Nature Reviews

Drug Discovery , 1 , 211 – 219 .

Marzola , P. , Osculati , F. and Sbarbati , A. ( 2003 ) High ﬁ eld MRI in preclinical research . Euro-

pean Journal of Radiology , 48 , 165 – 170 .

Beckmann , N. , Mueggler , T. , Allegrini , P. R. , et al. ( 2001 ) From anatomy to the target: contribu-

tions of magnetic resonance imaging to preclinical pharmaceutical research. Anatomical Record ,

265 ( 2 ), 85 – 100 .

Payne , G. S. , Collins , D. J. and Loynds , P. , et al. ( 2004 ) Quantitative assessment of the hepatic

pharmacokinetics of the antimicrobial sitaﬂ oxacin in humans using in vivo

F magnetic reso-

nance spectroscopy . British Journal of Clinical Pharmacology , 59 ( 2 ), 244 – 248 .

Brix , G. , Bellemann , M. E. , Haberkorn , U. , et al. ( 1996 ) Assessment of the biodistribution and

metabolism of 5 - ﬂ uorouracil as monitored by

F PET and

F MRI: a comparative animal study .

Nuclear Medicine and Biology , 23 , 897 – 906 .

Park , B. K. , Kitteringham , N. R. and O ’ Neill , P. M. ( 2001 ) Metabolism of Fluorine - containing

Drugs . Annual Reviews of Pharmacology and Toxicology , 41 , 443 – 470 .

Danielson , M. A. and Falke , J. J. ( 1996 ) Use of

F NMR to probe protein structure and con-

formational change . Annual Reviews of Biophysics and Biomolecular Structure , 25 , 163 – 195 .

Dzik - Jurasz , A. S. K. , Leach , M. O. and Rowland , I. J. ( 2004 ) Investigation of microenviron-

mental factors inﬂ uencing the longitudinal relaxation times of drugs and other compounds .

Magnetic Resonance Imaging , 22 ( 7 ), 973 – 982 .

Harris , R. K. ( 1983 ) NMR Spectroscopy: A Physicochemical View , Pitman , London .

Bruno , T. J. and Svoronos , P. D. N. ( 2003 ) CRC Handbook of Basic Tables for Chemical Analy-

sis , 2nd edn. , CRC Press, Taylor and Francis Group , Boca Raton, FL .

Jynge , P. , Skjetne , T. , Gribbestad , I. , et al. ( 1990 ) In vivo tissue pharmacokinetics by ﬂ uorine

magnetic resonance spectroscopy: A study of liver and muscle disposition of ﬂ eroxacin in

humans . Journal of Clinical Pharmacy and Therapeutics , 48 , 481 – 489 .

Klomp , D. W. J. , van Laarhoven , H. W. M. , Kentgens , A. P. M. and Heerschap , A. ( 2003 ) Opti-

mization of localized

F magnetic resonance spectroscopy for the detection of ﬂ uorinated drugs

in the human liver . Magnetic Resonance in Medicine , 50 , 303 – 308 .

Seddon , B. M. , Payne , G. S. , Simmons , L. , et al. ( 2003 ) A Phase I study of SR - 4554 via intra-

venous administration for noninvasive investigation of tumor hypoxia by magnetic resonance

spectroscopy in patients with malignancy . Clinical Cancer Research , 9 , 5101 – 5112 .

Kamm , Y. J. L. , Heerschap , A. , van den Bergh , E. J. and Wagener , D. J. T. ( 2004 )

F - magnetic

resonance spectroscopy in patients with liver metastases of colorectal cancer treated with 5 -

ﬂ uorouracil . Anti - Cancer Drugs , 15 , 229 – 233 .

Bolo , N. R. , Hode , Y. , Nedelec , J. F. , et al. ( 2000 ) Brain pharmacokinetics and tissue distribution

in vivo of ﬂ uvoxamine and ﬂ uoxetine by ﬂ uorine magnetic resonance spectroscopy . Neuropsy-

chopharmacology , 23 , 428 – 438 .

Miner , C. M. , Davidson , J. R. T. , Potts , N. L. S. , et al. ( 1995 ) Brain ﬂ uoxetine measurements

using

F MRS in social phobia . Biological Psychiatry , 38 , 696 – 698 .

Strauss , W. L. , Layton , M. E. , Hayes , C. E. and Dager , S. R. ( 1997 )

F MRS of acute and steady

state brain ﬂ uvoxamine levels in obsessive compulsive disorder . American Journal of Psychia-

try , 151 , 516 – 522 .

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

Study of Metabolism of Fluorine-containing Drugs 521

Strauss , W. L. , Layton , M. E. and Dager , S. R. ( 1998 ) Brain elimination half life of ﬂ uvoxamine

F MRS . American Journal of Psychiatry , 155 , 380 – 384 .

Christensen , J. D. , Yurgelun - Todd , D. A. , Babb , S. M. , et al. ( 1999 ) Measurement of human

brain dexfenﬂ uramine by

F MRS . Brain Research , 831 , 1 – 5 .

Komoroski , R. , Newton , J. E. , Cardwell , D. , et al. ( 1994 ) In vivo

F spin relaxation and localized

spectroscopy of ﬂ uoxetine in human brain . Magnetic Resonance in Medicine , 31 , 204 – 211 .

Renshaw , P. , Guimaraes , A. R. , Fava , M. , et al. ( 1992 ) Accumulation of ﬂ uoxetine and norﬂ uox-

etine in human brain during therapeutic administration . American Journal of Psychiatry , 149 ,

1592 – 1594 .

Schenck , J. F. ( 2000 ) Safety of strong, static magnetic ﬁ elds . Journal of Magnetic Resonance

Imaging , 12 , 3 – 19 .

Bottomley , P. A. and Andrew , E. R. ( 1978 ) RF magnetic ﬁ eld penetration, phase shift and power

dissipation in biological tissue: implication for NMR imaging . Physics in Medicine and Biology ,

23 , 630 – 643 .

Hoult , D. I. ( 2000 ) Sensitivity and power deposition in a high - ﬁ eld imaging experiment . Journal

of Magnetic Resonance Imaging , 12 , 46 – 67 .

Hoult , D. I. and Lauterbur , P. C. ( 1979 ) The sensitivity of the zeugmatographic experiment

involving human samples . Journal of Magnetic Resonance , 34 , 425 – 433 .

van Laarhoven , H. W. M. , Klomp , D. W. , Kamm , Y. J. L. , et al. ( 2003 ) In vivo monitoring of

capecitabine metabolism in human liver by

Fluorine magnetic resonance spectroscopy at 1.5

and 3T ﬁ eld strength . Cancer Research , 63 , 7609 – 7612 .

Bolo , N. R. , Hode , Y. and Macher , J. P. ( 2004 ) Long - term sequestration of ﬂ uorinated compounds

in tissues after ﬂ uvoxamine or ﬂ uoxetine treatment: a ﬂ uorine magnetic resonance spectroscopy

study in vivo . Magnetic Resonance in Materials Physics, Biology and Medicine , 16 , 268 – 276 .

Purcell , E. M. , Torrey , E. C. and Pound , R. V. ( 1946 ) Resonance absorption by nuclear magnetic

moments in a solid . Physics Review , 69 , 37 – 38 .

Karson , C. N. , Newton , J. E. , Livingston , R. , et al. ( 1993 ) Human brain ﬂ uoxetine concentra-

tions . Journal of Neuropsychiatry and Clinical Neurosciences , 5 , 322 – 329 .

Strauss , W. L. and Dager , S. R. ( 2001 ) Magnetization transfer of ﬂ uoxetine in the brain with

F MRS . Biological Psychiatry , 49 , 798 – 802 .

Schlemmer , H. P. , Bachert , P. , Semmler , W. , et al. ( 1994 ) Drug monitoring of 5 - ﬂ uorouracil:

in vivo

F NMR study during 5 - FU chemotherapy in patients with metastases of colorectal

adenocarcinoma . Magnetic Resonance Imaging . 12 , 497 – 511 .

Christensen J. D. , Babb , S. M. , Cohen , B. M. and Renshaw , P. F. ( 1998 ) Quantitation of dexfen-

ﬂ uramine/d - norfenﬂ uramine concentration in primate brain using

F NMR spectroscopy . Mag-

netic Resonance in Medicine , 39 , 149 – 154 .

Henry , M. E. , Moore , C. M. , Kaufman , M. J. , et al. ( 2000 ) Brain kinetics of paraxetine and

ﬂ uoxetine . American Journal of Psychiatry , 157 , 1506 – 1508 .

Martino , R. , Malet - Martino , M. and Gilard , V. ( 2000 ) Fluorine nuclear magnetic resonance, a

privileged tool for metabolic studies of ﬂ uoropyrimidine drugs . Current Drug Metabolism 1 ( 3 ),

271 – 303 .

Wolf , W. , Waluch , V. and Presant , C. A. ( 1998 ) Non - invasive

F NMRS of 5 - ﬂ uorouracil in

pharmacokinetics and pharmacodynamic studies . NMR in Biomedicine , 11 , 380 – 387 .

Haase , A. , Frahm , J. , H ä nicke , W. and Matthaei , D. ( 1985 )

H NMR chemical shift selective

(CHESS) imaging . Physics in Medicine and Biology , 30 , 341 – 344 .

Frahm , J. , Haase , A. , H ä nicke , W. , et al. ( 1985 ) Chemical shift selective MR imaging using a

whole - body magnet . Radiology , 156 , 441 – 444 .

Gonen , O. , Murphy - Boesch , J. , Li , C. W. , et al. ( 1997 ) Simultaneous 3D NMR spectroscopy of

proton - decoupled ﬂ uorine and phosphorus in human liver during 5 - ﬂ uorouracil chemotherapy .

Magnetic Resonance in Medicine , 37 , 164 – 169 .

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.