Comparative biodistribution study of the new tumor tracer [123I]-2-iodo-l-phenylalanine with [123I]-2-iodo-l-tyrosine

www.elsevier.com/locate/nucmedbio

Nuclear Medicine and

Comparative biodistribution study of the new tumor tracer

[123I]-2-iodo-l-phenylalanine with [123I]-2-iodo-l-tyrosine

Veerle Kersemansa,T, Bart Cornelissena,b, Ken Kersemansc, Rudi A. Dierckxd,

Bart De Spiegeleera, John Mertensc, Guido Slegersa

aLaboratory for Radiopharmacy, Universiteit Ghent, B-9000 Ghent, BelgiumbLaboratory for Molecular Imaging and Targeted Radiotherapy, University of Toronto, Canada M5G 2C4

cLaboratory for Medical Imaging and Physics, Vrije Universiteit Brussel, Brussel, B-1090 BelgiumdDivision of Nuclear Medicine, Ghent University Hospital, Ghent, B-9000 Belgium

Received 16 July 2005; received in revised form 29 August 2005; accepted 29 August 2005

Abstract

Introduction: Both A- and l-type amino acid transport are increased in tumor cells relative to normal tissue; these transport systems have

been the major focus of the development of amino acid tumor tracers to overcome the limitations of [18F]-fluorodeoxyglucose (18F-FDG).

The newly developed tracer 2-amino-3-(2-[123I]iodophenyl)propanoic acid ([123I]-2-iodo-l-phenylalanine) showed high and specific tumor

uptake, slow renal elimination and low brain uptake. We compared [123I]-2-iodo-l-phenylalanine with 2-amino-3-(4-hydroxy-2-

[123I]iodophenyl)propanoic acid ([123I]-2-iodo-l-tyrosine), an l-tyrosine analogue that has recently entered clinical trials.

Methods: [123I]-2-iodo-l-phenylalanine and [123I]-2-iodo-l-tyrosine were evaluated in rhabdomyosarcoma tumor-bearing athymic mice by

means of dynamic planar imaging (DPI) and dissection. A displacement study with l-phenylalanine was performed to prove the specificity of

tracer tumor uptake, and kinetic modeling was applied to the DPI results. Moreover, the biodistribution of both tracers was compared with

that of 18F-FDG.

Results: Both [123I]-2-iodo-l-phenylalanine and [123I]-2-iodo-l-tyrosine showed fast, high and specific tumor accumulation with

no significant difference. However, [123I]-2-iodo-l-phenylalanine was cleared faster from the blood to the bladder in comparison

with the tyrosine analogue. Moreover, [123I]-2-iodo-l-phenylalanine tumor uptake equilibrated faster with blood. Dissection showed that

[123I]-2-iodo-l-tyrosine slightly accumulated in the liver, which was not the case for the phenylalanine analogue. In contrast to 18F-FDG,

both tracers showed low uptake in the heart and normal brain tissue, which is advantageous for tumor detection in these organs.

Conclusions: [123I]-2-iodo-l-phenylalanine showed more promising characteristics for oncological imaging as compared with [123I]-2-iodo-

l-tyrosine. The former tracer not only demonstrated faster blood clearance but also showed that the tracer uptake in the tumor reached its

equilibrium with the blood pool activity faster, which led to faster and better tumor contrast. Moreover, both tracers could overcome an

important limitation of 18F-FDG—its high normal brain uptake.

D 2006 Elsevier Inc. All rights reserved.

Keywords: [123I]-2-iodo-l-phenylalanine; Radiolabeled amino acid analogue; Tumor imaging; [123I]-2-iodo-l-tyrosine; 18F-FDG; SPECT

1. Introduction

During the last decade, amino acid analogues gainedmuch

more appeal in metabolic tumor imaging. Their beneficial

properties over [18F]-fluorodeoxyglucose (18F-FDG), such as

their high and fast tumor uptake and rather low uptake in gray

matter and inflammatory lesions, have been put forward. As a

0969-8051/$ – see front matter D 2006 Elsevier Inc. All rights reserved.

doi:10.1016/j.nucmedbio.2005.08.007

T Corresponding author. Tel.: +32 9 264 80 65; fax: +32 9 264 80 71.

E-mail address: [email protected] (V. Kersemans).

result, they may help in imaging areas in which 18F-FDG

imaging is limited [1].

Malignant tumors are characterized by a hypermetabolic

state: not only glucose metabolism but also protein synthesis

and amino acid transport are enhanced in cancer cells [2,3].

Both A- and l-type amino acid transport have been shown

to be up-regulated in tumor cells as compared with normal

tissue, and these transport systems have been the major

focus of the development of amino acid tumor tracers

for oncological imaging. Earlier studies showed that the

enhanced amino acid transport across the cell membrane

Biology 33 (2006) 111–117

V. Kersemans et al. / Nuclear Medicine and Biology 33 (2006) 111–117112

correlated with the malignancy of the tumor. Therefore,

amino acid transport tracers are valuable tracers to reflect

the malignancy of tumors [1].

2-Amino-3-(4-hydroxy-3-[123I]iodophenyl)-2-methylpro-

panoic acid (3-[123I]-iodo-a-methyltyrosine; 123I-IMT) is

currently the main and only routinely used amino acid tumor

tracer for SPECT. However, its application in the abdominal

area is limited by its rapid renal elimination and accumu-

lation in the urinary tract [4,5]. Numerous amino acid

analogues were investigated to overcome these limitations.

Most attention was given to 2-iodo-l-tyrosine, which has

shown no accumulation in the kidneys and demonstrated

high tumor specificity [6]. This amino acid has recently

entered the clinical trial process.

To overcome the limitations of 123I-IMT, our research

group focused on 2-amino-3-(2-[123I]iodophenyl)propanoic

acid ([123I]-2-iodo-l-phenylalanine) and described its prom-

ising properties as a SPECT tumor diagnosticum in a variety

of tumors [7,8]. The aim of this study was to compare the in

vivo behavior of [123I]-2-iodo-l-phenylalanine with that of

2-amino-3-(4-hydroxy-2-[123I]iodophenyl)propanoic acid

([123I]-2-iodo-l-tyrosine) to evaluate whether the phenylal-

anine analogue exhibited better diagnostic imaging charac-

teristics than the tyrosine analogue.

2. Materials and methods

All the conventional products mentioned were at least

of analytic or clinical grade and obtained from Sigma-

Aldrich. The solvents were of high-performance liquid

chromatography quality (Chemlab, Belgium). The precur-

sor 2-iodo-l-tyrosine was obtained from ABX (Belgium).

2.1. Synthesis of precursor 2-iodo-l-phenylalanine

2-Iodo-l-phenylalanine was prepared from 2-bromo-

l-phenylalanine (PepTech, USA) as described before, using

the Cu1+-assisted nucleophilic exchange under acidic and

reducing conditions [9]. After optimization, the mean

reaction yield was 74% and no contamination of the

d-isomer was detected by chiral chromatography.

Fig. 1. Overall mean uptake of [123I]-2-iodo-l-phenylalanine (A) and [123I]-2-iod

function of time by DPI expressed as DUR.

2.2. Radiochemistry

2.2.1. [123I/125I]-2-iodo-l-phenylalanine and [123I/125I]-2-

iodo-l-tyrosine

Radioiodination of 1.0 mg of 2-iodo-l-phenylalanine or

2-iodo-l-tyrosine with 123I� (222 MBq; 10–20 ll) or 125I�

(37 MBq; 10 ll; Nordion Europe, Belgium) was performed

by Cu1+-assisted isotopic exchange (0.2-mg CuSO4, 2.5-mg

citric acid, 0.5-mg SnSO4, 1.3-mg gentisic acid in 565 ll;60 min at 1008C). Radiolabeling of both tracers with 123I�

or 125I� resulted in a radiochemical purity of N99% and a

specific activity of 65 GBq/mmol ([123I]-labeling) and

11 GBq/mmol ([125I]-labeling) for 2-iodo-l-phenylalanine

and a specific activity of 68 GBq/mmol ([123I]-labeling) and

11 GBq/mmol ([125I]-labeling) for 2-iodo-l-tyrosine. Chiral

chromatography showed no transformation to the d-isomer.

2.2.2. [123I]-iodo-human serum albumin

Radioiodination of human serum albumin (HSA) with123I� was performed by electrophilic substitution using the

Iodogen technique: 50 lg of HSA together with 111 MBq of123I� (10 ll) and 140 ll of 0.1 M KH2PO4 buffer at pH 8.5

was added to a vial coated with 1, 3, 4, 6-tetrachloro-3a,6a-diphenylglycouracil (Pierce, Belgium) for 15 min. The

reaction mixture was sent through an Ag filter to remove the

free 123I� from [123I]-iodo-HSA. The specific activity was

4.104 GBq/mmol.

2.3. In vivo experiments

2.3.1. Laboratory animals

All in vivo studies were carried out in accordance with

Belgian legislation, including the approval of the study

protocol by the ethical committee for animal studies of the

University of Ghent. Guidelines of the National Institute of

Health principles of laboratory animal care were followed.

Water and feed (SSNIFF special treated, Bio-Services,

The Netherlands) were freely available during the experi-

mental period.

Studies were performed in female Swiss nu/nu mice

(N=36; weight range=20–25 g) obtained from Bio-Services.

o-l-tyrosine (B) in the tumor and the contralateral background region as a

Fig. 3. Plot of inverse relationship of the tumor-to-blood ratio as a function

of time for [123I]-2-iodo-l-phenylalanine ( y =2.14 [95% confidence inter-

val=2.08–2.21] x+1.28; R2= .997) and [123I]-2-iodo-l-tyrosine ( y =0.67

[95% confidence interval=0.61–0.73] x+1.68; R2= .974).

Table 1

Displacement study of [123I]-2-iodo-l-phenylalanine and [123I]-2-iodo-l-

tyrosine activities in the tumor using planar gamma scintigraphy

Amino acid analogue % Pa % Db

[123I]-2-iodo-l-phenylalanine 68 8.5F1.0

[123I]-2-iodo-l-tyrosine 63 7.0F1.4

a Percentage of all animals in which a significant displacement of

radioactivity was observed (n =12).b Amount (%) of tracer decrease by administering the amino acid

(meanFS.D.).

V. Kersemans et al. / Nuclear Medicine and Biology 33 (2006) 111–117 113

The mice were implanted with an R1M (rhabdomyosarco-

ma) tumor by subcutaneous injection of 5.106 R1M cells

(BEFY-VUB, Brussels, Belgium) in the right armpit region.

Normal tumor growth curves were obtained using sliding

caliper measurement and the estimate volume formula

V=0.4 a2.b, with a and b being the short axis and the long

axis of the tumor, respectively [9,10]. All mice grew tumors

with a volume of approximately 1 ml after 28 days. The

imaging experiments were started when the tumor reached a

volume of 0.5 ml. This R1M tumor model has been used

previously to study the behavior of radioiodinated amino

acids [6].

Imaging experiments with [123I]-iodo-HSA, [123I]-2-iodo-

l-phenylalanine and [123I]-2-iodo-l-tyrosine were carried

out on the same group of mice. The following scanning order

was followed with an interval of 6 days in which the injected

radioactivity decayed: [123I]-iodo-HSA, [123I]-2-iodo-l-ty-

rosine and [123I]-2-iodo-l-phenylalanine. All tracers were

injected intravenously in the lateral tail vein.

During all imaging experiments, anesthesia was induced

by intraperitoneal injection of pentobarbital [1.5 mg in 75 Alper animal; prepared by dilution (1:3) of 60 mg/ml

Nembutal from Ceva Sante Animale, Belgium]. For the

biodistribution experiments by dissection, the animals were

killed by cervical dislocation without sedation and the

organs of interest were dissected.

2.3.2. Dynamic planar imaging

Imaging was performed using a gamma camera

(Toshiba GCA-9300A/hg) in planar mode equipped with

a high resolution parallel-hole collimator. After the intrave-

Fig. 2. Overall mean uptake of [123I]-2-iodo-l-phenylalanine (A) and [123I]-2-iodo

through the kidneys to the bladder.

nous bolus injection of the [123I]-labeled product in the

lateral tail vein, the data were recorded in a 128�128 matrix

(field of view=23.5�12.5 cm) and with a photopeak

window set at 15% around 159 keV. The same time schedule

as described for [123I]-2-iodo-l-phenylalanine was used [9].

The tracer uptake was recorded as differential uptake ratio

(DUR), taking into account a dose calibration and back-

ground correction following the guidance of Thie [14].

The DUR was calculated as [(countstissue*pixelstotal body)/

(pixelstissue*countstotal body)].

Briefly, a [123I]-iodo-HSA study was performed to

measure the relative blood pool distribution to correct the

uptake of [123I]-2-iodo-l-phenylalanine or [123I]-2-iodo-l-

tyrosine for blood pool activity. Ten tumor-bearing mice

were injected with 7.4 MBq of [123I]-iodo-HSA. Ten

dynamic planar images of 1 min were acquired starting

10 min postinjection (p.i.). ROIs were drawn around the

tumor and the contralateral background area. The tumor-to-

contralateral background ratio (RTB) was calculated and the

overall mean for all animals was determined.

Subsequently, immediately after the injection of 18.5MBq

of [123I]-2-iodo-l-phenylalanine or [123I]-2-iodo-l-tyrosine,

the steady state of tumor uptake for both tracers was

determined followed by a displacement study with

l-phenylalanine (200 ll iv of a 145-mM solution). Tumor

-l-tyrosine (B; DUR) as a function of time by DPI: clearance of the tracer

Table 2

Kinetic parameters obtained for [123I]-2-iodo-l-phenylalanine and [123I]-2-

iodo-l-tyrosine by two-compartment modeling

Kinetic parametersFS.D.

Unit [123I]-2-iodo-l-

phenylalanine

[123I]-2-iodo-l-

tyrosine

V1 (activitytotal body)/(DUR) 9.65F0.20 9.89F0.33

k1,0 1/min 0.035F0.003 0.019F0.002

k1,2 1/min 0.80F0.04 1.13F0.11

k2,1 1/min 0.42F0.02 0.92F0.05

V1=apparent distribution volume of central compartment; k1,0=elimi-

nation velocity; k1,2=distribution velocity from the central to the

peripheral compartment; k2,1=velocity from the peripheral to the central

compartment.

V. Kersemans et al. / Nuclear Medicine and Biology 33 (2006) 111–117114

[123I]-2-iodo-l-phenylalanine or [123I]-2-iodo-l-tyrosine up-

take was compared with the uptake in the contralateral

background area and the RTB was calculated. Other organs

quantitatively assessed by dynamic planar imaging (DPI)

were the kidney, bladder, and heart, the latter organ being

representative of blood activity as similar kinetic profiles

were obtained by dissection. The significance of the

displacement of [123I]-2-iodo-l-phenylalanine or [123I]-2-

iodo-l-tyrosine activity by l-phenylalanine was calculated at

a 95% confidential interval.

Time–activity curves were obtained from ROI analysis

using an MRI maximum intensity projection as described

before [9].

The data (ROI of the heart) obtained by DPI for

[123I]-2-iodo-l-phenylalanine and [123I]-2-iodo-l-tyrosine

were fit to a two-compartment model with intravenous

bolus injection, without a lag time and with first-order

elimination, using WinNonlin 4.0.1. The primary para-

meters V1, k1,0, k1,2 and k2,1 were determined.

The behavior of both tumor tracers was studied in the same

animal to reduce variability between animals. The results

were submitted to paired Student’s t statistical analysis.

Table 3

Biodistribution study by dissection of [125I]-2-iodo-l-tyrosine in R1M-bearing at

Differential absorption ratio (DARFS.D.)

2 min 5 min 10 min 15 min

Blood 1.98F0.04 1.64F0.29 1.57F0.11 1.46F0.15

Brain 0.43F0.02 0.69F0.03 0.72F0.19 0.65F0.09

Heart 2.01F0.14 1.61F0.26 1.37F0.11 1.35F0.01

Lung 1.54F0.04 1.42F0.38 1.22F0.10 1.04F0.01

Stomach 0.64F0.28 0.86F0.34 1.53F0.43 1.02F0.25

Milt 1.68F0.19 1.50F0.34 1.40F0.19 1.30F0.29

Liver 2.66F0.48 2.17F0.88 1.99F0.64 1.92F0.66

Kidneys 4.06F0.50 3.67F0.42 3.29F0.59 3.06F0.41

Small intestine 1.18F0.04 1.11F0.25 0.94F0.22 0.87F0.06

Large intestine 0.65F0.10 0.55F0.10 0.57F0.12 0.54F0.19

Contralateral

background

1.01F0.11 1.09F0.15 0.89F0.06 0.98F0.12

Pancreas 6.68F3.84 11.21F2.88 10.40F6.07 10.56F8.34

Tumor 0.34F0.06 0.36F0.18 0.55F0.11 0.79F0.05

RTB 0.34F0.11 0.33F0.16 0.61F0.14 0.81F0.06

2.3.3. Dissection: tissue distribution in time of

[125I]-2-iodo-tyrosine

The injected activity was calculated by weighing the

syringes before and after the injection of the tracer and by

using a dilution series of the injected tracer solution, which

was also weighed and counted for radioactivity using an

auto gamma-counting system (CobraII Series, Canberra

Packard, Meriden, CT, USA).

Of the 36 animals used for imaging, 27 R1M-bearing

athymic mice were injected with 7.4 kBq of [125I]-2-iodo-

l-tyrosine 6 days after the last imaging experiment was

performed. At various time points (2, 5, 10, 15, 30, 45, 60,

120 and 180 min) p.i., 3 animals per time point were

killed. The organs and tissues were removed, washed,

dried and weighed. The blood was collected and weighed.

The radioactivity of the samples was counted using the

auto gamma-counting system CobraII Series (Canberra

Packard). The amount of radioactivity in the organs and

tissues was calculated as the differential absorption rate

(DAR): [(IA/g)tissue]/[(IA/g)total body].

2.3.4. Organ clustering

The data obtained by biodistribution by dissection were

used for nonlinear regression in SPSS. The parameters of

the exponential fitted curves in SPSS were applied for

hierarchical clustering of the organs using the average

linkage algorithm. This analysis was performed to discover

which organs exhibit the same behavior for [123I]-2-iodo-l-

phenylalanine or [123I]-2-iodo-l-tyrosine tracer biodistribution.

2.3.5. Dissection: comparison of [125I]-2-iodo-l-

phenylalanine with [125I]-2-iodo-tyrosine and18F-FDG at 90 min p.i.

Of the 36 mice used for imaging, 3 R1M-bearing

athymic mice per tracer were injected with 7.4 kBq of

[125I]-2-iodo-l-phenylalanine, [125I]-2-iodo-l-tyrosine or18F-FDG 6 days after the last imaging experiment was

hymic mice expressed as DAR (n =3)

30 min 45 min 60 min 120 min 180 min

1.41F0.07 1.41F0.03 1.40F0.01 1.23F0.05 1.15F0.08

0.91F0.29 0.96F0.20 0.78F0.02 0.84F0.23 0.72F0.04

1.23F0.01 1.26F0.16 1.30F0.08 1.16F0.05 0.93F0.03

1.02F0.11 1.21F0.14 1.25F0.05 1.01F0.07 0.99F0.11

1.33F0.87 1.20F0.80 1.58F0.49 1.89F0.62 1.57F0.58

1.24F0.20 1.23F0.25 1.12F0.29 0.89F0.09 1.07F0.18

1.79F0.58 1.75F0.68 1.71F0.39 1.77F0.45 1.95F0.43

2.97F0.28 2.77F0.38 2.84F0.47 2.74F0.48 2.41F0.10

1.08F0.30 1.00F0.23 1.05F0.32 1.02F0.23 1.08F0.32

0.55F0.14 0.50F0.15 0.37F0.13 0.45F0.10 0.34F0.07

0.83F0.14 0.99F0.15 0.89F0.16 0.62F0.04 0.66F0.19

8.15F4.25 10.55F6.88 9.38F4.59 10.85F6.31 10.53F5.80

0.97F0.06 1.38F0.14 1.33F0.03 1.31F0.18 1.14F0.15

1.16F0.22 1.40F0.21 1.49F0.19 1.67F0.23 1.74F0.24

Table 4

Biodistribution study by dissection: comparison of [125I]-2-iodo-l-phenyl-

alanine, [125I]-2-iodo-l-tyrosine and 18F-FDG biodistribution in R1M-

bearing athymic mice (DAR values; meanFS.D.; n =3; t =90 min p.i.)

Tissue MeanFS.D.

18F-FDG [125I]-2-iodo-l-

tyrosine

[125I]-2-iodo-l-

phenylalanine

Blood 0.19F0.02 1.76F0.13 1.42F0.36

Brain 3.76F0.26 1.05F0.13 1.08F0.15

Heart 7.01F1.16 1.37F0.09 1.42F0.04

Lung 1.48F0.13 1.22F0.06 1.18F0.08

Stomach 1.04F0.11 1.96F0.17 1.49F0.27

Milt 1.10F0.11 1.66F0.18 1.33F0.06

Liver 0.58F0.03 2.06F0.05 1.32F0.08

Kidneys 0.48F0.06 4.93F0.18 2.26F0.17

Large intestine 0.97F0.08 1.48F0.06 1.25F0.02

Small intestine 1.86F0.03 0.82F0.19 0.78F0.05

Contralateral

background

1.60F1.11 2.11F0.13 1.19F0.12

Pancreas 1.62F0.17 24.72F1.75 16.81F1.59

Tumor 1.69F0.29 3.69F0.55 2.18F0.70

RTB 1.18F0.59 1.92F0.26 2.06F0.78

V. Kersemans et al. / Nuclear Medicine and Biology 33 (2006) 111–117 115

performed. At 90 min p.i., when the steady state for all

tracers was reached, the animals were killed. The organs and

tissues were removed, washed, dried and weighed. The

blood was collected and weighed. The radioactivity of the

samples was counted using the auto gamma-counting system

CobraII Series (Canberra Packard). The amount of radioac-

tivity in the organs and tissues was calculated as DAR.

3. Results

3.1. Biodistribution by DPI

DPI with [123I]-iodo-HSA showed no significant

difference (P b.05) between the blood flow in the

tumor and the contralateral reference leg: overall mean

Fig. 4. Hierarchical clustering using average linkage algorithm for the exponential

both 2-iodo-l-phenylalanine and 2-iodo-l-tyrosine.

RTB[123I]-iodo-HSA=1.1 (S.D.=0.2; n=10; 10 min p.i.;

independent of tumor volume).

Both [123I]-2-iodo-l-phenylalanine and [123I]-2-iodo-l-

tyrosine showed high and fast tumor accumulation: the

equilibrium for tumor uptake for both tracers was reached

within 10 min p.i., with an overall mean tumor DUR uptake

value of 1.19 (S.D.=0.05; n=12) and 1.24 (S.D.=0.06;

n=12) for [123I]-2-iodo-l-phenylalanine and [123I]-2-iodo-l-

tyrosine, respectively, at equilibrium between 15 and 30 min

p.i. (Fig. 1). Paired Student’s t statistical analysis showed no

significant difference in tumor tracer concentration at

equilibrium. Administration of l-phenylalanine resulted in

a significant (Pb.05) displacement of the radioactivity from

the tumor for both tracers, as represented in Table 1.

Both amino acid analogues showed fast clearance from

the blood through the kidneys to the bladder (Fig. 2),

but [123I]-2-iodo-l-phenylalanine showed higher bladder

accumulation at the same time point in comparison with

[123I]-2-iodo-l-tyrosine. Blood clearance of both tumor

tracers could be described by first-order kinetics.

The tumor uptake of [123I]-2-iodo-l-phenylalanine

and [123I]-2-iodo-l-tyrosine equilibrated with blood: both

tumor-to-blood ratios as a function of time reached a steady

state approximately at 15 min. However, the phenylalanine

analogue tumor uptake equilibrated faster with the blood as

compared with the tyrosine analogue: equilibrium was

obtained from 9.7 and 15.3 min, respectively, as measured

by the first derivatives of the tumor-to-blood ratio as a

function of time becoming zero (Fig. 3).

The results obtained by two-compartment modeling are

given in Table 2. Both [123I]-2-iodo-l-phenylalanine and

[123I]-2-iodo-l-tyrosine biodistribution by DPI fitted the

calculated curve for the proposed kinetic model (both

R2N .95). Moreover, the pharmacokinetic model illustrated

that [123I]-2-iodo-l-tyrosine was cleared almost two times

curve parameters obtained by nonlinear regression of the dissection data of

V. Kersemans et al. / Nuclear Medicine and Biology 33 (2006) 111–117116

slower from the blood in comparison with the phenylalanine

analogue. Although [123I]-2-iodo-l-tyrosine was taken up

faster in the peripheral compartment, this tracer also

disappeared faster from this compartment in comparison

with [123I]-2-iodo-l-phenylalanine.

3.2. Biodistribution by dissection

The results of the biodistribution by dissection of [125I]-

2-iodo-l-tyrosine are shown in Table 3; those of the

phenylalanine analogue were described earlier by Kerse-

mans et al. [9].

The biodistribution by dissection confirmed the results

obtained by DPI and additionally showed that the plateau for

tumor uptake for both tracers was constant, between 45 min

and 3 h p.i. Slightly elevated liver [125I]-2-iodo-l-tyrosine

uptake was detected. No significant accumulation of

radioactivity was observed in other abdominal organs such

as the lungs, stomach, small intestine and large intestine;

neither was any observed in the brain for both tracers.

The results of the comparative biodistribution by

dissection at equilibrium (90 min p.i.) are shown in

Table 4. Both amino acid tracers showed better character-

istics relating to the brain and heart uptake in comparison

with 18F-FDG, which showed elevated levels in the organs.

The results of the initial physiological modeling are

shown in Fig. 4. Five groups can be defined: tumor,

remainder of the body, stomach, blood group and other

organs group.

4. Discussion

Recent animal studies have pointed out [123I]-2-iodo-l-

tyrosine as a promising new amino acid tracer. Moreover,

this tracer could overcome the limitations of IMT (rapid

renal elimination and accumulation in the urinary tract) and18F-FDG (high and fast tumor uptake and rather low uptake

in gray matter and inflammatory lesions). Thus, [123I]-2-

iodo-l-tyrosine could be considered as one of the most

promising amino acid tumor tracers for SPECT. However, a

recent clinical evaluation of [123I]-2-iodo-l-tyrosine resulted

in one false-positive result [11]. Our research group

extensively evaluated the characteristics of radioiodinated

[123I]-2-iodo-l-phenylalanine in R1M-bearing athymic mice

and obtained promising results [7]. As a consequence, a

direct comparison between [123I]-2-iodo-l-tyrosine and

[123I]-2-iodo-l-phenylalanine was performed in detail in

larger animal groups, allowing statistical evaluation to be

able to detect (and confirm with greater statistical signifi-

cance) differences between the two tracers. Moreover,

studying more time points will allow deriving pharmacoki-

netic data and, thus, conclusions on elimination velocity

could be formulated.

Both amino acid analogues showed high and fast tumor

accumulation. This fast tumor accumulation implicates

that imaging with both amino acid tumor tracers could be

performed within a very short period, which is beneficial for

patients. Moreover, the displacement study with l-phenyl-

alanine demonstrated that, on one hand, the tumor uptake of

both tracers was specific and that, on the other, both amino

acid analogues were transported through LAT1, an obliga-

tory amino acid exchanger.

Concerning the general tracer characteristics, both amino

acid tumor tracers demonstrated a renal clearance to the

bladder without accumulation in the kidneys because both

curves showed a parallel decline (Fig. 2). Although they

followed the same excretion route, [123I]-2-iodo-l-phenyl-

alanine reached higher accumulation values in the bladder at

the same time point p.i. (Fig. 2). Moreover, [123I]-2-iodo-l-

phenylalanine tumor uptake equilibrated faster with the

blood as compared with the tyrosine analogue. Both

observations are important concerning radiation burden

and imaging.

The results obtained by two-compartment modeling

confirmed the first-order fittings of the excretion. Moreover,

our kinetic analysis confirmed a faster influx into the tumor

and a higher tumor-to-background contrast for [123I]-2-iodo-

l-phenylalanine than for [123I]-2-iodo-l-tyrosine.

The biodistribution results not only confirmed the

general tracer characteristics as observed by DPI but also

demonstrated a slightly elevated liver [125I]-2-iodo-l-tyro-

sine uptake. The latter finding could be caused by the slower

blood clearance of the tracer in comparison with [125I]-2-

iodo-l-phenylalanine together with the high blood perfusion

of this organ.

The comparative biodistribution study demonstrated that

both amino acid analogues are more suitable for brain tumor

imaging because of their low uptake in comparison with18F-FDG. Moreover, earlier experiments illustrated that

[123I]-2-iodo-l-phenylalanine and [123I]-2-iodo-l-tyrosine

showed only a minor accumulation in inflamed lesions

[6]. Comparison of [125I]-2-iodo-l-tyrosine with [125I]-2-

iodo-l-phenylalanine revealed once more the slower blood

clearance of [125I]-2-iodo-l-tyrosine, which is reflected by

the longer kidney retention of this tracer.

Besides [123I]-2-iodo-l-tyrosine, another iodo-l-phenyl-

alanine analogue is being investigated by Samnick et al.

[12,13]: [123I]-4-iodo-l-phenylalanine. However, the litera-

ture reports a high blood pool activity for this tracer,

resulting in a lowering of the tumor uptake-to-background

ratio. Moreover, [123I]-4-iodo-l-phenylalanine was taken up

in relative high amounts in inflammatory tissue, which

lowers its tumor specificity [6]. As a consequence, our

research group focused on the development of an iodo-l-

phenylalanine analogue, substituted on the 2 position,

allowing the 4 position to be free for enzymatic interaction;

our results showed a fast blood clearance of [123I]-2-iodo-l-

phenylalanine and a high tumor specificity.

The results of the first physiological model indicated that

not every organ is to be considered separately in full

pharmacokinetic modeling for both tracers because of similar

pharmacokinetic behaviors within the groups. Five groups

can be defined: tumor, remainder of the body, stomach, blood

V. Kersemans et al. / Nuclear Medicine and Biology 33 (2006) 111–117 117

group and other organs group. The stomach as an individual

group can be explained by the uptake of free 123I� after

dehalogenation (shown in Table 3). Although differences

between both tracers were observed concerning blood elimi-

nation, both tracers showed the same clustering patterns.

5. Conclusion

The aim of this study was to compare the recently

developed tumor tracer [123I]-2-iodo-l-phenylalanine with

[123I]-2-iodo-l-tyrosine to check whether the phenylalanine

analogue exhibited better in vivo characteristics over the

tyrosine analogue, which has recently entered the clinical

trial process.

This study confirmed the more promising characteristics

of [123I]-2-iodo-l-phenylalanine for oncological imaging as

compared with [123I]-2-iodo-l-tyrosine. The former amino

acid analogue not only demonstrated faster blood clearance

but also showed that the tracer uptake in the tumor reached

its equilibrium with the blood pool activity faster. Both

observations lead to faster and better tumor contrast for

[123I]-2-iodo-l-phenylalanine. Moreover, high tumor uptake

and no significant uptake of [123I]-2-iodo-l-phenylalanine

in the abdominal region and brain favor the use of the

phenylalanine analogue for tumor imaging with SPECT,

whereas [123I]-2-iodo-l-tyrosine tumor imaging shall be

slightly hindered by its liver uptake. Both tracers showed no

accumulation in the kidneys and, thus, overcome the

limitations of IMT.

References

[1] Jager PL, Vaalburg W, Pruim J, et al. Radiolabelled amino acids: basic

aspects and clinical applications in oncology. J Nucl Med 2001;42:

432–45.

[2] Isselbacher KJ. Sugar and amino acid transport by cells in culture:

differences between normal and malignant cells. N Engl J Med

1972;286:929–33.

[3] Ishiwata K, Vaalburg W, Elsinga PH, Paans AM, Woldring MG.

Comparison of l-[1-11C]methionine and l-[methyl-11C]methionine for

measuring in vivo protein synthesis rates with PET. J Nucl Med

1988;29:1419–27.

[4] Langen KJ, Pauleit D, Coenen HH. 3-[123I]iodo-alpha-methyl-l-

tyrosine: uptake mechanisms and clinical applications. Nucl Med Biol

2002;29:625–31.

[5] Jager PL, Franssen EJ, Kool W, et al. Feasibility of tumour imaging

using l-3-[iodine-123]-iodo-alpha-methyl-tyrosine in extracranial

tumours. J Nucl Med 1998;39:1736–43.

[6] Lahoutte T, Mertens J, Caveliers V, et al. Comparative biodis-

tribution of iodinated amino acids in rats: selection of the optimal

analog for oncologic imaging outside the brain. J Nucl Med 2003;

44:1489–94.

[7] Kersemans V, Cornelissen B, Kersemans K, et al. Detection of various

tumour types in athymic mice using [123I]-2-iodo-l-phenylalanine

planar SPECT. Eur J Nucl Med 2004;31(S2):S380.

[8] Mertens J, Kersemans V, Bauwens M, et al. Synthesis, radiosynthesis,

and in vitro characterization of [125I]-2-iodo-l-phenylalanine in a

R1M rhabdomyosarcoma cell model as a new potential tumour tracer

for SPECT. Nucl Med Biol 2004;31:739–46.

[9] Kersemans V, Cornelissen B, Kersemans K, et al. In vivo character-

isation of 2-iodo-l-phenylalanine in a R1M rhabdomyosarcoma

mouse model as a potential tumour tracer for SPECT. J Nucl Med

2005;46:532–9.

[10] Waterton JC, Alott CP, Pickfort R, et al. Assessment of mouse tumour

xenograft volumes in vivo by ultrasound imaging, MRI and calliper

measurement. In: Faulkner K, Carey B, Crellin A, Harisson RM,

editors. Proceedings of the 19th LH Gray Conference: Quantitative

imaging in oncology. London; 1997. p. 146–9.

[11] Keyaerts M, Lahoutte T, Everaert H, et al. Initial clinical evaluation of

123-I-2-iodo-tyrosine in patients with brain tumours. Eur J Nucl Med

2005 [Abstract 101 at EANM 2005].

[12] Samnick S, Hellwig D, Bader JB, et al. Initial evaluation of the

feasibility of single photon emission tomography with p-[123I]iodo-l-

phenylalanine for routine brain tumor imaging. Nucl Med Commun

2002;23:121–30.

[13] Samnick S, Richter S, Romeike BF, et al. Investigation of

iodine-123-labelled amino acid derivatives for imaging cerebral

gliomas: uptake in human glioma cells and evaluation in stereo-

tactically implanted C6 glioma rats. Eur J Nucl Med 2000;27:

1543–51.

[14] Thie JA. Understanding the standardized uptake value, its methods

and implications for usage. J Nucl Med 2004;45:1431–4.