Imaging Herpes Virus Thymidine Kinase Gene Transfer and … · (CANCER RESEARCH 58. 4333-4341. October I. I998| Imaging Herpes Virus Thymidine Kinase Gene Transfer and Expression

(CANCER RESEARCH 58. 4333-4341. October I. I998|

Imaging Herpes Virus Thymidine Kinase Gene Transfer and Expression byPositron Emission Tomography1

Juri G. Tjuvajev, Norbert Avril,2 Takamitsu Oku, Toshio Sasajima, Tadashi Miyagawa, Revathi Joshi,

Michelle Safer, Bradley Beattie, Gene DiResta, Farhad Daghighian, Finn Augensen, Jason Koutcher, Jamal Zweit,John llumin, Steven M. Larson, Ronald Finn, and Ronald Blasberg3

Department of Neurology Â¡J.G. T.. N. A.. T. O., T. S., T. M., R. J., M. S.. B. B., R. B.Â¡.Radinchemislry/Cyclotron Core Facilily [R. F.Â¡,Department of Radiology and MedicalPhysics Â¡J.K.. J. HJ. Nuclear Medicine Service Â¡C.D.. F. D.. F. A.. S. M. L.Â¡.Memorial Sloun-Kettering Cancer Center. New York. NeÂ« York 10021. and The Royal Marsden

Hospital. Summ. SM2 5PT United Kingdom {J. ZJ

ABSTRACT

We report a series of studies that assess the feasibility and sensitivity ofimaging of herpes virus type one thymidine kinase (HSVl-tk) gene transfer and expression with [l24I]-5-iodo-2'-fluoro-l-ÃŸ-D-arabinofuranosyl-uracil ([I24IJ-FIAU) and positron emission tomography (PET) and theability of [I24I]-FIAU-PET imaging to discriminate different levels of

HSVl-tk gene expression. Studies were performed in rats bearing multiple

s.c. tumors derived from W256 rat carcinoma and RG2 rat glioma cells.In the first set, we tested the sensitivity of [I24I]-FIAU-PET imaging to

detect low levels of HSVl-tk gene expression after retroviral-mediatedgene transfer. HSVl-tk gene transduction of one of preestablished wild-

type W256 tumor in each animal was accomplished by direct intratumoralinjection of retroviral vector-producer cells (W256â€”Â»W2S6TK*tumors).

Tumors produced from W256 and W256TK+ cells served as the negativeand positive control in each animal. Highly specific images of [I24I]-FIAU-

derived radioactivity were obtained in W256TK* tumors (that were transduced in vivo) and in W256TK+ tumors but not in nontransduced wild-type W256 tumors. The level of "specific" incorporated radioactivity in

transduced portions of both W256TK* and W256TK+ tumors was rela

tively constant between 4 and 50 h. In the second set, we tested whether[I24I]-FIAU and PET imaging can measure and discriminate between

different levels of HSVl-tk gene expression. Multiple s.c. tumors wereproduced from wild-type RG2 cells and stably transduced RG2TK celllines that express different levels of HSVl-tk. A highly significant relationship between the level of [I24I]-FIAU accumulation [% injected dose/g

and incorporation constant |A/>| and an independent measure ofHSVI-tkexpression (sensitivity of the transduced tumor cells to ganciclovir, l( '^,1

was demonstrated, and the slope of this relationship was defined as asensitivity index. We have demonstrated for the first time that highlyspecific noninvasive images of HSVl-tk expression in experimental animaltumors can be obtained using radiolabeled 2'-fluoro-nucleoside [I24I]-

FIAU and a clinical PET system. The ability to image the location (distribution) of gene expression and the level of expression over time provides new and useful information for monitoring clinical gene therapyprotocols in the future.

INTRODUCTION

In vivo imaging of transgene expression is a new and rapidlydeveloping field. In the original papers from our group by Tjuvajev,et al. (l, 2), we described a paradigm for noninvasive imaging oftransgene expression that requires the appropriate combination of"marker gene" and "marker substrate". The marker gene product (an

enzyme) converts marker substrate to a metabolite that is selectively

Received 4/29/98; accepted 8/3/98.The costs of publication of this article were defrayed in part hy the payment of page

charges. This article must therefore he hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1This work was supported in part hy NIH Grants ROI CA60706. ROI CA69769. and

DOE 86ER60407 and the Gershel Foundation.~ Current address: Department of Nuclear Medicine. Technical University of Munich.

Klinikum rechts der Isar, Ismaningcrstrasse 22. 81675. Munich. Germany.1To whom requests for reprints should be addressed, at Department of Neurology,

K923. Memorial Sloan-Keltering Cancer Center, 1275 York Avenue, New York, NY

10021.

trapped within the transduced cell. The paradigm is essentially an invivo radiotracer enzyme assay.

To demonstrate the feasibility and implementation of the paradigm,we described model systems in tissue culture and in experimentalanimals using the HSVl-tk4 gene as a marker gene and radiolabeled

FIAU as a marker substrate. We have demonstrated that highlyspecific images of HSVl-tk expression in experimental animal tumorscan be obtained using radioiodinated ["'l|-FIAU and a clinical

gamma camera system (2) and a quantitative relationship betweenFIAU accumulation in //SV7-/Â£-transduced cells and independentmeasures of HSVl-tk expression ( 1).

Several groups have synthesized and tested a number of differentradiolabeled marker substrates for imaging HSVl-tk gene expression

and have reported on similar results. Morin, et al. (3) described thesynthesis and in vitro uptake of 2'-substituted analogues of (E)-5-(2-['25I]iodovinyl)-2'-deoxyuridine in tumor cells transduced with

HSVl-tk gene. These compounds are similar to FIAU but have theiodovinyl substitution in the 5-position of uracil moiety. It was demonstrated that (E)-5-(2-['2SI]iodovinyl)-2'-fluoro-2'-deoxyuridine(['"IJ-FIVRU) accumulated in //SW-rA--transduced KBALB murinesarcoma cells in vitro better than (E)-5-(2-['25I]iodovinyl)-2'-fluoro-2'-arabinouridine ([12SI]-FIVAU) and (E)-5-(2-['25I]iodovinyl)-2'-ar-abinouridine ([I25I]-IVAU), but no comparison was done with radio

iodinated FIAU. Recently, the same group demonstrated successfulimaging of HSVl-tk expression in retrovirally transduced murinetumors using a gamma camera and (E)-5-(2-[l:11I]iodovinyl)-2'-fluoro-2'-deoxyuridine (["'I]-FIVRU: Ref. 4).

Acy Io-derivati ves of guanosine labeled with li!F were also sug

gested as marker substrates for imaging HSVl-tk expression with

PET. Srinivasan, et al. (5) reported on the synthesis and PET imagingof the in vivo uptake of 1KF-labeled acyclovir in rodent liver trans

duced with adenoviral vector bearing the HSVl-tk gene. Recently, thesame group reported on the synthesis of lxF-labeled GCV (6) and'"F-labeled PCV and demonstrated that lxF-labeled PCV had betterimaging properties (~2-fold greater accumulation) than IKF-labeledGCV for imaging HSVl-tk expression (7). The '"F-labeled acyclovir,lxF-labeled GCV, and '"F-labeled PCV are radiofluorinated in 8-position of guanosine. Another approach to '"F-labeling of GCV in

the acyclo-portion was explored by Goldman, et al. (8), who reportedon the synthesis and in vivo uptake of lxF-labeled 9-l{(l-["<F]fluoro-

3-hydroxy-2-propoxy)methyl}guanine in rats with intracerebral//SV7-/Â¿--transduced 9L gliomas (8). A different radiolabeling proce

dure was developed by Alauddin, et al. (9), who reported on thesynthesis of lxF-labeled 9-(3-[lsF]fluoro-l-hydroxy-2-propoxymeth-

yl)guanine and its uptake in HT-29 transduced with the HSVl-tk gene(10). The pharmacokinetics of the latter compound, [ISF]FHPG, has

4 The abbreviations used are: HSVl-tk, herpes virus type one thymidine kinase: PET.positron emission tomography; FIAU. 5-iodo-2'-fluoro-t-ÃŸ-D-arabinofuranosyl-uracil;

PCV, pencyclovir: GCV. ganciclovir: CD. cytosine deaminase; 5FC. 5-fluoro-cytosine;

Ki. incorporation constant; MR. magnetic resonance; MRI. magnetic resonance imaging:HPLC, high-performance liquid chromatography; ROÃ•.region of interest: %\D. percent ofinjected dose; 1RES, internal ribosome entry site; lUdR, 5-iodo-2'-deoxyuridine.

4333

Research. on February 14, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

PET IMAGING OF A REPORTER GENE EXPRESSION IN VIVO

been also studied in primates and was found to be suitable for imagingHSVl-tk expression in vivo (11).

A different marker gene-marker substrate combination has also

been explored. Haberkorn, et al. (12) examined the potential forimaging the Escherichia coli CD expression with radiolabeled 5FC.This effort was unsuccessful because E. coli CD converts 5FC to5-fluorouracil (5FU) which is able to diffuse out of the cell. The E.coli CD-5FC reaction product (5FU) is not trapped within the cell,

and, thus, the measured activity in the tissue does not reflect the levelof E. coli CD expression.

In this study, we: (a) assessed the feasibility and sensitivity of theimaging of HSVl-tk gene transfer and expression in vivo with [124I]-

FIAU and PET and extended our initial reports (13, 14); (b) tested theability of PET to discriminate different levels of HSVl-tk gene expression; (c) demonstrated stable accumulation of [124I]-FIAU intransduced tumors; (d) calculated a steady-state [I24I]-FIAU incorporation (plasma clearance) constant, Ki; (e) determined the [I24I]-

Fl&U-HSVl-tk sensitivity index in RG2TK+ tumors; and (/) assessedthe radiation dosimetry of [I24I]-FIAU after i.v. administration.

MATERIALS AND METHODS

Cell Cultures. W256 rat mammary carcinoma cells were obtained fromAmerican Type Tissue Culture Collection (Rockville, MD); the RG2 ratglioma cell line was kindly provided by Dr. D. Bigner (Duke UniversityMedical Center, Durham, NC). Both cell lines were grown as monolayers inMEM supplemented with 10% FCS. The gp-STK-A2 vector-producer cell line,

which produces the recombinant replication deficient STK retrovirus containing the NeoR gene and HSVl-tk gene was described previously (15). Thegp-STK-A2 cells also were grown as monolayers in DMEM supplemented

with 10% FCS. The in vitro transduced W256TK+ and RG2TK+ cells as wellas the single cell clones of RG2TK+ cells (RG2TK#4, RG2TK#I6, andRG2TK#32) were grown in MEM supplemented with 10% PCS and 250fig/mi G418. These stably transduced cell lines have various levels of HSVl-ik

expression and were characterized by us previously (1, 2). The cells werereleased from the culture plates by treatment with 0.5% trypsin in PBS for5-10 min, resuspended in the growth media to neutralize the trypsin, and

passaged. To produce a cell suspension for injection, cells were centrifuged toobtain the cellular pellet after neutralization of trypsin and then resuspended inMEM (without FCS) at concentration of IO6 viable cells in 100 Â¿il.

s.c. Tumors. The experimental protocol involving animals was approvedby the Institutional Animal Care and Use Committee of the Memorial Sloan-Kettering Cancer Center. Tumor cells (IO6 cells in 100 /Â¿I)were injected s.c.

into RNU-rnu rats (HarÃanSprague Dawley. Indianapolis, IN) weighing 200-

250 g.Two groups of rats were studied. In the first group, three s.c. tumors were

produced in each rat: the W256TK+ cells (5 X IO6 in 100 Â¿il)were injected

s.c. in the dorsal part of the left thigh; and two wild-type W256 tumors wereproduced by s.c. injection of W256 cells (5 x IO6 cells in 100 Â¡u)into the

dorsal part of the right thigh and the dorsum of the neck, respectively. Thewild-type s.c. tumors in the neck were allowed to grow until 0.6-0.7 cm in

diameter, and then they were transduced in vivo by intratumoral injection ofgp-STK-A2 vector-producer cells (10" cells in 100 Â¿il).

In the second group, five s.c. tumors were produced in each rat: theRG2TK+ cells (5 x IO6 in 100 fil) were injected s.c. in the central region of

the back; a similar injection of the RG2TK#4 and wild-type RG2 cells was

performed in the dorsal side of the left and right thighs, respectively; similarly,the RG2TK#16 and RG2TK#32 cells were injected into the left and rightshoulders, respectively.

MRI Studies. MRI was performed on a Bruker Omega CSI system(Bruker, Fremont, CA) with a 4.7-tesla, 33-cm bore magnet equipped with

shielded gradients and a standard imaging coil as described previously (16).Briefly, the rats were anesthetized with the mixture of 1.5% isoflurane in 70%nitrous oxide and 30% oxygen (vol/vol) and placed in the radiofrequency coil;the s.c. tumors were positioned at the center of the solenoid resonator. Theanesthetic gas mixture was constantly supplied to the animal via the tubingattached hermetically to the holder system throughout the imaging session.

T2-weighted images were obtained through the area of the s.c. tumor using a256 X 128 matrix, 100-mm field of view, and standard echo sequences (TR,

3500 ms; TE, 40 ms; and 2 excitations/phase encoding step). The MRI studieswere performed 1 day before radiotracer injection. Tumors were detected asthe areas of hyperintense T2 signal.

No-carrier-added Synthesis [124I]-FIAU. No-carrier-added I24Iwas pro

duced in U.K. (by J. Z.) from an enriched tellurium oxide target that afterirradiation was subjected to dry distillation to release the trapped radionuclide.The isotope was collected in dilute sodium hydroxide solution and shipped toMemorial Sloan-Kettering Cancer Center. No-carrier-added [I24I]-FIAU wasprepared (R. F.) by reacting 2'-fluoro-2'-deoxy-l-/3-D-arabinofuranosyl-uracil

(FAU; Moravek Biochemicals, Brea, CA) with carrier-free I24I using iodogen

followed by isolation of the product by column chromatography. The [124I]-

FIAU injection solution was prepared by evaporation of the methanol-elutedfraction from the CIS cartridge and dissolving the residue in a sterile pyrogen-

free physiological saline solution and passage through a sterile nonpyrogenic0.22 firn Millipore filter. Radiochemical purity of [124I]-FIAU was assessed

using HPLC system consisting of a HPXL Pump (Rainin, Woburn, MA) andFlo-One Beta detector Series 100 (Radiomatic, Meriden, CT). Data were

collected and analyzed using online Dynamax software (Rainin). HPLC conditions included the following: (a) a 250 X 4.6-mm reverse phase 10-fim CIS

Maxsil column (Phenomenex. Terranee, CA); (b) the isocratic mobile phase of10% methanol in water and at a flow rate of 1.5 ml/min. The results ofradio-HPLC analysis of the [124I]-FIAU synthesis yielded a >95% purecompound. The theoretical specific activity of [I24I]-FIAU produced by a

no-carrier-added synthesis can be calculated from:

Specific Activity = Av/(R X tl/2/ln2). (A)

where Av is Avogadro's number, R is the conversion factor 3.7 X 10'Â°Bq/Ci,

and /'/.. is the physical half-life of the radionuclide in seconds. For [I24I]-FIAU,the theoretical specific activity is ~3.2 X IO7 Ci/mol.

PET. No-carrier-added [124I]-FIAU (25-50 /xCi/animal) was injected i.v.

21 days after s.c. inoculation of //SW-//t-transduced tumor cells (e.g.,W256TK+ or RG2TK+) and wild-type W256 and RG2 tumor cells. In vivo

transduction of the preestablished s.c. W256 tumors growing in the neck wasperformed 14 days before [I24I]-FIAU administration. One day before [I24I]-

FIAU administration, all animals received a 2-ml i.p. injection of a 0.9% Nal

solution to block thyroid uptake of radioactive iodide.PET was performed using an Advance PET tomograph (General Electric,

Milwaukee, WI). Resolution of this PET scanner is about 5 mm (full with halfmaximum). Transmission and emission scans were obtained in three-dimen

sional mode; the emission scans were corrected for random counts, dead time,and attenuation. Duration of the transmission scans was 10 min, and theduration of the emission scans ranged from 40-60 min. Images were reconstructed with filtered back-projection (Hanning filter with cutoff frequency of

0.5 cycles/pixel) yielding a slice thickness of 4.2 mm. Sequential images wereobtained at 4, 24, and 36 or 48 h after [124I]-FIAU injection to assess thekinetics of [I24I)-FIAU accumulation and washout. A set of '24I reference

standards of different volumes (1-20 ml) and different radioactivity concen

tration was placed within the field of view of the PET scanner. These standardswere used for quantitation of the images and to account for differences inimaging time and radioactive decay.

ROI measurements were performed on the transaxial slices of the PETimages. Tumor radioactivity concentration (%ID/g) was determined by dividing total tumor radioactivity by the tumor mass. The total tumor radioactivitywas determined from ROIs placed around the tumor slices where the tumorwas visible, and total radioactivity was calculated by summing the ROIactivities obtained from each slice and corrected for background activitymeasured in remote area. Tumor mass was determined postmortem by measuring the tumor weight. This method for determining mean tumor radioactivityconcentration was used to avoid partial-volume effects (17).

In RG2 tumor-bearing animals, venous blood samples were obtained at 1min, IO min, and at 1,4, 20, and 32 h after [124I]-FIAU administration and

plasma radioactivity measured in a Packard 5500 gamma spectrometer (Packard, Meriden, CT). An estimate of the plasma radioactivity input function(lower limit) was obtained and an approximate tissue incorporation constant(Ki, /il/g/min) of FIAU was calculated using the following formula:

4334




Ki Cdr (B)

where the A, is tissue radioactivity at time / and Cp is plasma radioactivity.To estimate organ dosimetry of radioiodinated ['^IJ-FIAU and to confirm

the validity of the PET measurements of tumor radioactivity, the tumors andselected organs were sampled, weighed, and directly assayed for radioactivityusing a Packard 5500 gamma spectrometer (Packard).

RESULTS

PET Imaging of HSVl-tk Expression in Transduced W256Tumors. In the first group of animals (n = 5). PET images of[124I]-FIAU accumulation in W256, W256TK+, and W256TK* tu

mors were obtained at 4, 30, and 50 h after the administration of 25fxCi of [124I]-FIAU per animal. These images define the early distri

bution and washout of FIAU from the body of the rat and provide adirect comparison between in vivo transduced W256TK*, in vitrotransduced W256TK+ (positive control), and wild-type W256 tumors

(negative control). Highly specific accumulation and retention ofFIAU-derived radioactivity was observed in the areas of W256TK*

and W256TK+ tumors at 30 h and 50 h after the administration of[124I]-FIAU. The pattern of distribution of retained radioactivity was

heterogeneous in both W256TK* and W256TK+ tumors (Fig. 1). Theaverage level of radioactivity retention was high (~0. l-0.2%ID/g) inW256TK* and W256TK+ tumors at 50 h after [124I]-FIAU admin

istration, whereas the wild-type W256 tumors showed substantiallyless retention (~0.01%ID/g) of radioactivity (Fig. 2A).

The washout of [124I]-FIAU and [l24I]-FIAU-derived radioactivity

from nontransduced (wild-type) W256 tumors was rapid and essen

tially monoexponential between 4 and 50 h: the clearance half-timewas 15 h (Fig. 2A). In contrast, the washout of |I24I]-FIAU and[124I]-FIAU-derived radioactivity from transduced W256TK* and

W256TK+ tumors was much slower and appeared to be bicxponen-

tial (Fig. 2A). If the level of radioactivity measured in nontransducedtumors is assumed to be nonspecific and a similar level of nonspecificlevel of radioactivity is also present in transduced tumors as well, itcan be subtracted from the total radioactivity measured in transducedtumors to obtain an estimate of "specific" radioactivity due to

HSV1-TK (thymidine kinase enzyme) activity. The specific (HSVl-/A:-dependent) component of measured radioactivity in W256TK* and

W256TK+ tumors was essentially constant between 4 and 50 h after[I24I]-FIAU administration; the levels were 0.11 Â±0.03%ID/g and

0.05 Â±0.01%ID/g for W256TK+ and W256TK*. respectively (Fig.

2ÃŸ).Also of note, the specific (//5W-/Â£-dependent) component of

measured radioactivity accounted for 64.9 Â±5.0% and 75.1 Â±13% oftotal radioactivity measured at 30 and 50 h in W256TK* tumors,

respectively, and the values were even higher for W256TK+ tumors(75.8 Â±7.4 and 89.9 Â±13.0%. respectively). At earlier times, thespecific component of measured radioactivity was substantially lower;only 27.3 Â±7.1% of total radioactivity at 4 h in W256TK* tumors

(51.1 Â±4.8% in W256TK+ tumors at 4 h; Fig. 2C).To more precisely assess radioactivity distribution in W256TK +

and W256TK* tumors. T2-weighted MR images of the rats were used

as a morphological reference for coregistration with the PET images(Fig. 3). The focal areas of high radioactivity retention in W256TK +tumors colocalized with the viable portions of these tumors: there wasno accumulation of radioactivity in the necrotic portions of thesetumors. In W256TK* tumors, the focal areas of high radioactivity

accumulation corresponded to the central areas of the tumors, the siteof implantation of the retroviral vector-producer cells. The level of

In VivoTransducedW256

In VitroTransducedW256TK+

Fig. I. PET imaging of M.ST/-/ÃŒgene expression al 48 h after i.v. administration of [ I|-FIAl'. Panel A. pholograph of a tumor-hearing rat. Panel H. iranvmal Phi images ot

wild-type W256 and in viiro transduced W256TK + s.e. tumors located in bolh thighs of the rat. Panel C. summed coronal (pseudo-planar) PET image of the tumor-bearing rat. Theoutline of the rat is shown in white. Small Imri-oniul lines on the left and right sides of the outline indicate the planes in which the Iransaxial images were obtained. Panel D. transaxialPET images of a W256TK* tumor in the neck region (produced by retroviral-mediated HSVl-tk gene transduction of preestablished wild-type s.e. W25o tumor). The images arecolor-coded to a range of [''4l]-FIAU-derived radioactivity levels, expressed as %lD/g (% Diise/a in the figure).

4335




O.I

0.01

Â»*Â»â€¢¿�i ' *â€¢'*â€”aâ€” W256TK.

â€¢¿�*-W2S6TK---â€¢-â€¢W2SÂ»bâ€¢â€”'--â€¢â€¢;^^

^â€¢^-^"^

~-~"""1*Â»"-s.-

â€¢¿�H"~N...0

10 20 30 40 SO6Time

(hours)

|124I|-FIAURÂ»dioÂ»ctivity(%dose/g)1

Â£â€”

Bâ€” WÃŽHTKttorr 1-0â€” W256TK"torr1fâ€”h,.

.=_.â€”

1â€”â€ži1:â€”41

10 20 30 40 506Time

(hours)

B

20 30 40

Time (hours)

Fig. 2. Time course of |'-4Il-FIAU-derived radioactivily in differenl s.c. lumors as measured by PET imaging. W256TK+ (D). W256TKÂ« (O). and wild-type W256 lumors (â€¢).Dala arc presented as mean Â±SD. Panel A. time course of radioactivity concentration (% dase/g in the figure). Panel 3. time course of "specific" radioactivity (% dose/g) reflectingHSVl-tk enzyme activity. "Specific" radioactivity (SR) was calculated as SR = measured radioactivity (total) - "nonspecific" radioactivity, where nonspecific radioactivity wasmeasured in the nontransduced wild-type tumors. Panel C, time course of percent "specific" radioactivity CM/f) where %SR = \SR ^ measured (total) radioactivity] X 100.

Fig. 3. Coregistration (fusion) of T2-weighted MR and [ I241]-FIAL'-PET Â¡magesof HS\'l-tk cene expression. Panels .t and B show the T2-weighted MR Â¡mages.Panels C and 1)shun the [I:JI1-FIAU PET images. Panels Â£and F demonstrate the co-registered (fused) PET/MR1 images. The in 171/1transduced W256TK* tumor is located in the dorsal portion

nl the neck i.\i and the W256TK+ and wild-type W256 s.c. tumors are located in the left and right thighs, respectively (fl). The tumors are usu.ili/eil as an increased T2 signal. The[ l24Il-FIAl'-derived radioactivity is largely locali/ed to the viable-appearing areas of W256TK+ tumor ili. />. /â€¢*).whereas the necrotic portions ol this tumor are visuali/.ed as the areas

with substantially lower levels of retained radioactivity. The necrolic regions were also confirmed on histolÃ³gica! examination of the tumor. The level of radioactivity of the wild-typeW256 tumor was very low and similar to that in the rest of the body. The focal areas of high I'^IJ-FlAU-derived radioactivily accumulation in the W256TK* tumor (A. C, Â£1corresponded to the cenlral region of the W256TK* tumor, the site of implantation of the retroviral vector-producer cells.

4336




Fig. 4. PET imaging of various levels of HSVl-ik gene expression at 3fi h after i.\. administration of [ Ij-FIAl'. Panel A. photograph of a rat hearing multiple s.c. tumors \\ ith

differenl levels of HSVI-ik gene expression (the s.c. tumors are outlined in black). Panel B, summed coronal (pseudo-planar) PET image of the rat bearing multiple s.e. tumors withdifferent levels of HSVl-ik gene expression. Smalt hori-onral lines on the right sitie of the image indicate the planes in which the transaxial images were obtained (see panel O. inpanel C. these PET images are color-coded to a range of values for [t24I]-FlAU accumulation (Ki. Â¿tl/g/min)and to levels of predicted GCV sensitivity (GCVIC50, // :â€¢ml. extracellular

fluid level) of the s.c. tumors.

activity gradually decreased toward the periphery of these tumors.Microscopic examination of the transduced W256TK* neck tumorsfailed to reveal any evidence of the viable vector-producer cells

(murine fibroblasts) that could contribute to the accumulation ofFIAU in W256TK* tumors.

PET Imaging of HSVl-tk Expression in Transduced RG2 Tumor Clones. To assess whether PET imaging with [I24I]-FIAU of the

HSVl-tk gene-transduced tumors is sufficiently sensitive to discriminate between different levels of HSVl-tk gene expression, a secondgroup of animals (n = 4) was studied. Each animal contained fivedifferent s.c. tumors produced from: (a) wild-type RG2 cells: (b) a

mixed population of in vitro transduced RG2TK+ cells; and, fromthree single cell-derived clones, (c) RG2TK#4; (</) RG2TK#16: and

(e) RG2TK#32. The implantation of multiple tumors per animalprovides the opportunity for a better comparison because all of thetumors are exposed to the same levels of [I24I]-FIAU in blood plasma(same input function). The animals were injected with [124I]-FIAU

when the tumors grew to approximately 1-1.5 cm in diameter, andPET images were obtained at 32 h after [124I]-FIAU administration

(Fig. 4). Highly specific levels of FIAU accumulation (>0. l%ID/g)were observed in the areas of the RG2TK+, RG2TK#4, andRG2TK#16 tumors. RG2 wild-type tumors and tumors derived fromthe low-expressor clone RG2TK#32 showed no substantial accumu

lation above background radioactivity (<0.01%ID/g).Radioactivity concentration (%ID/g) in tumors was also assessed

by tissue sampling and by gamma spectroscopy after the animals werekilled to confirm the validity of the PET measurements. There was a

reasonably good correlation between the levels of [124I]-FIAU-

derived radioactivity (%ID/g) obtained by quantitation of the PETimages and those obtained by the direct tissue sampling and gammaspectroscopy (Fig. 5). PET imaging of W256 tumors tended to produce somewhat lower values than tissue sampling and gamma spectroscopy, which may reflect the large size and central necrosis of theW256, W256TK+. and W256TK* tumors and because the gammaspectroscopy measurements included only viable-appearing portions

of the tumors. The comparison between the PET and well countermeasurements was better for the higher activity RG2TK+ tumorclones than the low activity clone RG2TK#32 and wild-type RG2

tumors.Sequential blood samples were obtained in these animals and

plasma radioactivity (%ID/ml) was plotted against time after [I24I]-FIAU administration (Fig. 6/1). The kinetics of ['24I)-FIAU elimina

tion from blood was biexponential. The initial plasma clearancehalf-time of [124I]-FIAU was 1.1 h: the second clearance half-time

was 4.4 h.The 32-h tissue radioactivity measurements were normalized to the

plasma concentration time course (input function) to obtain tissueincorporation constants, Ki (see Equation B). The range of [124I]-

FIAU accumulation, expressed as Ki in RG2TK + , RG2TK#4, andRG2TK#16 tumors varied from 0.375 Â±0.071 to 1.03 Â±0.325/xl/min/g. In the RG2TK#32 (the low expressor clone) and wild-type

RG2 tumors, the Ki values were significantly and substantially lower,0.028 Â±0.008 and 0.034 Â±0.009 jil/min/g, respectively (Fig. 4). Agood correlation was also observed between the Ki values for FIAU

4337




Il"IfEx. c

â€¢¿�it 0.01

0.0010.001 0.01 0.1 l

11241-1 IAl Accumulation

(%dose/g by tissue sampling)Fig. 5. Comparison of |':4I]-F1AU radioactivity CJt Â¡Ime/gin the figure) in HSVI-tk-

transduced and in wild-type W256 and RG2 s.c. tumors (obtained by tissue sampling andgamma spectroscopy) with those obtained by PET imaging (mean Â±SD). The W256 andlow activity points fall below the line of idenlity. indicating that the PET measurementsunder estimate the levels of radioactivity assessed by tissue sampling.

calculated using from the PET imaging and tissue sampling data(Fig. 65).

Relationship between the FIAU Accumulation and anIndependent Measure of HSVI-tk Expression. To further assesswhether PET imaging of the HSVI-tk gene expression in transducedtumors reflects the level of HSVI-tk gene expression, we comparedthe level of [124I]-FIAU accumulation measured at 32 h C/HD/g) and

the FIAU incorporation constant (Ki) to an independent measure ofHSVl-ik expression, GCV sensitivity (IC5()), in the cell lines that were

used to generate these tumors. Highly significant relationships wereobserved between the level of [124I]-FIAU accumulation (%ID/g and

Ki) and the GCV sensitivity (IC5()) of the corresponding cell lines(Fig. 7). The slope of the relationships in Fig. 7 constitute a sensitivityindex for the FIAU-PET image, and this slope can be used to generatea color-coded scale bar for a parametric image that reflects GCV

sensitivity (ICM,) of the tumors (Fig. 4).

DISCUSSION

In the present study, we demonstrated that retroviral-mediatedHSVI-tk gene transfer (by intratumoral injection of retroviral vector-producer cells into the preestablished wild-type tumors) results inlevels of HSVI-tk expression that can be measured by noninvasive

imaging. These results were obtained with a clinical PET camera anda dose of [124I]-FIAU similar to that which would be used in patients.Highly specific images of [l24I]-FIAU-derived radioactivity, localized

to the areas of HSVI-tk expression, were obtained in W256TK*

tumors (which were transduced in vivo) and in W256TK+ tumors(which served as a positive control). In contrast, the level of radioactivity retained was low in nontransduced wild-type W256 tumors

(which served as a negative control).One important observation from the serial imaging studies was that

the level of specific incorporated radioactivity in transduced portionsof both W256TK* and W256TK+ tumors was relatively constantbetween 4 and 50 h (Fig. 2B). This illustrates the advantage of "late"imaging (24 h or more after [I24I]-FIAU administration) and is con

sistent with our earlier results demonstrating that FIAU-derived radioactivity in A/SW-/&-transduced tumor tissue is acid insoluble (i.e.,

incorporated into macromolecules such as DNA; Ref. l). In contrast,the level of [124I]-FIAU-derived radioactivity in wild-type W256

tumors and nontransduced tissues of the body followed plasma radioactivity and declined exponentially.

Because of relatively low levels of radioactivity in nontransducedportions of the W256TK* tumors, wild-type W256 tumors and otherbody tissues, a more precise localization of retained [I24I]-FIAU-

derived radioactivity could be obtained by coregistration of corresponding PET and MR images. The T2-weighted MR images of the

rats were used as a morphological reference for the coregistered PETimages. The coregistered images revealed that focal areas of highradioactivity retention in W256TK+ tumors corresponded to viable-

appearing portions of these tumors on the MR image. As expected,little or no radioactivity accumulated in the necrotic portions of theW256TK+ tumors. In W256TK* tumors, focal areas of high radio

activity accumulation corresponded to the site of implantation of theretroviral vector-producer cells in the central areas of the tumors. The

level of radioactivity gradually decreased toward the periphery ofW256TK* tumors. This pattern of radioactivity distribution (HSVI-tk

10Â£

a-t!E-o

oCB"^*

Â¿0.1U

m7E1

â€¢¿�-o.oi0.001C1â€¢

\'*....%Â».rnV

^x.

."'k. . i-x*â€¢..510 15 20 25 303.A

Time (hours)

10

tuAi

Si

0.1

0.010.01 0.1 l

ÃŸ |124I|-FIAUKi(ul/min/g,Sampling)10

Fig. 6. Panel A. lime course of |124I)-FlAU-derived radioaetivity concentration (^ dose/ml) in arterial blood plasma in rats bearing the wild-type and HSVI-/^'-transduced s.c. RG2tumors (mean Â±SD). A biexponential clearance of radioactivity from blood plasma is observed. Panel B, comparison of ['-4I]-F1AU accumulation (Ki, /Â¿1/g/min)values (see Equation

B) using the data obtained by PET imaging and tissue sampling (see Fig. 5). The regression line (not shown) deviates slightly from the line of identity.

4338



PET IMAGING OF A REPORTER GENE EXPRESSION IN VIVOy= 0.089779â€¢*â€¢(-0.42761)R= 0.64605

y = 0.59194â€¢*X-059091) R= 096829

1J

Accumulaa

Â»Â«|0.-,i-

T"01

0â€¢<.1iâ€¢risi ix^10UM);'*'1000

lU"S3asm

' nin0.r-

f IM!M|01

0r

Â» iuni]{V.1â€” >-"â€¢\j\i1

1-i

â€”¿�r-T"i"\sf0

1â€¢;"i

a10

1000

Ganciclovir IC50 (ug/ml) Ganciclovir IC50 (ug/ml)

BFig. 7. Relationship between the [I24IJ-FIAU accumulation and an independent measure of the HSVI-tk expression (GCV sensitivity). The level of [I24I]-FIAU accumulation (%

dose/g; Panel A) and the F1AU incorporation constant (Ki: Panel ÃŸ)were measured at 32 h after [I24I]-FIAU administration and compared to the GCV sensitivity (ganciclovir IC50)of corresponding W256- and RG2-derived tumor cell lines. Highly significant relationships were observed between the level of [I24I)-FIAU accumulation (% dose/g and Ki) in vivoand GCV sensitivity (ganciclovir IC50) of corresponding cell lines in vitro. A "sensitivity index" for the FIAU-PET image is obtained from the slope of the relationships presented

in both panels.

gene expression) reflects a more limited (focal) transduction of tumortissue after implantation of retroviral vector-producer cells. The

coregistration of PET and MR images provides a better opportunity tolocalize HSVI-tk gene expression in particular tumor or body struc

tures.We also verified that the high levels of radioactivity in the center of

W256TK* tumors were not due to accumulation of [124I]-FIAU by

surviving vector-producer cells. Microscopic examination of transduced W256TK* neck tumors failed to reveal any evidence of the

viable vector-producer cells that may have contributed to such patternof [124I]-FIAU accumulation. Our previous studies (1,2, 14) and other

reports (18) have demonstrated that retroviral vector-producer cells(murine fibroblasts) do not survive more than 7-10 days after intra-

tumoral injection into the rat host.Another important issue addressed in this study is whether [124I]-

FIAU and PET imaging can quantitatively discriminate between different levels of HSVI-tk gene expression; specifically, does the levelof [I24I]-FIAU radioactivity in the transduced tumors correspond to an

independent measure of HSVI-tk gene expression. We demonstrated

previously ( 1) a relationship between the level of FIAU accumulationin transduced cells in culture and two independent measures (HSVI-tkmRNA level and GCV sensitivity) of HSVI-tk gene expression. Asimilar relationship between the level of [124I]-FIAU accumulation in

tumors derived from stably transduced cell lines and GCV sensitivity(IC5()) of these cell lines was demonstrated in Fig. 7. The slope of therelationship shown in Fig. 7 provides a sensitivity index that relatesthe radiotracer assay to an independent functional measure of HSVI-tk

expression. Although this relationship compares an in vivo measurement of radioactivity with an in vitro assessment of sensitivity toGCV, the slope provides a useful number for comparing studiesinvolving different HSVI-/Â¿-expressing tumors as well as differentradiolabeled substrates for measuring HSVI-tk expression. More im

portantly, the sensitivity index provides a measure that can be compared to future studies using different //SW-rt-containing vectors and

different gene expression cassettes.A question we addressed previously (2, 19) was whether PET and

gamma camera (or single-photon emission computed tomography)instrumentation could image clinically relevant levels of HSVI-tkexpression. Establishing the lower limit sensitivity of [124I]-FIAU-

PET imaging helps to define whether clinically relevant levels ofHSVI-tk gene transduction and expression can be imaged. A clinicallyrelevant level of HSVI-tk expression in a transfected tumor tissue

would be one that results in a significant antitumor response to GCVtherapy. We calculated that the level of [124I]-FIAU radioactivity at

24 h in transduced tumor tissue expressing clinically relevant levels ofHSVI-tk will be approximately 0.07-0.12 jxCi/cc for a 2-mCi dose (2,19). Based on these tissue concentrations of [l24I]-FIAU-derived

radioactivity, we estimated that the lower limit sensitivity for imagingwith the present generation of PET tomographs with capability ofsepta-in and septa-out imaging would be at least 100-fold lower. To

illustrate this point, we have generated parametric images of predictedGCV sensitivity (IC50, extracellular fluid level) for the s.c. tumorsshown in Fig. 4. This parametric image is based on the color-coded

GCV IC50 scale bar. The GCV IC5(, scale bar assumes a correspondence between the response of transduced cells to different GCV

Table I Levels of radioactivity in organs and tissues at 36 h after i.v. administration

Organ/TissueBlood

plasmaBrainMuscleGonadsHeartStomachBladderLiverKidneyLarge

intestineSpleenSmall

intestineBonemarrowThyroid

(blocked)Average0.00890.00040.00190.00230.00350.01

110.01180.01290.01660.01790.01840.01920.02111.1133%

ID/gSD0.00640.00010.00050.00130.00110.00610.01350.00190.00200.00630.00330.00260.00261.3034

4339




Table 2 Dosimetri estimates for Â¡ '/1-lUilR

' From Daghighian el ill. (27).

Radiation Exposure (cSv/37MBq or mrem/mCi)

OrganBloodBrainThyroid

(blocked)HeartLiverSpleenStomachLarge

intestineKidneyGonadsMuscleBone

marrow[124I]-IUdR0.433"0.0530.610.160.210.150.170.340.290.230.210.173"[I24I]-FIAU0.1670.0650.0440.1010.220.240.0350.0520.110.0380.2260.067[I23I]-FIAU0.02100.00790.00600.01200.03120.02740.00800.00520.00920.00440.02790.0084[I25I]-FIAU0.01720.00620.00500.00950.01980.02260.01000.00330.01390.00230.02040.0069['â€¢"IJ-FIAU0.1640.0310.0480.0580.1560.1400.0330.0240.0730.0180.1290.066

concentrations in the tissue culture medium of the GCV-sensitivity

assays and the tumor response to comparable GCV concentrations inplasma. The GCV IC5I, scale bar is based on a comparison of the threedata sets: (a) PET imaging of FIAU Ki values in //SW-f/c-transducedtumors (Fig. 4); (b) GCV-sensitivity assays in the cell lines used to

produce the tumors (Fig. 7); and (c) plasma GCV concentration levelsthat have been achieved in clinical trials. The parametric image ofGCV IC50 concentrations (Fig. 4) suggests that RG2TK#16 tumorswould be highly susceptible to modest, clinically administered dosesof GCV (resulting in plasma concentrations of ~1.0 /Â¿g/ml).Treat

ment of RG2TK#4 and RG2TK+ tumors would be expected torequire higher doses of GCV (resulting in plasma levels of ~10

fig/ml) that are tolerable in patients.These results point toward the potential for a wider application of

HSVl-tk imaging with [I24I]-FIAU and PET. A wider application forHSVl-tk imaging may be necessary because all "therapeutic genes" donot have an appropriate "marker substrates" that can be labeled for

imaging (as is the case for HSVl-tk). Therefore, we are exploring analternative approach by using the HSVl-tk as a marker gene for

imaging other therapeutic genes. This approach involves the use ofmultigene vector constructs where the therapeutic gene is linked to amarker gene (HSVl-tk) by an 1RES sequence (20-22) or is expressedin the form of a fusion (chimeric) protein. We and others (23-26) have

recently shown that genes linked by an 1RES sequence are expressedin a proportional relationship such that the magnitude of expression ofone gene, the marker gene, reflects the magnitude of expression of thesecond gene, the therapeutic gene. We suggest that gene therapyvectors that include the 1RES sequence for bicistronic coexpression oftherapeutic and marker genes are the most likely candidates to bedeveloped for the wider application of monitoring gene therapy in thefuture. Based on results presented here, HSVl-tk may be an excellent

marker gene and is likely to find wider application in future clinicalgene therapy studies.

In conclusion, we have demonstrated for the first time that highlyspecific noninvasi ve images of HSVl-tk expression in experimentalanimal tumors can be obtained using radiolabeled 2'-fluoro-nucleo-side [124I1-FIAU and a clinical PET system. PET imaging with [124I]-

FIAU can discriminate and quantitate different levels of HSVl-tk gene

expression. The ability to image the location (distribution) of geneexpression and the level of expression over time will provide new anduseful information for monitoring clinical gene therapy protocols inthe future.

APPENDIX

Preliminary dosimetry estimates were obtained from the distribution of[l24I]-FIAU-derived radioactivity in various organs and tissues (Table 1). With

the exception of thyroid, the levels of radioactivity in proliferative and non-

proliferative tissues were low (0.01-0.02%ID/g) and were similar to that of

blood plasma. The levels of retained radioactivity increased in the followingorder: brain < muscle < gonads < heart < stomach < bladder < liver < kidney < large intestine < spleen < small intestine < bone marrow <3Cthyroid(blocked). Dosimetry estimates (cSv/37MBq or mrem/mCi) for [124I]-FIAU,[123I]-FIAU, [125I]-FIAU, and [I3II]-FIAU, are compared to [l24I]-lUdR and

presented in Table 2.The radiotoxicity from administering [*l]-labeled 2'-fluoro nucleosides

arises from two components: (a) the traditional MIRD contribution (S valuesare available for [I23I], [124I], and ['â€¢"!]):and (b) the Auger electron contribution from radioiodinated |*I]-FIAU incorporated into the DNA of sensitive

normal tissues. The radiation exposure in humans from diagnostic doses of[I24I]-FIAU (2 mCi) is expected to be small and well within acceptable limits.In the case of [124I]-FIAU, it will be less than that of radioiodinated [*I]-IUdR,

which is presently undergoing clinical evaluation as an imaging agent fortumor proliferation (28, 29). The bulk of [I24I]-FIAU derived radioactivity

would be excreted in the urine as unchanged parent compound, because of thestability and resistance to metabolic degradation of 2'-fluoro-arabino-nucleo-

sides. The major radiolabeled metabolite is expected to be iodide, although thefraction and amount of radiolabeled iodide in blood and tissues are expected tobe much less compared to that after [l24I]-IUdR administration. Radiotoxicitydue to the effects of Auger electrons from [ ' 24I]-FIAU radioactivity on rapidly

dividing normal (nontransduced) tissues (e.g., bone marrow and intestinalmucosa) is also expected to be much less than from lUdR because the FIAUis only minimally phosphorylated by mammalian thymidine kinase and is onlyminimally incorporated into DNA of normal (nontransduced) proliferatingmammalian cells. Similarly, the accumulation of [l24I]-FIAU-derived radiolabeled [I24l]-iodode in thyroid and gastric and duodenal mucosa is expected tobe significantly less than the accumulation of [I24l]-iodode after [124l|-IUdR

administration.

REFERENCES

1. Tjuvajev. J.. Stockhammer, G.. Desai, R., Uehara, H.. Watanabe, K.. Gansbacher. B.,and Blasberg. R. Imaging the expression of transfected genes in mv;. Cancer Res.. 55:6126-6132. 1995.

2. Tjuvajev. J.. Finn. R.. Watanabe. K.. Joshi. R., Oku, T., Kennedy, J., Beattie, B.,Koutcher. J.. Larson. S.. and Blasberg, R. Noninvasive imaging of herpes virusthymidine kinase gene transfer and expression: a potential method for monitoringclinical gene therapy. Cancer Res.. 56: 4087-4095, 1996.

3. Morin. K. W.. Artazheva. E. D.. Knaus. E. E., and Wiebe, L. I. Synthesis and cellularuptake of 2'-substituted analogues of (E)-5-(2-['25I]iodovinyl)-2'-deoxyuridine in

tumor cells transduced with the herpes simplex type-1 thymidine kinase gene.Evaluation as probes for monitoring gene therapy. J. Med. Chcm., 40: 2184-2190,1997.

4. Morin. K. W.. Knaus. E. E.. and Wiebe. L. I. Non-invasive scintigraphic monitoringof gene expression in a HSV-] thymidine kinase gene therapy model. NucÃ.Med.Commun., 18: 599-605, 1997.

5. Srinivasan. A.. Gambhir. S. S.. Green, L. A., Cherry, S. R.. Sharfstein. S., Barrio.J. R., Satyamurthy. N., Namavari. M.. Wu, L., Berk, A. J.. Phelps, M. E., andHerschman, H. A PET reporter gene (PRG)/PET reporter probe (PRP) technology forrepeatedly imaging gene expression in living animals. J. NucÃ.Med.. 37 (Suppl.):107P, 1996.

6. Green. L. A.. Gambhir. S. S., Barrio. J. R., Bauer. E.. Nguyen. K.. Namavari. M.,Cherry, S. R.. Herschman. H., and Phelps. M. E. Tracer kinetic modeling of 8-(F18)-fluoroganciclovir PET data: A new tracer for measuring reporter gene expression.J. NucÃ.Med.. 39 (Suppl.): 10P, 1998.

4340




7. Gambhir. S. S.. Barrio. J. R.. Bauer. E.. Iyer. M.. Namavari, M.. Salyamurthy, N..Shah. P. Toyokuni. T., Wu. L.. Berk, A. J.. Phelps. M. E.. and Herschman. H.Radiolabeled penciclovir: a new reporter probe with improved imaging propertiesover ganciclovir l'or imaging herpes-simplex virus type 1 thymidine kinase reporter

gene expression. J. NucÃ.Med.. 39 (Suppl.)/ 53P, 1998.8. Goldman. S., Monclus, M.. Cool, V.. Damhaut. P.. Luxen, A.. Lcvivier. M.. Brotchi.

J.. Hildebrand. J., and Velu. T. A novel PET tracer for evaluation of gene therapy.J. NucÃ.Med., 37 (Suppl.). 53P. 1996.

9. Alauddin, M. M., Conti, P. S., Mazza, S. M., Hamzeh, F. M., and Lever. J. R.9-[(3-[18F]-nuoro-l-hydroxy-2-propoxy)methyl]guanine (|I8F]-FHPG): a potentialimaging agent of viral infection and gene therapy using PET. NucÃ.Med. Bio]., 23:787-792, 1996.

10. Alauddin. M. M.. Raman. R. K., Kundu, R.. Gordon. E. M.. and Conti, P. S.Evaluation of F-18 9-[3-[l8F]fluoro-l-hydroxy-2-propoxy)melhyl)guanine (F-18FHPG) in HT-29 cells. J. NucÃ.Med.. 38 (Suppl.). I76P. 1997.

11. Bading. J. R.. Alauddin. M. M.. Fissekis, J. H.. Kirkman. E.. Raman. R. K., and Conti,P. S. Pharmacokinetics of F-18 fluorohydroxypropoxymelhylguanine (FHPG) inprimates. J. NucÃ.Med.. 38 (Suppl.). 43P. 1997.

12. Haberkom, U.. Oberdorfer. F.. Geben, J.. Morr. I.. Haack. K.. Weber. K.. Lindauer. M.,van Kaick. G.. and Schackert. H. K. Monitoring gene therapy with cytosine deaminase:in vitro studies using triliated-5-fluorocytosine. J. NucÃ.Med.. 37: 87-94. 19%.

13. Tjuvajev. J., Finn, R.. Watanabe, K.. Joshi. R.. Beatlie. B.. and Blasberg. R. Imagingin vivo HSVI-tk gene transfer and expression wilh (124-I]-F1AU and PET. CancerGene Ther., 3: S30. 1996.

14. Tjuvajev. J.. Sasajima, T.. Oku. T.. Miyagawa, T., Joshi. R.. Safer. M.. Beattie, B..Daghighian. F.. Augensen. F.. Di Resta. G.. Koutcher. J., Zweit. J.. Finn. R., Larson.S., and Blasberg. R. Quantitative imaging of HSVI-lk gene transfer and expression invivo studies with [I241]-FIAU and PET. J. Cereb. Blood Flow Metab.. 17(Sappi. 1):

S180. 1997.15. Moolten. F. L., and Wells. J. M. Curability of tumors bearing herpes thymidine kinase

genes transfected by retroviral vectors. J. Nati. Cancer Insi.. 82: 297-300. 1990.16. Tjuvajev, J. G., Gansbacher, B.. Desai, R.. Beattie. B.. Kaplitt. M.. Matei, C.,

Koutcher. J.. Gilboa, E., and Blasberg. R. G. RG-2 glioma growth attenuation andsevere brain edema caused by local production of interleukin-2 and Â¡merferon-y.Cancer Res.. 55: 1902-1910. 1995.

17. Martin. W. R.. Palmer, M. R.. Patlak. C. S.. and Calne. D. B. Nigrosirial function inhumans studied wilh positron emission tomography. Ann. Neurol.. 26: 535-542.

1989.18. Short. M. P., Choi, B. C., Lee, J. K., Malik. A., Breakfield, X. O.. and Martuza. R. L.

Gene delivery to glioma cells in rat brain by grafting of a retrovirus packaging cellline. J. Neurosci. Res.. 27: 427-433. 1990.'

19. Blasberg. R. G.. and Tjuvajev. J. G. In vivo monitoring of gene therapy by radiotracerimaging. In: W. Semmler and M. Schwaiger (eds.). Impact of Molecular Biology andNew Technical Developments in Diagnostic Imaging, pp. 161-199. Berlin. Heidelberg: Springer-Verlag. 1997.

20. Morgan. R. A., Couture. L.. Elroy-Stein, O.. Ragheb. J.. Moss. B.. and Anderson.

W. F. Retroviral vectors containing putative internal ribosome entry sites: development of a polycistronic gene transfer system and applications to human gene therapy.Nucleic Acids Res., 20: 1292-1299, 1992.

21. Ramesh, N., Kim, S. T., Wei. M. Q.. Khalighi. M.. and Osbome. W. R. High-literbicistronic retroviral vectors employing foot-and-mouth disease virus internal ribosome entry site. Nucleic Acids Res., 24: 2697-2700. 1996.

22. Maga. J. A.. Widmer. G., and LeBowitz. J. H. Leishmania RNA virus I-mediatedcap-independent translation. Mol. Cell. Biol.. 15: 4884-4889. 1995.

23. Sugimoto. Y.. Aksentijevich, I.. Murray. G. J.. Brady. R. O.. Pastan, !.. and Gottes-

man. M. M. Retroviral coexpression of a multidrug resistance gene (MDRh andhuman a-galactosida.se A for gene therapy of Farby disease. Hum. Gene Ther.. 6:905-915. 1995.

24. Tjuvajev, J., Sharma. S.. Miller, P.. Paul. R.. Joshi, R., Dubinett. S.. and Blasberg. R.Monitoring the expression of therapeutic genes linked to a marker gene by an 1RESsequence. J. Neuro-oncol., JO (Suppl.): 101. 19%.

25. Tjuvajev, J., Gallardo, H.. Joshi, A., Joshi, R.. Blasberg, R.. and Sadelain. M.Bi-cistronic co-expression of a cell surface marker gene and the HSVl-ik gene in

jurkat cells: implications lor noninvasive imaging in vivo. Cancer Gene Ther.. 4(Suppl.): S44. 1997.

26. Tjuvajev. J.. Avril, N.. Lindsley, L.. Safer, M.. Joshi. R.. Beattie. B., Larson, S.,Sadelain. M.. and Blasberg. R. Monitoring the expression of therapeutic genes bynoninvasive imaging of the HSVl-lk marker gene using double-gene vector constructs. Cancer Gene Ther.. 4 (Suppl.): S43-S44. 1997.

27. Daghighian. F.. Humm. J. L.. Macapinlac. H.A.. Zhang. J.. Iz/o. J.. Finn. R.. Kemeny.N.. and Larson. S. Pharmacokinetics and dosimetry of iodine-125-lUdR in thetreatment of colorÃ©ela!cancer metaslalic lo liver. J. NucÃ.Med. 37 (Suppl): 29S-32S.

1996.28. Tjuvajev. J., Macapinlac. H.. Scott. A.. Daghighian. F.. Ginos. J.. Finn. R.. Kothari.

P.. Desai. R.. Zhang. J.. Beattie. B.. Graham. M.. Larson. S.. and Blasberg. R. Imagingof the tumor activity with ['"ll-iododeoxyuridine. J NucÃ.Med. 35: 1407-1417.

1994.29. Roelcke, U.. Blasberg. R., Missimer. J.. Weinreich, R., and Lecnders. K. L. I-I24

lododeoxyuridinc (lUdR) retention in peritumoral edema of glioblastomas. J. Neuro-

oncol.. 35 (Suppl.). S38, 1997.

4341



1998;58:4333-4341. Cancer Res Juri G. Tjuvajev, Norbert Avril, Takamitsu Oku, et al. Expression by Positron Emission TomographyImaging Herpes Virus Thymidine Kinase Gene Transfer and

Updated version

http://cancerres.aacrjournals.org/content/58/19/4333

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/58/19/4333To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Imaging Herpes Virus Thymidine Kinase Gene Transfer and … · (CANCER RESEARCH 58. 4333-4341. October I. I998| Imaging Herpes Virus Thymidine Kinase Gene Transfer and Expression