Evaluation of a radiolabelled cyclic DTPA-RGD analogue for tumour imaging and radionuclide therapy

Evaluation of a Radiolabelled CyclicDTPA-RGD Analogue for Tumour Imaging and

Radionuclide Therapy

P.M. van Hagen, M.D., Ph.D.,1,4 W.A.P. Breeman, M.D., Ph.D.,2 H.F. Bernard,2M. Schaar,2 C.M. Mooij, M.D., Ph.D.,3 A. Srinivasan, Ph.D.,5

M.A. Schmidt, Ph.D.,5 E.P. Krenning, M.D., Ph.D.,2,4 and M. de Jong, Ph.D.21Department of Immunology, Erasmus Medical Centre, Rotterdam, The Netherlands

2Department of Nuclear Medicine, Erasmus Medical Centre Rotterdam,Rotterdam, The Netherlands

3Department of Pathology, Erasmus Medical Centre Rotterdam, Rotterdam, The Netherlands4Department of Internal Medicine, Erasmus Medical Centre Rotterdam,

Rotterdam, The Netherlands5Mallinckrodt Medical Inc., St. Louis, Missouri, USA

SUMMARY Tumours depend on sufficient blood supply for their growth. They are ableto promote new blood vessel formation (neoangiogenesis) via angiogenic factors. Inhibi-tion of this process results in tumour involution or necrosis. RGD (Arg-Gly-Asp) peptidesare described to antagonise neoangiogenesis, e.g., by binding to avb3 receptors on bloodvessels. In order to visualise neoangiogenesis in tumours in vitro and in vivo, we intro-duced and tested an RGD analogue [c(Arg-Gly-Asp-D-Tyr-Lys)], coupled to the chelatordiethyleletriamepentaacetic acid (DTPA). This analogue can be radiolabelled with both111In and 125I. In autoradiography and immunohistochemistry studies, the 125I-labelledanalogue appeared to bind specifically and with high affinity to avb3 receptors on neovas-cular blood vessel sections of different major human cancers, like prostate and breastcancer, which express these receptors. This radioiodinated radiopharmaceutical alsobound to and internalised in human carcinoid Bon cells and rat pancreatic CA20948tumour cells. Internalisation was receptor-specific and appeared to be time and tempera-ture dependent. In vivo in rats, we investigated administration of different peptideamounts (0.1, 0.5, and 100 µg). The best amount of the radiolabelled analogue to beadministered to rats appeared to be 0.1 µg/rat, as uptake decreased with increasing peptideamount. We also found receptor-specific accumulation of the 111In-labelled analogue inthe transplantable pancreatic tumour CA20948. The introduction of the DTPA group inthis peptide resulted in renal clearance of the radiopharmaceutical, in contrast to thenon-DTPA-conjugated compound that is cleared predominantly via the liver. 111In emitsAuger and conversion electrons besides gamma radiation, therefore, this radiopharma-ceutical is suitable not only for tumour scintigraphy but also has potential for radionuclidetherapy of major human cancers as well. Moreover, after coupling to the chelator DOTA,the analogue could be radiolabelled in a stable way with b-emitters, e.g., 90Y and 177Lu,enlarging its potential. Int. J. Cancer (Radiat. Oncol. Invest.) 90, 186–198 (2000).© 2000 Wiley-Liss, Inc.

Key words: RGD; angiogenesis; tumour imaging; radionuclide therapy

*Correspondence to: P.M. van Hagen, M.D., Ph.D., Department of Immunology, Erasmus Medical Centre Rotterdam, Dr.Molewaterplein 40, 3015 GD Rotterdam, The Netherlands. Phone: 31-10-4639222; Fax: 31-10-4633268; E-mail: [email protected]

Received 29 February 2000; Revised 10 July 2000; Accepted 10 July 2000

Int. J. Cancer (Radiat. Oncol. Invest):90, 186–198 (2000)© 2000 Wiley-Liss, Inc. Publication of the International Union Against Cancer

INTRODUCTIONRadiolabelled derivatives of peptides have been ap-plied with great success for imaging of tumoursand/or inflammatory processes. These peptideshave a number of properties in common, such asstable radiolabelling with gamma-emitting radionu-clides, in vivo stability of the radiopharmaceutical,high receptor affinity, and sufficient accumulationof radioactivity in tumours or inflammation pro-cesses.

In cell-cell and cell-matrix interactions,avb3

receptors play an important role as adhesion mol-ecules. They are composed of two transmembraneglycoproteins, thea and b subunit. The bindingregion in thisavb3 receptor binds essentially to thesequence of three amino acids: arginine (R), gly-cine (G), and aspartate (D), this sequence beingcommonly present in the naturalavb3 receptor li-gands. This sequence is commonly called RGD,after the single-letter codes assigned to these aminoacids. A number of highly active, low-molecular-mass peptides have been designed to antagonise theavb3 receptor. These are synthetic RGD moleculesthat bind with high affinity to this receptor andhave been shown to be inhibitors of cell adhesion,migration, growth, and differentiation [1–4]. Theseagents also inhibit growth of new blood vessels andinduce tumour regression in animal models, pre-sumably by restricting the blood supply to the tu-mour [5–8]. Theseavb3 receptors are expressed ata lower density on the endothelial cells of vesselsthat are not growing in comparison with newlygrowing endothelial cells [9]. Recently a radioio-dinated cyclic RGD peptide for receptor-targetedimaging has been described that accumulates intransplantable rat tumours [10]. It was concludedthat this radiolabelled cyclic RGD compound ispromising for tumour imaging and radionuclidetherapy. In this article, we describe a newly syn-thesised diethyleletriamepentaacetic acid (DTPA)-conjugated variant of this peptide: a cyclic DTPA-RGD peptide [c(Arg-Gly-Asp-D-Tyr-Lys)-«-DTPA,hereafter referred to as DTPA-RGD] that can bothbe radiolabelled with111In and be radioiodinated(Fig. 1). Using this radiolabelled peptide, we inves-tigated binding to and the density ofavb3 receptorsin different human tumours. We also looked at in-ternalisation in different tumour cell lines in vitroas well as at tumour uptake in rats in vivo.

MATERIALS AND METHODS

RadiopharmaceuticalSolid-phase peptide synthesis of c(Arg-Gly-Asp-D-Tyr-Lys)-«-DTPA was performed using a PE Bio-

systems Pioneer synthesizer employing Fmoc strat-egy. The target peptide was prepared on a 0.1 mmolscale with Fmoc-Asp(Peg-PS)-OAl (PE Biosys-tems; 0.16 mmol/g loading) as the starting resin.Fmoc-protected amino acids (0.4 mmol) were ac-tivated with N-[(dimethylamino)-1H-1,2,3-tri-azolo[4,5-b]pyridin-1-ylmethylene]-N-methyl-methan-aminium hexaflurophosphateN-oxide(HATU). All the amino acids were purchased fromPE Biosystems in prepacked tubes except for theFmoc-D-Tyr-OH and Fmoc-Lys(Mtt)-OH, whichwere supplied by Novabiochem (San Diego, CA).The other peptide synthesis reagents were also ob-tained from PE Biosystems. On-board cyclizationof the peptide was achieved using the Allyl De-block protocol followed by PyAOP activation. Theresin containing the protected, cyclized peptide wasthen removed from the instrument, and the Mtt pro-tecting group of the lysine was removed by treat-ment with 5% TFA/5% triisopropylsilane/ 0% di-chloromethane (2 ×30 min). The resin was washedwith dichloromethane and tetrahydrofuran beforeresuspending in DMF (5 mL) containing DIEA (70ml, 0.4 mmol). In a separate vessel, tri-t-butylDTPA (224 mg, 0.4 mmol) was dissolved in DMFcontaining HBTU/HOBt (0.4 mmol, 2.0 ml of a 0.2mmol/ml solution) and DIEA (70mL, 0.4 mmol) togive a 10 ml solution. After agitating for 1 hr, theactivated DTPA derivative was added to the previ-ously suspended resin. The reaction was permittedto continue overnight before washing the resin withDMF and THF. Cleavage and deprotection werefinally accomplished using 85% TFA/5% thioanis-ole/5% phenol/5% water (overnight). The TFA waspurchased from Pierce Chemical while the othercleavage reagents were obtained from Aldrich. Thecrude peptide was isolated by precipitation witht-butyl methyl ether (Sigma, St Louis, MO) andpurified by reverse phase HPLC using an acetoni-trile/water gradient containing 0.1% TFA (retentiontime4 10.1 min). Solvent A: 0.1% TFA/H2O; Sol-

Fig. 1. Structure of c(Arg-Gly-Asp-D-Tyr-Lys)-DTPA,referred to as DTPA-RGD.

Van Hagen et al.: Radiolabelled Cyclic DTPA-RGD Analogue 187

vent B: 0.1% TFA/10% H2O/CH3CN; Gradient:Hold at 95% A/5% B for 2.0 min before ramping to30% A/70% B over 15 min). Molecular weightdetermination was accomplished by mass spec-trometry operating in the electrospray mode (ESI)[calc. 994.5; found 995.4 (M+H)+]. Amino acidanalysis yielded a peptide content of more than95%.

Radiolabelling111InCl3 (DRN 4901, 370 MBq/mL in HCl, pH1.5–1.9) was obtained from Mallinckrodt (Petten,The Netherlands). Radiolabelling with111In to aspecific activity of 150 MBq111In per mg DTPA-RGD and consecutive quality control by high-performance liquid chromatography (HPLC) wasperformed with a Waters 600E multisolvent deliv-ery system, connected to amBondapak C18 re-versed phase column (300 × 3.9 mm, particle size10mm)(Waters, Etten-Leur, The Netherlands). Elu-tion was carried out at a flow of 1.5 mL per minwith a linear gradient of 10% to 50% acetonitrile in0.1% TFA in 30 min. The latter composition waskept constant for another 5 min.

Radioiodination was carried out essentially asdescribed earlier [11]. In short, 140mg DTPA-RGD and 2mg chloramine T in 0.05 M phosphatebuffer (pH 7.5) and 100 MBq (50mL Na125I (spe-cific activity 0.62 TBq 125I/mg, Amersham, UK).The mixture was vortexed for 1 min. The radioio-dination was stopped by adding 1 mL 10% humanserum albumin (Merieux, Lyon, France). After vor-texing for 30 sec, 2 mL 5 mM ammonium acetatewas added. The labelling efficiency of DTPA-RGDwith 125I was typically >90%. HPLC quality con-trol of the radioiodinated DTPA-RGD reveals twopeaks within a narrow “elution volume range.” Thisis a consistent finding of DTPA coupled peptides(Fig. 2).

AutoradiographyThe presence of RGD binding receptors in tumourswas investigated by receptor autoradiography,mostly as described earlier [12]. Shortly, tumourbiopsies were embedded in TissueTek and pro-cessed for cryosectioning. Tissue sections weremounted on glass slides and stored at −20°C for atleast two days to improve adhesion of the tissue tothe slide. Sections were air-dried, pre-incubated in170 mM Tris-HCl buffer, pH 7.6, for 10 min atroom temperature (RT) and then incubated for 120min at RT with the iodinated ligand (5 ×10−11 M).The incubation solution was 170 mM Tris-HClbuffer, pH 7.6, containing 1% (w/v) BSA, 1 mgbacitricin and 5 mM MgCl2 to inhibit endogenous

proteases. Nonspecific binding was determined byincubating a sequential section in the presence of 1mM of unlabelled peptide. After incubation, thesections were washed twice for 5 min in cold in-cubation buffer including 0.25% BSA, then inbuffer alone, after that once with cold MilliQ (Wa-ters), and dried quickly. The sections were exposedto Fuji films (New RX, Tokyo, Japan) for threeweeks in x-ray cassettes. In selected cases, dis-placement experiments were performed in adjacentsections by means of increasing concentrations ofDTPA-RGD (range 5 × 10−11–10−6 M). The auto-radiograms were quantified using a computer-assisted image processing system (OptiQuant03.00, Packard Instruments, Groningen, The Neth-erlands). A tissue section was defined as RGD re-ceptor-positive when the optical density measuredover an area in the total binding was at least twicethe optical density of the nonspecific binding. IC50

values were calculated from competitive bindingcurves using GraphPad Prism (GraphPad PrismSoftware, San Diego, CA).

Immunohistochemistry

Sinceavb3 receptors are the major RGD bindingreceptors, immunohistochemical analysis ofavb3

expression in tumour biopsies was performed aswell. Cryostat sections from tumour biopsies weremounted on glass slides and air-dried for 30 min.The sections were fixed in acetone for 10 min. Im-

Fig. 2. High-pressure liquid chromatography elution pat-tern of 125I-DTPA-RGD.

188 Van Hagen et al.: Radiolabelled Cyclic DTPA-RGD Analogue

munohistochemistry was performed according tostandard methods using mouse monoclonal anti-avb3 antibodies (anti-CD51/61; Sanver Tech, CO)in PBS in 5% BSA (100mL/slide) with 2 hr incu-bation. Nonspecific binding was determined by in-cubation with normal goat serum in a dilution 1:10in PBS/BSA 5%. Biotinylated goat anti-mouse an-tiserum (dilution 1:50) in PBS containing 5% BSAwas used as the second antibody. Alkaline phos-phatase was used as a label (1:50 in PBS/BSA 5%)with new fuchsine as substrate, which gives a redcolour. The sections were counterstained with hae-matoxylin for 30 sec. Vascularisation was assessedas preexistent when the vessel wall includingsmooth muscle cells was intact.

Radioligand bindingand internalisation

Receptor-mediated and internalisation of125I-DTPA-RGD was measured in a rat and human cellline. The rat pancreatic cell line CA20948 was cul-

tured in DMEM (Gibco BRL, Life Technologies),whereas Bon cells (human carcinoid cell line) werecultured in RPMI 1640 (Gibco). All media weresupplemented with 2 mM glutamine (glutamax I), 1mM sodium pyruvate, fungizone (0.1 mg/mL, peni-cillin/streptomycin 50 IU/mL, and heat-inactivatedFBS (10%). Subconfluent cell cultures were trans-ferred to six-well plates.

The binding of radiolabelled peptides to thesehuman and rat cells and subsequent internalisationwere studied as described previously [13]. Briefly,cells were washed in PBS. Incubation was startedafter addition of 1 mL internalisation medium(RPMI supplemented with 20 mM Hepes and 1%BSA) with 80 kBq of 125I-DTPA-RGD. The cellswere incubated at 37°C or 5°C for indicated peri-ods. To determine specific vs. nonspecific uptake,cells were incubated with 0.1 nM radioiodinatedRGD analogue with or without excess of unlabelledanalogue. Cellular uptake was stopped by removingthe medium from the cells, followed by washing

Table 1. Autoradiography With Radioiodinated DTPA-RGD and Endothelial avb3Expression on 20 Sections From Different Human Cancers

Tissue N Autoradiography avb3 expression (immunohistochemistry)

Prostate cancer 21 Pos PEBVa, majority2 Pos No evaluationb

Breast cancer 43 Pos Neovasc.c, moderate4 (scirrheus) Pos Neovasc. and PEBV, sporadic5 (scirrheus) Pos PEBV, sporadic6 Pos Neovasc., sporadic, tumor cells 11%–25%

Retinoblastoma 17 Neg PEBV, moderate

Choroid melanoma 28 Neg Neovasc., sporadic, tumor cells 0%–10%9 Neg Neovasc., sporadic

Bladder carcinoma 110 Pos PEBV, majority

Undifferentiated lung carcinoma 211 Pos Neovasc., majority12 Pos Neovasc., moderate

Carcinoid tumor 313 Pos Neovasc., majority14 Pos Neovasc., majority15 Neg Negative

Colon/rectum adenocarcinoma 216 Pos Stromal PEBV and myoepithelium17 Pos Stromal PEBV and myoepithelium

Esophagus adenocarcinoma 118 Pos Neovasc. and PEBV, majority, tumor cells 0%–10%

Neuroblastoma 119 Pos Neovasc., sporadic

Adrenal cell carcinoma 120 Pos Neovasc., majority

aPreexistent blood vessels.bNo assessment because of technical reasons.cNeovascularization.


twice with ice-cold 2 mL PBS. To discriminatebetween internalised and noninternalised (surface-bound) ligand, intact cells were incubated with 1mL 20 mM sodium acetate [13]. Radioactivity wasdetermined in a well-type LKB-1282-Compu-gamma system. The uptake was expressed as a per-centage of the applied dose per mg cellular protein.The protein content was measured using a commer-cially available kit (BioRad, Woerden, The Neth-erlands). The data were expressed as the mean withstandard deviation for incubations assayed in trip-licates.

Tissue distribution

Animal experiments were performed in compliancewith the regulations of our institution and with the

generally accepted guidelines governing suchwork. The CA20948 pancreatic tumours weregrown at the flank of male Lewis rats (Harlan, TheNetherlands; rat body weight was 150–220 g).Male Lewis rats were injected subcutaneously inthe flank with 500mL of a cell suspension ofCA20948 tumour, prepared by mechanical disper-sion from 5 g crude tumour tissue in 100 mL saline.About 15 days after inoculation, experiments werestarted. 111In-DTPA-RGD analogue (3 MBqcoupled to 0.1 or 0.5mg peptide) was injected intothe dorsal vein of the penis under ether anaesthesia.A parallel group of four rats was co-injected simi-larly with an excess of RGD peptide, 100mgc(Arg-Gly-Asp-D-Phe-Val) (Bachem, Switzerland)in 0.05 M acetic acid in 154 mM NaCl. Twenty-

Fig. 3. (A) Autoradiography of a tumour section (undifferentiated lung carcinoma) using125I-DTPA-RGD analogue in aconcentration of 5 × 10−11M, with and without aavb3-receptor blocking dose of 1mM DTPA-RGD analogue, indicating thereceptor specificity.(B) Immunohistochemistry using anti-CD51/61 MAb of an adjacent section showingavb3 expression bythe neovascular blood vessels.


four hours post-injection, the rat organs, blood, andtumour were collected. In a different experiment, invivo tissue distribution was investigated as functionof time. In this experiment, 3 MBq of111In coupledto 0.1mg peptide was injected in CA20948 tumour-bearing rats. One, four, and 24 hr post-injection, ratorgans, blood, and tumour were collected. Thesamples obtained in both experiments were deter-mined by measuring radioactivity in a well-typeLKB-1282-Compugamma system. The followingorgans and tissues were isolated and counted forradioactivity: blood, lungs, thymus, pancreas,spleen, adrenals, kidneys, stomach, thigh (soft tis-sue), liver, femur, and the CA20948 tumour.

RESULTS

Autoradiography

To determine specific binding of DTPA-RGD tohuman tumour tissue, we started autoradiographyusing125I-labelled analogue on 20 tumour sectionsfrom different human cancers (Table 1). In onlyfour of 20 tumour sections, i.e., retinoblastoma, twochoroid melanomas, and one carcinoid tumour, nospecific binding was found, whereas in all othersections specific binding of the peptide was found.Simultaneous immunohistochemistry ofavb3 ex-pression with anti-avb3 (anti-CD51/61) monoclo-nal antibody (MAb) in adjacent sections revealedpositive staining of the preexistent and neovascularendothelium and to a lesser degree of the tumourcells. In Figure 3a, an example is shown of a posi-tive autoradiogram of an undifferentiated lung car-cinoma, and in Figure 3b, the correspondent immu-nohistochemical staining is shown.

Autoradiography results corresponded verywell with the endothelialavb3 expression in thetumour sections as shown by immunohistochemis-try. Exceptions were autoradiography-negative sec-tions of two choroid melanoma tumours that hadsporadic immunohistochemistry-positive bloodvessels. Autoradiography using radioiodinated pep-tide of suitable tumour biopsies with increasingconcentrations of unlabelled DTPA-RGD revealeddisplacement, as shown in Figure 4. The affinity ofthis compound in the different tissue sections wasin the nanomolar range (mean of different sections:Kd about 10 nM).

Internalisation

Binding and internalisation of radioiodinatedDTPA-RGD analogue was studied in two cell lines,rat pancreatic CA20948 cells, and human carcinoidBon cells. Figure 5 shows that internalisation of theradiopharmaceutical in both cell lines was receptor-

specific as internalisation decreased significantly inthe presence of excess unlabelled peptide. Nonin-ternalised (membrane-bound) radioactivity waslow. In the Bon cells,avb3 expression was con-firmed with FACS analysis. Radioactivity in thecells increased as a function of time during incu-bation at 37°C, while at 5°C cellular radioactivityremained constant (Fig. 6).

Tissue distributionIn the first animal experiment, tumour-bearing ratswere injected with different amounts of111In-DTPA-RGD analogue with or without excess pep-tide and sacrificed 24 hr post-injection (Table 2).Data for blood, some organs, and tumour radioac-tivity are visualised in Figure 7a as well. The pep-tide was rapidly cleared from the blood, as shownby the very low amount of residual blood radioac-tivity. In most organs and in the tumour, receptor-specific binding was found, as shown by a signifi-cant decrease in radioactivity in the presence of 100mg compared with the lower peptide amounts (0.1and 0.5mg). Furthermore, uptake in most organsand tumour was higher for 0.1 compared with 0.5mg 111In-DTPA-RGD. Table 3 shows the in vivotissue distribution data obtained as a function oftime. Data for some organs, blood, and tumour arevisualised in Figure 7b. Radioactivity in the tu-mour, liver, pancreas, and spleen remained almostconstant in the period from 4 to 24 hr, whereas inother organs radioactivity decreased rapidly withtime. The tumour to blood ratio was nine at 4 hr,increasing to 23 at 24 hr.

From both in vivo experiments, the high up-take of radioactivity in the kidneys after injectionof 111In-DTPA-RGD was evident. The urinary ex-cretion was the major route of clearance of radio-activity, since >85% of the injected dose was re-covered within 24 hr post-injection.

DISCUSSIONAngiogenesis or the development of new bloodvessels plays an important role in physiologicalprocesses like wound healing. However, uncon-trolled neovascularization may contribute to a num-ber of pathological circumstances like proliferativeretinopathy and tumour growth and metastasis[6,7,14–16]. Several naturally occurring angiogenicfactors have been identified, with basic fibroblastgrowth factor (bFGF) and vascular endothelialgrowth factor (vEGF) being the most commonlyup-regulated in malignant growth. Cyclic RGDpeptides block angiogenesis induced by growthfactors in a variety of human tumours and also theinvasive behaviour of human breast carcinoma


cells [5,6]. A number of antiangiogenic drugs beingdeveloped and are now entering clinical trials, in-cluding specific anti-avb3 antibodies [17-19]. An-other approach may be radionuclide radiotherapyof tumoural blood vessel cells using radiolabelledpeptides that have a high affinity for the neovascu-lar endothelium. Recently, a radioiodinated cyclicRGD peptide was reported that exerts a high affin-ity for the avb3 receptor in vitro as well as in vivo[10]. The peptide structure was based on the selec-tive avb3 antagonist c(-Arg-Gly-Asp-D-Phe-Val-)with IC50-values in the nanomolar range. Radioio-dinated c(-Arg-Gly-Asp-D-Tyr-Val-) accumulatedin transplantable melanoma, mammary carcinomaand osteosarcoma models, but a major drawbackwas the high accumulation of radioactivity in andclearance via the liver (10). Radioiodinated Tyr-containing peptides are frequently used, for in-stance with125I-labelled analogues for in vitro re-ceptor studies, and123I- or 131I-labelled analoguesmay be applied in vivo for receptor-targeted scin-tigraphy and radiotherapy, respectively. However,these radioiodinated ligands often have the draw-

back of being rapidly degraded in vivo, with releaseand subsequent deiodination of radioiodotyrosine,hampering their diagnostic and/or therapeutic use.This has also been reported for [Tyr4]bombesin and[Tyr3]octreotide [11,20–23].

In our study, we presented a DTPA-coupledcyclic RGD molecule that can be stably radiola-belled with 111In. Autoradiographic analysis withthis compound revealed specific binding in varioussections of different human cancers, includingbreast, prostate, and colon cancer. Simultaneousimmunohistochemical staining of adjacent sectionsshowedavb3 expression in the neovascular vesselsand to a lesser degree in tumour cells. Theavb3-positive staining correlated well with autoradiogra-phy. In vivo, specific accumulation of radioactivitywas found in rats in the transplantable CA20948tumour in tissue distribution studies. The low bloodand muscle levels of radioactivity contributed to ahigh target-background ratio.

The 111In-DTPA-RGD analogue has severaladvantages compared with the radioiodinated ana-logue. First, it is more convenient for clinical ap-

Fig. 4. Displacement curve of125I-DTPA-RGD analogue in five tumour tissue sections. The sections were incubated with125I-DTPA-RGD (5 × 10−11M) analogue and increasing concentrations of unlabelled DTPA-RGD analogue. Each pointrepresents the optical density of binding measured in a region rich in blood vessels. Binding of 5 × 10−11M 125I-labelled peptidewas considered as 100%.


Fig. 5. Internalised (“intern”) and noninternalised (“nonintern”) radioactivity after incubation of rat pancreatic CA20948cells and human carcinoid Bon cells with indicated concentrations of125I-DTPA-RGD analogue during 120 min. Each pointwas determined in triplicate in two independent experiments. Data are expressed in percent of the dose per mg cellular protein(mean ± SD).


Fig. 6. Internalised (“intern”) radioactivity after incubation of rat pancreatic CA20948 cells and human carcinoid Bon cellswith indicated concentration of125I-DTPA-RGD analogue during 120 min at 37°C or 5°C. Each point was determined intriplicate in two independent experiments. Data are expressed as percent of the dose per mg cellular protein (mean ± SD).


plications because of easier labelling procedures,and second, the excretion route of111In-DTPA-RGD is via the kidneys, while the radioiodinatedpeptide is predominantly cleared via liver and in-testines [10]. The latter situation leads to a highconcentration of radioactivity in the large abdomi-nal area. Third, after internalisation of the (radio)ligand, 111In has a much longer cellular retentiontime in the target cells, being a residualising radio-nuclide. This is in contrast to the radioiodinatedligand, where radioiodine is quickly released fromthe cells, resulting in a short half-life of the radio-iodine.

From our in vivo data, it is evident that renaluptake of 111In-DTPA-RGD is relatively high.Small peptides in the blood plasma are filteredthrough the glomerular capillaries in the kidneysand subsequently reabsorbed almost completely($90%) by the proximal tubular cells by carrier-mediated endocytosis. The membranes of renal tu-bular cells contain negatively charged sites towhich positively charged residues of peptides orproteins are thought to bind. So, it was found in thissame rat model that [111In-DOTA0]CCK8 analogueuptake in the kidneys was much lower (about 0.3%of the injected dose per gram kidney 24 hr post-injection)[24] than that of [111In-DTPA0]octreotide(about 2% of the injected dose per gram kidney 24hr post-injection)[25]. [111In-DOTA0]CCK8 con-tains two Asp moieties, giving anionic charge to themolecule, whereas octreotide contains the posi-tively charged Lys moiety. Negatively charged

peptides therefore will have a lower renal re-uptakein the tubular cells than neutral or cationic ones,and although our DTPA-RGD analogue also con-tains an Arg and Lys moiety, the latter one is notcharged because of the coupling to DTPA. There-fore, the renal uptake found is lower than that of theoctreotide analogue mentioned. These qualitiesmake111In-DTPA-RGD most promising as a newradiopharmaceutical for the imaging ofavb3-positive tumours and in particular tumour neovas-cularization.

Moreover, as soon as the success of peptide-receptor scintigraphy using somatostatin analogues,binding to somatostatin-receptor subtypes on neu-roendocrine tumours for example, became clear,the next logical step was to label these peptideswith radionuclides emittinga- or b-particles or Au-ger electrons and to perform radionuclide therapywith these radiolabelled peptides. Moreover, pre-clinical results show radiotherapeutic effects ofboth 111In- and 90Y-labelled somatostatin ana-logues in rats bearing the CA20948 tumour afterinoculation in either the flank or the liver [26,27].We also treated 30 end-stage patients with mostlyneuroendocrine tumours with the somatostatin ana-logue [111In-DTPA0]octreotide [28]. Twenty-onereceived a total cumulative dose of at least 20 GBq[111In-DTPA0]octreotide. All 21 patients had pro-gressive disease, i.e., unequivocal increase in tu-mour volume according to CT or MRI before thestart of [111In-DTPA0]octreotide therapy. In eightpatients, this treatment resulted in stable diseaseand in another six patients in actual tumour shrink-age, making this new tumour therapy modalitymost promising.

Neovascularization-targeted radionuclidetherapy with 111In-labelled cyclic DTPA-RGDmolecules alone or additional to tumour chemo-therapy may therefore be an interesting option forthe future. Internalisation of the peptide, as shownin the internalisation experiments, is important notonly for successful scintigraphy at later time pointsbut also is essential for radionuclide therapy of tu-mours as111In emits conversion and Auger elec-trons with a very short particle range. Also, cou-pling to b-emitting radionuclides like90Y or 177Luusing the chelator DOTA may be interesting forradiotherapy.

CONCLUSIONIn our study, we present a stable cyclic DTPA-RGD analogue with intact high affinity for theavb3

receptor. This ligand can be radiolabelled with both111In and radioiodine. This radiolabelled analoguebinds to and internalises specifically in human and

Table 2. Radioactivity, Expressed as Percentof the Injected Dose Per Gram Tissue inDifferent Organs and Tumour of CA20948Tumour-Bearing Rats, 24 Hr After Injectionof Different Amounts (3 MBq, 0.1, or 0.5µg/Rat) of 111In-Labelled DTPA-RGDAnalogue With or Without Excess(100 µg) Peptide*

100 mg 0.5 mg 0.1 mg

Blood 0.0031 ± 0.0002 0.0028 ± 0.0001 0.0034 ± 0.0001Spleen 0.056 ± 0.004 0.15 ± 0.01 0.16 ± 0.01Pancreas 0.014 ± 0.001 0.037 ± 0.005 0.044 ± 0.003Adrenals 0.028 ± 0.003 0.052 ± 0.005 0.072 ± 0.004Kidney 0.47 ± 0.03 0.50 ± 0.08 0.50 ± 0.08Liver 0.032 ± 0.001 0.089 ± 0.011 0.117 ± 0.014Stomach 0.025 ± 0.002 0.11 ± 0.01 0.13 ± 0.01Muscle 0.0048 ± 0.0005 0.016 ± 0.003 0.016 ± 0.001Femur 0.017 ± 0.004 0.044 ± 0.008 0.045 ± 0.011Thymus 0.016 ± 0.001 0.041 ± 0.007 0.056 ± 0008Lung 0.023 ± 0.001 0.071 ± 0.006 0.078 ± 0.009CA20948

tumour 0.047 ± 0.009 0.17 ± 0.04 0.22 ± 0.03

*For each group: n$3, data expressed as mean ± SD.


Fig. 7. (A) Radioactivity, expressed as percent of the injected dose per gram tissue (%ID/g), in different organs and tumourof CA20948 tumour-bearing rats 24 hr after injection of different amounts (3 MBq, 0.1, or 0.5mg/rat) of 111In-labelledDTPA-RGD analogue with or without excess (100mg) peptide. For each group: n$ 3, data expressed as mean ± SD.(B)Radioactivity, expressed as percent of the injected dose per gram tissue (%ID/g), in different organs and tumour of CA20948tumour-bearing rats, 1, 4, or 24 hr after injection of111In-DTPA-RGD analogue (3 MBq, 0.1mg/rat). For each group: n$ 3,data expressed as mean ± SD.


rat tumour cells expressing theavb3 receptor.Moreover, it binds specifically to various humantumour sections with a high affinity and accumu-lates in a receptor-specific way in a rat tumour invivo, making this peptide highly promising for vi-sualisation and radionuclide therapy of major hu-man cancers.

REFERENCES

1. Ruoslahti E, Pierschbacher MD. Arg-Gly-Asp: a versa-tile cell recognition signal. Cell 1986;44:517–518.

2. Aumailley M, Gurrath M, Muller G, Calvete J, TimplR, Kessler H. Arg-Gly-Asp constrained within cyclicpentapeptides. Strong and selective inhibitors of celladhesion to vitronectin and laminin fragment P1. FEBSLett 1991;291:50-54.

3. Pfaff M, Tangemann K, Muller B, Gurrath M, MullerG, Kessler H, Timpl R, Engel J. Selective recognitionof cyclic RGD peptides of NMR defined conformationby alpha IIb beta 3, alpha V beta 3, and alpha 5 beta 1integrins. J Biol Chem 1994;269:20233–20238.

4. Ruoslahti E. RGD and other recognition sequences forintegrins. Annu Rev Cell Dev Biol 1996;12:697–715.

5. Brooks PC, Montgomery AM, Rosenfeld M, ReisveldRA, Hu T, Klier G, Cheresh DA. Integrin alpha v beta3 antagonists promote tumor regression by inducingapoptosis of angiogenic blood vessels. Cell 1994;79:1157–1164.

6. Brooks PC, Stromblad S, Klemke R, Visscher D, SarkarFH, Cheresh DA. Antiintegrin alpha v beta 3 blockshuman breast cancer growth and angiogenesis in humanskin. J Clin Invest 1995;96:1815–1822.

7. Hammes HP, Brownlee M, Jonczyk A, Sutter A,Preissner KT. Subcutaneous injection of a cyclic pep-tide antagonist of vitronectin receptor-type integrins in-

hibits retinal neovascularization. Nat Med 1996;2:529–533.

8. Ellerby HM, Arap W, Ellerby LM, Kain R, AndrusiakR, Rio GD, Krajewski S, Lombardo CR, Rao R, Ruo-slahti E, Bredesen DE, Pasqualini R. Anti-cancer activ-ity of targeted pro-apoptotic peptides. Nat Med 1999;5:1032–1038.

9. Brooks PC. Role of integrins in angiogenesis. Eur JCancer 1996;32A:2423–2429.

10. Haubner R, Wester HJ, Reuning U, Senekowitsch-Schmidtke R, Diefenbach B, Kessler H, Stocklin G,Schwaiger M. Radiolabeled alpha(v)beta3 integrin an-tagonists: a new class of tracers for tumor targeting [InProcess Citation]. J Nucl Med 1999;40:1061–1071.

11. Breeman WA, Hofland LJ, van der Pluijm M, vanKoetsveld PM, de Jong M, Setyono-Han B, BakkerWH, Kwekkeboom DJ, Visser TJ, Lamberts SW. Anew radiolabelled somatostatin analogue [111In-DTPA-D-Phe1]RC-160: preparation, biological activ-ity, receptor scintigraphy in rats and comparison with[111In-DTPA-D-Phe1]octreotide. Eur J Nucl Med1994;21:328–335.

12. Reubi JC, Kvols L, Krenning E, Lamberts SW. Distri-bution of somatostatin receptors in normal and tumortissue. Metabolism 1990;39:78–81.

13. de Jong M, Bernard BF, de Bruin E, Van Gameren A,Bakker WH, Visser TJ, Macke HR, Krenning EP. In-ternalisation of radiolabelled [DTPA0]octreotide and[DOTA0,Tyr3]octreotide: peptides for somatostatin re-ceptor-targeted scintigraphy and radionuclide therapy.Nucl Med Commun 1998;19:283–288.

14. Fidler IJ, Ellis LM. The implications of angiogenesisfor the biology and therapy of cancer metastasis. Cell1994;79:185–188.

15. Ellis LM, Fidler IJ. Angiogenesis and metastasis. Eur JCancer 1996;32A:2451–2460.

16. Gasparini G, Brooks PC, Biganzoli E, Vermeulen PB,Bonoldi E, Dirix LY, Ranieri G, Miceli R, Cheresh DA.Vascular integrin alpha(v)beta3: a new prognostic in-dicator in breast cancer. Clin Cancer Res 1998;4:2625–2634.

17. Pluda JM, Cheson B.D, Phillips PH. Clinical trials re-ferral resource. Clinical trials using bryostatin-1. On-cology (Huntingt) 1996;10:740–742.

18. Pluda JM, Parkinson DR. Clinical implications of tu-mor-associated neovascularization and current antian-giogenic strategies for the treatment of malignancies ofpancreas. Cancer 1996;78:680–687.

19. Molema G, Griffioen AW. Rocking the foundations ofsolid tumor growth by attacking the tumor’s blood sup-ply. Immunol Today 1998;19:392–394.

20. Davis TP, Crowell S, Taylor J, Clark DL, Coy D, StaleyJ, Moody TW. Metabolic stability and tumor inhibitionof bombesin/GRP receptor antagonist. Peptides 1992;13:401–407.

21. Rogers BE, Rosenfeld ME, Khazaeli MB, Mikheeva G,Stackhouse MA, Liu T, Curiel DT, Buchsbaum DJ.Localizatiom of iodine-125-mIP-Des-Met14 bombesin(7-13)NH2 in ovarian carcinoma induced to express thegastrin releasing peptide receptor by adenoviral vector-

Table 3. Radioactivity, Expressed as Percentof the Injected Dose Per Gram Tissue inDifferent Organs and Tumour of CA20948Tumour-Bearing Rats, 1, 4, or 24 Hr AfterInjection of 111In-DTPA-RGD Analogue(3 MBq, 0.1 µg/Rat)*

1 hr 4 hr 24 hr

Blood 0.088 ± 0.007 0.012 ± 0.004 0.004 ± 0.001Spleen 0.19 ± 0.00 0.15 ± 0.00 0.16 ± 0.03Pancreas 0.06 ± 0.00 0.041 ± 0.00 0.042 ± 0.003Adrenals 0.14 ± 0.01 0.088 ± 0.003 0.080 ± 0.011Kidney 1.13 ± 0.16 0.83 ± 0.10 0.70 ± 0.08Liver 0.14 ± 0.03 0.14 ± 0.01 0.15 ± 0.06Stomach 0.23 ± 0.01 0.15 ± 0.00 0.10 ± 0.01Muscle 0.028 ± 0.003 0.014 ± 0.002 0.013 ± 0.002Femur 0.053 ± 0.003 0.039 ± 0.005 0.034 ± 0.001Thymus 0.074 ± 0.003 0.053 ± 0.003 0.048 ± 0.004Lung 0.15 ± 0.00 0.08 ± 0.00 0.060 ± 0.008CA20948 tumour 0.18 ± 0.01 0.11 ± 0.02 0.10 ± 0.01

*For each group: n$3, data expressed as mean ± SD.


mediated gene transfer. J Nucl Med 1997;38:1221–1229.

22. Bakker WH, Krenning EP, Breeman WA, Kooij PP,Reubi JC, Koper JW, de Jong M, Lameris JS, Visser TJ,Lamberts SW. In vivo use of a radioiodinated somato-statin analogue: dynamics, metabolism, and binding tosomatostatin receptor-positive tumors in man [pub-lished erratum appears in J Nucl Med 1991 Oct;32(10):1999]. J Nucl Med 1991;32:1184–1189.

23. Breeman WAP, Hofland LJ, de Jong M, Bernard BF,Srinivasan A, Kwekkeboom DJ, Visser TJ, KrenningEP. Evaluation of radiolabelled bombesin analogs forreceptor-targeted scintigraphy and radiotherapy. Int JCancer 1999;81:658–665.

24. de Jong M, de Bakker WH, Bernard HF, Valkema R,Kwekkeboom DJ, Reubi J-C, Srinivasan A, Schmidt M,Krenning EP. Preclinical and initial clinical evaluationof 111In-labeled unsulfated CCK8 analog: a peptide forCCK-B receptor-targeted scintigraphy and radionuclidetherapy. J Nucl Med (in press).

25. de Jong M, Breeman WAP, Bakker WH, Kooij PPM,

Bernard HF, Hofland LJ, Visser TJ, Srinivasan A,Schmidt MA, Erion JL, Bugaj JE, Ma¨cke HR, KrenningEP. Comparison of111In-labeled somatostatin ana-logues for tumor scintigraphy and radionuclide therapy.Cancer Res 1998;58:437–441.

26. Slooter GD, Breeman WAP, Marquet RL, KrenningEP, van Eijck CHJ. The antiproliferative effect of ra-diolabelled octreotide in a metastases model in rat liver.Int J Cancer 1999;81:767–771.

27. de Jong M., Breeman WAP, Bernard HF, Kooij PPM,Slooter GD, van Eijck CHJ, Kwekkeboom DJ,Valkema R, Krenning EP. Therapy of neuroendocrinetumors with radiolabeled somatostatin-analogues.Quart J Nucl Med (in press).

28. Krenning EP, de Jong M, Kooij PPM, Breeman WAP,Bakker WH, de Herder WW, van Eijck CHJ, Kwekke-boom DJ, Jamar F, Pauwels S, Valkema R. Radiola-belled somatostatin analogue(s) for peptide receptorscintigraphy and radionuclide therapy. Ann Oncol1999;10:S23–S29.


Evaluation of a radiolabelled cyclic DTPA-RGD analogue for tumour imaging and radionuclide therapy

Documents