and costs, remain and have stimulated researchon radiolabeling with many alternative radionuclides, including 90Y andl6l@ (4) for therapeutic applications; 68Ga (5), @Cu (6), and

‘8F(7) for positron imaging; and, in particular, 99mTc.Thus,99mTc labeling of somatostatin analogues has been extensively studied, including attempts at direct labeling afterreduction of the disulfide bridge (8,9) and the use ofbifunctional chelators such as propyleneaminoxime (10),triamidomonothiols (11), and tetramines (12,13). One analogue based on a carbocyclic peptide and a 99mTc@N3S@chelating moiety (P829) has been studied extensively inclinical trials (14). Recently, Maecke and Behe (15) described the use of the hydrazinonicotinyl (HYNIC) core toprepare @°‘Tc-labeledoctreotide derivatives with bindingaffinity to somatostatin receptors and identified [Tyr3J-octreotide as a particularly promising analogue. We havetherefore evaluated this labeling approach in some detail(16,17) using different coligands based on aminocarboxylates (18) and tricine ternary ligand systems (19). In thisarticle, we describe the preclinical evaluation of 99mTc@[HYNIC-l'yr3}-octreotide (HYNIC-TOC) labeled through 3different coligands: tricine, ethylenediaminediaceticacid(EDDA), and a tricine—nicotinicacid ternary ligand system,including a comparison of the biodistribution of the @Tclabeled peptides with that of 1@In-DTPA-octreotide.

MATERIALS AND METHODS

PeptldesTyr@-octreotide(TOC) [H-(D)Phe-Cys-Tyr-(D)Trp-Lys-Thr-Cys

Thr(ol)] and Lys5(BOC)-TOC (where BOC is butoxycarbonyl)werepurchasedfrom BachemLtd., SaffronWalden,UK.

HPLCA solvent module 125 (Beckman Coulter, Inc., Fullerton, CA)

with a 166-nm ultraviolet detector (Beckman) and radiometricdetection was used for reverse-phase high-performance liquidchromatography(HPLC) analysisandpreparation.An UltrasphereODS 5-jim system(Beckman),4.6 X 250 mm column, I mL/minflow rate, and 220-nm ultraviolet detection,was used with thefollowing solvent systems:method 1—acetonitrile(ACN):0.l%trifluoroacetic acid:water (0—3mm, 0% ACN; 3—10mm, 0—40%ACN; 10-20 mm, 40% ACN; 20-23 mm, 40-70% ACN; 26-27

In this paper we describe the preclinical evaluation of @TchydrazinonicotinyI-Tyr@-octreotide(HYNIC-TOC)using differentcoligandsfor radiolabelingand a comparisonof their in vitro andin vivo properties with 1111n-diethylenetriaminepentaacetic acid(DTPA)-octreotide.Methods: HYNIC-TOCwas radiolabeledathighspecificactivitiesusingtricine,ethylenediaminediaceticacid(EDDA),andtricine—nicotinicacidascoligandsystems.Receptorbinding was tested usingAR42J rat pancreatictumor cell membranes. Internalization and protein binding studies were performed,and biodistributionand tumor uptakeweredeterminedinAR42J tumor-bearing nude mice. Results: All @Tc-labeledHYNICpeptidesshowedretainedsomatostatin-receptorbindingaffinities(Kd< 2.65 nM). Proteinbindingand internalizationrateswere dependenton the coligandused. Specifictumoruptakebetween5.8 and 9.6 percentageinjecteddose per gram (%ID/g)was foundfor the @Tc-labeIedpeptides,comparedwith 4.3%ID/g for 1111n-DTPA-octreotide.Tricine as coligand showedhigher activity levels in muscle, blood, and liver, whereas tricinenicotinicacid producedsignificantlevelsof activity in the gastrointestinaltract. EDDAshowedthe most promisingoverall biodistributlon profile, with tumor-to-liverand tumor-to-gastrointestinaltract ratios similar to those obtainedwith 1111n-DTPA-octreotide,lower ratios in blood and muscle, but considerably highertumor-to-kidney ratios. Conclusion: TOG can be radiolabeled tohigh specific activities using HYNIC as a bifunctional chelator.The high specific tumor uptake, rapid blood clearance, andpredominantlyrenal excretionmake @‘@‘Tc-EDDA-HYNIC-TOCapromising candidate for an alternative to 1111n-DTPA-octreotidefor tumor imaging.Key Words: somatostatin;octreotide;hydrazinonicotinamide;

@Tc

J NucI Med2000;41:1114-1119

ver the last few years, @In-diethylenetiiaminepentaacetic acid (DTPA)-octreotide has found widespread clinicalapplicability, especially in oncology (1—3).However, limitations, especially concerning availability, imaging properties,

Received Jul. 6, 1999; revision accepted Sep. 23, 1999.Forcorrespondenceor reprintscontact:StephenJ. Mather,PhD,Nuclear

Medicine Researth Laboratory, St. Bartholomew's Hospital, West Smithfield,LondonEC17BE,UK.

1114 THEJoui@i. OFNUCLEARMEDICINE•Vol. 41 •No. 6 •June 2000

99mTCHyNIC [Tyr3]-Octreotide for ImagingSomatostatin-Receptor—PositiveTumors:Preclinical Evaluation and Comparisonwith@@ 11n-OctreotideClemens Decristoforo, Laura Melendez-Alafort, Jane K. Sosabowski, and Stephen J. Mather

Nuclear Medicine Research Laboratory, St. Bartholomew's Hospital, West Smithfield, London, United Kingdom

by on May 17, 2020. For personal use only. jnm.snmjournals.org Downloaded from

mm,70—100%ACN) or method2—ACN:0.OlN phosphatebuffer,pH 6.2 (0—3mm, 0% ACN; 3—10rain, 0—25%ACN; 10—20mm,25% ACN; 20-23 mm, 25-70% ACN; 26-27 ,nin, 70-100%ACN).

SPE PurificationFor purification of the radiolabeled peptide for stability and

proteinbindingstudies,asolid-phaseextraction(SPE)methodwasused. The radiolabeling mixtu@ was passed through a preactivatedC18SepPakmini cartridge(Waters,Milford, MA). The cartridgewas washed with 5 mL water, the radiolabeled peptide was elutedwith 80% acetonitrile, and the organic solvent was evaporatedunder a vacuum. This method efficiently removed all hydrophilicimpurities (@“TcO4, @“Tc-coligand)and @Tc-colloidto aconcentrationof lessthan2% whentestedby HPLC or thin-layerchromatography.

Synthesis of HYNIC-TOCFive micromoles 6-BOC-hydrazinopyridine-3-carboxylicacid

(BOC-HYNIC), 6pmol O-(7-azabenzotriazolyl)-1,1,3,3-tetramethyluroniumhexafluorophosphate,and20pmol diisopropylethylaminein 300 i.ILdimethylformamidewereallowedto reactfor 15mm atroom temperature.Sixty microlitersof this solutionwereaddedto1 @.imol[Lys5-BOCJ-protectedTOC (Bachem) in a mmxtureof 20 p1dimethylformamide(DMF):5 pL waterandallowedto reactfor 1h.The resultingsolution waspurified on a SepPakcolumn, and thepeptide was deprotectedwith trifluoroacetic acid containing 2%thioanisole and purified on HPLC using method 1.

Radlolabeling of HYNIC-TOCTricine as Coligand. In a rubber-sealed vial, 10 pg HYNIC-TOC

were incubatedwith 0.5 mL tricine solution (100 mg/mL in 25mmol/L succinatebuffer, pH 5.0), 0.5 mL @TcO4solution(100—1000MBq), and 25 piLtin(ll) solution (10 mg SnCl2 . 2H20in 10 mL nitrogen-purged 0.1 N HC1) for 30 mm at roomtemperature.

EDDA as Coligand. Ten micrograms HYNIC-TOC were incubatedwith 0.5 mL EDDA solution (10 mg/mL, pH 7.0), 0.5 mL

@“TcO4solution (100—1000MBq), and 5—10pL tin(II) solution(10 mg SnCl2 . 2H20 in 10 mL nitrogen-purged 0.1 N HC1)for 60mm at roomtemperature.

Tricine and Nicotinic Acid as Coligands (Ternary ColigandSystem). Ten micro@@s HYNIC-TOC, 0.4 mL tricine solution(100 mg/mL in 25 mmol/L succinate buffer, pH 5.0), 0.1 mLnicotinic acid (20 mg/mL in 25 mmol/L succinatebuffer,pH 5.0),0.5 mL @‘@‘TcO4solution (100—1000MBq), and 25 pL tin(ll)solution (10 mg SnCl2. 2H20 in 10 mL nitrogen-purged0.1 NHC1)were heated for 15 mm at 100°C.

“In-DTPA-octreotidewas prepared from a commercial kit(Octreoscan;Mallinckrodt Medical BV, Petten,The Netherlands).

Plasma Protein BindingProtein binding of the SepPak-purified @‘@‘Tc-labeledpeptide

wasdeterminedafter 15mm and 1,2, 3, and4 h of incubationinhuman plasma at 37°Cusing size exclusion chromatography withSephadex G-50 mithcolu@s (Microspin 0-50; Pharmacia Biotech,Piscataway,NJ). The columnswereprespunat 2000gfor 1mm. A 25-j.d plasma sample was added, and the column wascentrifugedagainat 2000gfor 2 mmn.The collectedeluateandthecolumn were counted in an Na! scintillation counter, and proteinbound peptidewas calculatedas the percentageeluted from thecolumn. Controls were studied after incubation in phosphatebuffered saline for 1h at 37°C.

Somatostatin Receptor BindingThe binding affinity of peptide conjugates was tested in a

competition assay against ‘@‘I-somatostatin-l4as described previ

ously (20). Rat pancreatictumor cell (AR42J) membraneswereused as a source for somatostatin receptors, bound radioligand wasseparatedfrom free radioligand by filtration through glass fiberfilters (grade GF/C; Whatman, Fairfield, NJ), and inhibitoryconcentration ofSO% (IC@o)values were calculated using nonlinearregressionwith version5.0Origin software(Microcal, Northampton, MA). The specificbinding of the @Tc-labe1edpeptideswasdetermined by competition against unmodified TOC in a similarassay.Kd valuesof carrier-free @“Tc-labeledpeptideconjugates(HPLC-purified,method2) weredeterminedin radioligandsaturation assaysusing AR42J cell membranesas the receptorsourcewith increasingamountsof peptideand 1 pmol/L cold peptidetodeterminenonspecificbinding.

Internalization StudiesStudiesof the internalizationof receptor-bound @“Tc-labeled

peptides were performed by adapting methods described previously (21,22).Live AR42J cells were washedtwice with culturemedium (Roswell Park Memorial Institute 1640/2% bismuthsulfite agar).Cells equivalentto 0.5 mg proteinwereincubatedintriplicate with l50,000-cpm carrier-free @Tc-labeledpeptide(HPLC-purified, method 2) with and without an excess of coldpeptide(1 pmol/L) at37°C.After 20,40,60,and90mm,cellswereseparatedby centrifugation and washed with cold culture medium.Surface-boundactivitywasremovedbyincubationwithacidbuffer(50 mmol/L glycine-HCI/l00 mmol/L NaCl, pH 2.8) at roomtemperature for 20 mm, followed by 2 washing steps. Surfaceboundactivity andinternalizedactivity weremeasuredandrelatedto thetotal activity added.

In VIvo Tumor Uptake StudiesAnimal experimentswere carried out in compliancewith the

Animals (ScientificProcedures)Act1995,accordingto British law.AR42J rat pancreatic tumor cells (5—10X l0@ cells) were

injected subcutaneouslyinto the flank of nu/nu mice (ImperialCancerResearchFund, London, UK). Between 10 and 30 dayslater, tumors had grown to a size of 0.5—1mL. Eight animals werestudied simultaneously, 4 of them with intraperitoneal pretreatmentof 50 pgoctreotide30 mmbeforeinjectionof theradiopharmaceutical. The animals were killed 4 h after 1 MBq @Tc-labeledpeptide was injected into the tail vein, and samples of differentorgansand the tumor were dissectedand counted.Uptakeof theradiopharmaceutical in terms of percentage injected dose (%ID)/gand %ID/organ was calculatedby referenceto standardspreparedfrom dilutions of the injected preparation.Specific uptake wasdeterminedby comparisonofblocked andunblockedanimalsusingthe Student t test.

RESULTS

Radiolabellng, In Vitro Stabllfty, and Protein BindingRadiolabeling of HYNIC-TOC was performed at specific

activities greater than 37 GBq/pmol. Quantitative labeling(>98%) was achieved with tricine and tricine—nicotimc acidas coligand, whereas with EDDA as coligand an averagelabeling yield of 63% was observed (Table 1). The resulting

@°@Tccomplexes showed a high in vitro stability. Labeling

99MTc@HYNIC@[TYR3J@OcTREoTiDE Decristoforo et al. 1115

by on May 17, 2020. For personal use only. jnm.snmjournals.org Downloaded from

KdofIC@ofcold@Tc

conjugatesLabelingpeptides(nmol/L)Coligandyield(%)(nmovL)HYNIC-TOC0.65Tricine

Tricine—nicotinicacid

EDDA98.298.763.61.14

2.112.65TOG0.50———

TABLE 1In Vitro Data for @Tc-IabeIedHYNIC-TOC

and Nonconjugated TOC

and saturation assayswith corresponding Scatchard analysisfor EDDA and tricine—nicotinic acid as coligand.

Receptor binding studies on intact cells showed a rapidinternalization of the 99mTc@labeledpeptides. Figure 3 showsthe internalization of @Tc-HYMC-TOC using EDDA andtricine—nicotinicacid as coligands.Surface-boundactivityremained less than 1% per milligram of protein, andinternalized activity rapidly increased over time, with afaster internalization for EDDA (5% per milligram ofprotein after 80 mm) compared with tricine—nicotinic acid ascoligand (2% per milligram of protein after 80 mm). Withtricine alone as coligand, a high nonspecific binding wasobserved, and no quantitative comparison could be made.

Biodistribution and Tumor Uptake

Biodistribution data of @Tc-labeledHYNIC-TOC usingdifferent coligands in AR42J tumor-bearing nude mice incomparison with ‘@In-DTPA-octreotideare summarized inTable 2. Significant differences (t test, P < 0.05) in tumoruptake between blocked and unblocked animals were found

for all compounds tested. A higher specific, but alsononspecific, tumor uptake was found for all @Tc-labeledpeptides (EDDA as coligand, 9.6 and 1.8 %IDIg,respectively; tricine as coligand, 9.6 and 3.0 %ID/g, respectively;and tricine—nicotinicacid as coligand, 5.8 and 1.0 %ID/g,respectively)comparedwith @In-DTPA-octreotide(4.3 and0.8 %ID/g, respectively).

The highest residual activity levels in all organs werefound for tricine as coligand, especially in blood (1.14%IDIg), muscle (0.92 %ID/g), and liver (2.08 %ID/g).EDDA and tricine—nicotinicacid as coligands showedhigher activity levels compared with @In-DTPA-octreotidein blood (>0.26 versus 0.07 %ID/g), liver (>0.75 versus0.47 %IDIg), and gut (>1.58 versus 0.55 %ID/g) butconsiderably lower levels in kidneys (<4.7 versus 22.1%ID/g). EDDA showed higher levels in muscle, liver, andspleen compared with tricine—nicotinicacid but lower levelsin gut.

Tumor-to-organ ratios are also shown in Table 2. Thelowest ratios were found for tricine as coligand. ‘@In-DTPAoctreotide showed the highest ratios, especially for blood(62.5) and muscle (52.3); tumor-to-liver (9.14) and tumor-togastrointestinaltract (7.7) ratios were similar to thosewithEDDA as coligand (9.11 and 6.11, respectively). Tricinenicotinic acid showed high tumor-to-muscle ratios (51.14)but low tumor-to-gastrointestinal tract ratios (1.78). Kidneyuptake was considerably lower for EDDA and tricinenicotinic acid compared with “In-DTPA-octreotide,withtumor-to-kidney ratios of 2.05, 1.59, and 0.19, respectively.

DISCUSSION

Our aim was to prepare a @Tc-labeledanalogue that is atleast comparable with @In-DTPA-octreotide in its ability toimage somatostatin receptors in vivo. Such an aim requirespreparationof a @°‘Tctracerwith good in vitro and in vivostability, high affinity for somatostatin receptors, and afavorable pattern of biodistribution. This last parameter

and stability studieshave been describedin greater detailelsewhere (16).

Plasma protein binding depended on the coligand used.Resultsof time-coursestudiesare shownin Figure 1. Whentricine was used as coligand, an increase in protein binding

over time from 12% to greater than 30% was observed.Using the ternary coligand system tricine—nicotinic acid, asteady level of protein binding of approximately 17% wasachieved. EDDA showed the lowest level of protein binding—less than 10%—with no significant increase over time.

Receptor Binding and InternalizationThe resultsof receptorbindingstudiesare summarizedin

Table 1. In displacementstudiesusing @I-somatostatin-14as the radioligand, an IC50 value for HYNIC-TOC of 0.6nmol/L, compared with 0.5 nmolfL for unmodified TOC,was determined. Displacement studies with TOC using the

@Tc-labeledpeptide as radioligand showed specific binding for all preparations independent of the coligand used.Saturation studies with the @Tc-labeledpeptide HYNICconjugates resulted in K@values in the nmol/L range: 1.1nmol/L for tricine, 2. 1 nmol/L for the tricine—nicotinic acidternary complex, and 2.6 nmol/L for EDDA as coligand.Figure 2 shows binding curves from displacement studies

FIGURE 1. Protein binding versus timeTOC injection.NA= nicotinicacid.

I.eC‘5

0 1 2 3 4

timeIhouisI

after @Tc-HYNlC

1116 THEJOuRNALOFNUCLEARMEDICINE•Vol.41 •No. 6 •June2000

by on May 17, 2020. For personal use only. jnm.snmjournals.org Downloaded from

2001506 8 10 12

A2oo B@00

8150

C.;100

0

C

2

C

I=2

I0,010

0

0,005

0,000

bound Ifmol/mg proteininM ‘Tc-compleiD2o

10

8

6

4

2

15

e

.5CC2

5

1E-4 1E-3 0,01 0,1 1 10

nM HYNIC-TOC

1E-4 ff3 0,01 0,1

nM TOC

100 1000

FIGURE2. Receptorbindingof@Tc-labeledpeptidesonAR42Jcellmembranes.(A)SaturationcurvesofHYNIC-TOGwithEDDAand HYNIC-TOCwithtricine—nicotinicacid. (B) Scatchardplotscorrespondingto saturationcurves.(C) DisplacementcurvesofHYNIC-TOGwith EDDAand HYNIC-TOGwithtricine—nicotinicacid. Competitoris unmodifiedTOC. (D) Displacementcurveof1251-somatostatin-14usingHYNIG-TOCas competitor.

requires that the compound show a rapid blood clearanceand low uptake in receptor-negative tissues and organs ofexcretion. In particular, the radiopharmaceutical should beexcreted predominantly through the renal system to avoidaccumulation in the gastrointestinal tract and the consequenthindrance of detection of pelvic tumor deposits. In previousstudies, we showed that both the peptide sequence and thebifunctional chelate system can profoundly influence thebiodistribution pattern and excretory route (17) and that thisin vivo behavior can be predicted somewhat by in vitromeasurementof parameterssuchas stability, lipophilicity,and protein binding (23). In the studies, HYNIC showed anumber of advantages over @“Tclabeling approaches usingN3S-based ligands as bifunctional chelators for labeling ofsmall peptides for tumor imaging (17). In the currentinvestigation, we have shown that HYNIC-TOC can belabeled at high specific activities (>37 GBq/jimol), resultingin stable complexes with retained binding affinity to somatostatin receptors in the nmol/L range. This finding agrees wellwith the findings of Maecke and Behe (15), who alsoshowed that derivatization with spacers between the bifunc

tional chelatorandthepeptidedoesnot improvethe imagingpropertiesof octreotidederivatives. The selectionof coligands for use in the labeling is an important part ofpreclimcal evaluation. Babich and Fischman (24) firstshowed that the nature of the coligand can influence thebiologic properties of the radioconjugate, whereas Liu et al.(18,19) described an increased complex stability with EDDAand ternary ligand systems using N-heterocycles comparedwith tricine. The current study confirms that complexstability depends on the coligand used and also verifies ourprevious observation that patterns of biodistribution can bepredicted in part by the plasma protein binding behavior.Thus, the use of tricine as a coligand showed the highestdegree of plasma protein binding, corresponding with highlevels of activity in blood, liver, spleen, and muscle. The 2other coligands tested in this study showed much lowerlevels of protein binding, indicating a higher complexstability and correlating with considerably lower levels ofactivity, especially in blood and muscle. The ternary ligandsystem tricine—nicotinic acid showed some advantages inthe labeling process in that it gave quantitative labeling

•Decristoforo et al. 1117

by on May 17, 2020. For personal use only. jnm.snmjournals.org Downloaded from

0 20 40 60 80

HYNIG-TOC HYNIG-TOCwithwfth EDDA tricine—nicotinicacidHYNIG-TOC withtncine1111n-octreotideSite

Unblocked Blocked Unblocked BlockedUnblockedBlockedUnblockedBlockedMean

±SDBlOOd0.28 ±0.04 0.36 ±0.06 0.26 ±0.08 0.28 ±0.031.14 ±O.1O@1.31±0.090.07 ±0.010.06 ±0.03Liver1.06±0.40 0.99±0.40 0.75 ±0.67 0.42 ±0.192.08 ±0.40 2.20 ±0.390.47 ±0.090.47 ±0.19Kidney4.71 ±1.38 6.67 ±3.05 3.64±0.52 4.40 ±0.7514.57 ±3.42 18.15±5.3422.12 ±6.5323.5 ±14.39Spleen0.40±0.15 0.31 ±0.12 0.18±0.04 0.18 ±0.061.07 ±0.16 0.98 ±0.040.16 ±0.020.19 ±0.07Pancreas0.45 ±0.23 0.26 ±0.09 0.25 ±0.03* 0.11±0.021 .66 ±0.0& 0.80±0.130.16 ±0.07*0.05 ±0.02Gut1.58±0.39 1.46±0.73 3.25±0.94 2.37 ±0.512.32 ±0.42 2.82 ±0.860.55 ±0.150.70 ±0.50Adrenals0.86 ±0.25* 0.43 ±0.26 0.63±0.23* 0.35±0.081 .80 ±0.20* 1.24±0.140.24 ±0.07*0.11 ±0.08Muscle0.31 ±0.28 0.41 ±0.46 0.11 ±0.08 0.08 ±0.020.92 ±0.14 0.73 ±0.150.08 ±0.070.03 ±0.02Tumor9.65 ±2.16* 1.82±0.81 5.80 ±2.31* 1.02±0.199.58 ±0.90 3.04±0.754.26 ±1.00*0.79 ±0.25RatioTumortoblood

33.9722.678.3862.54Tumortoliver 9.117.774.619.14Tumortogut 6.111.784.137.68Tumor

to muscle 31.29 51.1410.4652.28Tumortokidney2.051.590.660.19*Significent

difference (P < 0.05) between blocked and unblockedanimals.Valuesare%ID/g4 hafterinjectionof @Tc-labeIedpeptide,withblockingwith50pgoctreotide30mmbeforeinjection.

did the tricine—nicotinic acid conjugate (Fig. 3), comparablewith levels observed for “In-labeled octreotide analogs(22). This increased rate of internalization ofthe radioligand

receptor complex, with subsequent retention of the radionuclide within the lysosomal compartment, seems likely to bein some degree responsible for the higher tumor retention 4 hafterinjection.

The useof EDDA ascoligandproducedthe mostpromising pattern of biodistribution of all the @°@Tc-TOCcomplexes explored, combining the highest degree of specifictumor uptake with the fastest blood clearance and the lowestlevels of uptake in soft tissue and the gastrointestinal tract.In comparison with “In-DTPA-octreotide,a higher tumoruptake was seen, but also higher levels in some receptornegative tissues, especially blood, muscle, liver, and spleen.HYNIC-TOC with EDDA thus produced tumor-to-organratios almost identical to those of “In-DTPA-octreotide inthe liver and gastrointestinal tract, approximately half thoseof “In-DTPA-octreotide in blood and muscle, but 10-foldhigher than those of “In-DTPA-octreotide in the kidney.

The question of how well HYNIC-TOC with EDDA willcompare with octreotide in clinical studies remains to beanswered, because studies in animal models can onlypartially predict the likely pattern of biodistribution inhumans. However, some patient studies recently performedwith @°@Tc-tricine—HYNIC-TOC(25) have had promisingresults showing its ability to image somatostatin-positivetumors in humans.The faster clearance and lower soft-tissueandgastrointestinaluptakeof the EDDA complex in animalstudies raise hopes that its performance in humans will alsobe better than that of the tricine complex. The lower renalaccumulation of 99mTc4abeled conjugates may also prove an

--0—EDDAsurfac bound—U---EDDAlnternaftmd—0- Tricine/NA .urface bound—.— TrIcIn&NA internalbod

C@8

0.0@

Imini

FIGURE3. Specificinternalizationandsurfacebindingversustime, as percentage of @Tc-labeledpeptides per milligram ofprotein(intactAR42J cells). NA = nicotinicacid.

yields, whereas EDDA typically gave only a 60% yield andrequired a purification step before use. The great improvement in complex stability compared with tricine aloneindicates the potential of ternary coligand systems for @Tclabeling for peptides in general, but the pattern of biodistribution for this complex was not ideal, with significant levelsof activity appearing in the gastrointestinal tract. Despite acomparablein vitro binding affinity, tricine—nicotimcacidproduced a lower tumor uptake (5.8 %IDIg) than did theother coligand systems (9.6 %ID/g for EDDA and tricine). Apossible reason for the difference may be found in therelative rates of internalization of these compounds. In invitro studies on AR42J cells, HYNIC-TOC with EDDAshowed a significantly higher level of internalization than

TABLE 2Biodistribution and Tissue Ratios in AR42J Tumor-Bearing Nude Mice

1118 THE JouRNALOF NUCLEARMEDICINE •Vol. 41 •No. 6 •June 2000

by on May 17, 2020. For personal use only. jnm.snmjournals.org Downloaded from

advantage, because high kidney retention of “In-DTPAoctreotide is a significant problem in imaging this region.

Recently, another @Tc-labeled somatostatin analogue,P829 (26) (depreotide), has been approved for human use.Depreotide is basedon the carbocyclic structure of seglitide,which is known to have a different subtype specificitycompared with octreotide (27). In clinical trials, depreotidehas shown a pattern of tumor specificity different from thatof octreotide, and a high binding affinity to somatostatinreceptor subtype 3 has been found for P829 in a recent study(28). Depreotide itself cannot, therefore, be considered adirect substitute for octreotide. Because the amino acidsequence of 1@yr3-octreotidebears a closer resemblance tooctreotide itself, HYNIC-TOC with EDDA and “In-DTPAoctreotide might be expected to share a common profile ofreceptor subtype specificity, but this possibility remains tobeproven.

CONCLUSION

We haveshownthatthehydrazinonicotinamideconjugateof Tyr3-octreotide can be labeled with @Tcto high specificactivities and somatostatin binding affinity. EDDA wasfound to be the most promising of all the coligands tested.

Although the biodistribution of HYNIC-TOC with EDDAis different from that of “In-DTPA-octreotide, the high

specific tumor uptake, low gastrointestinal activity, andrapid renal elimination with low renal retention make this99mTclabeled peptide a promising alternative to “In-DTPAoctreotide for imaging somatostatin receptor—positivetumors in humans. Clinical studies to explore this possibilityare under way.

ACKNOWLEDGMENTS

The authors thank Helmut R. Maecke and Martin Behe fortheir advice on peptide conjugation and for sharing theirexperience in HYNIC labeling. In addition, the authorsthank Riccabona for his support. This study was supportedby an EC Marie Curie ResearchTraining Grant, proposal EB4001GT963891; the IAEA Co-ordinated Research Programme “99mTcLabelled Peptides for Imaging of PeripheralReceptors―; and the Imperial Cancer ResearchFund.

REFERENCES1. Rieger A, Rainov NG, Elfrich C, et al. Somatostatin receptor scintigraphy in

patients with pituitary adenoma. Neumsurg Rev. 1997;20:7—12.2. Krenning EP, Kwekkeboom Di, Bakker WH, et al. Somatostatin receptor

scintigraphy with [“-In-DTPA-D-Phel and [123-I-tyr]-octreotide: the Rotterdamexperience with more than 1000 patients. EurfNuclMed. 1993;20:716-731.

3. Eising EG, Bier D, Knust EJ, Reiners C. Somatostatin-receptor scintigraphy:methods, indications, results. Radiologe. 1996;36:81—88.

4. De Jong M, Breeman WAP, Bernard BF, et al. Evaluation in vitro and in rats of‘61ThD@A@@oflde, a somatostatin analogue with potential for intraoperativescanning and radiotherapy. EuriNuclMed. 1995:22:608—616.

5. Stolz B, Smith Jones PM, Albert R, Reist H, Macke H, Bruns C. Biologicalcharacterisation of[Ga-67] or [Ga-68] labeled DFO-octreotide (SDZ-216-927) forPETstudiesof somatostatinreceptor-positivetumors.HornsMetabRes. 1994:26:453—459.

6. Anderson Ci, Pajeau TS, Edwards WB, Sherman ELC, Rogers BE, Welch Mi. Invitro and in vivo evaluation of copper-64-octreotide conjugates. J Nuci Med.1995;36:23l5—2325.

7. Guhlke5, WesteraHi, BrunsC, StöcklinG. (2@[I*F]fiuoropropionyl@(D)Phel)@octreotide, a potential radiopharmaceufical for quantitative somatostatin receptor

imaging with PET: synthesis, radiolabeling, in vitro validation and biodistributionin mice. NuclMedBiol 1994:21:819—825.

8. Kolan H, Li J, Thakur ML. Sandostatin labeled with w@@Tc:in vitro stability, invivo validity and comparison with “ln-DTPA-octreotide.Pept Res. 1996:9:144—150.

9. Haberberger T, Zamora P. Hosono M, et al. Initial studies on @‘Tc-and‘88Re-somatostatinanalogues: stannous mediated direct-labelling and biodistribution in mice. In: Nicolini M, Banoli G, Mazzi U, eds. Technetium and Rhenium inChemistry and Nuclear Medicine. 4. Padova, Italy: 5G Editorali: 1995:367—371.

10. Mama T, Stolz B, Albert R, Bruns C, Koch P. Macke H. Synthesis and biologicalevaluation of a new somatostatin analogue (SDZ 219-387) labelled withtechnetium-99m. EurJ Nucl Med. 1994:21:437-444.

11@ Guhike 5, Schaffland A, Zamora P0, et al. “8Re-and @Tc-MAG3as prostheticgroups for labeling amines and peptides: approaches with pre- and postconjugatelabeling. NucI Med Biol. 1998:25:621—632.

12. Mainal, Stolz B, Albert R, Nock B, Brunt C, Macke H. Synthesis, radiochemistryand biological evaluation of w@Tc[N4@(D)PHE1)@octreotide, a new octreotidederivative with high affinity for somatostatin receptors. In: Nicolini M, Banoli G,

Mazzi U, eds. Technetium and Rhenium in Chemistry and Nuclear Medicine. 4.Padova, Italy: SG Editorali: 1995:395—400.

13. Thakur ML, Kolan H, Wiaderkiewicz R, Pallela VR, Duggaraju R, Schally AV.Radiolabeled somatostatin analogs in prostate cancer. NuclMedBiot 1997:24:105-113.

14. Valabhajosula 5, Moyer BR, Lister-James J, et al. Precinical evaluation oftechnetium-99m-labeled somatostatin receptor-binding peptides. J Nucl Med.1996:37:1016—1022.

15. Maecke HR. Behe M. New octreotide derivatives labelled with technetium-99m[abstracti. JNucIMed. 1996;37(suppl):29P.

16. Decristoforo C, Mather SJ. Preparation. w@@Tc@labelingand in vitro characterization of hynic and N(3)S modified RC-160 and [TYR(3)] octreotide. BioconjugChem. 1999;10:70l—702.

17. Decristoforo C, Mather Si. Technetium-99m somatostatin analogues: effect oflabelling methods and peptide sequence. EurJ Nuci Med. 1999:26:869—876.

18. Liu 5, Edwards DS, Looby Ri, et al. Labeling a hydrazino nicotinamide-modifiedcyclic lib/IlIa receptor antagonist with w@@Tcusing aminocarboxylates as coligands. Bioconjug Chem. 1996:7:63—71.

19. LiuS.EdwardsD5,HarrisAR.Anovelternaryligandsystemfor @Tc-labeingofhydrazino nicotinamide-modified biologically active molecules using imine-Ncontaining hetemcycles as coligands. Bioconjug Chem. 1998:9:583—595.

20. Holland U, van Knoetsveld PM, Wouters N, Waaijers M, Reubi JC, LambertsSW. Dissociation of antiproliferativeand antihormonaleffects of somatostatinanalogue octreotide on 7315b pituitary tumorcells. Endocrinology. l992;l3l:57l—577.

21. Casalini P, Luison E, Menard 5, Colnaghi MI, Paganelli 0, Canevari S. Tumourpretargeting: role ofavidin/streptavidin on monoclonal antibody internalization. JNucI Med. 1997:38:1378—1381.

22. De Jong M, Bernard BF, D@ Bniin E, et al. Internalization of radiolabelled[DTPA°]octreotide and [DOTA, Tyr3)octreotide: peptides for somatostatin receptor targeted scintigraphy and radionuclide therapy. Nuci Med Commun. 1998:19:283—288.

23. Decristoforo C, Mather SJ. 99m-technetium-labelled peptide-HYNIC conjugates:effects of lipophilicity and stability on biodistribution. Nucl Med Biol. 1999:26:389—396.

24. Babich JW, Fischman Al. Effect of “co-ligand―on the biodistribution of‘@“Tc-Iabeledhydrazino nicotinic acid derivatized chemotactic peptides. NuciMed Biol.1995:22:25—30.

25. Bangard M, Behe M, Bender H, et al. Technetium-99m-octreotide for thedetection of somatostatin receptor positive (SSTR+) tumors: preliminary results[abstractj. EuriNuclMed. 1998:25:837.

26. Valabhajosula 5, Moyer BR, Lister-James J, et al. Preclinical evaluation oftechnetium-99m-labeled somatostatin receptor-binding peptides. J Nucl Med@l996;37: 1016—1022.

27. Patel YC, Greenwood MT. Panetta R, Demchyshyn L Nizuik H, Srikant CB. Thesomatostatin receptor family. Life Sci. 1995:57:1249—1265.

28. \‘irgoliniI, Leimer M, Handmaker H, et al. Somatostatin receptor subtypespecificity and in vivo binding of a novel tumor tracer, w@@Tc@P829.Cancer Res.1998:58:1850-1859.

99MTC@yC[yR3]O@D@O@DE •Decristoforo et al. 1119

by on May 17, 2020. For personal use only. jnm.snmjournals.org Downloaded from

2000;41:1114-1119.J Nucl Med.   Clemens Decristoforo, Laura Melendez-Alafort, Jane K. Sosabowski and Stephen J. Mather 

In-Octreotide111Preclinical Evaluation and Comparison with ]-Octreotide for Imaging Somatostatin-Receptor-Positive Tumors:3Tc-HYNIC-[Tyr99m

http://jnm.snmjournals.org/content/41/6/1114This article and updated information are available at:

  http://jnm.snmjournals.org/site/subscriptions/online.xhtml

Information about subscriptions to JNM can be found at:  

http://jnm.snmjournals.org/site/misc/permission.xhtmlInformation about reproducing figures, tables, or other portions of this article can be found online at:

(Print ISSN: 0161-5505, Online ISSN: 2159-662X)1850 Samuel Morse Drive, Reston, VA 20190.SNMMI | Society of Nuclear Medicine and Molecular Imaging

is published monthly.The Journal of Nuclear Medicine

© Copyright 2000 SNMMI; all rights reserved.

by on May 17, 2020. For personal use only. jnm.snmjournals.org Downloaded from