clinical outcome of the patients treated with MIBG ispoorly correlatedwith valuesof the doseabsorbedby thetumor and derived from gamma camera estimates of MIBGkinetics and uptake (4). This probably stems from the factthat the calculationof these values is based on the assumption that MIBG is homogenously distributedin the tumor.Precise determination of intracellular as well as intratumordrug localization is essential for a reliable microdosimetricevaluation of this targeted radiotherapy.

Although little is known about the intratumorpatternofMIBG biodistribution, its intracellular localization has already been achieved. In vitro experiments indicate thatMIBG uptake is heterogeneous and that this tracer is almost totally located in the cytosol of neuroblastomacells(5,6). However, its exact subcellular localization remainscontroversial since routine fixation procedures using organic solvents are unable to prevent the drug's diffusionfrom its originalsite of uptake (7).

To optimize MIBG therapy, the biodistributionof thistracerwas determinedboth in its reference targetedtissue,the adrenomedulla, and in neuroblastomaxenografts intonude mice with the help of secondaiy ion massspectrometiy (SIMS) microscopy. This technique, developed in theearly 1960s (8), provides images of distribution of manystable or radioactivenucides, within biological tissues (9),has benefited from recent technological advances and mayprove useful in solving certain biomedical problems (10).Special ciyotechniques adapted to tissue sample processing andSIMS analysisrequirementswere appliedin orderto achievein situ immobilization of ‘27I-MIBGand subsequent ion microscope mappingrepresentativeof its in vivobiodistribution.

MATERIALSAND METHODS

Cells and EXpO,Imental TumorsTwo cell lines were tested for MIBG accumulation. The first

was theSK-N-SHneuroblastomacell linewhichhasbeenpreviouslydescribed(11)andshownto activelytake up MIBGinvitro(12,13). Cells were propagated in RPMI medium containing 12%fetal calf serum and peniciffin/streptomycin (100 lU/mi) andtrypsinized on average eveiy five days. The second cell line was

Mlcrodoalmethcevaluationsof targetedradiotherapyof neurob@stomawfth metaiodobenzylguanidine(MIBG)requirepreciseassessmentof the intracellularand Intratumordistributionof thedrug.We reportthe useof secondaryion massspectrometry(SIMS)microscopy, a technique capable of mapç@ngany them

@alelementwithina biOlOgicalspecimen,to determine @@iMIBG biodistribution in human neurob@stomaSK-N-SH Xenografted into nude mice. Highly specific images of @i-MlBGbiodistributionwere mappedwfthinthe tumor afterin vhioadministrationof the drug and sample processingw@icryotechniques(high-speedfreezingand cryo-embedding),which preventMIBGdiffusionfrom onginal sites of uptake.We showed that the biodistributionof the tracer was highly nonuniformwithin the tumor.At the cellular level,most ofthe drug accumulatedin the cytosolandpennudearareas.Incontrast,chemicalsampleproces@ngprovidednotonly a considerableloss in sensitivitydueto passivediffusionof thedrugintheorganicsolvents,butalsoartefactualimages mainly due to MIBG redistiibu@ononto the cell nuclei.Based on our findings in this SK-N-SH experimental tumormodel,wesuggestthatMIBGshouldbeattachedto long-rangeemitters,inthehopeof irradiatingthemanytumorousareasthatremain carder-free.

JNuciMed1993;34:1565-1570

ue to its high sensitivity and specificity, radioiodinelabeled MIBG has been used successfullyduring the lastdecade to localize neuroblastoma and other neural cresttumors (1,2). Indeed, positive @I-MffiGscans can evenbe obtained in patients bearing neuroblastoma with veiylow uptake values of MIBG by the tumor (3). In contrast,targeted radiotherapy of neuroblastomausing 131I-MIBGremains ratherdisappointingas only 30%—50%of patientswill achieve some form of remission. Unfortunately, this isboth incomplete and transitory in most cases.In addition,

ReceivedNov.17,1992;revisionacceptedMay25,1993.For correspondenceor reprintscon@ Dr. JérômeClerc, INSERMU 66,

L@oratoirede MicroscopieIonk@ueMa@que, InstitutGustaveRoussy,39 rueCafT@Desmoulins,94805ViflejUifCedex@France.

1565SIMS Microscopyand NeuroblastomaMIBG Detection•Clercet al.

SIMS Microscopy Imaging of the IntratumorBiodistribution of Metaiodobenzylguanidine inthe Human SK-N-SH Neuroblastoma Cell LineXenografted into Nude MiceJ. Clerc, S. Halpern, C. Fourré,F. Omri, C. Bnançon,J. Jeusset and P. Fragu

Equ4pe de Microscopie Ion@ue - INSERM U 66, Institut Gustave-Roussy, VillejuIfCede@ France

by on April 12, 2018. For personal use only. jnm.snmjournals.org Downloaded from

the IGR-N-835 cell line, derived from a Stage IV abdominal neuroblastomacapableof secretingdopamine,whichhadbeenohtained from a 2-yr-old girl (14). To obtain experimental tumors,the flanksof 6- to 8-wk-oldpre-irradiated(5 Gy) nudemice(SPSwiss) were inoculated either with 5 x iO@cells (SK-N-SH Uimors) or with tumor fragments (IGR-N-835 tumors), obtainedfrom a previously xenografted mouse. It took 8-10 wk to obtainSK-N-SH tumorsmeasuringapproximately0.5 cmi, while ashorter latencyperiod (about4 wk) was requiredfor IGR-N-835tumors.

DwgsIodine-127-MIBG(1300 @g/4OOz1)was purchasedfromUS

Bioindustries Company (Gif-sur-Yvette, France). Iodine-125-MIBG (specific activity 1110 MBq/g) and ‘@I-MIBG(specificactivity 185 MBijmg) used respectively for autoradiography orcountingpurposeswere kindlydonatedby the samecompany.Todeterminewhetherinvivo M1BGaccumulationwas possiblycellcycle dependent,we co-injectedbromodeoxyuridine(BrdU, 100;zg/g,SIGMA Chemical, St. Louis, MO), a thymidine analog usedas a proliferationmarker(15). All drugswere injectedintraperitoneally.

MIBGUptakeby theAdrenaLcand Tumors.Iodine-123-MIBGuptake by the adrenals was first determined in 10 mice injectedwith about 1.85 MBq and results expressed as a percentage of theinjecteddosepergramoftissue (%ID/g).Likewise,MIBGuptakeby tumors was determined both in mice bearing IGR-N-835 xcnografts (n = 3) and SK-N-SHxenografts (n = 4). Mice werealways killed for counting purposes 24 hr after administration of

@ilBG.Micmautoiudiogruphies. Live animals (n = 2) were injected

with about 185 MBq of 1@I-MIBG and killed 24 hr later. Tumorswerecut insmallfragments(n = 40)thatfirstweretransferredtoplastictubescontaining0.5 mlof fixativeandcounted.Adrenalswereexcisedandcountedinsimilarconditions.Chemicalsampleprocessing (see below) was carried out and semi-thin sections ofthespecimenweremountedonslidesandcoatedwithAmershamLM1 photographic emulsion. Exposure time was 10-30 days.Finally, preparations were stained with hematoxylin-eosin. In addition, loss of ‘@I-MTfBGradioactivityby the fragmentswassystematicallycounted at each step of the chemicalprocedure.

SIMSAna@ysis@ForSIMSexperiments,‘@I-MIBGwas administeredintraperitoneallyto the mice by escalatingdose increments(0, 50, 200, 400 and 600 @sg)and tumorswere excised 24 hr later.

FLr.ativeP@vcedw@s,EPnbeddÜ@gand Sectionà@.Both chemical and physical procedures were used for sample processing.Withthe chemicalprocedure,the tissue specimenwas cut into 1—2mm3 fragments and fixed by immersion in a solution of 0.1%glutaraldehydeplus 2%paraformaldehydein 0.1 M cacodylatebufferat a pH of 7.4. Fragmentswerethenrinsedin thebuffer,dehydratedin ethanolandembeddedin methacrylateresin(Historesin,Pharmacia,Sweden).Thephysicalprocedure(cryotechnique) was used because MIBG has been reported to be a highlydiffusible molecule, especially in organic solvents. With thismethod, tumors were ciyofixed by ultra rapid immersion (5000 Ksee-') in liquid propane, subcooled (77 K) by liquid nitrogen,ciyosubstituted in acetone (183 K) and then embedded in Lowicryl K11M (Chemische Werke Lowi, Germany) at 213 K. Whatever the procedure, semi-thin sections (3 @m)werecut and laid onultrapure gold holders at room temperature for SIMS analysis.

SIMSMicroscopy:‘271-MIBGMapping.We used the IMS-3Fmicroscope (CAMECA, Courbevoie, France) fitted with a digitalimagingsystemdesignedto dealwithspecificbiologicalproblems,aspreviouslydescribed(16,17).Aprimaiycesiumionbeam(Cs@)with an intensity of 10—25 nA and an energy of 10 KeV was

focused on the surface (60 or 102 sm) of the resin-embeddedspecimen.Atomssituatedin thesuperficiallayers(1—5am)wereprogressivelysputtered and possibly ionized. This techniqueisbasicallyerosiveand destroysthe specimenwhileanalyzingit. Inroutineworkingconditions,speed of erosionis about 1.5nm persec. Secondaryions, characteristicof theatomiccompositionofthe analyzed area were then focused, energy filtered and separated by a mass spectrometer. The analytical ion images werefinally displayed on a fluorescent screen coupled with a sensitiveS.I.Tvideocamera(LHESASIT4036).Inabiologicalspecimen,manyionizedatomsareemittedaspolyatomicclusterionswhosenumberof nucleonsmay be identicalto that of the singleion ofinterest. Actual weight of these cluster ions is slightly differentfrom that of the element to be mapped and can be eliminatedbythe mass spectrometerif a highmass resolution(M/i@M> 2000)isused. Unfortunately,working at high mass resolution reducesdetection sensitivity. This drawback was attenuated by usinghigh-speedintegrationfor the elemental ion images,which improved the signal-to-noiseratioandoptimizedfinalimagequality.Inaddition,localconcentrationscouldbeestimatedbymeasuringintensityof the secondaryion beam of interest with an electronmultiplierand using highlyhomogenoustissue equivalentstandards (18).

Biodistribution of the drugs was delineated on detection of‘@I-MIBGand 81Br(BrdU) while the histological structure of thespecimen was visible throughmappingof 31P,which is especiallyabundantin the cellnuclei.The lateralresolutionof the IMS-3Fis0.5 @m.

RESULTS

MIBG Uptake by Adrenals and TumorsMouse adrenals were found to have an ‘@I-MIBGup

take of 2.3 ±0.7 %ID/g at 24 hr. The IGR-N-835 tumorsfailed to accumulate MIBG. Values of uptake were O.3%ID/g which corresponded to “background―levels ofnontargettissuesuchasthe liver or lungs.In contrast,withSK-N-SH xenografts, we found a value of 0.82 ±0.30%ID/g.

Iodlne-125-MIBGImaging of Tumors byAutoradlography

At the end of chemical processing, about 75% of the‘@I-MIBGactivity was lost due to passive diffusionin theorganic solvents. Images of MIBG biodistribution in theSK-N-SH xenografted tumorswere obtained after3—4wkof exposure. A dramaticallyheterogeneous signal was obtamedwithin the tumor. A few clustersof small roundcellswere intensely stained in the tumor. Positive stainingwassometimesobservedeither in cellssurroundingthe vesselsor in isolated large cells (Fig. 1). No significant @I-MIBGtissue bindingwas found in the slides obtained from IGRN-835 xenografted tumors.

1566 The Journal of Nuclear Medicine •Vol. 34 •No. 9 •September 1993

by on April 12, 2018. For personal use only. jnm.snmjournals.org Downloaded from

,u@,s, —

I

SIMS MappIngof 1@I-MlBGIodine-127-MIBG in the Adrenomedulla. A specific@

signal was observed in the adrenals of mice regardless ofthe level of MIBG administered (50—600@g)with thechemical preparation. Drug biodistribution in the referencetarget organ involved the cytosol of the cells (Fig. 2), aspreviously reported (19).

Iodine-127-MIBG in the Tumo,@c:Influence of SamplePreparation Technique. A specific SIMS signal of@ wasalways detectable for the 600-pg injected dose and@MIBG images of biodistribution could be visualized, regardless of the sample preparation procedures. Elementalmapping of 1271revealed considerable heterogeneity ofMIBG uptake among the cells and even more within thetumor. However, with the chemical processing procedure,large areas in tumor samples were found to be iodine-free.

At a dose of 400 @g,1271signal recovery was hardlyeverconstantin the analyzed areasof chemicallypreparedfragments. For this dose level, MIBG biodistributionwas onlydisplayedby SIMS in cell clusterscapableof intensedrugaccumulation. Moreover, SIMS images of MIBG were ob

,. .. *@. @4. .@

..1.@@ e@@@@

@4* wI4@@.I:%@I@ I*%@@

@ ..@‘@ ,@ @4. .- .

.@.

•!@b@:@ •.

,

FIGURE2. SIMSimagesof 1@I-MlBGwithinthe adrenalmedullsof miceafterinvivoadminietraflonof thedrug(400 @g)andchemical sample processing.Left view: The tissue structure ismappedthrough31P,which is especiallyabundantin cell nuclei.Rightview:@ bIOdiStribUtiOnshowedcytosolicbiodistilbutlonofMIBG.Imagefields:60 ian.

served both in the cytosol and on the nuclei of tumor cellswith the chemical processing procedure (Fig. 3). In contrast, with cryo-prepared fragments, the@ signal wasonly emitted from the cytosol and perinuclear areas.Within the tumor, MIBG biodistributionwas found to behighly nonuniform (Fig. 4). In addition, ‘27I-MIBGcouldalways be detected by processing high-resolution iodinespectra even when the secondary ion beam lacked sufficient intensity to provide reliable imagescorrespondingtoa reasonable depth range (500 nm) for SIMS analysis,which graduallydestroys the specimen.

Basedon comparisonof 81Br-BrdU and @I-MIBGbiodistribution within the tumor, no clear-cut relationshipcould be found between tumor cell proliferation and acapacity for MIBG uptake in vivo. Indeed, although themost frequent feature encountered was either a simultaneous positivity or negativity for both 81Br and 1271,singlestainingcould alsobe observed(Fig. 5).

DISCUSSION

The SIMS microscopeIMS-3F proved to be effective inmapping ‘@I-MIBGand its analytical capacity was sufficient to demonstratea dramaticallynonuniform pattern ofMIBG biodistributionamong tumor cells.

Choice of adequate model systems to study MIBG uptake is of paramount importance. Mechanisms of MIBGuptake have been extensively studied in the human neuroblastoma SK-N-SH cell line, in vitro (7,12). Two uptakesystemsare involved. The first one fulfills all the characteristics of the neuronal uptake-i system (described forcatecholamines),namely high specificity and affinity andsaturability at extracellular concentration in the range ofiO_6 M. This specific uptake-i is mainly responsible forMIBGuptakeinvivo andallows MIBGincorporationup to30 times greater than the incorporation achieved by thesecondsystemwhich is likely to be simple diffusion (13).However, specific uptake depends on structuralorganization of the cells which conditionsthe number of cell contacts and on the preferential growth of certain cell subtypes, such as neuroblastic or Schwann-like cells, whichcoexist and may transdifferentiate (5,20,21). This prefer

4,..@j@ ‘@ .; @, .

@, :@@

‘p.@ .:@•*‘@. -@@ ...@

%., @7 ‘@@ I

,‘@ 7@ ,

@4@c@ , @w

@@ r'@.

FiGURE1. Autoradk@graphsof SK-N-SHneurot@astomaxenograftedintonudemiceinjected(ip)with 1850kBqof [1@I1-MIBG.Animalswerekilled24hrpostinjection.Sampleswereexposedfor3wk. Bar = 8 j@m.Tumor biodisthbutlon of MIBG was dramaticallynonuniform(Topview).Sometimes,onlypenvascular(Bottomview)neuroblastomacellswere stained.Dueto a chemicalsampleproceasingmethodthat cannot preventMIBG d@fusionfrom onginalsitesofuptake,noreliableinformationcanbedrawnregardingexactsubcellularlocalizationofthedrug.

4.

1567SIMSMk@roscopyandNeuroblastomaMIBGDetection•Clercatal.

.@ ..‘

I@. :‘@&@•e *@

w@@

p' •@@•.

by on April 12, 2018. For personal use only. jnm.snmjournals.org Downloaded from

•k@.

@. _

@‘?@b@@

I

—UN@@

FIGURE3. SIMSImagesof 1@I-MlBGInI@@ •processing.TissuestructureIsdelineatedthrough31PQeftview).Twocellshadstronglyincorporatedbromodecxyurldlne,whichIsmappedthroughstableel& (upperpartofmiddleview).The1@Idistribution(rightIm@e)IsexclusivelyInthefewcellssituatedIntheupperpartofthe Imagefield.Dueto Inedequateuseof a chemicalsampleprocessing,bothImagesof uptake,in the cytosol,andof redistribution,ontothenuclei,areevidenced.Imagefields:102sun.

ential growth probably occurs during xenografting and cxplains why most of these neuroblastoma cell lines partly(SK-N-SH) or totally (IGR-N-835) lose ability to take upMIBG, once inoculated in mice. Notwithstanding, highervalues of uptake were found with the subcutaneous 5K-N-SH tumor model when compared to those reported inMIBG positive human tumors (3,22). In addition, a variantof this tumor model has recently proved to be effective instudying 131I-MIBGtargeted radiotherapy (23). Xenograftsare likely to mimic what actually occurs in human tumorsbecause these models take into account drug accessibilityto tumor cells. This may be critical for MIBG uptake bytumors since many positive cells were seen close to vessellumens on the autoradiographs.

Sample processing techniques appearedto be crucial forassessing reliable images of MIBG biodistributionwithinbiological materials. Indeed, drug diffusion in organic solvents is only limited in tissues that possess a specific in

tragranularstorage system, such as the adrenalsand pheochromocytoma. This corroborates the feasibility of MIBGdetection by SIMS in the adrenals of mice even after injection of the lowest dose (50 @g)and chemical sampleprocessing. Unfortunately, it is likely that MIBO storage inneuroblastoma is not dependent on the presence of specialized neurosecretoiygranulesbut only reflects a re-uptakemechanism of the drug (24). Thus, with chemical sampleprocessing, most of the drug that had accumulated in thetumor was lost by diffusion in the solvents in such a waythat the remainingstaining only represented what had cccurred in the cells exhibiting the highest MIBG uptake.Moreover, the intracellular localization of MIBG should beinterpretedwith great caution when using thesemethods,because we found that both uptake and redistributionimages could be obtained.

Mappingof diffusible molecules such as MIBG in neuroblastoma must only be performed using cryoprepared

;%@

- - . .. .

,_ . .

.s: ‘@

. I a.@@ •.@

..@‘-I@ •@ . •@ ..

. •:‘@

FIGURE4. SIMSlm@esof1@I-MIBGint fthedrug(400 @)andprocessingsamplesusing high-speedfreezing,cryo-substltution erved and tumor nucleiwere stil mappedthrough31PQeltview).Bromodeox@udolneit@ (m@dleview).The 1@l-MIBGbkdstdbutlonappearedhighlynonuniformin the tumorand involvedthe cytosolof the cells (rightview).Imagefields: 102 sm.

1568 The Journalof NuclearMedicine•Vol.34 •No. 9 •September1993

.-.

@.

‘t. :.:

by on April 12, 2018. For personal use only. jnm.snmjournals.org Downloaded from

FiGURE5. SIMSImagesofthesupe.impositionof the elemental @nimagescorrespondingto Figures3 and 4, in pseudocoburs Tumor nuclei are Imaged In green@31P),81 bromIneof the BrdUInblueand

1271of the MIBG in red to yellow. Using thechemical sample processingmethod (leftvIew),MIBGwasmappedbothInthecytosolandartefactuallyontocell nuclei(redistribution), while with the cryotechniques(rightview),MIBGwas only distributedIn the cytoed. MIBG accumulationcould be ohservedboth In proliferative(BrdU positive)and nonproliterativecells (BrdU negative).Imagefields:102pin.

@:

•1@ Il@@::i.: ..

reasons,it is clear that SIMS imagesof ‘@I-MIBGin theSK-N-SH tumors mainly represent images of active uptake. However, the new generation of instruments thatuses microprobes should further enhance detection sensitivity and provide subcellular scaled quantitative images(10,28).

Using innovative SIMS microscopy techniques, thepresent study addresses fundamentalquestions concerningoptimization of metabolic radiotherapy with M!BG in cxperimental models. Our results on MIBG biodistribution inthe SK-N-SH tumor model strongly suggestthat MIBGshould be attached to long-rangeparticle emitters, in thehope of irradiatinglarge tumor areas found to be more orless carrier-free by crossfire. Use of the ‘@Ito label MIBO(29) would be of limited interest, since its decay processresultsin a cascadeofAuger electronswhosepath rangeistoo short for DNA targetingandbecauseMIBG is localizedin the cytosol. In addition, due to lack of uniformity inMIBG distribution within the tumor, 1@I labeling wouldlead to unchecked regions from which tumor regenerationcould proceed (30).

In human beings, SIMS detection of tracers labeled withstable halogens, such as ‘271-MIBG,should be proposedfor excised soft-tissue neuroblastoma, obtained after biopsy or surgery and previous administration of the “cold―drug. SIMS mapping of the tracer within the tumor wouldthen provide a reliablebiological model, which is a prerequisite for calculating the absorbed dose by computationalMonte Carlo methodswithout neglectingintratumor heterogeneity of uptake. In addition, the amount of cold tracerinjectedwould be identicalto that used in therapysince wedo not know whether use of test “hot―doseswould reiably mimic the behavior of the therapeutic dose in MIBGtherapy. Finally, the assessment of tissue specificity ofnew imaging or therapeutic agents, such as meta-bromobenzylguanidine, which can be labeled with the positronemitter 76Br,are promising new directions for developmentof SIMS in improving nuclear imaging of neural crest tumors.

ACKNOWLEDGMENTS

TheauthorsthankDr.J. Benard(Laboratoirede Pharmacologie Clinique et Moléculaire,IGR) for providing the IGR-N-835

tumor fragments (25). Unfortunately these methods aretime consuming but were previously shown to be adaptedto SIMSanalysisof evenhigh-speeddiffusingelementssuch as calcium, potassium or sodium (26). Using the cryotechniques, we showed that the@ signal emanatedmainly from the cytosol at the cellular level. However,minute traces of iodine distribution were, at times, cvidenced at the peripheryof cell nuclei, when superimposedwith the image of nuclear phosphorus distribution(Fig. 5).This minor distribution may simply reflect the poor lateralresolution of the IMS-3F for imaging molecules situatedclose to the nucleus; this phenomenon may have beenaggravatedby the fact that many neuroblastomacells hada small cytosol width (0.i—1pin). However, the main rcason this phenomenon occurred is that elemental ion imagesare acquired successively and represent deeper levels ofanalysis of the specimen, causing a further deterioration ofthe lateralresolution of up to 100—300nm, especially whenmapping trace elements for which longer analysis times(100—200see) are required.

Failure to obtain reliable ‘@I-MIBGimages in the 5K-N-SH tumors for lower injected doses (50 and 200 pg),partly stemmed from the decision to work with high massresolution (M@@M> 2000), which undoubtedly affectedamplitudeof the specificsignal(27). However, thiswas theonly way to guarantee signal specificity, since a phosphorus cluster ion (126.980 uma) is always detected close to theiodine emission (126.907 uma). The detectable 400-pg injected doseof MIBG in the mice may haveled to levelsofconcentration of the drug in the blood, for which bothactive and passive mechanisms of uptake could be involved. However, given that tumor uptake was borderingon 1 %ID/g,expected concentrations due to passive mechanisms ofuptake would be below the threshold of detectionof IMS-3F, which is of about 5 ppm (5 p@giodine per gramof tissue, assuming that the isotope is homogenously distributed in its matrix) (18). Mapping 1271was achievablebecause it correspondedin positive cells to local concentrations of iodine in the range of detectability. Finally, wewere unable to detect@ in the nontargetneuroblastomaIGR-N-835, even when larger doses of 1@I-MIBG(up to800 gig/mouse) were injected although simple diffusion ofthe drug was bound to occur in the tumor. Given these

1569SIMSMicroscopyand NeuroblastomaMIBGDetection•Clercat al.

(4.

f .

by on April 12, 2018. For personal use only. jnm.snmjournals.org Downloaded from

16. Olivo JC, K.hanE, Halpern S, BriançonC, Fragu P. Di Paola R. Microcomputer system for ion microscopy digital imaging and processing. IMicmscopy 196956:105—114.

17. FourréC, Halpern S, Jeusset .1, Qeic J, Fragu Ph. Significance of SIMSmicroscopyfor Technetium-99mmappingin leucocytes.I NuciMed 199233:2162-2166.

18.JeussetJ,Halpern5, BriançonC,GarciaF, FraguPh.Halogensstandardization by SIMS microscopyfor applicationsin biologicalsamples [Abstract].Bid Cell 1992;75:8a.

19. TelenczakP, RicardM, }Ialpern 5, FraguP. Amethodtoevaluate the tissueabsorbeddose in metabolicradiotherapy preliminaiyresults with mIBO.In: BenninghovenA, Evans C, McKeeganK, Storms H, Xerner H, eds.Seconda,y ion mass spectromet,y, SIMS VII Chichester:Wiley and Sons,1988:315—318.

20. BiedleriL, SpenglerBA,ChangTD,RossR. Transdifferentiationof humanneuroblastomacells results in coordinate loss of neuronal and malignantproperties.In: Evans AE, D'AngioGJ, SeegerRC, eds.Advancesin neumblastoma research 2. New York: Wiley-Usa; 1988:265-276.

21. TrOJanOWSkiJO, Molenaar WM, Baker DL, Pleasure D, Lee VM. Neuraland neuroendocrinephenotypeof neumblastomas,ganglioneuroblastomas,ganglioneuromasand mature versus embiyonichumanadrenal medullaiycells. In: EvansAE, D'AngioGi, SeegerRC, CdS.AdWZnCeSin newoblastoma resewvh 3. New York: Wiley-Usa; 1991:335-341.

22. KemsheadJT, Pizer BL, Patel K. In: PochedlyC, ed. Neumblastoina:twnorbiologyandtherapy. neumblastotna: perspective fo.fiaw@research.Boca Raton:CRC Press, 1990:383-385.

23. RutgersM, GubbelsAAT,HoefnagelCA, VoütePA, SmetsLA A humanneuroblastomaxenograftmodelfor [“IJ-metaiodobenzylguathdine(MIBO)biodistributionand targetedradiotherapy.In: EvansAE, D'An@o03, Seeger RC, eds. Advancesin newvblasto,na,@sea,rh3. New York:WileyLiss;1991:471—478.

24. MontaldoPG. LanciOttiM, CasaraloA, Cornaglia-FerrarisP, PonzoniM.Accumulationof m-iodobenzylguanidineby neuroblastomacells resultsfrom independentuptake and storage mechanisms.Cancer Res 199151:4342—4346.

25. KellenbergE.Thepotentialofcryofixationandfreezesubstitution:observationsand theoriticalconsiderations.IMicrosc 1991;161:183-203.

26. Stelly N, Halpern 5, NIcOIaS0, Fragu P. Moutte A. Ion imaginginciyofixedparamecium:high Ca2@concentrationin the cortex [Abstractj.Bid Cell 1990;69:20a.

27. Halpern5, FraguP, BriançonC, Larras-RegardE. Contributionofthe highmass resolutionin trace elementimagingin biology.In: BenninghovenA,Evans C, McKeeganK, Storms H, Xerner H, eds. Seconda,y ion ma&cspectro@netiySIMS VI. Chichester:Wileyand Sons; 1987:897-900.

28. Slodzian0, DaigneB, GirardF, BoustF, HillionF. Scanningsecondaiyionanalyticalmicroscopywith paralleldetection.Bid Cell 1992;74:43—50.

29. SissonJC,HutchinsonRI, ShapirOB, Ctal. Iodine-125-MIBGto treatneuroblastoma:preliminaryreport.INuci Med 1990;31:1479-1485.

30. O'DonoghueJA, WheldonTE, Babich3W, Mcyes SE, Barett A, MellerST.Implicationsof theuptakeof 1311-radiolabeledmetaiodobenzylguanidine (MIBO)for the targeted radiotherapyof neuroblastoma.BrI Radiol1991;64:428—434.

tumor, Dr. J. Lumbmso (Service de MédecineNucléaire,IGR)forprovidingradiolabeledMIBG;Ms. L. Saint-Angeforeditingthe manuscript.Thisworkwas supportedby a grantfromtheCNAMTS No. 998853.

REFERENCES1. LumbrosoJ, GuermaziJ, Hartmann0, Ctal. Metaiodobenzylguanidine

(M1BG)scans in neuroblastoma: sensitivity and specificity, a review of 115scans. Pmg ClinBid Res 1988271:689-705.

2. HoefnagelCA, VoutePA,Dc KrakerJ, MarcuseHR. Radionuclidediagnosis and therapyof neuralcrest tumorsusingiodine-131metaiodobenzylguanidine.INudMed 1987;28:308-314.

3. MoycsSE,BabichJW,CarterR, MdllerST.AgrawalM, McElwainTJ.Quantitativestudy of radiolodinatedmetaiodobenzylguathdineuptake inchildrenwithneuroblastoma:correlationwithtumorhistopathology.JNuc!Med 199930:474—480.

4. Haitmann0, LumbrosoJ,LemerteL, CtaLThetherapeuticuseofiodine131metaiodobenzylguanidine(MIBG)inneurobinstomaaphaseIIstudyinnine patients.Med Pediatr Oncol 1987;15:205-211.

5. GuerreauD, Thedrez Ph, Fritsch P. Ctal. In vitro therapeutictargetingofneuroblastomausing ‘@‘I-labeledmetaiodobenzylguanidine.mt I Cancer199045:11M—1168.

6. GazeMN,HuxhamIM, MairsRi, BarrettA. Intracellularlocalizationofmctaiodobenzylguanidinein neuroblastoinacellsby electronspectroscopicimaging.Intl Cancer 1991;47:875—880.

7. Smets LA, Janssen M, Rutgers M, Ritzen K, Buitenhuis C. Pharmacokinetics and intracellulardistributionof the tumor-targetedmdiopharmaceuticalm-iodo-benzylguanidinein SK-N-SHneuroblastomaand PC-12pheochromocytoinacells. Intl Cancer 1991;48:609—615.

8. CastaingR,SlodzianG.Microanalyseparemissionioniquesecondaire.IMOms1%2;1:395—414.

9. GalleP.Tissuelocalizationofstableandradioactivenuclidesbysecondaiyion microscopy.INucIMed 1982;23:52—57.

10. FraguPh,BriançonC,FourréC,etaLSIMSmicroscopyinthebiomedicalfield.Bid Cell 19fl;74:5—18.

11. BicdlcrJL,HelsonL,SpenglerB.Morphologyandgrowth,tumorigenicity,andcytogenesisofhumanneuroblastoniacellsincontinuousculture.CancerRes 197333:2M3—2652.

12. Buck J, Bruchelt G, Girgert R, Treuner J, Niethammer D. Specific uptakeof m.['2@IJIodobenzylguanidinein the humanneuroblastomacell lineSKN-SH. Cancer Res 1985;45:6366-6370.

13. Smets LA, Loesberg C, Janssen M, Metwally EA, Huiskamp R. Activeuptakeand extravesicularstorageofm-Iodobenzylguanidinein humanneuroblastomaSK-N-SHcells. CancerRes 1989;49:2941-2944.

14.Bettan-RenaudL,BayleCh,TeyssierJR,BenardJ.Stabilityofphenotypicand genotypic traits during the establishments of a human neuroblastomacell line, IGR-N-835. Intl Cancer 1989;44:460-466.

15. Van Oostrum lEA, Rozemulcr E, Kuol RFG, Erkens-Schuize S, RutgersDH.Cellproliferationkinedcsofsixxenograftedhumancervixcarcinomas:comparisonof autoradiographyand bromodeoxyuridinelabellingmethods.Cell Tissue Kinet 199O@,23:523—544.

I 570 The Journal of Nuclear Medicine •Vol. 34 •No. 9 •September 1993

by on April 12, 2018. For personal use only. jnm.snmjournals.org Downloaded from

1993;34:1565-1570.J Nucl Med.   J. Clerc, S. Halpern, C. Fourré, F. Omri, C. Briançon, J. Jeusset and P. Fragu  into Nude MiceMetaiodobenzylguanidine in the Human SK-N-SH Neuroblastoma Cell Line Xenografted SIMS Microscopy Imaging of the Intratumor Biodistribution of

http://jnm.snmjournals.org/content/34/9/1565This 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 1993 SNMMI; all rights reserved.

by on April 12, 2018. For personal use only. jnm.snmjournals.org Downloaded from