CANCER BIOTHERAPY & RADIOPHARMACEUTICALSVolume 21, Number 2, 2006© Mary Ann Liebert, Inc.

Pharmacokinetics and Tumor Localization of 111In-Labeled HuCC49�CH2 in BALB/c Mice andAthymic Murine Colon Carcinoma Xenograft

Paul C. Chinn,1 Ron A. Morena,1 Denise A. Santoro,1 Timothy Kazules,1 Syed V.S. Kashmiri,2Jeffrey Schlom,2 Nabil Hanna,1 and Gary Braslawsky1

1Biogen Idec, Cambridge, MA2National Cancer Institute, Bethesda, MD

ABSTRACT

The primary limitation of IgG antibodies for radioimmunotherapy of solid tumors is their prolonged serumhalf-life, leading to dose-limiting bone marrow toxicity at doses providing inadequate radiation to the tu-mor. A humanized CH2 domain-deleted variant of the anti-TAG-72 antibody CC49 (HuCC49�CH2) hasfaster blood clearance, compared to the IgG, while retaining tumor targeting. We compared the phar-macokinetics and tumor uptake of 111In-HuCC49�CH2 in BALB/c mice and a colon carcinoma (LS-174T)mouse xenograft with that of 111In-labeled chimeric CC49 (cCC49), an antibody with pharmacokineticssimilar to the humanized CC49 parent. Immunoconjugates of HuCC49�CH2 and cCC49 prepared withthe 111In chelator Mx-DTPA (1-isothiocyantobenzyl-3-methyldiethylenetriaminepentaacetic acid) retainedlow nM affinity and radiolabeling protocols provided greater than 95% radioincorporation with 111Inwhile retaining greater than 80% immunoreactivity. Blood clearance of 111In-HuCC49�CH2 in BALB/cmice was monoexponential (t1/2 5.4 hours) and faster than 111In-cCC49 (biexponential clearance; t1/2�

1.5 hours; t1/2� 162 hours). The 111In-HuCC49�CH2 also cleared more rapidly from the blood in themurine xenograft. At 1 hour postinjection, blood concentrations for 111In-HuCC49�CH2 and 111In-cCC49were comparable (25.5 injected dose per g [%ID/g] and 21.3 %ID/g, respectively); tumor uptake for 111In-HuCC49�CH2 was 7.9 %ID/g, compared to 7.5 %ID/g for 111In-cCC49. However, at 24 hours, bloodconcentration for 111In-HuCC49�CH2 was less than 111In-cCC49 (0.9 %ID/g versus 5.2 %ID/g, respec-tively) with comparable tumor retention (14.4 %ID/g versus 19.0 %ID/g, respectively). Faster blood clear-ance of 111In-HuCC49�CH2 and tumor localization comparable to that of 111In-cCC49 provided a four-fold improved tumor-to-blood ratio for 111In-HuCC49�CH2 at 24 hours postinjection.

Key words: radioimmunotherapy, radioisotopes, 111In, monoclonal antibody, carcinoma, TAG-72, CC49,domain-deleted, chelate, Mx-DTPA, CH2, murine

106

Address reprint requests to: Paul C. Chinn; Biogen Idec;5200 Research Place, San Diego, CA 92122; Tel.: (858)401-4000; Fax: (858) 431-8715E-mail: Paul.Chinn@biogenidic.com

INTRODUCTION

The U.S. Food and Drug Administration (FDA)approval in 2002 of 90Y-labeled Zevalin® (ibri-

tumomab tiuxetan) for the treatment of relapsedor refractory low-grade, follicular, or CD20�

transformed B-cell non-Hodgkin’s lymphoma(NHL) and rituximab-refractory follicular NHLwas the first radioimmunotherapeutic marketedfor the treatment of cancer. The clinical successof Zevalin,1–3 and of other radiolabeled antibod-ies in treating NHL,4–7 was the culmination of 20years of efforts in the field to implement ra-

6138_02_p106-116 4/28/06 1:50 PM Page 106

dioimmunotherapy (RIT) as a regimen in thetreatment of cancer. However, in spite of the nu-merous clinical studies evaluated to date, thetreatment of solid tumors using radioimmuno-conjugates has remained challenging, with clini-cal results continuing to be much less impressive.

The success of RIT in treating NHL may be at-tributed to the greater radiosensitivity of lym-phoma compared to solid tumors, coupled withgreater accessibility of the targeting antibody tothe tumor cells. To date, the majority of anti-bodies evaluated for RIT have been intact IgGs,with the marrow as the dose-limiting organ. Be-cause marrow toxicity is primarily attributable tothe long blood retention half-life of the radiola-beled IgG, and higher radiation doses required tosterilize the tumor, the use of IgG has not enabledadequate doses of radiation to solid tumors. Al-though the use of faster-clearing IgG fragments,such as Fabs and F(ab�)2 constructs, enablehigher input doses of radiation, these antibodyfragments are not retained within the tumor aslong as IgG. Therefore, less radiation is deliveredto the tumor.8–10 Also, improved efficacy of RITfor solid tumors might be attained using target-ing constructs that clear rapidly from the bloodwhile retaining accumulation within the tumor to the same extent as IgG. Genetically modifiedantibody constructs, such as scFv-CH3 minibod-ies, have provided improved pharmacokineticproperties, resulting in reduced myelotoxicity.11

The removal of the CH2 domain of IgG results ina significantly faster blood clearance and morerapid tumor uptake, compared to the intactIgG.12,13

The CC49 monoclonal antibody targets the tu-mor-associated glycoprotein (TAG)-72 overex-pressed on a wide spectrum of carcinomas, in-cluding colon, ovarian, pancreatic, breast, andprostate.14 Murine CC49 (mCC49) has been eval-uated in several carcinoma RIT clinical trials. Ithas shown efficient tumor targeting of colorec-tal,15–20 breast,21,22 prostate,19,23,24 and ovarian25

carcinomas in clinical trials. Some patients, in anovarian cancer trial using 177Lu-labeled mCC49,were free of disease for more than 18 months,25

and patients with metastatic breast and prostatecancer showed some partial responses whentreated with 131I-labeled mCC49.22,26 Overall,objective clinical efficacy of radiolabeled mCC49has been limited.19 Murine preclinical studies us-ing CH2 domain-deleted constructs of the CC49antibody radiolabeled with several different iso-topes have demonstrated faster blood clearance,

107

compared to the IgG parents.27–31 A preliminaryclinical study in colon carcinoma patients, us-ing 131I-labeled humanized CH2 domain-deletedCC49 (HuCC49�CH2), showed faster bloodclearance and a concomitant reduction in radia-tion to the marrow, compared to murine CC49.32

The pharmacokinetic properties exhibited byHuCC49�CH2 may improve radiation delivery tosolid tumors and enhance the efficacy of RIT.

Although several radioisotopes have been eval-uated clinically over the years for use in RIT, themost extensively studied have been 131I and 90Y.90Y has several advantages over 131I for use inRIT, including a higher-energy beta emission anda longer pathlength, enabling more energy to bedelivered to larger tumors.33–35 Additionally, be-cause 90Y has a shorter half-life (64 hours) andlacks the significant penetrating gamma emissionof 131I, 90Y-labeled antibodies are easier to ad-minister in a clinical setting and are moreamenable to outpatient-based therapy, even athigh radioactivity doses. Because 90Y emits onlybeta radiation, the gamma emitter 111In can beused as a surrogate for 90Y when imaging is re-quired; 111In can be incorporated into antibodies,using the same chelators suitable for use with90Y.36,37 This paper describes the preparation and in vitro characterization of 111In-labeledHuCC49�CH2 (111In-HuCC49�CH2) antibodyusing the high-affinity 90Y and 111In chelator,Mx-DTPA (1-isothiocyantobenzyl-3-methyldi-ethylenetriaminepentaacetic acid). Additionally,the results from pharmacokinetic and tumor lo-calization studies, using 111In-HuCC49�CH2 innormal and colon tumor-bearing mice, were com-pared to 111In-labeled chimeric CC49 (cCC49)IgG. The humanized CC49 IgG (HuCC49) wasnot available to us at the time these studies wereconducted. Therefore, cCC49 was used as a sur-rogate for the parent antibody; the cCC49 anti-body has been shown to exhibit pharmacokineticproperties comparable to HuCC49.38,39

MATERIALS AND METHODS

Cell Lines, Antibodies, and Reagents

The TAG-72-expressing LS-174T cells were ob-tained from the American Type Culture Collec-tion (Rockville, MD) and cultured in RPMI-1640containing 10% fetal bovine serum (FBS) at 37°Cand 5% CO2.

Purified anti-TAG-72 HuCC49�CH2 andcCC49 antibodies were obtained from Dr. J.

6138_02_p106-116 4/28/06 1:50 PM Page 107

Schlom at the National Cancer Institute(Bethesda, MD); these antibodies have been pre-viously described.28,38 111In chloride was ob-tained from Nycomed Amersham (ArlingtonHeights, IL). Mx-DTPA was obtained fromHauser Chemical Co. (Boulder, CO). DELFIA™europium (Eu) labeling and assay reagents wereobtained from Perkin Elmer (Boston, MA), andEu2O3 standard dissolved in 5% nitric acid wasobtained from EM Sciences (Gibbstown, NJ).Bovine serum albumin (BSA), bovine submaxil-lary mucin (BSM), and diethylenetriaminepen-taacetic acid (DTPA) were from Sigma Chemi-cal Co. (St. Louis, MI). Human serum albumin(HSA) was obtained from Baxter-Hyland (LosAngeles, CA). Agarose bead Reacti-Gel™ wasobtained from Pierce Biotechnology (Rockford,IL). All other reagents were obtained from stan-dard sources. Reagents used for conjugation andradiolabeling protocols were processed to removecontaminating heavy metal ions by batch-treat-ing with Chelex 100-ion exchange resin (BioRadIndustries; Richmond, CA). Deionized water orSterile Water for Irrigation (SWFI) was used forall preparations and dilutions.

Competitive Binding Assay

Competitive binding assays were performed us-ing Eu-labeled HuCC49�CH2 as a tracer. Tracerwas prepared by incubating Mx-DTPA-conju-gated HuCC49�CH2 for 24 hours with a 100-foldmolar excess of Eu in metal-free 50 mM sodiumacetate. Binding was determined using a fluores-cence-based europium detection system. A 96-well plate was coated with BSM at 0.8 �g/wellovernight at 2–8°C. The plate was blocked withphosphate-buffered saline (PBS) containing 1%(w/v) BSA. The antibody samples were dilutedin DELFIA Assay Buffer and serially diluted 1:3on the plate in triplicate. Europium-labeledHuCC49�CH2 was added to each well as tracerand the plate incubated with circular agitation for2 hours at ambient temperature; the fluorescencesignal was developed using DELFIA Enhance-ment Solution and quantified using a Wallac Vic-tor fluorometric plate reader (Perkin Elmer LifeSciences, Boston, MA). Mean triplicate valueswere determined for bound (B) and no inhibitoradded (B0) wells. Percent inhibition values ([1-B/B0] � 100) were plotted versus inhibitor anti-body concentration; molar concentrations werecalculated using molecular weights of 121 and145 kD for HuCC49�CH2 and cCC49, respec-

tively, and protein concentrations determined byabsorbance at 280 nm (1.5 OD � 1 mg/mL). Plotswere generated using a four-parameter curve fitby plotting the inhibitor antibody concentration(x) versus the percent inhibition (y), and IC50 val-ues were determined. Apparent affinity constantswere calculated using the method of Müller.40

Preparation of BSM-Coated Resin

Bovine submaxillary mucin (1 g) was added to100 mL of 100 mM carbonate buffer, at a pHof 9.5, and thoroughly mixed for 3 hours usingend-over-end rotation. Agarose bead Reacti-Gelwas washed exhaustively with the ice-coldSWFI, using a 0.2-�m filter. The BSM/carbon-ate buffer was added to the resin and the resindegassed; resin suspension was at a pH of 7.5–8.0 using pH paper. The suspension was de-gassed and mixed end-over-end overnight andthen washed with 1M Tris at a pH of 11. Theresin was resuspended in Tris buffer and mixedend-over-end for 2 hours. The resin was thenwashed with 10 L of Dulbecco’s PBS (DPBS),using a glass-filter vacuum flask, and then de-gassed. Sodium azide was added to the resinsuspension (final concentration, 0.01%) andstored at 2–8°C.

Immunoreactivity Assays

Immunoreactivity was determined using themethod of Lindmo et al.41 Radiolabeled antibod-ies were diluted to 30 ng/mL in PBS containing1% (w/v) BSA (dilution buffer). Increasing vol-umes (0.05, 0.10, 0.20, 0.40, 0.80, and 1.6 mL)of washed and resuspended (1:2 slurry) BSM-coated resin were transferred into a duplicate se-ries of tubes. The volume in each tube was ad-justed to 1.6 mL by adding the dilution buffer,and diluted radiolabeled antibody (0.5 mL) wasadded to all tubes. To determine total radioactiv-ity added, 0.5 mL of the diluted sample was addedto duplicate tubes containing 1.6 mL of dilutionbuffer only. Tubes were incubated for 60 minutesat ambient temperature with end-over-end mix-ing. Tubes were centrifuged to pellet resin, andradioactivity of the supernates (0.80 mL) was de-termined. Bound counts were expressed as a per-centage of the total counts added. Immunoreac-tivity values were calculated by plotting 1/resinsuspension volume versus total counts/boundcounts, and immunoreactivity was calculatedfrom the y-intercept value (% immunoreactiv-ity � 1/y-intercept � 100).

108

6138_02_p106-116 4/28/06 1:50 PM Page 108

109

Radioincorporation Assay

The percent of radioisotope bound to the conju-gate was determined by instant thin-layer chro-matography (ITLC), using a commercially avail-able kit (Tec-Control Radiochromatographic Kit,Part# 151-770; Biodex Medical Systems; Shirley,NY). Briefly, 1-�L samples were spotted at theorigin (triplicate strips), and the strips were de-veloped in 0.9% (w/v/) NaCl. The strips were cut crosswise in half, and radioactivity of top andbottom halves was determined using a gammacounter (100–500-KeV window settings); radio-incorporation was calculated by dividing theamount of radioactivity in the top half of the stripby the total radioactivity found in both the topand bottom halves. This value was expressed asa percentage, and the mean value was determined.

Mx-DTPA per Antibody Assay

Antibody conjugates (1–3 mg/mL) were mixed 1:1(v/v) with 50 �M Eu3� (Eu2O3) standard solutionin 40 mM of acetate buffer at a pH of 4.5; metal-free 0.9% (w/v) NaCl was used as a backgroundcontrol. After incubating for 20–28 hours at roomtemperature in the dark, Eu3� incorporation wasquenched by adding one half volume of 600 mMTris, 1 mM DTPA, at a pH of 7.5, and incubatingfor 10 minutes. Conjugated antibody containingMx-DTPA-bound Eu3�, equivalent to the molaramount of Mx-DTPA attached to the antibody, wasseparated from excess Eu3� bound to DTPA byITLC (in triplicate). Samples (2.0 �L) were spot-ted on ITLC strips (Biodex Tec-Control Radio-chromatographic ITLC kit) and developed using0.9% NaCl. Strips were cut crosswise in half, andthe bottom and top halves of each strip were incu-bated with end-over-end mixing in 1 mL each of100 mM HCl for 5–60 minutes to extract Eu3�.After extraction, 40 �L of solution was added to 1mL of buffered europium fluorescence enhance-ment solution (DELFIA Enhancement Solutionand 400 mM acetate, pH 4.5 [9:1 (v/v]) and mixedend-over-end for 15 minutes; fluorescence associ-ated with the bottom and top halves of each ITLCstrip was determined by measuring 0.1 mL in du-plicate wells on a low-fluorescence, 96-well plateusing a Wallac Victor fluorometric plate reader.Fluorescence values were corrected for back-ground, and for each strip the molar amount ofEu3� bound to conjugated antibody (bottom halfof strip) was determined by comparison to the to-tal fluorescence of top and bottom strip halves, cor-responding to the known amount of Eu3� standard

added to the antibody conjugate sample. Molar Mx-DTPA-to-antibody ratios were calculated usingprotein concentrations, determined by absorbanceat 280 nm and the molecular weights of 121 and145 kD for HuCC49�CH2 and cCC49, respectively.

Preparation of Mx-DTPA CC49 Conjugates

Stock solutions (5–10 mM) of the Mx-DTPA wereprepared in 50 mM borate, at a pH of 8.6, and storedat �70°C. Antibodies were prepared for conjuga-tion by repetitive diafiltration at 2–8°C into metal-free 0.9% (w/v) NaCl using Centricon 30 spin fil-ters (Amicon Corporation; Beverly, MA). The filterretentate was assayed for protein using absorbanceat 280 nm (1.5 absorbance unit � 1 mg/mL), di-luted to 8 mg/mL, and stored at 2–8°C until con-jugated. Antibody solutions at 8 mg/mL were ad-justed to a pH of 8.6 by adding one tenth volumeof metal-free 0.5 M sodium borate, at a pH of 8.6,containing 0.9% (w/v) NaCl and reacted with Mx-DTPA at a molar ratio of chelator-to-protein of 3:1and 6:1 for HuCC49�CH2 and cCC49, respec-tively. The conjugation mixture was incubatedovernight (16–20 hours) at ambient temperature.Unreacted Mx-DTPA was removed by repetitivediafiltration into metal-free NaCl (0.9% [w/v]). Theprotein concentration was adjusted to 1.5–3.0mg/mL and stored at 2–8°C or �70°C until radi-olabeled.

Preparation of Radiolabeled CC49 Antibodies

Antibodies were radiolabeled with 111In at a spe-cific activity of 2–4 mCi/mg. In a typical reac-tion, 0.5 mCi of 111Indium chloride solution(111InCl3) was transferred to a polypropylene mi-crofuge tube, and the solution was adjusted to ap-proximately a pH of 6 by adding an equal vol-ume of metal-free 250 mM of sodium citrate andmixed by vortexing. Conjugated antibody (0.17mg) in 0.9% (w/v) NaCl was transferred to thetube, the solution mixed, and incubated for 15minutes at ambient temperature. Reaction mix-tures were quenched by adding 1�PBS, at a pHof 7.4, containing 7.5% (w/v) HSA and 1 mMDTPA (formulation buffer) to a final antibodyconcentration of 0.1–0.2 mg/mL; the formulatedsample was stored at 2–8°C.

Pharmacokinetic Studies in BALB/c Mice

All animal study protocols were approved byBiogen Idec’s institutional animal care and use

6138_02_p106-116 4/28/06 1:50 PM Page 109

110

committee (IACUC). The blood clearance ratesof 111In-HuCC49�CH2 and 111In-cCC49 wereevaluated in 8- to 10-week-old female BALB/cmice. For each antibody, mice were injectedintravenously via tail with 100 �L of 111In-HuCC49�CH2 (2 �Ci; 1.3 �g of antibody) or111In-cCC49 (7.5 �Ci; 2 �g of antibody). At se-lected times, blood from groups of 3 or 5 miceeach was removed by cardiac puncture or col-lected from sacrificed animals, and radioactivityassociated with a weighed sample of blood wasdetermined by gamma counting. Radioactivityvalues were corrected for radioactive decay, andthe percent injected dose per gram (%ID/g) bloodvalues were calculated.

Tumor Localization Studies in Mice

The blood clearance and tumor uptake of 111In-HuCC49�CH2 and 111In-cCC49 were evaluatedin 8- to 10-week-old female athymic nu/nu miceinoculated subcutaneously with LS-174T cells(0.1 mL at 3 � 106 cells/mL) in DPBS. Micewere used for localization studies when tumorsmeasured approximately 10 mm in diameter. Foreach antibody, mice were injected intravenouslythrough the tail with 100 �L (10 �Ci) of 111In-HuCC49�CH2 or 111In-cCC49, and groups of 3mice each were sacrificed at 1 and 24 hourspostinoculation. After animals were sacrificed,radioactivity associated with weighed samples ofblood and tumors was measured by gammacounting, and the %ID/g values were calculated.

RESULTS

Preparation of Radiolabeled HuCC49�CH2and cCC49 Antibodies

The conjugation and radiolabeling protocols usedensured high radioincorporation levels of 111In,while maintaining significant retention of bind-

ing to TAG-72 antigen (Table 1). Based on pre-vious experience with Zevalin,42 we optimizedconjugation conditions to incorporate 1–2 molesof the Mx-DTPA chelator per mole of antibody.Using a three- or six-fold molar excess of Mx-DTPA to antibody and an incubation time of16–20 hours at ambient temperature resulted in1.1 and 1.7 mol of Mx-DTPA per mol of anti-body for HuCC49�CH2 and cCC49, respectively(Table 1). Solid-phase competition immunoas-says with BSM as the TAG-72 antigen source andEu-labeled HuCC49�CH2 as a tracer were usedto compare the binding of the Mx-DTPA-conju-gated antibodies (Fig. 1). The slopes of the com-petition profiles for both conjugates were similarwith IC50 values of 15 � 10�9 M and 5.8 � 10�9

M, respectively, for HuCC49�CH2 and cCC49.No inhibition of binding was observed using anantigen-negative control IgG antibody (murineanti-CD20; data not shown). The competitivebinding data were analyzed using the method ofMüller40 to estimate apparent affinity constantsof 4.2 � 10�9 M and 0.62 � 10�9 M forHuCC49�CH2 and cCC49 conjugates, respec-tively (Table 1).

Radiolabeling specific activities in the range of2–4 mCi/mg antibody were used for 111In incor-poration of Mx-DTPA-conjugated HuCC49�CH2and cCC49. At these specific activities, a 15-minute incubation at approximately pH 6 routinely provided radioincorporations greaterthan 95%, as measured by ITLC, for bothHuCC49�CH2 and cCC49 conjugates (Table 1).Immunoreactivity values for the radiolabeled an-tibodies on TAG-72 expressing LS-174T humancolon cancer cells were greater than 80% (Fig.2). Specificity of HuCC49�CH2 and cCC49 anti-gen binding was demonstrated by the negligibleTAG-72 binding observed for 111In-labeled anti-CD20 antibody (Zevalin; 2B8-MX-DTPA; Fig.2). In anticipation of future studies with 90Y, ra-diolabeled antibodies were formulated in 1 �

Table 1. Characterization of Mx-DTPA-Conjugated and 111In-Labeled HuCC49�CH2 and cCC49

Mx-DTPA: Apparent 111InAntibody antibody affinity radioincorporation Immunoreactivityconjugate ratio (nM) (%) (%)

HuCC49�CH2 1.1 4.20 97.1 83.8cCC49 1.7 0.62 96.8 80.5

Mx-DTPA, 1-isothiocyantobenzyl-3-methyldiethylenetriaminepentaacetic acid.

6138_02_p106-116 4/28/06 1:50 PM Page 110

111

PBS containing human serum albumin as a ra-dioprotectant.42,43

Pharmacokinetics of RadiolabeledHuCC49�CH2 and cCC49 in BALB/c Mice111In-labeled antibodies were injected intraven-ously into normal BALB/c mice, and, at selected

times, the radioactivity associated with the bloodwas determined. Radioactivity was decay cor-rected and expressed as %ID/g of blood and rep-resents biological clearance only. It was assumedthat the 111In label remains attached to the anti-body and, therefore, radioactivity was directlyproportional to the amount of antibody present.As shown in Figure 3, the blood clearance of111In-HuCC49�CH2 was dramatically faster thanthat of the 111In-labeled cCC49 IgG construct. At1 hour postinoculation, radiolabeled HuCC49�CH2and cCC49 antibodies were 43.3 and 43.5 %ID/gblood, respectively. However, at 24 hours, the%ID/g values for 111In-HuCC49�CH2 had fallento �1%, whereas the blood concentration of thechimeric antibody remained significantly higherat 25.3 %ID/g. Pharmacokinetic analysis of theantibody blood clearance profiles was used to fitthe data to simple exponential models, using the curve fit feature of the Organ Level InternalDose Assessment (OLINDA) radiation dosime-try program.44 The clearance pattern for 111In-HuCC49�CH2 closely fit (R2 � 0.98) a singleexponential decay curve with a biological t1/2 of5.4 hours. The clearance trend for 111In-cCC49was reasonably fit with a biexponential elimina-tion equation with the resulting equation, %ID/g(t) � 23.2e�0.46t � 28.8e0.0043t, and indicated that23.2 %ID/g of the cCC49 antibody was clearedwith a biological t1/2 of 1.5 hours, whereas 28.8%%ID/g cleared at a much slower rate, with a t1/2of 162 hours.

Figure 1. Competitive immunoassay of Mx-DTPA-conju-gated HuCC49�CH2 and cCC49 to TAG-72 antigen.HuCC49�CH2 (closed circles) and cCC49 (open circles)were titrated against a fixed concentration of Eu-labeledHuCC49�CH2 tracer on BSM-coated 96-well microtiterplates (each value is the mean of triplicate data points). Mx-DTPA, 1-isothiocyantobenzyl-3-methyldiethylenetriamine-pentaacetic acid; BSM, bovine submaxillary mucin.

Figure 2. Immunoreactivity of 111In-HuCC49�CH2 and 111In-cCC49 to TAG-72 antigen. Panel A: A fixed amount of radiolabeledHuCC49�CH2 (closed circles, solid line), cCC49 (open circles, dashed line), or 111In-Zevalin (squares) was incubated with increasingamounts of BSM-coated resin. The percent radioactivity bound to resin was plotted against the resin volume. Panel B: Reciprocal plotof the percentage of bound counts versus resin volume. The reciprocal of the extrapolated y-intercepts provided immunoreactivity val-ues of 83.8% and 80.5% for 111In-HuCC49�CH2 and 111In-cCC49, respectively. BSM, bovine submaxillary mucin.

A B

6138_02_p106-116 4/28/06 1:50 PM Page 111

112

Blood Clearance and Tumor Uptake of111In-HuCC49�CH2 and 111In-cCC49 inMurine Colon Carcinoma Xenografts

Tumor localization of 111In-HuCC49�CH2 and111In-cCC49 was evaluated in TAG-72 express-ing tumor-bearing athymic (nu/nu) mice. Ani-mals were injected with 111In-labeled antibody,and, at 1 and 24 hours postinjection, mice weresacrificed and radioactivity associated with sam-ples of blood and tumor measured and expressedas % ID/g tissue (Table 2). Blood clearance for111In-HuCC49�CH2 was faster than that of thecomparably labeled chimeric antibody (Table 2).At 1 hour postinjection, blood concentration for111In-HuCC49�CH2 was 25.5 %ID/g and was notsignificantly different from that of 111In-cCC49(21.3 %ID/g). After 24 hours, 0.9 %ID/g of 111In-

HuCC49�CH2 was present in the blood com-pared to 5.2 %ID/g for 111In-cCC49. We also examined tumor uptake and retention of the 111In-HuCC49�CH2 and 111In-cCC49 in the sameTAG-72-positive murine xenograft model. The111In-HuCC49�CH2 and 111In-cCC49 antibodiesdemonstrated comparable tumor uptake at 1 hourpostinjection (7.9 %ID/g versus 7.5 %ID/g; Table2). At 24 hours postinjection, tumor uptake for111In-HuCC49�CH2 and 111In-cCC49 were 14.4%ID/g and 19.0 %ID/g, respectively. The com-parable tumor uptake of the 111In-HuCC49�CH2and 111In-cCC49, in addition to the faster bloodclearance of the domain-deleted construct, pro-vided a four-fold improvement in tumor-to-bloodratio for 111In-HuCC49�CH2 compared to 111In-cCC49 at 24 hours postinjection (Table 2).

DISCUSSION

The therapeutic successes demonstrated using ra-dioimmunoconjugates in the treatment of NHLhave remained elusive for solid tumors. A keyreason for this lack of success is attributable tosolid tumors being less radiosensitive than lym-phomas. Therefore, higher radiation doses are re-quired to achieve demonstrable antitumor effects.Although a variety of antibody constructs havebeen evaluated for RIT, most clinical trials haveused full-length IgG. However, the relativelylong blood residence time of IgG antibodies con-tributes directly to marrow irradiation, resultingin dose-limiting myelotoxicity; this toxicity pre-cludes administering adequate doses of the radi-olabeled antibody to deliver sufficient radiationfor tumor sterilization.

Different antibody derivatives have been usedto overcome the pharmacokinetic limitations in-

Figure 3. Blood clearance of 111In-HuCC49�CH2 and111In-cCC49 in non-tumor-bearing mice. RadiolabeledHuCC49�CH2 (closed circles) or cCC49 (open circles) wasinjected intravenously in BALB/c mice and at the indicatedtimes blood collected, radioactivity determined and expressedas % injected dose/g (%ID/g) blood � standard errors.

Table 2. Blood Clearance and Tumor Uptake of 111In-HuCC49�CH2 and 111In-cCC49 in TAG-72-Expressing MurineXenografts

Percent injected dose per g postinjectiona

1 hour 24 hours

Tissue HuCC49�CH2 cCC49 HuCC49�CH2 cCC49

Tumor 7.9 � 0.7 7.5 � 2.0 14.4 � 2.3 19.0 � 3.4Blood 25.5 � 0.4 21.3 � 0.8 0.9 � 0.1 5.2 � 1.3Tumor�Blood 0.31 0.35 16.0 3.7

aTumor and blood values are % injected dose per (%ID/g) � standard error.

6138_02_p106-116 4/28/06 1:50 PM Page 112

herent in the use of IgG for targeting radioiso-topes and to improve the efficacy of RIT withsolid tumors. The use of more rapidly clearedF(ab�)2 antibody fragments does enable higher in-put doses of radioactivity.8,9 However, becausethe blood concentration of the targeting antibodyis the primary factor in determining tumor uptakeand retention, F(ab�)2 fragments are not retainedwithin the tumor as efficiently as intact IgG.10

The development of genetically engineered scFvconstructs has provided antibodies with fasterblood clearance and reduced marrow toxicity,while retaining significant tumor retention, com-pared to parent IgG.11,45,46 Another approach toincreasing the blood clearance rate of the target-ing antibody has been the removal of the CH2 do-main of the IgG. Generation of a CH2 domain-deleted chimeric anti-GD2 antibody resulted infaster blood clearance, compared to the IgG par-ent.12 Previously published studies in LS-174Ttumor-bearing athymic mice, using a 131I-labeledCH2 domain-deleted version of anti-TAG-72 chimeric B72.3 antibody, demonstrated a bloodclearance t1/2� of 7.8 hours, compared to 48.9hours for the 125I-labeled parent chimeric anti-body.13 Subsequent development of HuCC49�CH2provided an antibody construct with rapid bloodclearance, compared to the humanized parent.28

Recently, the pharmacokinetic advantage attrib-uted to HuCC49�CH2 for RIT has been corrob-orated by early phase 1 clinical results using 131I-HuCC49�CH2.32 In that study, the demonstrablyshorter blood half-life for HuCC49�CH2, com-pared to murine CC49, resulted in reduced mar-row toxicity for the domain-deleted construct.

The HuCC49�CH2 antibody has been evalu-ated preclinically and clinically for application toRIT using several isotopes, with the exception of90Y. 90Y exhibits several properties, making itsuse for RIT attractive, particularly when com-pared to 131I.47 Given the advantages of using 90Yfor RIT, we prepared and characterized the phar-macokinetic and tumor localization properties ofHuCC49�CH2 radiolabeled with 111In, a gammaemitter used clinically as a surrogate for 90Y.

For RIT to be maximally effective, the ra-dioisotope must remain attached to the targetingconstruct. For the studies described in this paper,we selected the Mx-DTPA chelator for the in-corporation of 111In. This chelator has previouslybeen demonstrated to have high affinity for both111In and 90Y, both in vitro and in vivo.48,49 Useof the phenylisothiocyanato derivative of Mx-DTPA enables facile attachment to antibodies via

lysine residues and the formation of a stablethiourea bond. We optimized conjugation condi-tions with Mx-DTPA to maintain low nM affin-ity binding to the TAG-72 antigen, based on ourprevious experience in preparing Mx-DTPA-con-jugated Zevalin.42 The radiolabeling process for111In was developed to routinely provide a 95%radioincorporation of the isotope while main-taining maximum retention of immunoreactivity(80%) to TAG-72-expressing human coloncarcinoma LS-174T cells.

At the time these studies were done, the human-ized CC49 parent IgG of the CH2 domain-deletedconstruct was unavailable. Therefore, we comparedthe pharmacokinetics of 111In-HuCC49�CH2 withthat of an intact IgG murine-human chimeric CC49antibody (cCC49). The cCC49 antibody has beendemonstrated previously to have similar bloodclearance and tumor localization properties in mice,compared to the humanized CC49 parent.38,39 Wewere able to conjugate the cCC49 antibody to Mx-DTPA and radiolabel with 111In, using the sameprotocols developed for the HuCC49�CH2 anti-body and attained comparable 111In radioincorpo-rations and retention of immunoreactivity to theTAG-72 antigen.

Pharmacokinetic studies in non-tumor-bearingBALB/c mice showed that 111In-HuCC49�CH2cleared from the blood more rapidly than 111In-cCC49. For 111In-HuCC49�CH2, blood clearancedata were best approximated as monoexponen-tial, with a biological t1/2 of 5.4 hours. The morerapid clearance of the 111In-HuCC49�CH2 is con-sistent with studies published by Slavin-Chioriniet al. that compared 131I-labeled HuCC49�CH2with 125I-labeled intact HuCC49 IgG injected in-travenously in non-tumor-bearing athymic miceand found greater than 99% of HuCC49�CH2cleared from the plasma at 24 hours postinjec-tion, compared to 74% of HuCC49.28 Studieswith 177Lu-labeled HuCC49�CH2 administeredintravenously to LS-174T tumor-bearing athymicmice demonstrated biexponential clearance, witha t1/2� of 13.3 minutes and a t1/2� of 5.3 hours.50

Based on these previously published studies, it isclear that the very rapid � elimination phase forHuCC49�CH2 represents a very small contribu-tion to the overall clearance rate. Given the timepoints measured in our study, the distinction be-tween � and � phases was not evident, resultingin a monoexponential fit of the data. The com-parably labeled 111In-cCC49 exhibited biexpo-nential clearance, with t1/2� and t1/2� values of and1.5 hours and 162 hours, respectively, and are

113

6138_02_p106-116 4/28/06 1:50 PM Page 113

consistent with the literature. For example, pre-vious studies with 125I- or 177Lu-labeled cCC49in non-tumor-bearing athymic mice gave t1/2�

values from 1.8 to 5.6 hours and t1/2� values inthe range of 77.2–179.4 hours.27 The consider-ably faster blood clearance of the domain-deletedconstruct, compared to the chimeric version, wasclearly observed with our data and corroborateswith previously published work. Our studies alsodemonstrated the more rapid blood clearance ofthe 111In-HuCC49�CH2, relative to 111In-cCC49,in athymic mice bearing LS-174T human colontumors. However, the clearance of 111In-cCC49in the tumor-bearing mice was faster than thatseen in the non–tumor-bearing BALB/c mice,possibly attributable to the presence of antigen-expressing tumor cells binding a portion of thecirculating antibody.

For the 1-hour and 24-hour timepoints evalu-ated, 111In-HuCC49�CH2 and 111In-cCC49 lo-calized to LS-174T human colon tumors in amurine xenograft model to the same extent usingMx-DTPA conjugates with low nM affinities andcomparable immunoreactivities after radiolabel-ing with 111In. The tumor uptake (14.9 %ID/g) at24 hours postinjection for 111In-HuCC49�CH2was in the range reported in the literature for the same construct in the LS-174T murinexenograft model, with values for intravenouslyadministered 131I, 125I, 177Lu, and 111In-labeledHuCC49�CH2 ranging from 8.39 to 22.3 %ID/g at24 hours postinjection.28,31,50 Tumor localizationfor 111In-cCC49, in our study, was 19.0 %ID/gat 24 hours postinjection, compared to values of27.8 %ID/g and 58.7 %ID/g for 125I- and 177Lu-labeled cCC49, respectively, reported by Slavin-Chiorini et al. for the same xenograft model.27

However, direct comparisons of different studies,using the LS-174T murine xenograft model, maybe problematic because of variable tumor uptakebeing dependent on several factors, including an-tibody immunoreactivity, radiolabel used, tumorsize, tumor vascularity, and antigen expression.

The faster blood clearance of 111In-HuCC49�CH2, with tumor uptake comparable to the in-tact 111In-cCC49, resulted in a four-fold highertumor-to-blood ratio for the domain-deleted con-struct. In general, our results are in agreementwith those reported in the literature, demonstrat-ing the rapid blood clearance of CH2 domain-deleted CC49 constructs compared to their chi-meric or humanized analogs.27–29,31 In particular,studies by Milenic et al.31 showed that 111In-HuCC49�CH2, prepared using a different chela-

tor than the one used in our study, was rapidlycleared from the blood and localized efficientlyto the tumor when administered either intra-venously or intraperitoneally in the same LS-174T tumor xenograft model used in our stud-ies. In that published study, 111In-HuCC49�CH2was retained in the blood at 0.5 %ID/g at 24 hours postinjection; tumor uptake was 22.3%ID/g, resulting in a tumor-to-blood ratio of44.6, compared to a tumor-to-blood ratio of 16.0 in our study. In another study using 177Lu-HuCC49�CH2, prepared with the CHx-A-DTPA chelator, the blood concentration in LS-174T murine xenografts was 0.85 %ID/g, with atumor uptake of 10.44 %ID/g at 24 hours postin-jection, resulting in a 12.3 tumor-to-blood ratio.50

Several scFv constructs derived from themurine CC49 antibody have been described in the literature and evaluated in the LS-174Tmurine xenograft model. Overall, the pharma-cokinetic and tumor localization properties of111In-HuCC49�CH2 providing improved tumortargeting advantages compare favorably to ra-diolabeled CC49 scFv constructs administeredintravenously. For example, a tetravalent scFvCC49 radiolabeled with 177Lu demonstratedblood concentration at 24 hours postinjection was0.5 %ID/g and tumor uptake was 5.5 %ID/g, re-sulting in a tumor-to-blood value of 11.0, com-pared to our value of 16.0 determined for 111In-HuCC49�CH2.51 Pavlinkova et al. reported that125I-labeled dimeric scFv CC49 constructs wererapidly cleared from the blood with 0.07 %ID/gand tumor concentration of 5.62 %ID/g at 24hours postinjection, resulting in a tumor-to-bloodratio of 80.3.45 Although the dimeric scFv in thatstudy provided a higher tumor-to-blood ratio thanthat observed for 111In-HuCC49�CH2 in ourstudy, overall tumor uptake was lower for thescFv construct (14.4 %ID/g versus 5.62 %ID/g).However, it may be difficult to draw definitiveconclusions regarding the potential advantages ofone construct over another given tumor uptakevariability in the LS-174T xenograft model. Clin-ical evaluation of HuCC49�CH2 and otherCC49-derived constructs will ultimately providedata determining the relative therapeutic advan-tages these constructs may provide in RIT.

CONCLUSIONS

In summary, we have demonstrated that 111In-HuCC49�CH2, prepared using the high-affinity

114

6138_02_p106-116 4/28/06 1:50 PM Page 114

115

chelator Mx-DTPA, exhibited pharmacokineticproperties consistent with those reported previ-ously in the literature. The results presented inthis paper, using 111In as a surrogate for 90Y, cor-roborate and extend the findings of others indemonstrating that the pharmacokinetic proper-ties of HuCC49�CH2 may provide improved ra-dioisotope delivery for solid tumor RIT and sup-port additional evaluation of the construct as an90Y-labeled antibody for cancer therapy.

ACKNOWLEDGMENTS

The authors thank Mr. Bill Carlisle, Ms. MargaretHogan, and Mr. Daniel Floyd for their excellenttechnical assistance with the murine pharmaco-kinetic and xenograft studies. The assistance ofMr. Richard Belanger (Ryan-Belanger Associ-ates; San Diego, CA), with the calculation ofblood clearance t1/2 values, is gratefully ac-knowledged.

REFERENCES

1. Knox SJ, Goris ML, Trisler K, et al. Yttrium-90-labeledanti-CD20 monoclonal antibody therapy of recurrent B-cell lymphoma. Clin Cancer Res 1996;2:457.

2. Witzig TE, Gordon LI, Cabanillas F, et al. Randomized,controlled trial of yttrium-90-labeled ibritumomab tiux-etan radioimmunotherapy versus rituximab immuno-therapy for patients with relapsed or refractory low-grade, follicular, or transformed B-cell non-Hodgkin’slymphoma. J Clin Oncol 2002;20:2453.

3. Witzig TE, Flinn IW, Gordon LI, et al. Treatment withibritumomab tiuxetan radioimmunotherapy in patientswith rituximab-refractory, follicular non-Hodgkin’slymphoma. J Clin Oncol 2002;20:3262.

4. White C, Halpern S, Parker B, et al. Radioimmunother-apy of relapsed B-cell lymphoma with yttrium 90 anti-idiotype monoclonal antibodies. Blood 1996;87:3640.

5. DeNardo GL, O’Donnell RT, Rose LM, et al. Mile-stones in the development of Lym-1 therapy. Hy-bridoma 1999;18:1.

6. Juweid M, Sharkey R, Markowitz A, et al. Treatmentof non-Hodgkin’s lymphoma with radiolabeled murine,chimeric, or humanized LL2, an anti-CD22 monoclo-nal antibody. Cancer Res 1995;55:5899S.

7. Kaminski MS, Estes J, Zasadny KR, et al. Radioim-munotherapy with iodine (131)I tositumomab for re-lapsed refractory B-cell non-Hodgkin’s lymphoma: Up-dated results and long-term follow-up of the Universityof Michigan experience. Blood 2000;96:1259.

8. Buchegger F, Pelegrin A, Delaloye B. Iodine-131-la-beled Mab F(ab�)2 fragments are more efficient and less

toxic than intact anti-CEA antibodies in radioim-munotherapy of large human colon carcinoma graftedin nude mice. J Nucl Med 1990;31:1035.

9. Yokota T, Milenic DE, Whitlow M, et al. Rapid tumorpenetration of single-chain Fv and comparison withother immunoglobulin forms. Cancer Res 1992;52:3402.

10. Brouwers A, Mulders P, Oosterwijk E, et al. Pharma-cokinetics and tumor targeting of 131-labeled F(ab�)2

fragments of the chimeric monoclonal antibody G250:Preclinical and clinical pilot studies. Cancer Biother Ra-diopharm 2004;19:466.

11. Hu S, Shively L, Raubitschek A, et al. Minibody: Anovel engineered anti-carcinoembryonic antigen anti-body fragment (single-chain Fv-CH3) which exhibitsrapid, high-level targeting of xenografts. Cancer Res1996;56:3055.

12. Mueller BM, Reisfeld RA, Gilles SD. Serum half-lifeand tumor localization of a chimeric antibody deletedof the CH2 domain and directed against the disialo-ganglioside GD2. Proc Natl Acad Sci USA 1990;87:5702.

13. Slavin-Chiorini DC, Horan-Hand P, Kashmiri SVS, etal. Biologic properties of a CH2 domain-deleted re-combinant immunoglobulin. Int J Cancer 1993;53:97.

14. Thor A, Ouchi N, Szpak CA, et al. Distribution of onco-fetal antigen tumor-associated glycoprotein-72 definedby monoclonal antibody B72.3. Cancer Res 1986;46:3118.

15. Murray JL, Macey DJ, Kasi LP, et al. Phase II radio-immunotherapy trial with 131I-CC49 in colorectal can-cer. Cancer 1994;73:1057.

16. Divgi CR, Scott AM, Gulec S, et al. Pilot radioim-munotherapy trial with 131I-labeled murine monoclonalantibody CC49 and deoxyspergualin in metastatic coloncarcinoma. Clin Cancer Res 1995;1:1503.

17. Mulligan T, Carrasquillo JA, Chung Y, et al. Phase Istudy of intravenous Lu-labeled CC49 murine mono-clonal antibody in patients with advanced adenocarci-noma. Clin Cancer Res 1995;1:1447.

18. Divgi CR, Scott AM, Dantis L, et al. Phase I radioim-munotherapy trial with iodine-131-CC49 in metastaticcolon carcinoma. J Nucl Med 1995;36:586.

19. Liu T, Meredith RF, Saleh MN, et al. Correlation oftoxicity with treatment parameters for 131I-CC49 ra-dioimmunotherapy in three phase II clinical trials. Can-cer Biother Radiopharm 1997;12:79.

20. Tempero M, Leichner P, Dalrymple G, et al. High-dosetherapy with iodine-131-labeled monoclonal antibodyCC49 in patients with gastrointestinal cancers: A phaseI trial. J Clin Oncol 1997;15:1518.

21. Murray JL, Macey DJ, Grant EJ, et al. Enhanced TAG-72 expression and tumor uptake of radiolabeled mono-clonal antibody CC49 in metastatic breast cancer pa-tients following �-interferon treatment. Cancer Res1995;55:5925S.

22. Macey DJ, Grant EJ, Kasi L, et al. Effect of recombi-nant �-interferon on pharmacokinetics, biodistribution,toxicity, and efficacy of 131I-labeled monoclonal anti-

6138_02_p106-116 4/28/06 1:50 PM Page 115

116

body CC49 in breast cancer: A phase II trial. Clin Can-cer Res 1997;3:1547.

23. Meredith RF, Bueschen AJ, Khazaeli MB, et al. Treat-ment of metastatic prostate carcinoma with radiolabeledantibody CC49. J Nucl Med 1994;35:1017.

24. Slovin SF, Scher HI, Divgi CR, et al. Interferon-� andmonoclonal antibody 131I-labeled CC49: Outcomes inpatients with androgen-independent prostate cancer.Clin Cancer Res 1998;4:643.

25. Meredith RF, Partridge EE, Alvarez RD, et al. In-traperitoneal radioimmunotherapy of ovarian cancerwith lutetium-177-CC49. J Nucl Med 1996;37:1491.

26. Meredith RF, Khazaeli MB, Macey DJ, et al. Phase IIstudy of interferon-enhanced 131I-labeled high-affinityCC49 monoclonal antibody therapy in patients withmetastatic prostate cancer. Clin Cancer Res 1999;5:3254S.

27. Slavin-Chiorini DC, Kashmiri SVS, Schlom J, et al. Bi-ological properties of chimeric domain-deleted anticar-cinoma immunoglobulins. Cancer Res 1995;1:5957S.

28. Slavin-Chiorini DC, Kashmiri SV, Lee HS, et al. ACDR-grafted (humanized) domain-deleted antitumorantibody. Cancer Biother Radiopharm 1997;12:305.

29. Safavy A, Khazaeli MB, Safavy K, et al. Biodistribu-tion study of 188Re-labeled trisuccin-HuCC49 andtrisuccin-HuCC49�CH2 conjugates in athymic nudemice bearing intraperitoneal colon cancer xenografts.Clin Cancer Res 1999;5:2994S.

30. Kennel SJ, Brechbiel MW, Milenic DE, et al. Actinium-225 conjugates of Mab CC49 and humanized�CH2CC49. Cancer Biother Radiopharm 2002;17:219.

31. Milenic D, Garmestani K, Dadachova E, et al. Ra-dioimmunotherapy of human colon carcinoma xeno-grafts using 213Bi-labeled domain-deleted humanizedmonoclonal antibody. Cancer Biother Radiopharm2004;19:135.

32. Forero A, Meredith RF, Khazaeli MB, et al. A novelmonoclonal antibody design for radioimmunotherapy.Cancer Biother Radiopharm 2003;18:751.

33. Anderson WT, Strand M. Radiolabeled antibody: Iodineversus radiometal chelates. Natl Cancer Inst Monogr1987;3:149.

34. Sharkey RM, Motta-Hennessy C, Pawlyk D, et al.Biodistribution and radiation dose estimates for yttrium-and iodine-labeled monoclonal antibody IgG and frag-ments in nude mice bearing human colonic tumorxenografts. Cancer Res 1990;50:2330.

35. Ugur O, Kostakoglu L, Hui ET, et al. Comparison ofthe targeting characteristics of various radioimmuno-conjugates for radioimmunotherapy of neuroblastoma:Dosimetry calculations incorporating cross-organ betadoses. Nucl Med Biol 1996;23:1.

36. Mirzadeh S, Brechbiel M, Atcher R, et al. Radiometallabeling of immunoproteins: Covalent linkage of 2-(4-isothiocyanatobenzyl) diethylenetriaminepentaaceticacid ligands to immunoglobulin. Bioconj Chem 1990;1:59.

37. Brechbiel M, Gansow O, Atcher R, et al. Synthesis of1-(p-isothiocyanatobenzyl) derivatives of DTPA andEDTA. Antibody labeling and tumor-imaging studies.Inorg Chem 1986;25:2772.

38. Kashmiri SVS, Shu L, Padlan EA, et al. Generation,characterization, and in vivo studies of humanized an-ticarcinoma antibody CC49. Hybridoma 1995;14:461.

39. Shu LM, Qi C-F, Schlom J, et al. Secretion of a single-gene–encoded immunoglobulin from myeloma cells.Proc Natl Acad Sci USA 1995;90:7995.

40. Müller, R. Calculation of average antibody affinity inantihapten sera from data obtained by competitive ra-dioimmunoassay. J Immunol Meth 1980;34:345.

41. Lindmo T, Boven E, Cuttitta F, et al. Determinationof the immunoreactive fraction of radiolabeled mono-clonal antibodies by linear extrapolation to bindingat infinite antigen excess. J Immunol Meth 1984;72:77.

42. Chinn PC, Leonard JE, Rosenberg J, et al. Preclinicalevaluation of 90Y-labeled anti-CD20 monoclonal anti-body for treatment of non-Hodgkin’s lymphoma. Int JOncol 1999;15:1017.

43. DeNardo GL, DeNardo SJ, Wessels BW, et al. 131I lym-1 in mice implanted with human Burkitt’s lymphoma(Raji) tumors: Loss of tumor specificity due to radiol-ysis. Cancer Biother Radiopharm 2000;15:547.

44. Stabin MG, Sparks RB, Crowe E. OLINDA/EXM—thesecond-generation personal computer software for in-ternal dose assessment in nuclear medicine. J Nucl Med2005;46:1023.

45. Pavlinkova G, Booth BJ, Batra SK, et al. Radioim-munotherapy of human colon cancer xenografts usinga dimeric single-chain Fv antibody construct. Clin Can-cer Res 1999;5:2613.

46. Schultz J, Lin Y, Sanderson J, et al. A tetravalent sin-gle-chain antibody-streptavidin fusion protein for pre-targeted lymphoma therapy. Cancer Res 2000;60:6663.

47. Chinn P, Braslawsky G, White C, et al. Antibody ther-apy of non-Hodgkin’s B-cell lymphoma. Cancer Im-munol Immunother 2003;52:257.

48. Roselli M, Schlom J, Gansow O, et al. Comparativebiodistribution studies of DTPA-derivative bifunctionalchelates for radiometal-labeled monoclonal antibodies.Nucl Med Biol 1991;18:389.

49. Brechbiel M, Gansow O. Backbone-substituted DTPAligands for 90Y radioimmunotherapy. Bioconj Chem1991;2:187.

50. Milenic DE, Garmestani K, Chappell LL, et al. In vivocomparison of macrocyclic and acyclic ligands for ra-diolabeling of monoclonal antibodies with 177Lu for radioimmunotherapeutic applications. Nucl Med Biol2002;29:431.

51. Chauhan SC, Jain M, Moore ED, et al. Pharmacokinet-ics and biodistribution of 177Lu-labeled multivalent sin-gle-chain Fv construct of the pancarcinoma monoclo-nal antibody CC49. Eur J Nucl Med Mol Imag2005;32:264.

6138_02_p106-116 4/28/06 1:50 PM Page 116