118
Bloconjugate Chem. 1002, 3, 118-125
Structure-Function Relationships in Indium4 11 Radioimmunoconjugates Kimberly D. Brandt and David K. Johnson' Abbott Laboratories, Department 90M, Abbott Park, Illinois 60064. Received September 19,1991
Conjugates formed by reaction of monoclonal antibody B72.3 with benzyl isothiocyanate derivatives of four amino polycarboxylate chelators (NTA, EGTA, EDTA, DTPA) were labeled with indium-111 and administered iv to athymic mice bearing antigen-positive (LS174T) and antigen-negative (A375) human tumor xenografts. Conjugate immunoreactivities, antibody dose, and xenograft size were controlled, so that the effects of varying chelate structure could be evaluated under conditions where immunologicaland physiological factors were effectivelyheld constant. Tissue distribution and excretion of the radiometal at 24and 48 h postinjection were shown to correlate directly with chelate thermodynamic stability (NTA < EGTA < EDTA < DPTA). Radioactivity levels in the blood and the LS174T xenograft increased, while kidney levels and excretion levels decreased, with increasing chelate stability. The kidney was the only normal organ that accumulated non-antibody-bound ll1In,uptake of radioactivity into all other tissues, and in particular the liver, being unaffected by changes in chelate structure. Mean transferrin saturation in the tumor-bearing athymic mice was found to be 65%. It is proposed that uptake of free lllIn by serum transferrin is precluded in this model, leading to the observed renal localization of unbound label. Kidney:blood and kidney:LS174T activity ratios at 48 h postinjection provided the most sensitive indices of conjugate instability in vivo, spanning 50- and 20-fold ranges, respectively, between the least stable and the most stable conjugate. It is concluded that this antigen/ antibody system and mouse model are well-suited to structure-function studies of immunoglobulin labels.
Antibody-chelator conjugates labeled with the radiometal indium-111 have been widely used in tumor radioimmunoscintigraphy (1-8). The results ofthese trials have been mixed, with high background activity in normal tissues, particularly the liver, often being a major factor that limits tumor detection. Although these findings may reflect inherent limitations on the tissue discrimination achievable in man with such immunoproteins, it is also likely that the true potential of these agents has been obscured by artifacts related to the radiolabeling process. The majority of human trials to date have employed antibody-DTTA' conjugates prepared by procedures (911) that have the potential to produce both antibody denaturation and unstable indium binding sites, and these effects would be expected to increase normal tissue background (12). Recognition of these shortcomings has prompted the development of alternative bifunctional chelators intended for use in indium-111 radioimmunoscintigraphy (13-1 8). An important objective is to obtain chelates that remain completely stable throughout the period required to conduct a radioimmunoscintigraphy study (typically several days when whole immunoglobulin is used). Nevertheless, there are few well-established and generally applicable methods for evaluating the in vivo stability of indium-111 radioimmunoconjugates. Incubation in human serum at 37 OC,with periodic size-exclusion HPLC analysis to determine the percentage of the radiolabel Abbreviations used BSA, bovine serum albumin; BSM, bovine submaxillary mucin; CEA, carcinoembryonic antigen; DTPA, diethylenetriaminepentaacetic acid; DTTA, diethylenetriaminetetraacetic acid;EDTA, ethylenediaminetetraaceticacid; EGTA, ethylene glycol tetraacetic acid; ELISA, enzyme-linked immunosorbent assay; HRPO, horseradish peroxidase; ID, injected dose; ND, not determined;NTA, nitrilotriacetic acid; OPD, o-phenylenediamine;PBS, phosphate-buffered normal saline (pH 7.4); RD, recovered dose; TAG, tumor-associated glycoprotein; TIBC, total iron binding capacity. 1043-1802/92/2903-0118$03.00/0
present in antibody-bound form, has been used to assess the likely stability of a chelate conjugate in the circulation (19). However, because such in vitro serum incubations do not mimic all of the conditions to which a conjugate may be exposed in vivo (e.g. the low pH environment within lysosomes),animal model studies have also generally been undertaken. The most sophisticated of these have used anti-chelate monoclonal antibodies to probe for the presence of intact chelate (20,211,while many investigators have also attempted to interpret biodistribution data from nude mouse/human tumor xenograft models in terms of the stability of the indium-111 chelate label (13,14,2224). In the latter situation, as in man, it has often been difficult to distinguish nontarget tissue uptakes that are due to chelate instability from those caused by intrinsic factors (e.g. immune complex formation, physiological processing of immunoglobulin by liver and kidney) or other label-related phenomena (e.g. reticuloendothelial scavenging of denatured protein). A case in point is the high liver background often seen when indium-11l-labeled antibodies have been administered to athymic mice implanted with a variety of tumor xenografts. This has been ascribed to immune mechanisms in some studies (251, to chelate instability in others (14,221,and, most recently, to an artifact in the labeling process (26). Such disparities point to the need for experimental designs that allow independent investigation of each of the various factors, intrinsic and label-related, that can potentially impact tissue distribution. The present study was designed to isolate the relationship between chelate stability and indium-111 biodistribution in a widely used nude mouse model of human colon cancer, employing the IgGl murine monoclonal antibody B72.3 (13,14,22-24,27). Theintentionwastouseasingle antibody and derivatization chemistry to prepare a series of different chelator conjugates of equivalent immunoreactivity and to study these at a fixed dose in athymic mice bearing antigen-positive and antigen-negative xe0 1992 American Chemical Society
Structure-Function Studles
NCS
SCN-Bz-NTA
NCS
SCN-Bz-EDTA
Bloconjugate Chem., Vol. 3, No. 2, 1992
NCS
SCN-Bz-EGTA
NCS
SCN-Bz-DTPA
Table I. Characteristics of Chelate Labels maximum log 6 for chelator denticity In(II1) chelate NTA 4 15.9 f O . l S , b EGTA a 15-20bvC EDTA 6 25.8 0.21a~b DTPA a 28.5 f 0.08"~~
Data taken from ref 28. Studies with model chelates of SCNBz-EDTA (29)suggestthat formation constants for complexesformed by the carboxymethyl-substituted bifunctional ligands are ca. 20 % lower than those for the correspondingunderivatized chelates. Data unavailable. The range shown is an estimate baaed on corresponding values for other trivalent metals.
nografts of fixed size. Under these conditions, physiologic and immunological factors were expected to remain uniform across all groups of animals studied. Conjugates were constructed using benzyl isothiocyanate derivatives of a series of structurally related chelators (Chart I) that span a range of indium chelate thermodynamic stabilities (Table I), with the expectation that the least stable system (NTA) would be likely to dissociate extensively in vivo while the most stable (DTPA) should show minimal dissociation. Complete metabolic balance studies were performed, in order to define the impact of chelate instability on both tissue distribution and excretion of the radiometal. EXPERIMENTAL PROCEDURES
Preparationof Antibody-Chelator Conjugates. The hybridoma producing B72.3 (American Type Culture Collection, Rockville, MD) was grown in tissue culture using a hollow-fiber bioreactor (Amicon Corp., Danvers, MA),the antibody being purified from the culture medium by affinity chromatography on protein A Sepharose 4B (12,131.Bifunctional chelating agents were synthesized as described elsewhere (13,30) and were coupled to the antibody following the same procedure previously developed for the preparation of B72.3-chelator conjugates (12, 13). All conjugations were carried out at pH 8.5 and 37 "C for 3 h, using an input stoichiometry of chelator: antibody of 3:l and an antibody concentration of 10 mg/ mL. Immunoassay of Antibodyxhelator Conjugates, Microtiter plate ELISA assays that employed BSM as the antigen were used to evaluate the chelator conjugates for differences in retained immunological activity. A detailed description of the assay procedure may be found elsewhere
110
(12,13). Briefly, 96-well microtiter plates (Dynatech Laboratories, Arlington, VA) were coated with BSM (Cooper Biomedical, Malvern, PA), overcoated with BSA, and stored at 2-8 "C until needed. Each plate was then washed with PBS and serial 2-fold dilutions of the B72.3chelator conjugates were applied, using conjugate stock solutions adjusted to an initial concentration of 4 pg/mL. After incubation at 37 "C for 1h, the plate was emptied and washed with PBS and a goat anti-mouse antibodyHRPO conjugate (Kirkegaard & Perry Laboratories, Gaithersburg, MD) was applied. After a further 1-h incubation a t 37 "C, the plate was again emptied and washed with PBS and then the color was developed by addition of OPD and, after quenching with HzS04, was read a t 490 nm using a microtiter plate reader (Dynatech). The chelator conjugates (B72.3-NTA, B72.3-EGTA, B72.3-EDTA, and B72.3-DTPA) were assayed, in duplicate, on the same plate. Indium-11 1 Labeling of Antibody+helator Conjugates. B72.3-chelator conjugates, at concentrations of 5-10 mg/mL in 0.05 M citrate buffer (pH 6.0), were incubated at 37 "C with sufficient ll1InC13 (Nordion International, Inc., Kanata, Ontario, Canada) to give a specific activity of 1mCi/mg if completely incorporated. Radiochemical yields were determined by ITLC following a brief challenge with excess DTPA (12,13).The EGTA, EDTA, and DTPA conjugates incorporated >90% of the radiolabel after overnight incubation and were used in subsequent animal studies without further purification. The radiochemicalyield of 111In-B72.3-NTA achievedafter overnight incubation was only 60% and this conjugate was therefore purified further by TSK-250 size-exclusion HPLC to give a preparation in which >90% of the radiolabel was bound to the antibody. For biodistribution studies, conjugates were diluted into normal saline to a final concentration of 10-50 pg/mL. The solution of indium-111 citrate that was used as a control in the biodistribution studies was prepared following the same procedure employed in labeling the conjugates; i.e. 111InCl3 was incubated in 0.05 M citrate buffer (pH 6.0) for 30 min at 37 "C and then diluted into normal saline. Biodistribution Studies. Female athymic mice (nu/ nu, BALB/c background, Charles River Biotechnology Services, Wilmington, MA) were injected subcutaneously with 7.5 X lo6 A375 human melanoma cells (American Type Culture Collection) in one flank. Two weeks later, 2.5 X lo6LS174T human colon carcinoma cells (American Type Culture Collection) were injected subcutaneously into the opposing flank. When the solid tumors that developed at the injection sites reached sizes of ca. 50-500 mg, the mice were divided into groups of five and 1-5 pg of lllIn-labeled B72.3-chelator conjugate in 100 pL of normal saline was administered to each mouse via a tail vein. Each group of animals was then housed in a separate metabolic cage (Bio-Serv., Inc., Frenchtown, NJ) and provided with food and drinking water. Control groups received iv indium-111 citrate (ca. 1 pCi) in 100 pL of normal saline and were similarly housed in metabolic cages. At either 24 or 48 h after administration of the conjugate, the mice were killed by cervical dislocation and the tumors and all internal organs were removed,weighed,and counted in a y-counter (LKB 1272 Clinigamma, Pharmacia LKB BiotechnologyInc., Gaithersburg, MD). Weighed aliquots of blood, muscle, and skin were also counted and the tail was counted separately to check for extravasation at the injection site. The residual carcass was counted and an estimate of whole body retention of radioactivity obtained, by totaling the carcass counts and all individual tissue
Brandt and Johnson
120 Bloconlugate Chem., Vol. 3, No. 2, 1992 h
U J
4.t 3
201
&
7.51
E c
0 d
b,
' O i
c
a 0 5 i I
02
r 40
I
I
I
I
20
10
05
025
I
0125
I
00625
Conjugate Concentration (pg/mL) Figure 1. ELISA titration curves for binding of B72.3-NTA (0- -o), B72.3-EGTA (0-01,B72.3-EDTA (e-- -e), and B72.3-DTPA (e--.) to a BSM-coated microtiter plate. Each data point represents the mean of two duplicate determinations.
counts. A 100-pL aliquot of each injectate was counted at the same time as the corresponding tissues and the radioactivity in each tissue was expressed as a percentage of this injected dose per gram of tissue. Stool and urine samples that had accumulated over the study period were removed and each cage was then rinsed with deionized HzO and wiped down with gauze. The stool and urine samples plus the rinse HzO and gauze (hereafter "washings") from each group were counted together with the corresponding injectate standard, the counts for each group being divided by 5 to arrive at the average output per animal. Differences in radioactivity distribution were evaluated for statistical significance by the two-tailed Student ttest. Transferrin Saturation Measurements. Four athymic mice bearing LS174T and A375 xenografts (prepared as for, but not used in, biodistribution studies) were bled via a retroorbital sinus and the serum was separated. The iron concentration and TIBC of each serum sample were determined using a commercially available colorimetric assay (A-GENT, Abbott Laboratories). Values for percent transferrin saturation were then calculated from the measured serum iron and TIBC levels. RESULTS
For the purposes of this study, it was important to exclude the possibility that differences in label biodistribution might be caused by differences in retained immunological activity between the various B72.3-chelator conjugates. All preparations were therefore tested and shown to differ by less than 20% in the midpoint absorbance of their ELISA titration curves (Figure l), indicating no significant differences in immunoreactivity between the different chelator conjugates (12). Other potential variables that were controlled included the size of tumor xenografts (31). Mean tumor weights were kept above 50 mg, to avoid unusually high uptakes due to facile vascular access in very small tumors, and below 500 mg, to avoid artifactually low uptakes reflecting necrosis and poor vascularization. The dose of antibody given to each mouse was also held in a narrow range (1-5 pg), so as to minimize any dependence of biodistribution on protein dose (20). To mimic the extreme case of a conjugate undergoing complete dissociation of the radiometal immediately on entering the circulation, control groups were given iv indium-111 citrate. As anticipated, these animals had only
low levels of radioactivity remaining in the blood by 24 h postinjection, with low (and equivalent) uptakes in both xenografts (Table 11). Indium-111 activity was concentrated in the kidneys, but not in the liver or spleen. Radioactivity levels at 48 h postinjection were unchanged from those at 24 h for all tissues, indicating that the observed distribution was achieved rapidly, as would be expected for a low molecular weight species. Tissue distributions of indium-111 administered in the form of the NTA conjugate were indistinguishable from those produced by indium-111 citrate, except for minor differences in activity levels in the blood and both tumors. Although low, blood levels at 24 h postinjection in animals given the NTA conjugate were significantly higher than those in both the 48-h lllIn B72.3-NTA group (p < 0.01) and the 24-h citrate controls @ < 0.01), while uptake into both tumors was some 1.5-2-fold higher for the NTA conjugate than for free indium-111. In some instances, the latter differences were statistically significant (e.g. 24h LS174T levels: citrate vs NTA conjugate, p < 0.001), although in others the large standard deviations characteristic of xenograft tissue in this model obscured the effect. This uptake most likely represented nonspecific trapping within xenografts of indium-111 reaching the tumor in macromolecular form, as levels in the antigen-negative xenograft were elevated (relative to indium-111 citrate) in all groups that received conjugated indium-111. These observations suggest that the ll1In-B72.3-NTA conjugate underwent complete loss of the indium-111 label by 48 h postinjection, but that some fraction of the radiometal remained bound to the antibody long enough to produce a nonspecific elevation of tumor levels and a low, but detectable, concentration of conjugated indium-111 still present in the circulation after 24 h. In contrast to animals given the NTA conjugate, those receiving the EGTA analogue showed clear evidence for specific tumor targeting of the radiolabel. The ratio of LS174T:A375 activity was ca. 2:l at both 24 and 48 h after administration of the antibody, and the absolute amounts of radioactivity in the LS174T xenograft were 3-fold higher than those seen with indium-111 citrate and 2-fold higher than those for the corresponding NTA groups. There was a 2-fold reduction in kidney uptake of the radiolabel and clearance from the circulation was markedly prolonged, with blood levels at 48 h being 8-fold higher than those seen with either indium-111 citrate or ll1In-B72.3-NTA. Viewed in isolation, these characteristics of the "'InB72.3-EGTA conjugate (specific tumor targeting, prolonged blood clearance,limited normal tissue uptake) could be taken as evidence for retention of conjugate integrity in vivo. However, when biodistribution data for the EGTA conjugate were compared to those for the EDTAanalogue, it was immediately apparent that the former underwent significant dissociation. Blood levels of indium-111 activity in animals given l1'In-B72.3-EDTA were significantly higher by 48 h postinjection than those in the corresponding EGTA group (p < 0.01), while LS174T levels were 2-fold higher and renal activity 2-fold lower in the EDTA animals than in those given the EGTA conjugate. Tissue levels of indium111following administration of the DTPA conjugate were not significantly different from those produced by "'InB72.3-EDTA, except that kidney uptake was lower in the DTPA group @ < 0.01). It is noteworthy that liver uptake of indium-111 was unaffected by chelate instability in this model. There was no statistically significant difference between liver activity in the citrate controls and that in any group
Structure-Function Studies
Bioconlugate Chem., Voi. 3, No. 2, 1992
121
Table 11. Tissue Distribution of Indium-111 Activity at 24 and 48 h after Intravenous Administration of Indium-111 Citrate and Indium-1ll-Labeled B72.3Chelator Conjugates to Athymic Mice Bearing Subcutaneous Antigen-Positive (LS174T) and Antigen-Negative (A375) Xenografts % injected dose of In-111 activity per gram of tissue [mean (*SD) for n = 51 indium-111citrate 111In-B72.3-NTA 24 h 48 h 24 h 48 h 24 h 24 h 48 h tissue 48 h 24 h 48 h 1.1(0.3) 13.5 (2.3) 8.5 (2.0) 13.6 (1.9) 13.9 (0.8) 3.4 (0.6) blood 1.9 (0.4) 1.1 (0.3) ND 12.9 (2.6) 7.5 (0.7) 6.0 (0.7) 13.7 (4.7) 14.4 (2.9) 17.6 (3.6) 30.5 (8.6) LS174T" 4.0 (0.4) 4.2 (0.5) ND 25.3 (5.0) 7.3 (1.4) 7.6 (3.8) 8.4 (3.8) 4.6 (0.6) 4.3 (1.1) 6.1 (1.9) 8.2 (1.2) 8.0 (1.2) A375b ND 8.7 (1.7) 4.7 (0.7) kidney 18.7 (3.8) 18.0 (1.7) 23.6 (3.2) 21.9 (2.5) 10.8 (3.6) 13.6 (2.9) 6.5 (0.6) ND 4.7 (0.8) 5.2 (0.4) 5.2 (0.7) 4.6 (0.9) 5.4 (1.5) 6.1 (0.8) liver 4.2 (0.3) 5.3 (0.8) 6.6 (1.3) ND 7.5 (2.7) 5.0 (0.7) 5.0 (1.6) 5.9 (1.3) 4.7 (0.6) 5.0 (0.7) spleen 4.6 (0.5) 5.4 (0.4) 5.2 (0.8) ND 6.5 (1.6) 5.8 (0.7) 6.7 (1.0) 5.3 (1.1) 3.8 (0.5) 3.3 (0.4) lungs 3.2 (0.3) 4.0 (1.3) 6.2 (0.6) ND 6.1 (1.2) 6.1 (1.1) 2.8 (0.3) 3.5 (0.3) 2.8 (0.3) heart 2.8 (0.1) 5.4 (0.9) 4.9 (0.6) 4.2 (0.7) ND 5.4 (1.8) 1.4 (0.4) 2.3 (0.6) 4.1 (0.4) 2.7 (0.2) 2.2 (0.4) GI tract 2.8 (0.1) 1.4 (0.4) 2.3 (0.5) ND 1.6 (0.2) 1.9 (0.4) 1.8 (0.3) 1.9 (0.5) 2.1 (0.5) muscle 1.6 (0.3) 1.9 (0.4) 2.3 (0.6) 2.2 (0.4) ND 2.0 (0.3) 4.1 (0.7) 3.6 (1.2) 4.0 (0.9) 3.8 (0.6) 3.4 (0.4) skin 3.3 (0.6) 3.4 (0.7) 4.3 (0.7) ND 3.4 (0.8) 0 Mean tumor weighta in grams (*SD): 24-h groups-indium-111 citrate, 0.32 (0.14);l1'In-B72.3-NTA, 0.10 (0.04);lllIn-B72.3-EGTA, 0.23 (0.04); 111In-B72.3-EDTA, 0.27 (0.09);48-h groups-indium-111 citrate, 0.28 (0.11); 1111n-B72.3-NTA, 0.12 (0.07); ll1In-B72.3-EGTA, 0.36 (0.13); 111In-B72.3-EDTA, 0.28 (0.06); ll1In-B72.3-DTPA, 0.36 (0.15). Mean tumor weights in grams (MD): 24-h groups-indium-111 citrate, 0.28 (0.08); 111In-B72.3-NTA, 0.05 (0.02); l1'In-B72.3-EGTA, 0.10 (0.06); l1'In-B72.3-EDTA, 0.20 (0.15); 48-h groups-indium-111 citrate, 0.22 (0.10); 111In-B72.3-NTA, 0.05 (0.03); ll1In-B72.3-EGTA, 0.11 (0.06); l111n-B72.3-EDTA, 0.33 (0.16); l1'In-B72.3-DTPA, 0.12 (0.04). Table 111. Distribution of Injected Indium-111 Activity between Tissues and Excreta at 24 and 48 h after Intravenous Administration of Indium-111 Citrate and Indium-lll-Labeled B72.3Chelator Conjugates to Athymic Mice Bearing Subcutaneous Antigen-Positive (LS174T) and Antigen-Negative (A375) Xenografts indium-111citrate ll1In-B72.3-NTA ll1In-B72.3-EGTA 1111n-B72.3-EDTA 111In-B72.3-DTPA 24h 48 h 24 h 48 h 24 d 48h 24 h 48h 24h 48h % ID in mouse [mean (*SD) for n = 51 60.4 (3.9) 61.9 (6.6) 80.3 (3.5) 67.1 (2.7) 81.0 (14.8) 75.6 (14.0) 67.1 (9.0) 70.6 (6.2) ND 80.0 (12.0) % ID in urine 12.0 11.2 3.6 5.2 3.0 3.0 4.5 5.7 ND 4.0 (mean for n = 5) % ID in stool 7.1 8.7 5.6 10.4 3.8 8.2 4.5 4.7 ND 3.8 (mean for n = 5) % ID in washings (mean for n = 5) 6.0 6.2 3.0 3.1 1.6 2.4 3.8 3.0 ND 2.4 total % ID recovered 85.5 88.0 92.5 85.8 89.4 89.2 79.9 84.0 ND 90.2 (mean for n = 5) Table IV. Tissue and Excreta Data from Table 111 Restated in the Form of Whole-Body Retention and Overall Excretion Values Expressed as a Percentage of the Radioactivity Recovered from Each Group indium-111citrate ll1In-B72.3-NTA l111n-B72.3-EGTA 1111n-B72.3-EDTA 1111n-B72.3-DTPA 24h 48h 24h 48h 24h 48 h 24h 48h 24 h 48h mean % RD in mouse (n = 5) 70.6 70.3 86.8 78.2 90.6 84.8 84.0 84.0 ND 88.7 mean % RD excreted 29.4 29.7 13.2 21.8 9.4 15.2 16.0 16.0 ND 11.3 (stool + urine + washings, n = 5)
receiving indium-111 in conjugated form @ > 0.1). The liver, heart, and lungs did show a trend toward slightly increased activity with increasing chelate stability, but this probably reflected the increasing radioactivity levels in blood trapped within these highly perfused organs. The only tissues that were responsive to changes in chelate structure were the kidney, the LS174T xenograft, and the blood. Blood levels of unsaturated transferrin, which is assumed to mediate distribution of indium-111 once lost from the immunoglobulin,were found to be lower in tumorbearing nude mice than in normal man. The mean serum iron concentration in athymic mice bearing LS174T and A375 xenografts was 250 pg/dL (range, 189-301 pg/dL) with a TIBC of 388 pg/dL (range, 312-435 pg/dL), giving a mean saturation of 65% (range, 51-81%). Patterns of excretion of the indium-111 label (Table 111)were consistent with the overall trends found in the tissue-distribution data, although metabolic studies proved to be of limited utility due to the relatively low levels of label excretion that were seen. There were no clear cut trends in the distribution of excreted activity among stool, urine, and washings, and given the inherent imprecision
of these measurements, the data are best combined and expressed as values for total label excretion in each group. While whole-body retention of radioactivity could be shown to be significantly higher for some conjugates than for indium-111 citrate (e.g. 24 h ll1In-B72.3-NTA vs indium-111 citrate, p < 0.001), this was not true in other cases where the standard deviations were large (e.g. 24 h ll1In-B72.3-EDTA vs indium-111 citrate, p > 0.1). The mean recovery of radioactivity over nine groups of animals was 87.2%, and as mechanical losses were minimal, this shortfall (relative to a 100pL counting standard) probably represents the actual mean volume of injectate received by the animals. It is therefore useful to restate the metabolic data in the form of whole-body retention and overall excretion values, expressed as a percentage of the counts recovered from each group rather than a percentage of the estimated counts injected (Table IV). Viewed in this way, a trend toward increased whole-body retention and decreased overall excretion of the radiometal with increasing chelate stability is more readily apparent. Comparison of the 24-h data with the corresponding 48-h values in Table IV also confirms other inferences
122
Bloconlugete Chem., Vol. 3, No. 2, 1992
drawn from the tissue-distribution measurements. Those groups where the label was either so unstable that distribution appeared complete by 24 h postinjection (indium-111 citrate) or so stable that nontarget distribution changed little during 24 h (l"In-B72.3-EDTA) showed no incremental excretion of the radiometal between 24 and 48 h after administration. In contrast, those groups in which tissue data suggested a progressive loss of indium-111 from the conjugate throughout the study period (ll1In-B72.3-NTA, ll1In-B72.3-EGTA) showed a decrease in whole-body activity and a corresponding increase in overall excretion between the 24- and 48-h time points. Although excretion data therefore reinforced the conclusions drawn from tissue-distribution studies, the narrow range between the least stable and the most stable conjugate (10% RD at 48 h, Table IV) made excretion a relatively insensitive index of conjugate instability invivo. This is well-illustrated by the EDTA and EGTA conjugates, which were indistinguishable on the basis of 48-h excretion values yet showed readily observable differences in tissue distribution of the radiometal at that time point. It is thus unlikely that excretion patterns alone could be used to differentiate conjugates of varying stability in this model. DISCUSSION
When a unique biological entity (e.g. a protein or polynucleotide having a particular specificity) is coupled to a unique reporter or effector molecule (e.g. an enzyme or fluorophore) and evaluated in a biological assay system that is also often unique, the singular nature of the results obtained makes comparison with those for any other conjugate, particularly between different laboratories, difficult if not impossible. Nevertheless, systematic intercomparisons of related conjugates in common assay systems are likely to be needed if the rules that govern construction of optimum reagents of this type are to become fully understood. In particular, research in which the reporter/effector molecule is a low molecular weight organic moiety (e.g. a drug or chelate) should lend itself to the structure-function approach often taken in medicinal chemistry when developing such molecules in nonconjugated form, provided that potential variables associated with the biological moiety and the bioassay can be held constant. This study was undertaken to test the feasibility of applying such an approach to antibodychelator conjugates for radioscintigraphic use, where the function of the reporter/effector molecule is to bind a metal in stable fashion in vivo. Control of all variables arising from the antibody and mouse model appears to have been adequate in this study, as the patterns of indium-111 distribution and excretion that were seen can be rationalized entirely on the basis of differences in the structure, and hence thermodynamic stability, of the various chelates employed. In order to retain indium-111 in vivo, a chelator must be able to compete effectively with endogenous metal-binding molecules and, in particular, with the iron transport protein transferrin, which has a high affinity for indium(II1) [log Kfor the 1:l In-transferrin complex was recently estimated to be 18.8 (32)l. Complete saturation of the coordination sphere of a metal, which has recently been shown to require an octadentate ligand in the case of indium (33),may also be needed for maximum chelate stability. It is therefore unsurprising that NTA, a tetradenate chelator with a relatively modest affinity for indium, gave a conjugate that appeared to completely lose the radiometal after 48 h in vivo. It is perhaps more surprising that this
Brandt and Johnson
dissociation process was relatively slow, the tissue data and metabolic studies together providing substantial evidence that breakdown of the ll1In-B72.3-NTA conjugate was still incomplete at 24 h postinjection. This finding is in keeping with previous observations (21)that transchelation reactions, even when favored thermodynamically, can be slow when substituted amino polycarboxylate ligands are used. It was therefore uncertain at the outset whether the EGTA ligand would give conjugates that were stable in vivo. The formation constant for the EGTA complex of indium is unavailable, but it is known that this chelator can function as an octadentate ligand (34) and it seemed possible that kinetic barriers to the dissociation of a coordinatively saturated indium chelate might be sufficient to produce acceptable inertness in vivo. That this proved not to be the case is evident from the data in Tables I1 and IV. In contrast, the EDTA and DTPA chelates gave conjugates that appeared to be highly stable in vivo, although a question remains as to whether meaningful performance differences exist between the latter two preparations. While the ll1In-B72.3-EDTA and l1lIn-B72.3-DTPA conjugates were indistinguishable on the basis of blood levels and LS174T uptake, the DTPA conjugate did produce a significantly lower level of activity in the kidney, which was the only normal organ to accumulate unbound indium. In addition, whole-body retention of indium-111 activity was higher, and excretion lower, in animals given l1lIn-B72.3-DTPA than in those receiving the EDTA analogue, although this difference was not statistically significant. A previous study in which the B72.3-EDTA and B72.3-DTPA conjugates were compared in nude mice bearing LS174T tumors (13)gave similar results (equivalent LS174T and blood levels, significantly lower kidney uptake and higher, but not statistically significant wholebody retention with the DTPA conjugate). Taken together, these findings suggest that the DTPA conjugate is probably slightly more stable than the EDTA analogue, but this difference in performance is small and could not be demonstrated unequivocally within the present study design. That chelate structure can be shown to correlate with function in this context is perhaps of less significance than is a clear definition of the physiological consequenceswhen that function (stable binding of indium to antibody) is not adequately fulfilled. Identification of the kidney as the sole normal organ responsive to changes in the stability of these chelates is advantageous, inasmuch as it simplifies the animal model, but also disadvantageous, as it indicates that the animal model does not mimic the way that unbound indium is handled in man. Trace amounts of indium in "ionic" form are rapidly taken up by the ironbinding sites of serum transferrin when the radiometal is administered iv in normal man (35). As most of the iron in transit in the bloodstream at any given time is destined to be stored as ferritin until needed for erythropoiesis, the primary organ of deposition of the "'In tracer becomes the primary organ responsible for iron storage, namely the liver. That this is not the case in the nude mouse suggests that distribution of free indium-111 in the latter may occur through alternative mechanisms. The rapid blood clearance and renal localization of radioactivity in animals given indium-111citrate have been previously described in similar models (24, 36) and are certainly consistent with glomerular trapping of indium111being filtered from the bloodstream in low molecular weight form. This, in turn, infers that there exists some type of barrier to binding of free indium-111 by serum
Structure-Function Studies
transferrin in the nude mouse. When such a barrier was intentionally created (361, by administering a saturating dose of iron 1h before giving indium-111 citrate, kidney activity at 48 h postinjection in MF1-nu/Ola mice bearing HX99 breast carcinoma xenografts was reported to increase 4-fold (from 10 f 3.1% ID/g to 40.0 f 14.4% ID/g) while liver activity was unaffected (4.1 f 1.2 % ID/g without presaturation vs 4.8 f 1.9% ID/g with presaturation). Although such intentional presaturation had not been performed in the present study, we speculated that ambient levels of transferrin saturation in the nude mouse might be significantly higher than in man. Serum iron and TIBC measurements confirmed that this was indeed the case. In normal man, ca. 35% of serum transferrin binding sites are occupied by iron ( 3 3 , whereas the analogous value measured in the mice used in this study was 65 5%. This disparity may reflect intrinsic differences in iron metabolism between man and the athymic mouse, or it may be that the presence of the rapidly growing xenograft tissue produces an elevated demand for iron [a growth factor essential for tumor proliferation (38)l that translates into elevated serum levels as dietary and storage iron is transported to the xenograft to satisfy this demand. Whatever its origin, the elevation is transferrin saturation that was documented in the tumor-bearing athymic mice used in this study should not have been sufficient to prevent binding of the injected dose of indium-111 on strictly stoichiometric grounds. However, the two ironbinding sites of transferrin are known to differ in their metal-chelating properties (39). If indium were to be able to bind with high affinity to only one of these sites in the nude mouse and if, at 65% saturation, that site were to be preferentially occupied by iron, uptake of indium by the transport protein would be preempted. Such a hypothesis offers the best explanation for the observed behavior of nonantibody-bound indium-111 in this paradigm. Recognition that the liver is not a site of accumulation of unbound indium-111 in this model narrows the range of possible explanations for the elevated liver activities that have certainly been produced by some preparations (22,23). Reticuloendothelial uptake of antibody that has become denatured during conjugation remains a likely source of high liver backgrounds when coupling is achieved using cross-linking agents such as the bicyclic dianhydride of DTPA (23). However, elevated liver activities have also been reported even when non-cross-linking conjugation methods were used [e.g. 14% ID/g for B72.3 labeled via an SCN-Bz-EDTAchelator (22)l. In such cases, the most likely source is first-pass hepatic clearance of immune complexes formed as a result of antigen being shed from the xenograft into the circulation. Such a mechanism has been proposed in the case of CEA (25) but not, to our knowledge,in the case of the TAG-72 antigen system. As leakage of antigen into the bloodstream is likely to depend both on the level of its expression and on physiological factors, such as the size and vascular integrity of the xenograft, elevations in liver background due to this source would be expected to be sporadic and difficult to control. Although the model is clearly imperfect in duplicating the fate of indium once released from a radioimmunoconjugate, the fact that this fate is fixation in a single organ system (as opposed to translocation to a serum protein and subsequent redistribution to a variety of possible tissues) simplifies the selection of biological endpoints. Kidney:blood (K.B) and kidney:LS174T ( K T ) activity ratios at 48 h postinjection provide the most
Biocon]ugate Chem., Voi. 3, No. 2, 1992
129
sensitive indices of in vivo instability, spanning 50- and 20-fold ranges, respectively, between the least stable (K:B = 20; K:T = 3.7) and most stable (K:B = 0.4; K T = 0.2) conjugate. It is unclear how low these ratios would become if a conjugate were to remain absolutely stable in vivo. Kidney activity produced by ll1In-B72.3-DTPA in this model has previously been shown to be the same as that produced by 1261-labeledB72.3 (121, inferring that the residual renal activity seen with the DTPA conjugate represents immunoglobulin and not free label. In clinical radioimmunoscintigraphy trials that have employed the SCN-Bz-DTPA chelator (Chart I), translocation of the indium-111 label to serum transferrin has not been detectable by SDS-PAGE, immunoprecipitation, and sizeexclusion HPLC analyses of patient sera (40),and hepatic metastases have been readily visualized in a high proportion of cases (40-43). These observations suggest that the apparent stability that the DTPA label shows in nude mice is also evident in man and, consequently, that the K:B and K:T values produced by l1'In-B72.3-DTPA in the mouse model probably approach the theoretical minima. Nevertheless, low levels of "'In-transferrin have been detected in patient sera when anti-transferrin affinity chromatography was used (441,indicating that the SCNBz-DTPA chelator certainly does not meet the criterion of absolute stability in vivo under all circumstances. It is possible that alternative chelators that more closely approach this ideal would produce significantly lower K: B and K T values, although, as with the EDTA and DTPA conjugates of this study, modified experimental designs are likely to be required if a small difference in performance between two high-affinity chelates is to be demonstrated with any degree of statistical confidence. Although access to many antibody/antigen systems is limited, the B72.3 hybridoma, the LS174TandA375 tumor cell lines, and the BSM antigen used in the immunoreactivity assay are all freely available to any investigator. This model is thus well-suited to serve as a generic testbed for evaluating different labeling chemistries in circumstances where the results can be readily compared with an existing body of data and where duplication of results between different laboratories could be undertaken. Without such studies, it is likely that attempts to understand the in vivo behavior of synthetic moieties used to label immunoglobulins will continue to be confounded by variables that, in reality, are unrelated to the label but rather arise from often unrecognized idiosyncracies of a particular antibody/antigen system and its host. LITERATURE CITED (1) Fairweather, D. S., Bradwell, A. R., Dykes, P. W., Vaughan, A. T., Watson-James, S. F., and Chandler, S. (1983) Improved tumor localization using indium-111 labeled antibodies. Br. Med. J. 287,167-170. (2) Murray, J. L., Rosenblum, M. G., Sobol,R. L., Bartholomew, R. M., Plager, C. E., Haynie, T. P., Jahns, M. F., Glenn, H. J., Lamki, L. M., Benjamin, R. S., Papadopoulos, N., Boddie, A. W., Frincke, J. M., David, G. S., Carlo, D. J., and Hersh, E. M. (1985) Radioimmunoimaging in malignant melanoma with lllIn-labeled monoclonalantibody 96.5. Cancer Res. 45,23762381. (3) Patt, Y. Z., Lamki, L. M., Haynie, T. P., Unger, M. W., Rosenblum, M. G., Shirkhoda, A., and Murray, J. L. (1988) Improved tumor localization with increasing dose of indium-111 labeled anti-carcinoembryonic antigen monoclonalantibody ZCE-025 in metastatic colorectal cancer. J. Clin. Oncol. 6,1220-1230. (4) Maguire, R. T., Schmelter, R. F., Pascucci, V. L., and Conklin, J. L. (1989) Immunoscintigraphy of colorectal adenocarcinoma: results with site-specifically radiolabeled
124
Bioconlugete Chem., Vol. 3, No. 2, 1992
B72.3 (11%-CYT-103). Antibody Zmmutwconjugates Radiopharm. 2, 257-269. (5) Beatty, J. D., Hyams, D. M., Morton, B. A., Beatty, B. G., Williams, L. E., Yamauchi, D., Merchant, B., Paxton, R. J., and Shively, J. E. (1989) Impact of radiolabeled antibody imaging on management of colon cancer. Am. J. Surg. 157, 13-19. (6) Siccardi, A. G., Buraggi, G. L., Callegaro, L., Colella, A. C., De Filippi, P. G., Galli, G., Mariani, G., Masi, R., Palumbo, R., Riva, P., Salvatore, M., Scassellati, G. A., Scheidhauer, K., Turco, G. L., Zaniol, P., Benini, S., Deleide, G., Gasparini, M., Lastoria, S., Mansi, L., Paganelli, G., Salvischiani, E., Seregni, E., Viale, G., and Natali, P. G. (1989) Immunoscintigraphy of adenocarcinomas by means of radiolabeled F(ab’)z fragments of an anti-carcinoembryonic antigen monoclonal antibody: A multicenter study. Cancer Res. 49, 3095-3103. (7) Chatal, J-F.,Saccavini, J-C.,Gestin, J-F.,ThBdrez,P.,Curtet, C., Kremer, M., Guerreau, D., NolibB, D., Fumoleau, P., and Guillard, Y. (1989) Biodistribution of indium-111 labeled OC 125 monoclonal antibody intraperitoneally injected into patients operated on for ovarian carcinomas. Cancer Res. 49, 3087-3094. (8) Patt, Y. Z., Lamki, L. M., Shanken, J., Jessup, J. M., Charnsangavej, C., Ajani, J. A., Levin, B., Merchant, B., Halverson, C., and Murray, J. L. (1990) Imaging with indiumlll-labeled anticarcinoembryonic antigen monoclonal antibody ZCE-025 of recurrent colorectalor carcinoembryonicantigen-producing cancer in patients with rising serum carcinoembryonicantigen levels and occult metastases. J. Clin. Oncol. 8, 1246-1254. (9) Krejcarek, G. E., and Tucker, K. L. (1977) Covalent attachment of chelating groups to macromolecules. Biochem. Biophys. Res. Commun. 77, 581-585. (10) Hnatowich, D. J., Layne, W. W., Childs, R. L., Lanteigne, D., Davis, M. A., Griffin, T. W., and Doherty, P. W. (1983) Radioactive labeling of antibody: A simple and efficient method. Science 220, 613-615. (11) Paxton, R. J., Jakowatz, J. G., Beatty, J. D., Beatty, B. J., Vlahos, W. G., Williams, L. E., Clark, B. R., and Shively, J. E. (1985) High specific activity 111In-labeledanticarcinoembryonic antigen antibody: Improved method for the synthesis of diethylenetriaminepentaaceticacid conjugates. Cancer Res. 45, 5694-5699. (12) Carney, P. L., Rogers, P. E., and Johnson, D. K. (1989) Dual isotope study of iodine-125and indium-111-labeled antibody in athymic mice. J. Nucl. Med. 30, 374-384. (13) Westerberg, D.A., Carney, P. L., Rogers,P. E., Kline, S. J., and Johnson, D. K. (1988) Synthesis of novel bifunctional chelators and their use in preparing monoclonal antibody conjugates for tumor targeting. J. Med. Chem. 32,236-243. (14) Brechbie1,M. W.,Gansow,O. A.,Atcher, R. W., Schlom, J., Esteban, J., Simpson, D. E., and Colcher, D. (1986) Synthesis of 1-@-isothiocyanatobenzyl) derivatives of DTPA and EDTA. Antibody labeling and tumor-imaging studies. Znorg. Chem. 25, 2772-2781. (15) Craig,A. S., Helps, I. M., Jankowski, K. J., Parker, D., Beeley, N. R. A., Boyce, B. A., Eaton, M. A. W., Millican, A. T., Millar, K., Phipps, A., Rhind, S. K., Harrison, A., and Walker, C. (1989) Towards tumor imaging with indium-111 labelled macrocycle-antibodyconjugates. J. Chem. SOC.Chem. Commun. 194-796. (16) Mathias, C. J., Sun, Y., Connett, J. M., Philpott, G. W., Welch, M. J., and Martell, A. E. (1990) A new bifunctional chelate, BrMe2HBED An effective conjugate for radiometals and antibodies. Znorg. Chem. 29, 1475-1480. (17) Mathias, C. J., Sun,Y., Welch, M. J., Connett, J. M., Philpott, G. W., and Martell, A. E. (1990) N,N’-Bis(2-hydroxybenzyl)-l-(4-bromoacetamidobenzyl)-1,2-ethylenediamineN,N’-diacetic acid: A new bifunctionalchelate for radiolabeling antibodies. Bioconjugate Chem. 1, 204-211. (18) Ruser, G., Ritter, W., and Maecke, H. R. (1990) Synthesis and evaluation of two new bifunctional carboxymethylated tetraazamacrocyclicchelating agents for protein labeling with indium-111. Bioconjugate Chem. 1 , 345-349. (19) Cole, W. C., DeNardo, S. J., Meares, C. F., McCall, M. J., DeNardo, G. L., Epstein, A. L., OBrian, H. A., and Moi, M.
Brandt and Johnson
K. (1987) Comparative serum stability of radiochelates for antibody radiopharmaceuticals. J. Nucl. Med. 28, 83-90. (20) Adams, G. P., DeNardo, S. J., Deshpande, S. V., DeNardo, G. L., Meares, C. F., McCall, M. J., and Epstein, A. L. (1989) Effect of mass of lllIn-benzyl-EDTA monoclonal antibody on hepatic uptake and processing in mice. Cancer Res. 49,17071711. (21) Deshpande, S.V., Subramanian,R., McCal1,M.J.,DeNardo, S. J., DeNardo, G. L., and Meares, C. F. (1990) Metabolism of indium chelates attached to monoclonalantibody: Minimal transchelation of indium from benzyl-EDTA chelate in vivo. J. Nucl. Med. 31, 218-224. (22) Esteban, J. M., Schlom, J., Gansow, 0. A., Atcher, R. W., Brechbiel, M. W., Simpson, D. E., and Colcher, D. (1987) New method for the chelation of indium-111 to monoclonal antibodies: Biodistribution and imaging of athymic mice bearing human colon carcinoma xenografts. J.Nucl. Med. 28, 861-870. (23) Brown, B. A,, Comeau, R. D., Jones, P. L., Liberatore, F. A., Neacy, W. P., Sands, H., and Gallagher, B. M. (1987)Pharmacokinetics of the monoclonal antibody B72.3 and ita fragments labeled with either lZsIor T n . Cancer Res. 47, 1149-1154. (24) Roselli, M., Schlom, J., Gansow, 0. A., Raubitschek, A., Mirzadeh, S., Brechbiel, M. W. and Colcher, D. (1989) Comparative biodistributions of yttrium- and indium-labeled monoclonal antibody B72.3 in athymic mice bearing human colon carcinoma xenografts. J. Nucl. Med. 30, 672-682. (25) Beatty, B. G., Beatty, J. D., Williams, L. E., Paxton, R. J., Shively, J. E., and OConnor-Tressel, M. (1989) Effect of specificantibody pretreatment on liver uptake of lllIn labeled anticarcinoembryonic antigen monoclonal antibody in nude mice bearing human colon cancer xenografts. Cancer Res. 49, 1587-1594. (26) Schuhmacher, J., Klivhyi, G., Matys, R., Kirchgebner, H., Hauser, H., Maier-Borst, W., and Matzku, S. (1990) Uptake of indium-111 in the liver of mice following administration of indium-111-DTPA-labeledmonoclonalantibodies: Influence of labeling parameters, physiologicparameters, and antibody dose. J. Nucl. Med. 31, 1084-1093. (27) Keenan, A. M., Colcher, D., Larson, S. M., and Schlom, J. (1984) Radioimmunoscintigraphy of human colon cancer xenografts in mice with radioiodinated monoclonal antibody B72.3. J. Nucl. Med. 25, 1197-1203. (28) Subramanian, K. M., and Wolf, W. (1990) A new radiochemical method to determine the stability constants of metal chelates attached to a protein. J. Nucl. Med. 31, 480-488. (29) Betebenner, D. A., Carney, P. L., Zimmer, A. M., Kazikiewicz, J. M., Brucher, E., Sherry, A. D., and Johnson, D. K. (1991) Hepatobiliary delivery of polyaminopolycarboxylate chelates: Synthesis and characterization of a cholic acid conjugate of EDTA and biodistribution and imaging studies with its indium-111 chelate. Bioconjugate Chem. 2,117-123. (30) Kline, S. J., Betebenner, D. A., and Johnson, D. K. (1991) Carboxymethyl-substituted bifunctional chelators: Preparation of aryl isothiocyanate derivatives of 3-(carboxymethyl)3-azapentanedioic acid, 3,12-bis(carboxymethyl)-6,9-dioxa3,12-diazatetradecanedioicacid, and 1,4,7,10-tetraazacyclododecane-N,N’,N”,N’’’-tetraacetic acid for use as protein labels. Bioconjugate Chem. 2, 26-31. (31) Williams, L. E., Duda, R. B., Proffitt, R. T., Beatty, B. G., Beatty, J. D., Wong, J. Y. C., Shively, J. E., and Paxton, R. J. (1988) Tumor uptake as a function of tumor mass: A mathematical model. J. Nucl. Med. 29, 103-109. (32) Bannochie, C. J., and Martell, A. E. (1989) Affinities of racemic and meso forms of N,N’-ethylenebis(2-(o-hydroxypheny1)glycine) for divalent and trivalent metal ions. J. Am. Chem. SOC.111,4735-4742. (33) Maecke, H. R., Riesen, A., and Ritter, W. (1989) The molecular structure of indium-DTPA. J. Nucl. Med. 30, 12351239. (34) Schauer, C. K., and Anderson, 0.P. (1987)Calcium-selective ligands. 2. Structural and spectroscopic studies on calcium 109, and cadmium complexes of EGTAd-. J. Am. Chem. SOC. 3646-3656.
Structure-Function Studles
(35) Hosain, F., McIntyre, P. A., Poulose, K., Stern, H. S., and Wagner, H. N. (1969)Binding of trace amounts of ionic indium113m to plasma transferrin. Clin. Chim. Acta 24, 69-75. (36) Ward, M. C., Roberta, K. R., Westwood, J. H., Coombes, R. C. C., and McCready, V. R. (1986) The effect of chelating agents on the distribution of monoclonal antibodies in mice. J. Nucl. Med. 27, 1746-1750. (37) Ramsay, W. N. M. (1957) The determination of the total iron-binding capacity of serum. Clin. Chim. Acta 2,221-226. (38) Weinberg, E. D. (1984)Iron withholding: A defense against infection and neoplasia. Physiol. Rev. 64, 65-102. (39) See, for example: Luk, C. K. (1971) Study of the nature of the metal-binding sites and estimate of the distance between the metal-bindingsites in transferrin using trivalent lanthanide ions as fluorescentprobes. Biochemistry IO,2838-2843. Donovan, J. W., and Ross, K. D. (1975) Non-equivalence of the metal-binding sites of conalbumin (ovotransferrin). Calorimetric and spectrophotometric studies of binding and displacement of aluminum. Fed. h o c . 34,593. Princiotto, J. V., and Zapolski, E. J. (1975) Difference between the two ironbinding sites of transferrin. Nature 255, 87-88. Cannon, J. C., and Chasteen, N. D. (1975) Nonequivalence of the metal binding sites in vanadyl labeled human serum transferrins. Biochemistry 14, 4573-4577. (40) Griffii,T. W.,Brill, A. B., Stevens,S.,Collins, J. A.,Bokhari, F., Bushe, H., Stochl, M. C., Gionet, M., Rusckowski, M., Stroupe, S. D., Kiefer,H. C., Sumerdon, G. A., Johnson, D. K.,
Bioconlugate Chem., Vol. 3, No. 2, 1992
126
and Hnatowich, D. J. (1991) Initial clinical study of indium111-labeledclone 110 anti-carcinoembryonicantigen antibody in patients with colorectal cancer. J . Clin. Oncol. 9,631-640. (41) Divgi, C. R., McDermott, K., Johnson, D. K., Schnobrich, K. E., Finn, R. D., Cohen, A. M., and Larson, S. M. (1991) Detection of hepatic metastases from colorectal carcinoma using indium-111 labeled monoclonal antibody: MSKCC experience with mAb lllIn-Cl10. Nucl. Med. Biol. 18, 705710. (42) Johnson, D. K., Seevers, R. H., Schnobrich, K. E., Golick, J. A., Carney, P. L., Vijayakumar, V., and Blend, M. J. (1991) Detection of hepatic metastases from colorectal carcinoma using indium-111 labeled antibody: Initial clinical results withB72.3. Antibody Immunoconjugates Radiopharm. 4,223. (43) Vijayakumar, V., Blend, M. J.,Johnson, D. K., Schnobrich, K. E., and Golick,J. (1992) Detection of recurrent colon cancer with In-111-labeled MoAb B72.3 in a patient who had normal CEA and TAG-72 levels. Clin. Nucl. Med. In press. (44) Hnatowich, D. J., Rusckowski, M., Brill, A. B., Siebecker, D. A., Misra, H., Mardirossian, G., Bushe, H., Rescigno, A., Stevens, S., Johnson, D. K., and Griffin, T. W. (1990) Pharmacokinetics in patients of an anti-CEA antibody labeled with indium-111 using a novel diethylenetriaminepentaacetic acid chelator. Cancer Res. 50, 7272-7278.
Registry No. SCN-Bz-EDTA, 117499-22-6;SCN-Bz-DTPA, 117499-23-7.
