Tumor Targeting of Radiometal Labeled Anti-CEA Recombinant T84

220

Bioconjugate Chem. 2001, 12, 220−228

Tumor Targeting of Radiometal Labeled Anti-CEA Recombinant T84.66 Diabody and T84.66 Minibody: Comparison to Radioiodinated Fragments Paul J. Yazaki,*,† Anna M. Wu,† Shih-Wa Tsai,‡ Lawrence E. Williams,§ David N. Ikle’,| Jeffrey Y. C. Wong,⊥ John E. Shively,‡ and Andrew A. Raubitschek# Department of Molecular Biology and Division of Immunology of the Beckman Research Institute of the City of Hope, Duarte, California 91010, Divisions of Radiology, Biostatistics, Radiation Oncology, and Radioimmunotherapy of the City of Hope National Medical Center, Duarte, California 91010. Received July 31, 2000

Recombinant antibody fragments offer potential advantages over intact monoclonal antibodies in the radioimmunoscintigraphy (RIS) of solid tumors. Due to their smaller molecular size, antibody fragments have shown rapid tumor targeting and blood clearance, a more uniform tumor distribution and a lower potential to elicit a human immune response. Previously, we have expressed two genetically engineered antibody fragments, the T84.66 diabody (scFv dimer) and the T84.66 minibody (scFv-CH3 dimer), specific to carcinoembryonic antigen (CEA). When radioiodinated, both antibody fragments exhibited rapid tumor targeting and rapid blood clearance in xenografted mice. To extend and optimize their future clinical RIS utility with radiometals, these antibody fragments were conjugated with the macrocycle 1,4,7,10-tetraazacyclododecane N,N′,N′′,N′′′-tetraacetic acid (DOTA) and labeled with 111In. Tumor targeting and biodistribution studies were carried out in athymic mice xenografted with a human colorectal tumor cell line, LS174T. The [111In]T84.66 diabody (55 kDa) exhibited very rapid tumor targeting with 12.5 ( 0.4% injected dose per gram (% ID g-1 ( standard error) at 2 h and reached a maximum of 13.3 ( 0.9% ID g-1 at 6 h. However, kidney uptake was observed to reached a peak of 183.5 ( 21.0% ID g-1 at 6 h, a result similar to that reported by others for other low molecular weight fragments labeled with radiometals. Preadministration of an oral dose of D-lysine resulted in a 59% lowering of the renal accumulation at 6 h, but was accompanied by a 31% reduction of tumor uptake to 9.2 ( 1.2% ID g-1. The second recombinant antibody fragment, the [111In]T84.66 minibody (80 kDa), displayed rapid tumor targeting of 14.2 ( 6.1% ID g-1 at 2 h, and reached a maximum activity of 24.5 ( 6.1% ID g-1 by 12 h. Renal uptake achieved a plateau of 12-13% ID g-1 which cleared to 7.2% ID g-1 at 72 h. However, hepatic uptake was elevated and reached a maximum of 26.0 ( 1.0% ID g-1 at 12 h in these xenograft-bearing mice. Experiments in nontumor bearing mice showed a reduction of hepatic activity at 12 h to 16.6 ( 1.5% ID g-1, indicative of an intrinsic hepatic accumulation of the [111In]DOTA-T84.66 minibody or metabolites. While the anti-CEA [111In]DOTAT84.66 diabody and T84.66 minibody retain the rapid tumor targeting properties of the radioiodinated form, the normal organ accumulation (kidneys and liver, respectively) of the [111In]DOTA forms appeared problematic for RIS and RIT applications. Development of alternative blocking strategies or new metabolizable chelates are under investigation to enhance the utility of the radiometal form of these and other promising recombinant antibody fragments.

INTRODUCTION

(Mabs)1

Radiolabeled monoclonal antibodies have been under continued development for the detection and imaging of primary and metastatic disease and have been extensively reviewed (1-3). However, intact Mabs have not proven to be ideal imaging agents due to their slow tumor targeting and extended blood clearance, the inability to uniformly penetrate large tumors, and the * To whom correspondence should be addressed. Phone: (626) 357-9711ext.64035.Fax: (626)301-8280.E-mail: [email protected]. † Department of Molecular Biology. ‡ Division of Immunology of the Beckman Research Institute of the City of Hope. § Division of Radiology. | Division of Biostatistics. ⊥ Division of Radiation Oncology. # Division of Radioimmunotherapy of the City of Hope National Medical Center.

elicitation of anti-murine or anti-chimeric human antibody responses (4). Improvements to the pharmacokinetics have come from the generation of antibody fragments by in vitro enzymatic digestion or the more easily produced single-chain Fv recombinant fragments (5). These antibody derived fragments have shown rapid tumor targeting and swift blood clearance. However, initial work has demonstrated that these truncated molecules may still require additional development (4, 6, 7). 1 Abbreviations: Mab, monoclonal antibody; RIS, radioimmunoscintigraphy; scFv, single-chain Fv protein; DOTA, 1,4,7,10-tetraazacyclododecane N,N′,N′′,N′′′-tetraacetic acid; CEA, carcinoembryonic antigen; DTPA, diethylenetriaminepentaacetic acid; HSA, human serum albumin; ANOVA, two way analysis of variance; CT computed tomography; pI, isoelectric point; microPET, micro positron emission tomography; RIT, radioimmunotherapy.

10.1021/bc000092h CCC: $20.00 © 2001 American Chemical Society Published on Web 03/01/2001

Tumor Targeting of T84.66 Diabody and T84.66 Minibody

Bioconjugate Chem., Vol. 12, No. 2, 2001 221

Figure 1. Schematic drawing shows the domain structure of the parental intact T84.66 antibody (left), T84.66 diabody (center), and the T84.66 minibody (right). In mammalian cell culture, the T84.66 diabody (scFv) is secreted as noncovalent cross-paired dimer (55 kDa), while the T84.66 minibody (scFv-CH3) self-assembles into a covalent dimer (80 kDa). Table 1. List of Potential Radionuclides for Radioimmunoscintigraphya radionuclide 123I 131I 99mTc 64Cu 67Cu 105Rh 111In 117mSn 153Sm 177Lu 186Re 188Re

half-life 13.2 h 8.1 d 6h 12.7 h 61 h 36 h 2.8 d 13.6 d 1.9 d 6.8 d 3.8 d 17 h

Eβ max (MeV)

Eγ (MeV)

0.13 0.61 0.12 0.57 (β-) 0.66 (β+) 0.57 0.96 0.24

0.159 (97%) 0.365 (81%) 0.140 (90%) 0.510 (38%)

0.13 0.16 0.81 0.50 1.07 2.12

0.180 (40%) 0.310 (24%) 0.174 (89%) 0.250 (94%) 0.158 (87%) 0.103 (29%) 0.21 (6%) 0.137 (9%) 0.16 (10%)

a A table of potential radionuclides for imaging based on halflife and emission spectra. Those above the horizontal break are negative ions; those below are positive ions. All β emissions are for electrons except in the case of 64Cu which emits both electrons and positrons. Percentages listed are the probability of a photoemission.

The development of antibody fragments for radioimmunoscintigraphy (RIS) can be categorized in terms of selection of radionuclide, metal chelate, production of sufficient quantities of protein, and reduction of the nontargeted uptake in normal tissues. Radiolabeling with 131I is the most commonly used radionuclide for RIS but its usage in cancer imaging is restricted by a high energy photon (360 keV), a long physical half-life (T1/2 ) 8 days), high energy beta emission and a significant loss of signal due to dehalogenation (3). Use of 123I has been a second possible isotope, but its relatively high cost, availability and short (13.2 h) half-life may restrict its clinical usage. 99mTc has been used for RIS, but with an even shorter half-life of 6 h, there is a substantial reduction of the amount of radioactivity deposited to the tumor tissue by the time of imaging (1). To overcome some of these limitations, a number of radiometals may be considered for RIS such as 64Cu, 67Cu, 105Rh, 111In, 111mSn, 153Sm, 177Lu, 186Re, and 188Re. These radionuclides are listed in Table 1, along with their half-lives and emission details. Developing new recombinant antibodies for RIS, two genetically engineered antibody fragments from the highly specific, anti-CEA murine antibody T84.66 have been previously expressed (Figure 1). The T84.66 diabody (55 kDa) is a single chain variable region (scFv) dimer

which when radioiodinated exhibited rapid tumor targeting and blood clearance in tumor bearing xenografted mice (8). The second construct, the T84.66 minibody, is a scFv-CH3 fusion protein that self-assembles into a covalent 80 kDa dimer, and when radioiodinated exhibited higher tumor targeting with a longer biological halflife in xenograft-bearing mice (9). To investigate the radiometal-labeled utility of these recombinant antibody fragments, the antibodies were conjugated with the macrocyclic chelating agent, 1,4,7,10-tetraazacyclododecane N,N′,N′′,N′′′-tetraacetic acid (DOTA) (10). This allowed labeling with 111In for biodistribution studies in tumor-bearing xenografted mice. To determine the merits and limitations of the [111In]form, a statistical analysis was made comparing the tumor targeting and biodistributions properties between the 111In and previous radioiodinated forms of these two promising antibody fragments. MATERIALS AND METHODS

Antibody Development. The T84.66 diabody previously expressed in Escherichia coli and characterized in radioiodinated biodistribution studies (8), has now been produced in one hundred milligram quantities from mammalian cell culture (Yazaki et al., manuscript submitted). The T84.66/212 Flex minibody (scFv-CH3) was expressed in Sp2/0 mammalian cells, radioiodinated, and characterized in animal biodistribution studies (9). A second minibody version, the T84.66/GS18 Flex minibody, was constructed from the T84.66/212 Flex minibody DNA template using the polymerase chain reaction/splice overlap extension as previously described (9). With an 18 amino acid linker (GSTSGGGSGGGSGGGGSS in single letter code), the GS18 Flex minibody was expressed in NS0 murine myeloma cells (Yazaki et al., manuscript submitted) (11). The expressed proteins were purified by hydrophobic interaction chromatography (Source ISO, Amersham Pharmacia Biotech, Piscataway, NJ) followed by anion-exchange chromatography (HQ50, PE Biosystems, Foster City, CA) using a BioCAD 700E chromatography system (PE Biosystems, Foster City, CA) (Yazaki et al., manuscript submitted). Purity was determined by electrophoresis on 10% acrylamide Ready Gels (Bio-Rad Laboratories, Hercules, CA) under nonreducing conditions on SDS-PAGE (12) with Kaleidoscope protein standards (Bio-Rad Laboratories, Hercules, CA) and staining with Coomassie Blue R-250. CEA-binding activity was determined by ELISA using microtiter plates

222 Bioconjugate Chem., Vol. 12, No. 2, 2001

coated with a recombinant N-A3 protein (N and A3 domains of CEA) (13) and an anti-human Fc alkaline phosphatase antibody for detection (Jackson Immunoresearch, West Grove, PA). Chimeric T84.66 (cT84.66) antibody was used as the standard (14). Conjugation and Radiolabeling. For radiometal studies, purified T84.66/GS18 Flex minibody and the T84.66 diabody were conjugated to DOTA using the water-soluble N-hydroxysuccinimide method (10). In a typical experiment, 400 µg of protein was reacted with a 1000:1 molar ratio of DOTA to protein for 18-24 h at 4 °C at pH 7.0. Following conjugation, the protein was dialyzed extensively in 0.2 M NH4OAc, pH 7.2, and concentrated to greater than 5 mg/mL. The DOTA: antibody conjugation ratio was determined by incubation of the DOTA-conjugated antibody with a labeling solution containing known quantities of 111In (Mallinckrodt, Hazelwood, MO) and cold indium (Aldrich, Milwaukee, WI). 111In labeling was analyzed by thin-layer chromatography using Monoclonal Antibody ITLC strips (Biodex Medical Systems, Shirley, NY). Percent bound indium was determined, and the DOTA:antibody ratio calculated. DOTA:antibody ratios ranged from 1:1 to 6:1 chelate per minibody while for the diabody the ratio was 0.5:1. DOTA-minibody or DOTA-diabody (400 µg of protein) were incubated with 1.3 mCi of pure [111In]chloride (Mallinckrodt, Hazelwood, MO) in 0.25 M NH4OAc pH 5.0 for 1 h at 43 °C. The reaction was stopped by the addition of DTPA to a final concentration of 1 mM. Conjugated monomeric proteins were purified by HPLC size-exclusion chromatography using a TSK-G2000SW column (30 cm × 7.5 mm ID; TosoHaas, Montgomeryville, PA) for the minibody and a Superdex 75 HR 10/30 column (Amersham Pharmacia Biotech, Piscataway, NJ) for the diabody. Radioiodination of the T84.66 minibody was by the Iodogen method as previously described (15). Radiolabeling efficiency was determined by integrating areas on the size-exclusion HPLC trace and determining the percentage of radioactivity associated in the 80 kDa peak for the minibody and the 55 kDa peak for the diabody. Immunoreactivity was determined by incubation of the labeled protein with a 20-fold excess (w/w) of either purified CEA or the recombinant N-A3 fragment of CEA (13) in 0.15 mL of PBS/1% HSA. Analysis by sizeexclusion HPLC was on tandem Superose 6 HR 10/30 columns (Amersham Pharmacia Biotech, Piscataway, NJ) to assess the formation of antibody:antigen complexes. Animal Studies. Groups of 7-8-week old female athymic mice (Charles River Laboratories, Wilmington, MA) were injected subcutaneously in the flank region with 106 LS174T human colon carcinoma cells obtained from ATCC (Manassas, VA). After 10 days, when tumor masses were in the range of 100-300 mg, 1-3 µCi of [111In]DOTA-antibodies/animal (2-3 µg of protein) were injected into the tail vein. Time points of analysis for the T84.66 diabody were 0, 2, 4, 6, 12, and 24 h while for the T84.66 minibody 0, 2, 6, 12, 24, 48, and 72 h were selected. For individual experiments, groups of five to six mice at the selected time points were euthanized, necropsy performed and organs weighed and counted for radioactivity. Data from several individual experiments of the same construct and radionuclide were compared and when there was not a significant statistical difference (p > 0.01), the data were combined. This resulted in pooled data sets of up to 16 mice/time point. All data are mean values and have been corrected for radiodecay back to the time of injection, allowing organ uptake to be reported as percent of the injected dose per gram (% ID/

Yazaki et al.

Figure 2. Radiochromatograms of size-exclusion HPLC analysis of [111In]DOTA-T84.66 diabody (A) and [111In]DOTA-T84.66 minibody (B). Lower trace was the starting sample used for evaluation of immunoreactivity. Upper trace, following incubation with CEA or N-A3 (a CEA mimetic) the bulk of the [111In]DOTA-T84.66 diabody and [111In]DOTA-T84.66 minibody are found in antibody:antigen complexes.

g-1) with standard errors. Blood curves were calculated using the ADAPT II software (16). D-Lysine renal-blocking experiments were carried out according to the protocol of Behr et al. (17). Briefly, D-lysine (Sigma, St. Louis, MO) was dissolved in PBS at a concentration of 200 mg/mL and 100 mg was administered by an oral intubation catheter, 30 min prior to the injection of the [111In]T84.66 diabody. Biostatistical Analysis. To compare changes in percent of the injected dose per gram over time among antibody constructs and radionuclides, two-way analysis of variance (ANOVA) was performed (16). Time (h), construct or radionuclide, and the interaction between these two factors were included in the statistical model. Dependent variables compared using this model included the percent of the injected dose per gram for organs blood, liver, spleen, kidney, lung, and intact bone as well as tumor and carcasses. Tumor-to-blood and tumor-to-liver ratios as well as tumor masses were recorded. To compare differences in mean percent of the injected dose per gram between antibody constructs at specific time points, the independent t-test was used. All significance testing was done at the 0.01 level, using the SAS/STAT software (SAS Inc., Cary NC).



Table 2. Biodistribution of [111In]T84.66 Diabody in Athymic Mice Bearing LS174T Xenograftsa time (h) organ tumor blood liver spleen kidney lung bone carcass ratios tumor/blood tumor/liver tumor tumor mass (g)

0

2

4

6

12

24

1.94 (0.53) 49.50 (4.42) 7.92 (0.70) 6.97 (0.57) 26.34 (10.74) 12.48 (0.85) 3.27 (0.70) 2.46 (0.06)

12.53 (1.00) 2.70 (0.38) 4.59 (0.59) 1.32 (0.09) 134.57 (19.29) 2.08 (0.30) 1.51 (0.47) 1.75 (0.18)

12.12 (1.60) 0.64 (0.14) 4.51 (0.78) 1.41 (0.40) 162.29 (17.66) 1.23 (0.18) 0.95 (0.20) 1.40 (0.14)

13.30 (2.06) 0.34 (0.06) 5.45 (1.61) 1.30 (0.40) 183.52 (20.97) 1.07 (0.13) 1.18 (0.35) 1.37 (0.14)

12.28 (3.09) 0.11 (0.03) 3.61 (0.45) 1.38 (0.35) 145.66 (16.43) 0.87 (0.22) 1.11 (0.32) 1.03 (0.22)

7.73 (1.01) 0.04 (0.01) 3.55 (0.51) 1.08 (0.30) 125.76 (11.63) 0.51 (0.06) 0.56 (0.09) 0.88 (0.12)

0.04 (0.00) 0.24 (0.03)

4.7 (0.4) 2.8 (0.2)

19.4 (1.6) 2.8 (0.3)

39.5 (3.4) 2.6 (0.4)

112 (11.2) 3.5 (0.6)

196 (11.1) 2.2 (0.1)

0.49 (0.13)

0.20 (0.04)

0.15 (0.04)

0.30 (0.10)

0.27 (0.08)

0.32 (0.10)

a [111In]T84.66 diabody was injected into xenograft-bearing mice and tumor targeting and biodistribution studies were carried out. Groups of five mice were analyzed at each time point. Tumor and normal organ uptake are expressed as percent injected dose per gram (% ID/g). Standard errors of the mean (SE) are given in parentheses. The ratios presented are the averages of the tumor/blood and tumor/liver ratios for the individual mice. Tumor masses are in grams.

RESULTS

[111In]DOTA-T84.66 Diabody. The T84.66 diabody was conjugated with the macrocycle DOTA to allow labeling of the protein with a variety of radiometals. The chelate-to-diabody ratio was 0.5:1 and when radiolabeled resulted in 67-100% incorporation of 111In into the T84.66 diabody. Size-exclusion chromatography displayed a single radiolabeled peak with a specific activity that ranged from 1.4 to 3.2 µCi/µg. The radiolabeled diabody displayed high immunoreactivity to N-A3 (CEA mimetic) with over 95% immunoreactivity (Figure 2A). For biodistribution studies, the tissues (tumor, blood, liver, spleen, kidney, lung, intact bone, and carcass) from groups of five mice per time point (0, 2, 4, 6, 12, and 24 h) were counted for radiolabel uptake and are shown in Table 2. The [111In]T84.66 diabody exhibited very rapid tumor uptake in the CEA expressing LS 174T xenografts. There was 12.5 ( 0.4% ID g-1 at 2 h and the deposition of the radiolabel was persistent with 12.3 ( 1.4% ID g-1 at 12 h. The diabody was found to clear quickly from the blood circulation with only 2.7 ( 0.2% ID g-1 remaining at 2 h with a terminal phase T1/2β of 3.04 h (Table 3). Excellent tumor-to-blood ratios were observed with 39.5:1 at 6 h rising to 196:1 at 24 h. The initial hepatic accumulation at T0 was low measuring 7.9 ( 0.3% ID g-1 and decreased to 3.6 ( 0.2% ID g-1 by 24 h. Other normal tissues showed low accumulation except for the kidney where a high level of accumulation was observed. This renal uptake reached a maximum of 183.5 ( 21.0% ID g-1 at 6 h with slow clearance during the 24 h time course of the experiment. Preadministration of cationic amino acids has been shown to block the renal proximal tubular cell readsorption of mAbs and antibody fragments without an effect on tumor uptake (18). Comparisons of the kidney and tumor uptake of the [111In]DOTA-T84.666 diabody with and without the administration of D-lysine are shown in Figure 3. Oral administration of D-lysine resulted in a

Table 3. Comparison of Parameters of the Two-Exponential Fit of the Radiolabeled Forms of the T84.66 Diabody and T84.66 Minibody Blood Dataa radiolabel and construct

A1

A2

k1

k2

[123I]T84.66 diabody (8) [111In]T84.66 diabody [131I]T84.66 minibody [111In]T84.66 minibody

33.0 1.3 48.2 0.3 28.3 2.0 23.9 0.9

6.4 1.2 1.3 0.3 14.6 2.0 22.9 0.9

2.7 0.2 1.6 0.03 0.58 0.1 1.6 0.3

0.2 0.1 0.2 0.1 0.1 0.01 0.15 0.01

T1/2R (h)

T1/2β (h)

0.25

2.89

0.43

3.04

1.19

6.99

0.42

4.51

a The T84.66 minibody and the T84.66 diabody were labeled with radioiodine or 111In. Current and previous (8) biodistribution studies were carried out in xenografted mice. Animal blood curves were calculated using the ADAPT II software (16) and the standard deviations are shown below the results. A1 and A2 are the amplitudes of the faster and slower clearance components. T1/2 are related to the inverse of k1 and k2 by the relationship: T1/2 ) 0.693/k.

59.2% reduction at the 6 h time point where renal uptake reached the maximum (183.5 ( 21.0 vs 75.0 ( 7.1% ID g-1 ). However at 6 h, there was a 31% reduction of tumor uptake, lowering the tumor activity to 9.2 ( 1.2% ID g-1. [111In]DOTA-T84.66 Minibody. As was performed for the diabody, the T84.66 minibody was conjugated with DOTA for radiolabeling with 111In. In three conjugations, the chelate-to-minibody ratio ranged from 1:1 to 6:1 and following radiolabeling resulted in 24-85% incorporation of 111In. Size-exclusion chromatography displayed a single radiolabeled peak with a specific activity that ranged from 2.3 to 4.7 µCi/µg. The [111In]DOTA-T84.66 minibody retained high immunoreactivity to CEA with a range of 70 to 100% immunoreactivity (Figure 2B). Two experimental studies of the biodistributions of the [111In]T84.66/ GS18 Flex minibody in xenograft-bearing mice were carried out with groups of five mice per time point. All tissues (see above) were counted for radionuclide uptake


Yazaki et al.

Table 4. Biodistribution of [111In]T84.66 Minibody in Athymic Mice Bearing LS174T Xenograftsa time (h) organ tumor blood liver spleen kidney lung bone carcass ratios tumor/blood tumor/liver tumor tumor mass (g)

0

2

6

12

24

48

72

1.70 (0.17) 46.69 (1.51) 7.64 (0.46) 6.18 (0.28) 10.96 (0.44) 14.92 (1.27) 3.13 (0.14) 2.08 (0.10)

14.19 (0.98) 17.74 (0.79) 17.82 (0.77) 5.52 (0.24) 13.26 (0.90) 6.88 (0.43) 3.48 (0.17) 2.70 (0.12)

20.67 (1.47) 9.14 (0.68) 20.73 (1.00) 5.81 (0.39) 12.47 (0.62) 5.11 (0.60) 3.87 (0.19) 2.84 (0.09)

24.47 (6.10) 3.61 (0.59) 18.95 (0.74) 6.05 (0.74) 11.58 (0.93) 3.11 (0.37) 3.88 (0.31) 2.60 (0.07)

18.84 (1.17) 0.60 (0.04) 25.97 (0.99) 6.05 (0.29) 12.48 (0.69) 2.06 (0.12) 3.50 (0.35) 2.37 (0.11)

11.33 (0.95) 0.17 (0.03) 20.22 (1.04) 5.34 (0.43) 7.59 (0.85) 1.34 (0.06) 2.72 (0.14) 1.75 (0.05)

8.33 (0.90) 0.08 (0.01) 17.21 (1.22) 3.89 (0.26) 7.16 (0.57) 1.09 (0.08) 2.33 (0.10) 1.58 (0.05)

0.04 (0.00) 0.23 (0.03)

0.81 (0.06) 0.81 (0.06)

2.30 (0.13) 1.02 (0.09)

7.19 (1.24) 1.36 (0.40)

32.17 (2.06) 0.73 (0.05)

109.48 (25.70) 0.57 (0.04)

140.38 (32.05) 0.49 (0.06)

0.15 (0.03)

0.22 (0.05)

0.16 (0.02)

0.16 (0.02)

0.21 (0.03)

0.36 (0.06)

0.36 (0.06)

a [111In]T84.66 minibody was injected into xenograft-bearing mice and tumor targeting and biodistribution studies were carried out. Groups of 10 mice were analyzed at each time point. Tumor and normal organ uptake are expressed as percent injected dose per gram (% ID/g). Standard errors of the mean (SE) are given in parentheses. The ratios presented are the averages of the tumor/blood and tumor/liver ratios for the individual mice. Tumor masses are in grams.

Figure 3. Comparison of renal uptake of the [123I]T84.66 diabody and the [111In]T84.66 diabody, with and without preadministration of D-lysine. The [123I]T84.66 diabody and [111In]T84.66 diabody with or without preadministration of D-lysine were injected into xenograft-bearing mice. Groups of five or more mice were sacrificed at each time point, and the tissues counted for radioactivity. Activity was expressed as percent injected dose per gram (% ID/g).

at 0, 2, 6, 12, 24, and 48 h. Statistical analysis showed no significant difference between the respective tissues from both studies and the 111In biodistributions were combined (Table 4). Tumor uptake of the [111In]T84.66 minibody achieved 14.2 ( 6.1% ID g-1 at 2 h and reached a maximum of 24.5 ( 6.1% ID g-1 at 12 h. Activity remained at the tumor site with 11.3 ( 0.9% ID g-1 at 48 h but decreased to 8.3 ( 0.1% ID g-1 by 72 h. The [111In]T84.66 minibody was cleared quickly from the circulation with a T1/2β of 4.51 h. It was found that the renal uptake was low, reaching a plateau of 12-13% ID g-1 over the first 24 h and clearing to 7.2% by 72 h. Intact bone activity reached a level of 3.9% ID g-1 during the first 24 h but was reduced to 2.3% ID g-1 by 72 h. The liver showed the highest normal organ accumulation,

reaching 26.0 (1.0% ID g-1 at 24 h with slow clearance to 17.2 ( 1.2% ID g-1 at 72 h. High hepatic uptake had been observed for the intact [111In]chimeric T84.66 (cT84.66) antibody in xenograft bearing mice (19). This accumulation had been determined to be the result of trafficking of CEA-antibody immune complexes into the hepatocytes (20). To investigate whether similar uptake of the antibody-antigen complex was occurring for the [111In]T84.66 minibody, biodistribution studies were also carried out in normal athymic mice. The only tissues that showed a significant difference compared to the tumor bearing animals were the liver and the spleen. The corresponding activity curves are shown in Figure 4, along with the activity curve of the radioiodinated form in tumor bearing mice. While the [111In]T84.66 minibody showed reduced hepatic uptake in the normal mice, the hepatic activity remained elevated in particular when compared to the radioiodinated form. For the spleen, normal athymic mice had a higher accumulation of the 111In form than the tumor bearing mice but again the radioiodinated form was much lower. No statistical significant difference was observed for blood (p ) 0.83) or kidney distributions (p ) 0.94). Indium vs Iodine Analysis. To understand the biodistribution properties of radiometal labeled fragments, the 111In data obtained was compared to the radioiodinated biodistribution data. A two way statistical comparison was made between the [111In]T84.66 diabody data given here and previously published [123I]T84.66 diabody data (8). Uptake in the tumor of the two radionuclides reached comparable levels within the first 4 h but after 6 h a higher level of activity remain for 111In labeled material (p < 0.0007). Initial blood uptake for the indium label was marginally greater (49.5 ( 4.4% ID g-1 vs 41 ( 2.8% ID g-1) with the overall blood clearance slightly slower for the indium labeled material. Calculations of the effective half-lives were made and are shown in Table 3. The 111In form showed a slower alpha



kidney (9.9 ( 0.3% ID g-1), liver (7.2 ( 0.3% ID g-1), and spleen (6.2 ( 0.3% ID g-1), and all were below 1% ID g-1 by 24 h. In an identical manner as to the T84.66 diabody, a comparison was made between the 111In and the radioiodinated forms of the T84.66 minibody. Activity in the tumor (p ) 0.18) and blood (p ) 0.11) showed no significant difference. There was significantly faster clearance of radioiodinated form from the liver and kidneys compared to the 111In form (p < 0.0001). DISCUSSION

Figure 4. Hepatic and splenic uptake of the [111In]T84.66 minibody in xenograft-bearing and normal mice and the radioiodinated T84.66 minibody in xenograft-bearing mice. [111In]T84.66 minibody and [123I] and [125I]T84.66 minibody were injected into tumor-bearing mice as well as the [111In]T84.66 minibody into normal athymic mice. Groups of five or more mice were sacrificed at each time point. Biodistribution of liver uptake (A) and spleen uptake (B) are shown.

phase but comparable beta phase. Hepatic uptake was significantly higher for the 111In-diabody with a slower clearance pattern compared to the radioiodinated form (p < 0.0001). Renal uptake was substantially higher for the indium form over all but the initial time point (p < 0.0001). Blocking of renal absorption by D-lysine substantially lowered the [111In]T84.66 diabody accumulation but the level of renal uptake was still highly elevated over the levels observed for the radioiodinated diabody (8). Two versions of the T84.66 minibody, the GS18 Flex and the 212 Flex were expressed differing only in the length of the scFv linker. Both versions exhibited identical in vitro biochemical properties and antigen binding activities (Yazaki et al., manuscript submitted) and when radioiodinated displayed similar biodistributions in xenograft-bearing animals (7). The data from these two studies plus a third minibody experiment were pooled, resulting in a data set of 16 mice/time point (Table 5). Tumor uptake was 14.1 ( 1.5% ID g-1 at 2 h, and reached a maximum of 26.2 ( 4.6% ID g-1 at 12 h. Blood clearance resulted in a T1/2β of 6.99 h (Table 3). The radioiodinated T84.66 minibody exhibited some initial uptake or blood pool activity into normal tissue, but there was rapid clearance of the radiolabel. Levels immediately after injection were the following: lung (13.6 ( 1.1% ID g-1),

Radiolabeled antibody fragments offer potential advantages over intact Mabs for the imaging of solid tumors. Due to their smaller molecular size, antibody fragments have exhibited rapid tumor targeting and blood clearance, increased tumor penetration and a lower potential of eliciting an immune response (4). Clinical studies with [99mTc]Fab′ (21) and [123I]scFv (22) antibody fragments have shown promise as cancer imaging agents, particularly in complement with CT scanning. Previously, we have expressed two anti-CEA recombinant antibody fragments, the T84.66 diabody (8) and the T84.66 minibody (9), which when radioiodinated, displayed rapid tumor targeting and rapid blood clearance in xenograft-bearing mice. To extend the clinical utility of the T84.66 diabody and T84.66 minibody, labeling with one of the radiometals available (Table 1) can arguably offer many potential advantages over the radioiodinated form. These include the lack of dehalogenation, a more efficient emission spectrum for γ camera imaging, and a potentially more suitable physical halflife. In addition, radiometals compared to 131I are less injurious to the patient due to a lack of 600 keV β particles produced in the decay scheme (3). In this present work, the biodistribution properties of the [111In]DOTA-T84.66 diabody displayed high level rapid tumor targeting of 12.5 ( 0.4% ID g-1 at 2 h and a rapid clearance from the circulation (T1/2β ) 3.04 h). This resulted in high tumor-to-blood ratios which reached 196:1 at 24 h compared to the 4.6:1 ratio observed for the parental cT84.66 at 24 h (23). Unfortunately, high renal uptake was observed for the [111In]T84.66 diabody as has been documented for other radiometal labeled antibody fragments (4, 6, 24). To overcome the renal accumulation problem, the T84.66 minibody was designed to have a molecular weight that exceeded the renal threshold (∼60 kDa) (18). Two versions of the minibody were expressed, the 212 Flex and GS18 Flex, which differed in the length of the scFv peptide linker (14 vs 18 amino acids, respectively). When radioiodinated both minibody versions showed comparable animal biodistributions in xenograft-bearing mice (7) as well as identical biochemical and affinity characteristics (Yazaki et al., manuscript submitted). For all subsequent studies, the T84.66/GS18 Flex minibody was used due to the ability to produce grams of crude material from mammalian cell culture production. The [111In]T84.66 minibody displayed rapid tumor targeting of 24.5 ( 6.1% ID g-1 at 12 h with considerable activity remaining at 72 h (8.3 ( 0.9% ID g-1) when the tumorto-blood ratio reached 140:1. While the 111In form had higher tumor uptake at 24 and 48 h compared to the radioiodinated form, overall the curves were not statistically different. Where a difference was noted between the radionuclide forms was in the normal tissue uptake. The hepatic uptake of the 111In form was comparable to the level of tumor activity. This hepatic accumulation may


Yazaki et al.

Table 5. Biodistribution of Radioiodinated T84.66 Minibody in Athymic Mice Bearing LS174T Xenograftsa time (h) organ tumor blood liver spleen kidney lung bone carcass ratios tumor/blood tumor/liver tumor tumor mass (g)

0

2

6

12

24

48

72

2.35 (0.83) 42.83 (0.84) 7.15 (0.25) 6.20 (0.34) 9.91 (0.32) 13.63 (1.06) 7.01 (3.77) 1.95 (0.16)

14.05 (1.49) 20.82 (1.11) 4.62 (0.19) 4.34 (0.23) 6.55 (0.27) 8.85 (0.43) 2.99 (0.24) 2.94 (0.09)

18.21 (1.42) 8.68 (0.55) 2.25 (0.13) 2.05 (0.10) 2.75 (0.14) 4.02 (0.24) 1.35 (0.11) 1.79 (0.07)

26.23 (4.57) 4.83 (0.31) 1.18 (0.07) 1.22 (0.08) 1.68 (0.08) 2.77 (0.21) 0.84 (0.07) 1.20 (0.09)

12.58 (1.77) 1.00 (0.17) 0.45 (0.03) 0.38 (0.03) 0.44 (0.04) 0.61 (0.07) 0.48 (0.12) 0.58 (0.03)

5.64 (0.85) 0.13 (0.01) 0.16 (0.01) 0.09 (0.01) 0.12 (0.01) 0.12 (0.01) 0.12 (0.07) 0.29 (0.02)

6.82 (1.67) 0.09 (0.02) 0.18 (0.02) 0.09 (0.01) 0.15 (0.01) 0.11 (0.02) 0.27 (0.15) 0.48 (0.08)

0.05 (0.02) 0.34 (0.13)

0.69 (0.07) 3.05 (0.30)

2.12 (0.12) 8.47 (0.79)

5.50 (0.78) 22.20 (3.12)

16.08 (2.13) 30.28 (4.82)

53.30 (7.28) 40.40 (7.52)

73.49 (9.09) 34.83 (7.48)

0.22 (0.48)

0.22 (0.52)

0.23 (0.04)

0.21 (0.06)

0.36 (0.08)

0.39 (0.08)

0.52 (0.08)

a Two versions of the T84.66 minibody were labeled with 131I and 125I, and tumor targeting and biodistribution studies were carried out in xenograft-bearing mice. Groups of 16 mice were analyzed at each time point except at 72 h where the group size was 11 mice. Tumor and normal organ uptake are expressed as percent injected dose per gram (% ID/g). Standard errors of the mean (SE) are given in parentheses. The ratios presented are the averages of the tumor/blood and tumor/liver ratios for the individual mice. Tumor masses are in grams.

be a result of circulating CEA which can complex with anti-CEA antibodies resulting in uptake by the hepatic immunoglobulin receptor system (20). Hepatic uptake with the intact [111In]cT84.66 antibody in mice had been observed to reach 9.8% ID g-1 at 96 h as the result of trafficking of CEA-antibody immune complexes into the hepatocytes (19). However, with only a slight reduction of hepatic accumulation noted in normal mice as compared to tumor-bearing mice, the possibility that a circulating CEA-minibody complex was the cause of the hepatic activity was excluded. Thus, the accumulation of the majority of the hepatic activity must be presumed to be due to a protein-radiometal-chelate effect of the [111In]T84.66 minibody and/or 111In labeled metabolites. It should be noted that similar tumor and hepatic uptake was observed for [64Cu]DOTA-T84.66 minibody by microPET imaging (25). However, hepatic accumulation was not observed with the radioiodinated minibody, but because of dehalogenation of the radiolabel, we do not know if the protein or metabolite was present. Several approaches have been reported to lower the uptake of the radiolabeled antibody fragments in normal organs (4, 6, 24). To lower the renal uptake these efforts have focused on the following areas: (1) charge modification of the protein by lowering the pI, (2) blocking of the renal reabsorption by administration of cationic amino acids, and (3) use of metabolizable linkers. Since the pI of the T84.66 diabody was predicted by cDNA analysis to be acidic (pI 4.9) and verified by isoelectric focusing gels (data not shown), we attempted to lower the renal uptake by blocking the renal proximal tubular cell readsorption process. Behr et al. (17) had reported the administration of the nonmetabolizable amino acid, Dlysine, before injection of a [90Y]Fab in xenografted mice resulted in a 5-fold reduction in renal uptake with no effect on tumor uptake. For the [111In]T84.66 diabody, this resulted in a 2.4-fold reduction of the kidney uptake, however this was accompanied by a 1.4-fold reduction of

the tumor uptake. This is the first report of a reduction of tumor activity with D-lysine using an immunoconjugate, although it had been noted for L-lysine in 111In- and 161Tb-labeled octreotide in neuroendocrine tumors (18). This conflicting result could be due to a number of factors, the main difference being the antibody and tumor system. While this lowered tumor activity (9.2 ( 1.2% ID g-1 at 6 h) still provided sufficient deposition of radionuclide for tumor imaging, a further reduction of the renal accumulation levels must be achieved. Alternative administration schedules of cationic amino acids in mice have been reported using multiple administrations (26). This high lysine dosage when translated to human studies would appear limited due to the induction of proteinuria by blocking renal tubular reabsorption (27). One clinical study was carried out in which five patients were infused with 2 L of a commercially available nutritive solution, which contained 2.25 g/L lysineglutamate, 2.5 g/L arginine, salts and xylitol, before injection of a [99mTc]anti-CEA Fab′ (28). In this small group of patients, this much lower dose had a 38% reduction of renal activity compared to the control group who was infused with saline. Another commercially available solution is Aminosyn, which showed a 78.5% lower renal activity of a [18F] fluoromethylbenzoyl disulfide stabilized Fv in baboons (29). In the case of the T84.66 minibody, while the renal levels appear acceptable, the elevated hepatic activity limits the use of the T84.66 minibody for RIS of CEA positive hepatic disease and subsequent radioimmunotherapy (RIT). However, this hepatic uptake should not preclude the scintigraphy of CEA-positive lesions in the colon, rectum, breast, bone, lung and other organs. Systematic mutagenesis could produce modified minibodies with further optimized pharmacokinetics. Alternatively, the largest advances for the use of radiometal labeled antibody fragments may come from the development of metabolizable chelate chemistry.


Intervention at two points in the metabolic process have been targeted for cleavage of the radiometal/chelate complex to be excreted from the body. The first approach would rely on the cleavage of a peptide linker before cellular uptake by the catabolic enzymes that are found at the lumen of the brush border region of the renal proximal tubule cells (30) and speculatively at the surface of hepatic cells. The second approach focuses on the design of chelates that will be cleaved by lysosomal proteases to liberate radiometabolites which would be exported from the cell and rapidly excreted in the urine (24). New chelate peptide linkers and administration schemes are currently under development in our laboratory and elsewhere. Such metabolizable chelates alone, or in combination with a clinically acceptable dose of a blocking solution, may reduce the normal organ uptake of the radiometal labeled antibody fragments to acceptable levels for highly sensitive RIS and RIT in all tissues. ACKNOWLEDGMENT

This work was supported by the NIH Program Project Grant P01 43904 and NIH Cancer Center Grant 33572. LITERATURE CITED (1) Moffat, F. L., Jr., Gulec, S. A., Serafine, A. N., Skakianakis, G. N., Pop, R., Robinson, S., Franceschi, D., Boggs, J., and Livingstone, A. S., (1999) A thousand points of light or just dim bulbs? radiolabeled antibodies and colorectal cancer imaging. Cancer Invest. 17 (5), 322-334. (2) Divgi, C. R. (1996) Status of radiolabeled monoclonal antibodies for diagnosis and therapy of cancer. Oncology 10 (6), 939-953. (3) Halpern, S. E. (1992) An overview of radioimmunoimaging. Targeted Diagn Ther. 6. 1-22. (4) Colcher, D., Goel, A., Pavlinkova, G., Beresford, G., Booth, B., Batra, S. K. (1999) Effects of genetic enegineering on the pharmacokinetics of antibodies. Q. J. Nucl. Med. 43, 132139. (5) Huston, J. S., George, A. J. T., Adams, G. P., Stafford, W. F., Jamar, F., Tai, M-s., McCartney, J. E., Oppermann, H., Heelan, B. T., Peters, A. M., Houston, L. L., Bookman, M. A., Wolf, E. J., and Weiner, L. M. (1996) Single-chain Fv radioimmunotargeting. Q. J. Nucl. Med. 40, 320-333. (6) Colcher, D., Pavlinkova, G., Beresford, G., Booth, B. J. M., Choudhury, A. and Batra, S. K. (1998) Pharmacokinetics and biodistribution of genetically engineered antibodies. Q. J. Nucl. Med. 42, 225-241. (7) Wu, A., and Yazaki, P. (2000) Designer genes: Recombinant antibody fragments for biological imaging. Q. J. Nucl. Med. 43, 268-283. (8) Wu, A. M., Williams, L. E., Zieran, L., Padma, A., Sherman, M., Bebb, G. G., Odom-Maryon, T., Wong, J. Y. C., Shively, J. E., and Raubitschek, A. A. (1999) Anti-carcinoembryonic antigen (CEA) diabody for rapid tumor targeting and imaging. Tumor Targeting 4, 1-12. (9) Hu, S.-z., Shively, L., Raubitschek, A. A., Sherman, M., Williams, L. E., Wong, J. Y. C., Shively, J. E., and Wu, A. M. (1996) Minibody: A novel engineered anti-carcinoembryonic antigen antibody fragment (single-chain FV-CH3) which exhibits rapid high-level targeting of xenografts. Cancer Res. 56, 3055-3061. (10) Lewis, M. R., Raubitschek, A., and Shively, J. E. (1994) A facile, water-soluble method for modification of proteins with DOTA. Use of elevated temperature and optimized pH to achieve high specific activity and high chelate stability in radiolabeled immunoconjugates. Bioconjugate Chem. 5, 565576. (11) Bebbington, C. R., Renner, G., Thomson, S., King, D., Abrams, D., and Yarranton, G. T. (1992) High-level expression of a recombinant antibody from myeloma cells using a glutamine synthetase gene as an amplifiable selection marker. Bio/technology 10, 169-175.

Bioconjugate Chem., Vol. 12, No. 2, 2001 227 (12) Laemmli, U. K. (1970) Cleavage of structural protein during the assembly of the head of bacteriophage T4. Nature 227, 680-685. (13) Young, H. Y., Hefta, L. J., Yazaki, P. J., Wu, A. M. and Shively, J. E. (1998) Expression, purification and characterization of a two domain carcinoembryonic antigen minigene (N-A3) in pichia pastoris. The essential role of the N-domain. Anticancer Res. 18, 3192-3202. (14) Neumaier, M., Shively, L., Fun-Shan, C., Franz-Josef, G., Ilgen, C., Paxton, R. J., Shively, J. E., and Riggs, A. D. (1990) Cloning of the genes for T84.66, an antibody that has high specificity and affinity for carcinoembryonic antigen, and expression of chimeric human/mouse T84.66 genes in myeloma and chinese hamster ovary cells. Cancer Res. 50, 21282134. (15) Wu, A. M., Chen, W., Raubitschek, A., Williams, L. E., Neumaier, M., Fisher, R., Hu, S-z., Odom-Maryon, T., Wong, J. Y. C. and Shively, J. E. (1996) Tumor localization of antiCEA single-chain Fvs: improved targeting by noncovalent dimers. Immmunotechnology 2, 221-236. (16) D′Argenio, D. Z., and Schumitzky, A. (1979) A program package for simulation and parameter estimation in pharmacokinetic systems. Comput. Prog. Biomed. 9, 115-134. (17) Behr, T. M., Sharkey, R. M., Sgouros, G., Blumenthal, R. D., Dunn, R. M., Kolbert, K., Griffiths, G. L., Siegel, J. A., Becker, W. S., and Goldenberg, D. M. (1997) Overcoming the nephrotoxicity of radiometal-labeled immunoconjugates. Cancer 80 (12), 2591-2610. (18) Behr, T. M., Goldenberg, D. M., and Becker, W. (1998) Reducing the renal uptake of radiolabeled antibody fragments and peptides for diagnosis and therapy: present status, future prospects and limitations. Eur. J. Nuc. Med. 25, 201-212. (19) Beatty, B. G., Beatty, J. D., Williams, L. E., Paxton, R. J., Shively, J. E., and O’Connor-Tressel, M. (1989) Effect of specific antibody pretreatment on liver uptake of 111-Inlabeled anticarcinoembryonic antigen monoclonal antibody in nude mice bearing human colon cancer xenografts. Cancer Res. 49, 1587-1594. (20) Beatty, J. D., Beatty, B. G., O’Conner-Tressel, M., Do, T., and Paxton, R. J. (1990) Mechanism of tissue uptake and metabolism of radiolabeled antibody-role of antigen: antibody complex formation. Cancer Res (Suppl.) 50, 840s-845s. (21) Moffat, F. L., Pinsky, C. M., Hammershaimb, L., Petrelli, N. J., Patt, Y. Z., Whaley, F. S., Goldenberg, D. M. (1996) Clinical utility of external immunoscintigraphy with the IMMU-4 technetium-99m Fab′ antibody fragment in patients undergoing surgery for carcinoma of the colon and rectum: results of a pivotal Phase III trial. J. Clin. Oncol. 14, 22952305. (22) Begent, R. H., Verhaar, M. J., Chester, K. A., Casey, J. L., Green, A. J., Napier, M. P., Hope-Stone, L. D., Cushen, N., Keep, P. A., Johnson, C. L., Hawkins, R. E., Hilson, A. J., and Robson, L. (1996) Clinical evidence of efficient tumor targeting on single-chain Fv antibody selected from a combinatorial library. Nat. Med. 9, 979-984. (23) Williams, L. E., Primus, F. J., Wong, J. Y. C., Wu, A. M., Odom-Maryon, T. L., Johnson, D. K., Hefta, L. J. F., Shively, J. E., and Raubitschek, A. A. (1996) Biodistribution of an indium-111- or yttrium-90-labeled chimeric anti-carcinoembryonic antigen monoclonal antibody (cT84.66) following its large scale production in a bioreactor. Tumor Targeting 2, 116-124. (24) Arano, Y. (1998) Stratagies to reduce renal radioactivity levels of antibody fragments. Q. J. Nucl. Med. 42, 262-270. (25) Wu, A. M., Yazaki, P. J., Tsai, S.-w., Nguyen, K., Anderson, A.-L., McCarthy, D. W., Welch, M. J., Shively, J. E., Williams, L. E., Raubitschek, A. A., Wong, J. Y. C., Toyokuni, T., Phelps, M., and Gambhir, S. S (2001) High-resolution microPET imaging of carcinoembryonic antigen-positive xenografts using a copper-64 labeled engineered antibody fragment. Proc. Natl. Acad. Sci. U.S.A 97, 8495-8500. (26) Rutherford, R. A. D., Smith, A., Waibel, R., and Schubiger, P. A. (1997) Differential inhibitory effect of l-lysine on renal accumulation of 67Cu-labeled F(ab)2 fragments in mice. Int. J. Cancer 72, 522-529.

228 Bioconjugate Chem., Vol. 12, No. 2, 2001 (27) Morgenson, C. E.,and Solling, K. (1977) Studies on the renal tubular protein reabsorption: partial and near complete inhibition by certain amino acids. Scand. J. Clin. Lab. 37, 477-486. (28) Behr, T. M., Becker, W. S., Sharkey, R. M., Juweid, M. E., Dunn, R. M., Bair, H.-J., Wolf, F. R., and Goldenberg, D. M. (1996) Reduction of renal uptake of monoclonal antibody fragments by amino acid infusion. J. Nuc. Med. 37, 829-833. (29) Carrasquillo, J. A., Lang, L., Whatley, M., Herscovitch, P., Wang, Q. C., Pastan, I., and Eckelman, W. C. (1998) Amino-

Yazaki et al. syn II effectively blocks renal uptake of 18F-labeled anti-tac disulfide-stabilized FV. Cancer Res. 58, 2612-2617. (30) Arano, Y., Fujioka, Y., Akizawa, H., Ono, M., Uehara, T., Wakisaka, K., Nakayama, M., Sakahara, H., Konishi, J., and Saji, H. (1999) Chemical design of radiolabeled antibody fragments for low renal radioactivity levels. Cancer Res. 59, 128-134.

BC000092H

Tumor Targeting of Radiometal Labeled Anti-CEA Recombinant T84

Recommend Documents