ORIGINAL ARTICLE 
Annals of Nuclear Medicine Vol. 12, No. 2,83-88, 1998 

Localization of colorectal carcinoma by rhenium-188-labeled B72.3 antibody in xenografted mice 

Masako N. HOSONO,* Makoto HosoNo,** Paul O. ZAMORA,*** Stefan GUHLKE,*** Thomas HABERBERGER,*** Hans BENDER,*** F.F. Russ KNAPP**** and Hans J. BIERSACK*** 

*Department of Radiology, Osaka City University Medical School **Department of Radiology, Saitama Medical Center, Saitama Medical School ***Department of Nuclear Medicine, University of Bonn, Germany ****Nuclear Medicine, Health Sciences Research Division, Oak Ridge National Laboratory, USA 

In order to evaluate the feasibility of 188Re-labeled antibodies for radioimmunotargeting, mono-clonal antibody B72.3_ , recognizing TAG-72, expressed on the surface membranes of colorectal cancer cells, was directly labeled with 188Re, obtained from a 188W/188Re generator, using stannous tartrate and compared with 125I-labeled B72.3. As a control, a human IgG was also radiolabeled with 188Re and 1251. Prepared antibodies for 188Re labeling could be stored as kits. Biodistribution was determined in nude mice inoculated with human colorectal carcinoma LoVo. Labeling efficiency and immunoreactivity of 188Re-B72.3_ were 80.3% and 64.7%, respectively. 188Re-B72.3 localized specifically in the LoVo tumors. Although the absolute tumor accumulation level of 188Re-B72.3 was lower than 125I-B72.3, 188Re-B72.3 demonstrated higher tumor-to-blood contrast than the 125I-labeled counterpart, 2.04+-0.44 vs. 1.05+-0.28 at 96 hours, because of fast clearance from the blood. 188Re-B72.3 seemed efficient for the imaging and therapy of colorectal carcinoma. 

Key words: rhenium- 188, xenograft, B72.3, colorectal carcinoma 

INTRODUCTION 
THE RADIOISOTOPES of rhenium, 186Re and 188Re, have currently been tested for radiolabeling of antibodies.1-4 186Re has a 1. I MeV p-emission complemented with a 137 keV gamma-emission and a half-life of 90.64 hours. 188Re has a maximum p-emission of 2.11 MeV and 155 keV of gamma photons, and the half-life is 16.98 hours. Both provide imageable gamma photons for scintigraphy, and appropriate p particles for radioimmunotherapy. Beaumier et al. reported the effectiveness of 186Re-labeled antibodies for the treatment of small-cell lung carcinoma in a mouse model.5 Further, 186Re-labeled 

Received October 16, 1997, revision accepted December 11, 1997. For reprint contact: Makoto Hosono, M.D., Department of Radiology, Saitama Medical Center. Saitama Medical School, Kamoda. Kawagoe. Saitama 350-8550. JAPAN. 

HEDP and EDTMP have been used as bone seeking agents for pain relief therapy in breast or prostate cancer patients with bone metastasis,6.7 but as 186Re contains a variable amount of carrier rhenium which can be an obstacle in labeling antibodies, 186Re is less desirable for the antibody-guided internal radiotherapy. On the other hand, 188Re is readily available as carrier-free sodium perrhenate in isotonic saline from a 188W/188Re generator system. In this study, a murine monoclonal antibody (Mab) B72.3 was labeled with 188Re by a direct method based upon reduction of the antibody molecule and binding of 188Re to the thiol group. B72.3, recognizing TAG (tumor associated glycoprotein)-72 antigen expressed on the cell surface of colorectal carcinomas,8,9 has been widely used in the diagnosis of colorectal cancer patients.10,ll In this study, the efficacy of 188Re-B72.3 was estimated by conducting biodistribution studies in nude mice inoculated with human colon carcinoma LoVo cells. MATERIALS AND METHODS 
Monoclonal antibody B72.3, a murine IgG1 , was obtained from Sterling Winthrop (Malvern, PA, USA). This antibody was raised by immunizing mice with membrane enriched fractions of a human breast carcinoma metastatic to the liver.8 It reacts with the human tumor antigen TAG-72, a high molecular weight glycoprotein with characteristics of a mucin, and expressed in colon cancers, breast cancers and ovarian cancers. On the basis of immunopathological examinations, TAG-72 is expressed in up to 85% of colon cancers, 70% of breast cancers, and 95% of ovarian cancers, but it shows minimal or no expression in normal adult tissue.12,13 

Antibody preparation 
B72.3 was prepared for 188Re labeling as previously reported.14 Briefly, the antibody was dialyzed overnight against either saline or buffer (maltose, 5%; potassium hydrogen phthalate, 40 mM; potassium sodium tartrate, 10 mM; Glycine, 0.3 M; pH = 5.6). The antibody was then reduced in this same buffer made to be 2 mM with stannous tartrate. The reduction reaction was carried out under the atmosphere for 21 hours at room temperature. The reduced or prepared antibody was then purified on PD- I O columns (Pharmacia LKB Biotechnology, Uppsala, Sweden). The protein concentration of the column eluate was determined by Bio-Rad Protein Assay (Bio-Rad Laboratories. Hercules, CA, USA), and the antibody solution was then diluted to 2 mg/ml with buffer. This was mixed 1 : 1 with buffer containing 8 mM stannous tartrate so that the final concentration was I mg reduced antibody per ml in 4 mM stannous tartrate. One-ml aliquots of this solution were lyophilized, sealed under argon and stored at 4dig.C until needed. As a control antibody, human IgG (Bayer, Leverkusen, Germany) was prepared by the same methods as B72.3. 

188Re labeling 
The 188W/188Re generator was supplied by Oak Ridge National Laboratory (Oak Ridge, TN, USA).15 The generator was eluted with saline, and the radioactivity of the perrhenate obtained was assayed with a dose calibrator. One ml of perrhenate solution was added to a vial contain-ing the lyophilized prepared antibody. The vial was gently swirled to dissolve the solids and initiate the labeling reaction. The reaction was allowed to proceed overnight (17 hours) at room temperature. 

125I labeling 
B72.3 and human IgG were radioiodinated by the Chloramine-T method. Briefly, 40 /ug of purified Mabs in 0.3 M phosphate buffer (PB), pH 7.5 and 11.1 MBq of 125I were mixed with 2.5 ug of chloramine-T dissolved in 0.3 M PB. After 5 minutes of reaction, radiolabeled Mabs were separated from free radioiodine by PD-1O gel chromatography . 

Immunoreactivity 
Immunoreactivity was determined by an affinity thin layer chromatography. The solution containing the radio-labeled antibody was diluted 1 : 20 with 50% newborn calf serum and 10 ul aliquots spotted onto a pair of thin layer chromatography strips (RhoChekllTM, RhoMed Inc. Albuquerque, NM, USA).16-18 One strip was positive and had a band of antigen on the strip immediately upstream from the origin. The second strip was a negative control, which did not have an antigen band. The strips were developed in PBS containing 4% ethanol. Radioactivity bound to the immunoreactive antibody remained within the band of antigen in the positive strip. Some preparations contained radiochemical impurities, such as radiocolloids, which remained at the origin on both the positive and negative strips. Thus to determine net immunoreactivity, the percentage of the radioactivity at the origin of the positive strip was corrected by subtracting the percentage of radioactivity at the origin of the negative strip. Immunoreactivity of 188Re-labeled and 125I-labeled antibodies was estimated by this method. 

Cell binding 
The human colorectal cell line LoVo 19 was cultured in RPMI 1640 culture medium (GIBCO BRL, Gaithersburg, MD, USA) supplemented with 1 mM glutamine and 10% fetal calf serum. To determine the binding to LoVo cells, 125I- and 188Re-B72.3 diluted with O.5% bovine serum albumin (Immuno GmbH, Heidelberg, Germany) in saline were incubated with increasing concentrations of LoVo cells in 5.7 x 46 mm microcentrifuge tubes for 1 hour at 4dig.C. After centrifugation at 1,500 x g, the tubes were washed with saline and cut. The radioactivity bound to cells was counted in a well-type gamma counter. Specific binding to cells was calculated by subtracting the nonspecific binding in the tubes in which 20 ug of unlabeled B72.3 was added.
 
Xenografts and mice 
For the study in mice, 5 to 7 week-old female NMRI nu/nu athymic mice were purchased from Harlan (Borchen, Germany). 1 x 10 7 of LoVo cells per mouse were implanted by s.c. inoculation into the flanks of mice. Xenografts were allowed to grow 3 to 4 weeks after inoculation. The weight of xenografts used in this study (n = 50) was 0.35 +- 0.19 g and there was no significant difference among groups. 
Biodistribution study 
Groups of 5 mice bearing LoVo xenografts per time point were given 200 kBq of 188Re-B72.3 and 37 kBq of 125I-B72.3 at the same time via the tail vein. At 30 min, 6 hours 1,2,3 and 4 days after injection, the mice were sacrificed. The radioactivity of tumors and selected organs was determined with a well-type gamma counter. Biodistribution of 188Re-labeled and 125I-labeled human IgG was also determined at 1 , 2, 3 and 4 days after injection. 

RESULTS 
Labeling efficiency and immunoreactivity, which was determined by thin layer chromatography, of 188Re-labeled B72.3 were 80.3% and 64.7%. On the other hand, labeling efficiency and immunoreactivity of 125I-labeled 72.3 were 53.8% and 72.4%. Final specific activities of 5.62 mCi/mg and 5.38 mCi/mg were achieved for 188Re-labeled B72.3 and 125I-labeled B72.3, respectively. The cell binding assay demonstrated specific binding of 188Re- and 125I-labeled B72.3 to LoVo cells (Figure 1), and there was no significant difference between the two counterparts . Biodistribution of 188Re-labeled B72.3 and human IgG showed that 188Re-labeled localized in the tumor well (Figure 2). 188Re-B72.3 in the tumor reached the maxi-mum accumulation of 3.49 +- 0.89%ID/g at 48 hours after injection. 125I-B72.3 Iocalized also in the LoVo tumor with the peak accumulation of 7.16 +- 1.96%ID/g at 48 hours (Figure 3 ). 
Both 188Re-labeled B72.3 and human IgG cleared more quickly from the blood and normal organs than 125I-labeled counterparts (Figures 2 and 3 ). Tumor-to-organ ratios are shown in Figures 4 and 5. Rapid clearance of 188Re-B72.3 resulted in the high tumor-to-blood ratio of 2.04 +- 0.44 at 96 hours in comparison with 1.05 +- 0.28 for 125I-B72.3, but 125I-B72.3 had higher tumor-to-liver and tumor to kidney ratios than 188Re-B72.3.
 
DISCUSSION 
In this study, Mab B72.3 was efficiently labeled with 188Re by a direct technique using reduction of the intrinsic disulfide bonds of the antibody. In addition, Mab B72.3 was prepared for 188Re-labeling and stored as a kit suitable for routine use. Immunoreactivity of 1888Re-B72.3 was similar to that of 125I-B72.3. The in vitro cell binding showed specific accumulation of 1888Re- and 125I-labeled B72.3 to LoVo cells. In the biodistribution studies, 188Re-B72.3_ specifically targeted the LoVo tumors as well as 125I-B72.3. Although absolute tumor accumulation of 188Re-B72.3 was lower than that of 125I-B72._3, rapid clearance of 188Re-B72.3 from the blood resulted in the higher tumor-to-blood ratio. Pimm et al.20 and Sakahara et al.21 suggested that the fast blood clearance of reduction-mediated 99mTc-labeled monoclonal antibodies may be attributed to the release of 99mTc or small fragments containing 99mTc from the antibodies. The similar release of 188Re from the reduction-mediated 188Re-labeled B72.3 may explain the rapid blood clearance in our study. Moreover, relatively high liver uptake may be attributable to the metabolism of 188Re-labeled B72.3 in the liver and to the excretion into the biliary system.21 And the urinary excretion of 188Re-B72.3 metabolites may account for high renal uptake of 3.96 +- 0.48 %ID/g at 24 hours as compared with 3.21 +- O.8 1 %ID/g for 125I-B72.3.21 Since 186Re is produced by neutron exposure to a 185Re-rich metal target, it is difficult to get high purity of 186Re. On the other hand, 188Re is easily obtained from a 188Wl 188Re generator without carrier. The mean beta energy levels of 188Re, O.73 and O.79 MeV, which represent mean ranges in the soft tissue of 2.7 and 3,1 mm, are an advantage for therapy of tumors which are often heterogeneously necrotic. Although the peak gamma energy of 188Re, 155 keV, is suitable for imaging, Eary et al. suggested that a medium-energy collimator is appropriate to get optimal image resolution and count rate for 188Re because of its higher energy peaks including 478, 633, 829, and 931 keV.3 Due to the short half life of 188Re (16.98 hours), a sufficient accumulation is needed to achieve effective irradiation of the target. Techniques aiming for more stable labeling, such as pretreatment of antibodies with 2-iminothiolane in order to provide a thiol reactive position for the attachment rhenium may enable higher tumor accumulation and tumor-to-normal tissue contrast.22.23 Another approach, which should improve the stability of rhenium-labeled antibodies, is conjugation of chelating agents. Goldrosen et al. reported that NR-LU-10 Mab, labeled stably with 186Re by using a tetrafluorophenylactivated ester derivative of the triamide thiolate as a chelate, showed blood, liver, kidney and tumor accumulation of 8.2, 2.3, 1.3, and 9.8%ID/g, respectively, at 24 hours in mice bearing human colon carcinoma.24 Visser et al. used S-benzoyl-mercaptoacetyltriglycine (S-benzoyl-MAG3) for 186Re-labeled E48 F(ab')2 and observed that in mice xenografted with squamous cell carcinoma, 10.4 %ID/g localized in the tumor, with only 2.8 %ID/g in the blood, and less than I .2%ID/g in the liver and kidney at 48 hours,25 so that chelate techniques seem efficient in achieving stable rhenium-labeling which enables high tumor-to-nornal tissue ratios. Nevertheless, direct-label-ing methods have the advantage of simple kit preparation suitable for clinical studies. In conclusion. Mab B72.3 was successfully labeled with 188Re by direct labeling and tumor localization was confirmed. Further studies are needed to obtain high tumor-to-normal tissue ratios by direct-labeling methods. 

ACKNOWLEDGMENTS 
The authors thank the German-Japanese Radiological Affiliation (Osaka, Japan) for the fellowship offered to Masako N. Hosono, and they also thank the Alexander von Humboldt Foundation (Bonn, Germany) for the fellowship awarded to Makoto Hosono. 

REFERENCES 
1 . Breitz HB, Weiden PL, Vanderheyden JL. Clinical experience with rhenium- 186-labeled monoclonal antibodies for radioimmunotherapy: results of phase I trials. J Nucl Med 33: 1099-1112, 1992. 
2. Breitz HB, Fisher DR, Weiden PL. Durham JS, Ratliff BA, Bjorn MJ, et al. Dosimetry of rhenium- 186-labeled mono-clonal antibodies: methods, prediction from technetium-99m-labeled antibodies and results of phase I trials. J Nucl Med 34: 908-917, 1993. 
3. Eary JF. Durack L, Williams D. Vanderheyden J-L. Consid-erations for imaging Re- 188 and Re- 186 isotopes. Clin Nucl Med 15: 911-916, 1990. 
4. John E, Thakur ML, DeFulvio J, McDevitt MR, Damjanov I. Rhenium-186-labeled monoclonal antibodies for radioimmunotherapy: preparation and evaluation. J Nucl Med 34: 260-267, 1993. 
5. Beaumier PL. Venkatesan P, Vanderheyden J-L. Burgua WD. Kunz LL. Fritzberg AR, et al. 186Re radioimmuno-therapy of small cell lung carcinoma xenografts in nude mice. Cancer Res 51 : 676-681, 1991 . 
6. de Klerk JMH, van het Schip AD. Zounenberg BA, van Dijk A, Quirijnen JMSP, Blijham GH, et al. Phase I Study of Rhenium-186-HEDP in Patients with Bone Metastases Originating from Breast Cancer. J Nucl Med 37: 244-249, 1996. 
7. Quirijnen JMSP, Han SH, Zonnenberg BA, de Klerk JMH. van het Schip AD, van Dijk A, et al. Efficacy of Rhenium-l 86-Etidronate in Prostate Cancer Patients with Metastatic Bone Pain. J Nucl Med 37: 1511-1515, 1996. 
8. Nuti M, Teramoto YA, Mariani-Costantini R. Horan Hand P, Colcher D, Schlorn J. A monoclonal antibody (B72.3) defines patterns of distribution of a novel tumor-associated antigen in human mammary carcinoma cell preparation. Int J Cancer 29: 539-545, 1982. 
9. Johnson V, Schlom J, Paterson A. Bennett J, Magnani JL, Colcher D. Analysis of a human tumor-associated glyco-protein (TAG-72) identified by monoclonal antibody B72.3 (human gammal). Cancer Res 46: 850-857, 1986. 
10. Carrasquillo JA, Sugarbaker P, Colcher D, Reynolds JC, Esteban J, Bryant G, et al. Radioimmunoscintigraphy of colon cancer with iodine-131-labeled B72.3 monoclonal antibody. J Nucl Med 29: 1022-1030, 1988. 
11. Mitchell EP, Singh A, Volkert W, Patterson WP. Holmes RA, Yarbro JW. Imaging of human adenocarcinoma with 111In labeled B72.3 monoclonal antibody (meeting abstract). Proc Annu MeetAm Soc Clin Oncol, 1990. 
12. Thor A, Gorstein F, Ohuchi N. Szpak CA. Johnston WW, Schlom J. Tumor associated glycoprotein (TAG-72) in ovarian carcinomas defined by monoclonal antibody B72.3. J Natl Cancer Inst 76: 995-1OO6, 1986. 
13. Thor A, Ohuchi N, Szpak CA, Johnston WW. Schlom J. Distribution of oncofetal antigen tumor-associated glyco-protein-72 defined by monoclonal antibody B72.3. Cancer Res 46: 3118-3124, 1986. 
14. Griffiths GL. Goldenberg DM, Diril H, Hansen HJ. Technetium-99m, rhenium- 186, and rhenium- 188 direct-labeled antibodies. Cancer supplement 73: 761-768, 1994. 
15. Knapp FFRJ, Mirzadeh S. The continuing important role of radionuclide generator systems for nuclear medicine. Eur J Nucl Med 21: 1151-1165, 1994. 
16. Zamora PO, Rhodes BA, Sass KA. Budd P, Lambert CA, Cardillo A, et al. A simple TLC assay to assess the percent immunoreactive fraction (IF) of radiolabeled antibodies. J Nucl Med 34: 23_ IP, 1993. 
17. Zamora PO, Sass KA, Budd P, Lambert CR, Rhodes BA. Affinity thin layer chromatography for determining the immunoreactive fraction or radiolabeled antibodies. Bio Techniques 16: 306-311 , 1994. 
18. Zamora PO, Domalewski M, Marek MJ, Budd P, Rhodes BA. Quantitation of radiolabeled antibody binding to cells by thin-layer chromatography. Nucl Med Biol 21 : 205-21O, 1994. 
19. Drewinko B, Romsdahl MM, Yang LY, Ahearn MJ, Trujillo JM. Establishment of a human carcinoembryonic antigen-producing colonadenocarcinoma cell line. Cancer Res 36: 467-475, 1976. 
20. Pimm MV, Rajput RS, Frier M, Gribben SJ. Anomalies in reduction mediated technetium-99m laleling of monoclonal antibodies. Eur J Nucl Med 18: 973-976. 199 1 . 
21. Sakahara H, Saga T. Endo K, Hattori N, Hosono M, Kobayashi H, et al. In vivo instability of reduction-mediated 99mTc-labeled monoclonal antiboy. Nucl Med Biol 20: 617-623, 1993. 
22. McCall MJ, Diril H, Meares CF. Simplified method for conjugating macrocyclic bifunctional chelating agents to antibodies via 2-iminothiolane. Bioconjugate Chemistry 1: 222-226, 1990. 
23. Srivastava SC, Mease RC. Progress in research on ligands, nuclides, and techniques for labeling monoclonal antibodies. Nucl Med Biol 18: 589-603, 1991. 
24. Goldrosen MH, Biddle WC, Pancook J, Bakshi S, Vanderheyden JL, Fritzberg AR, et al. Biodistribution, pharmacokinetic, and imaging studies with 186Re-labeled NR-LU-10 whole antibody in LS174T colonic tumor-bear-ing mice. Cancer Res 50: 7973-7978, 1990. 
25. Visser GWM, Gerretsen M, Herscheid JDM, Snow GB, van Dongen G. Labeling of monoclonal antibodies with rhe-nium- 1 86 using the MAG3 chelate for radioimmunotherapy of cancer, a technical protocol. J Nucl Med 34: 1953-1963, 1993.