Avidin-biotin system pretargeting radioimmunoimaging and radioimmunotherapy and its application in mouse model of human colon carcinoma

Abstract

AIM: To evaluate the multi-step pretargeting radioimm-unoimaging (RII) and radioimmunotherapy (RIT) in nude mice bearing human colon carcinoma with avidin-biotin system labeled with ¹⁵³Sm.

METHODS: Two- and three-step strategies for avidin-biotin system pretargeting techniques were established. In a three-step procedure, human colon carcinoma bearing nude mice were first injected with biotinylated monoclonal antibody (McAb-Bt) followed by cold avidin (Av) 48 h later and then ¹⁵³Sm-DB2 24 h thereafter; whereas the two-step procedure consisted of injection of ¹⁵³Sm-SA 48 h after pretargeting with biotinylated anti-CEA monoclonal antibody (CEA McAb-Bt). SPECT imaging and biodistribution were performed at 4, 24, 48, or 72 h after injection of ¹⁵³Sm-labeled compounds. Five groups of nude mice subcutaneously grafted with human colon carcinoma were treated 3 d after grafting. One group received the injection with 100 μg CEA McAb-Bt followed by cold avidin (80 μg) after 2 d and 11.1 MBq ¹⁵³Sm-DB2 after 1 d. Four control groups were treated respectively with 11.1 MBq ¹⁵³Sm-CEA McAb, 11.1 MBq ¹⁵³Sm-nmIgG, 11.1 MBq ¹⁵³Sm-DB2, 100 μL normal saline. Toxicity was evaluated by changes of leukocyte count, and the efficacy by variation in tumor volume. Histological analyses of tumors were performed.

RESULTS: The three-step procedure allowed faster blood clearance and yielded higher tumor blood ratios (5.76 at 4 h and 12.94 at 24 h) of the ¹⁵³Sm-DB2. The tumor was clearly visualized at 4 h in γ-imaging after the injection of ¹⁵³Sm-DB2, while a significant accumulation of ¹⁵³Sm-SA in the tumor was observed only 24 h after the injection and tumor blood ratios at 4 and 24 h were 1.00 and 2.03, respectively, in the two-step procedure. Pretargeting RIT and ¹⁵³Sm-CEA McAb had a strong tumor-inhibiting effect. The tumor inhibitory rate was 80.67% and 78.44%, respectively, five weeks after therapy. Histopathological evidence also indicated radioactive damage in tumor tissues as necrosis of tumor cells, while in the other organs such as liver and kidney no radioactive damage was observed. Leukocyte counts showed significant decrease after treatment in groups of ¹⁵³Sm-CEA McAb and ¹⁵³Sm-nmIgG.

CONCLUSION: The two kinds of pretargeting strategies can elevate the target-to-nontarget ratio, decrease the blood background and shorten the imaging time compared to ¹⁵³Sm-CEA McAb. Three-step pretargeting RIT is as efficient as ¹⁵³Sm-CEA McAb, but markedly less toxic. This study provides experimental evidence for the clinical application of pretargeting RII and RIT.

Key Words: Radioimmunoimaging; Radioimmunotherapy; Avidin-biotin; Colon carcinoma

INTRODUCTION

Radioimmunoimaging (RII) and radioimmunotherapy (RIT) are nowadays popular research realms in tumor nuclear medicine[1,2]. Various groups have investigated the concept of tumor pretargeting based on the avidin-biotin system[3-6], exploiting the high specificity and strong affinity (Ka = 10-15 mol/L) of avidin (or streptavidin, SA) for biotin, so as to accelerate the elimination of tumor-unbound labeled antibodies from blood and enhance the tumor-to-nontumor (T/NT) ratio. Two-step procedures were thus designed[7]. Paganelli et al[8], have further explored the potential of the avidin-biotin system by devising a three-step procedure, based on the administration of a cold biotinylated antibody, followed by an excess of cold avidin and finally by a radiolabeled biotin derivative. Currently, the pretargeting technique based on avidin-biotin system is widely applied to RII and radioimmunoguided surgery (RIGS) for tumor[9-11]. Avidin-biotin pretargeting RIT is also utilized to treat various tumors and has achieved promising clinical results[12-15].

Samarium-153 (¹⁵³Sm) is a nuclide that can be produced in high yield and has a high specific activity. It is a β-emitter [Emax = 640 (30%), 710 (50%), and 810 (20%) keV] with a half-life (T1/2) of 1.95 d and 2.5-mm peak path length in tissue and also emits a 103 keV γ-ray that is suitable for γ-camera detection. It possesses excellent property in nuclear physics and chemistry for diagnosis and therapy. Dosimetry estimates can therefore be made by quantification of γ-camera data. Compared to 131I, ¹⁵³Sm can result in minimal cross irradiation of normal organs and minimal radiation exposure to medical personnel and is a very attractive radioisotope for RII and RIT[16].

We, in this investigation, established the labeling method of SA and biotin with ¹⁵³Sm, and then the two- and three-step pretargeting RII and three-step RIT were adopted in nude mice bearing human colon carcinoma. The biodistribution and the SPECT imaging were observed and the feasibility of this three-step pretargeting RIT in tumor treatment was also explored.

MATERIALS AND METHODS

Biotinylation of anti-CEA McAb

mAb against carcinoembryonic antigen (anti-CEA McAb) and normal mouse IgG (nmIgG) were prepared by Shanghai Institute of Immunology. BNSH/dimethyl sulfoxide (DMSO) solution was added into bicarbonate buffer solution containing CEA McAb (1 g/L) at a ratio of 15-50:1. The mixture was gently stirred at room temperature for 2 h and finally purified by Sephadex G50 chromatography. Indirect ELISA was performed to assess the activity of biotin, and the purity of biotinylated antibody was determined by SDS-PAGE.

Conjugation of anti-CEA McAb or nmIgG with DTPA

Coupling of cyclic anhydride DTPA (cDTPAa) to anti-CEA McAb or nmIgG was carried out as previously described[19]. Briefly, cDTPAa was suspended in chloroform (1 g/L). An aliquot containing cDTPAa at a molar ratio of 20:1 DTPA:IgG was added to an acid-washed vial and evaporated to dryness under a stream of high-purity dry nitrogen. Anti-CEA McAb or nmIgG (200 μg) was added, the vial was vortexed for 1 min, and then allowed to stand at room temperature for 15-20 min. Acetic acid was added to the mixture to stop the reaction. Separation of the DTPA-CEA McAb or nmIgG conjugate from free DTPA was achieved by Sephadex G50 chromatography. Immunoreactivity of the DTPA-CEA McAb conjugate was assessed using indirect ELISA.

Radiolabeling of anti-CEA McAb, nmIgG, or SA

¹⁵³SmCl3 was supplied by Isotopes Center of China Atomic Energy Institute. SA was purchased from Sigma Chemical Co. (St. Louis, MO, USA). ¹⁵³SmCl3 at a dose of approximately 40 MBq (with specific activity of 22.2 GBq/mL) was mixed with purified CEA McAb, nmIgG, or SA-DTPA conjugate (0.1 mL) and incubated at room temperature for 20 min. Paper chromatography was carried out, using Xinhua No.1 filter paper (30% ammonium nitrate-treated) as the supporter and the mixture of tributyl phosphate, butanone, and acetic ether (at a proportion of 4:10:3) as the developing agent, to determine the labeling efficiency and radiochemical purity. Immunoreactivity of the labeled McAb was tested using indirect ELISA.

Radiolabeling of DB2

Diethylenetriaminepentaacetic acid d,ω-bis(biocytinamide) (DB2; Sigma Chemical Co., St. Louis, MO, USA) was labeled with ¹⁵³Sm as follows. ¹⁵³SmCl3 (37 MBq) was added into 10 μL DB2 solution (2 g/L) for reaction at room temperature for 20 min. Thin-layer chromatography was employed to determine the labeling efficiency and radiochemical purity, using 85% methanol as the developing agent. ¹⁵³Sm-DB2 had a specific activity of 37 MBq/mL, with a binding capacity to 80% biotin agarose beads.

Nude mouse models bearing human colon carcinoma

Balb/c nu/nu mice (female, 20 g) were xenografted subcutaneously in the thigh with 5×106 LoVo cells. Tumor growing to approximately 1 cm in diameter was cut into tiny pieces, which were suspended in normal saline, aspirated and injected (approximately 0.2 mL) subcutaneously into the forelimb of nude mice (4- to 5-wk old). The wounds healed in 12 h. When the tumor grew to a volume of 0.5-1.0 cm3, the mice were used for the subsequent study of pretargeting RII and biodistribution. Therapeutic study was initiated on the 3rd d following tumor inoculation.

Pretargeting radioimmunoimaging and biodistribution studies

Fifty nude mice bearing human colon carcinoma were randomly divided into five groups (three-step group, two-step group, directly labeling group, ¹⁵³Sm-DB2 group, and ¹⁵³Sm-SA group), with 10 in each group. In three-step group, the mice were subjected to injections via the tail vein with 100 μg McAb-Bt, then with 30 μg avidin 48 h later, followed by intra-peritoneal (ip) injection of 3.7 MBq (20 μg) ¹⁵³Sm-DB2. In two-step group, the mice were injected with 100 μg McAb-Bt followed by ip injection of 3.7 MBq (15 μg) ¹⁵³Sm-SA 48 h later. In directly labeling group, only ¹⁵³Sm-DTPA-CEA McAb (3.7 MBq, 20 μg) was administrated via ip injection 72 h after the initiation of the experiment. In ¹⁵³Sm-DB2 group, the mice received only ip injection of ¹⁵³Sm-DB2 (3.7 MBq, 20 μg). In ¹⁵³Sm-SA group, the mice received only ip injection of ¹⁵³Sm-SA (3.7 MBq, 15 μg). Static plane imaging with Siemens ZLC3700 SPECT was performed on the mice of each group at 4, 24, 48, and 72 h, respectively, after the injections. The mice were killed at each time point following the imaging, their organs were isolated and weighed, and the radioactivity counts were determined. The percentage of injected dose per gram of tissue (% ID/g) and T/NT ratio were calculated.

Pretargeting radioimmunotherapy studies

Grouping of the mice Twenty-five tumor-bearing mice were randomly divided into A-E groups (five in each group). The mice in group A received three-step treatment, consisting of injections via the tail vein with 100 μg McAb-Bt, 80 μg Av 48 h later and 11.1 MBq (100 μL) ¹⁵³Sm-DB2 ip injection after another 24 h. Groups B-D were therapeutic control groups, which were given ¹⁵³Sm-CEA McAb, ¹⁵³Sm-nmIgG, and ¹⁵³Sm-DB2, respectively, at a dose of 11.1 MBq (100 μL). As nontherapeutic control, mice in group E were only given 100 μL of normal saline by ip injection.

Observation of tumor inhibiting effect The length (a) and width (b) of tumor mass were measured double-blindly with a sliding caliper once a week for 5 wk. Tumor volume (V) was calculated according to the formula: V = (1/6) ab2. The inhibition rate (IR) of tumor growth was calculated according to the formula: IR = [(mean tumor volume of nontherapeutic control-mean tumor volume of therapeutic group)/(mean tumor volume of nontherapeutic control)]×100%.

Histological examination Pathological examination was also performed on these mice at the end of the treatment. The mice were killed. Their organs were isolated, weighed, and fixed in a 40 g/L formaldehyde solution, embedded in paraffin, cut into 4-μm-thick sections and then stained with hematoxylin and eosin (HE) for microscopic observation.

Observation of radiation toxicity Toxicity was evaluated by the change of the peripheral WBC counts. The number of WBC of all animals was determined on the day of injection and then once a week for 5 wk consecutively.

Statistical analysis

Data were expressed as mean±SD and analyzed by Student’s t-test. P<0.05 was considered statistically significant.

RESULTS

Quality control of labeled compounds

In the preparation of McAb-Bt, it was found that one molecule of antibody could conjugate with three molecules of biotin on average at the biotin/McAb molar ratio of 20:1. The resultant McAb-Bt possessed better capability of SA-binding and immunoreactivity. The optimal coupling condition of the McAb and DTPA was as follows: McAb to cDTPAa molar rate of 1:20, pH value of 7-8, and an antibody concentration at 20 mg/mL. The labeling efficiency of ¹⁵³Sm-CEA McAb was 56-60%, radiochemical purity was over 95% and immunoreactivity was about 50%. The labeling efficiency of ¹⁵³Sm-DB2 exceeded 80%, and radiochemical purity was over 95% after binding to SA. The labeling efficiency of ¹⁵³Sm-SA was over 95% and radiochemical purity was over 98%.

Biodistribution in tumor-bearing nude mice during multi-step pretargeting

The percent of ID/g and T/NT ratio in various organs at different time points during three- and two-step pretargeting are listed in Tables 1 and 2, respectively.

Table 1 Biodistribution of drugs in tumor-bearing nude mice during multi-step pretargeting (% ID/g, mean±SD).

Group	n	h	Organs
			Tumor	Liver	Spleen	Kidney	Bone	Lung	Blood
Three-step	3	4	1.78±0.11	1.40±0.16	1.60±0.48	4.60±2.22	0.67±0.27	0.27±0.10	0.33±0.09
pretargeting		24	1.36±0.17	0.92±0.09	0.78±0.12	5.50±1.55	0.58±0.14	0.43±0.26	0.12±0.06
		48	0.94±0.08	0.66±0.12	0.61±0.03	1.30±0.48	0.36±0.02	0.17±0.02	0.08±0.02
Two-step	3	4	1.35±0.22	3.05±0.07	2.31±0.55	3.37±0.15	0.64±0.12	0.50±0.03	1.41±0.45
pretargeting		24	2.10±0.16	2.02±0.30	1.82±0.19	4.56±0.42	0.33±0.04	0.37±0.03	1.05±0.13
		48	1.24±0.01	1.26±0.01	1.26±0.01	2.30±0.12	0.26±0.01	0.27±0.02	0.96±0.07
Directly	3	4	0.54±0.17	2.01±0.20	1.86±0.21	3.21±0.55	0.94±0.17	0.50±0.10	2.08±0.18
labeling McAb		24	0.99±0.16	1.53±0.10	1.00±0.12	2.42±0.24	0.61±0.01	0.44±0.11	1.29±0.48
		48	1.42±0.05	1.34±0.04	0.77±0.07	1.27±0.07	0.54±0.07	0.34±0.03	1.01±0.13
153Sm-DB2	3	4	0.21±0.01	1.07±0.10	1.14±0.10	5.76±0.68	0.38±0.04	0.37±0.10	0.22±0.05
		24	0.12±0.02	0.83±0.03	0.78±0.08	3.44±0.24	0.24±0.03	0.22±0.04	0.12±0.01
		48	0.10±0.01	0.62±0.11	0.53±0.02	2.60±0.84	0.14±0.04	0.18±0.05	0.10±0.02
153Sm-SA	3	4	0.32±0.01	2.95±0.87	2.27±0.46	3.79±0.48	0.40±0.05	0.36±0.12	0.94±0.08
		24	0.22±0.03	1.89±0.01	1.82±0.17	2.72±0.10	0.22±0.01	0.33±0.07	0.85±0.11
		48	0.19±0.03	1.53±0.16	1.38±0.13	1.03±0.08	0.11±0.01	0.20±0.04	0.59±0.05

Figure 1 Transplant tumor growth curves.

Table 2 T/NT ratio in tumor-bearing nude mice during multi-step pretargeting (mean±SD).

Group	n	h	Organs
			Liver	Spleen	Kidney	Bone	Lung	Blood
Three-step	3	4	1.28±0.18	1.21±0.51	0.46±0.23	2.93±1.05	7.12±2.64	5.76±1.64
pretargeting		24	1.47±0.11	1.76±0.12	0.27±0.08	2.39±0.27	4.06±2.47	12.94±5.05
		48	1.44±0.27	1.53±0.07	0.81±0.40	2.64±0.41	5.65±1.32	11.19±1.80
Two-step	3	4	0.44±0.08	0.60±0.10	0.40±0.08	2.18±0.60	2.71±0.60	1.00±0.22
pretargeting		24	1.05±0.15	1.16±0.07	0.46±0.08	6.48±1.30	5.78±0.82	2.03±0.37
		48	0.98±0.01	0.98±0.01	0.54±0.02	4.77±0.15	4.66±0.22	1.30±0.08
Directly	3	4	0.27±0.12	0.30±0.12	0.17±0.07	0.73±0.36	1.06±0.16	0.26±0.08
labeling McAb		24	0.65±0.11	1.01±0.23	0.41±0.09	1.62±0.27	2.31±0.57	0.85±0.37
		48	1.07±0.02	1.87±0.20	1.14±0.13	2.71±0.49	4.20±0.44	1.42±0.13
153Sm-DB2	3	4	0.20±0.01	0.19±0.02	0.04±0.01	0.56±0.04	0.61±0.15	1.01±0.22
		24	0.14±0.02	0.15±0.04	0.03±0.01	0.50±0.14	0.55±0.18	0.95±0.14
		48	0.17±0.03	0.20±0.01	0.04±0.01	0.75±0.21	0.59±0.17	1.04±0.19
153Sm-SA	3	4	0.11±0.04	0.14±0.03	0.08±0.01	0.80±0.07	0.93±0.22	0.34±0.04
		24	0.12±0.01	0.12±0.02	0.08±0.01	1.02±0.10	0.70±0.24	0.26±0.07
		48	0.13±0.03	0.14±0.02	0.19±0.04	1.77±0.34	1.01±0.30	0.33±0.02

Figure 2 Curves of growth inhibition rate of transplant tumors.

As shown in Tables 1 and 2, tumor uptake (% ID/g) was 1.78% and 1.36% in three-step pretargeting group, 4 and 24 h after the injection, and 1.35% and 2.10% in two-step pretargeting group. The tumor/blood ratio at the same time point was 5.76 and 12.94 in three-step pretargeting group, and only 1.00 and 2.03 in two-step pretargeting group. In directly labeling group, the tumor uptake (% ID/g) reached as high as 1.42% 48 h after the injection, but the level of blood background activity was also high (1.01%ID/g). The tumor uptake (% ID/g) and the ratios of tumor to other organs were all lower in ¹⁵³Sm-DB2 and ¹⁵³Sm-SA groups.

Pretargeting radioimmunoimaging

In the three-step procedure, higher radioactive uptake in the implanted tumor was observed 4 h after the injection, and better tumor display and higher T/NT ratio were also obtained at 24 h because the level of blood background activity was markedly reduced. In the two-step procedure, radioactivity accumulation in the tumor could be visualized at 24 and 48 h. However, higher radioactive uptake was observed in the liver, spleen, and kidney. In the group of directly labeled McAb with ¹⁵³Sm, radioactive uptake at the site of tumor implantation occurred 72 h after the injection; while in ¹⁵³Sm-DB2 and ¹⁵³Sm-SA groups, no significant radioactivity uptake in tumor was observed.

Observation of tumor inhibiting effect

In the 1st wk of treatment, the tumor volume showed little difference among the groups; while in the 2nd wk slower tumor growth rate in groups A and B was observed, indicating tumor inhibition. Till the 5th wk, the tumor volume was obviously smaller in groups A and B than in nontherapeutic control group (Figure 1), and the difference was statistically significant (P<0.01). It was also noted that the tumor volume was smaller in groups C and D than in group E, and the difference was also statistically significant (P<0.01). If the tumor inhibition rate in nontherapeutic control group was considered as zero, the tumor inhibition rate 5 wk post-therapy was as high as 80.67% and 78.44% in groups A and B, much higher than that in groups C and D (Figure 2). In the latter two groups, the tumor inhibition rate was only 29.78% and 23.99%, and no significant difference existed between them. However, significant difference existed between the former two and the latter two groups (P<0.01).

Histopathological evidence also showed that the tumor cell nuclei were pyknotic, karyoclastic, and autolytic in three-step pretargeting RIT and ¹⁵³Sm-CEA McAb groups, which were not seen in the other controls (Figure 3). In the non-treated group, tumor tissue was characterized by the absence of necrosis, and took on typical forms of carcinoma cells (Figure 3).

Figure 3 HE staining of tumor tissues in nude mice bearing human colon carcinoma after receiving three-step pretargeting RIT (A), ¹⁵³Sm-CEA McAb RIT (B) and normal saline (C).

Observation of side effects

Compared to non-treatment group, the number of peripheral blood WBC in treatment groups decreased after injection of 11.1 MBq of ¹⁵³Sm to some degree (Figure 4). However, WBC counts showed significant decrease after treatment in ¹⁵³Sm-CEA McAb and ¹⁵³Sm-nmIgG groups, and WBC recovery was also slower in both groups. Microscopic observation revealed no obvious tissue radiation damage in organs such as heart, liver, spleen, lungs, and brain in all nude mice at the end of the treatment.

Figure 4 WBC counts after injection of drugs.

DISCUSSION

Radiolabeled McAb specific for tumor-associated antigens is used for diagnosis and therapy of malignant tumors[1,2]. However, the blood clearance of antibodies is slow and T/NT ratio of radioactivity in the current system is not high. In order to increase the amount of radioactivity bound to cancer cells, a new approach in cancer therapy called pretargeting has been used. The pretargeting approach can be carried out as a two-step or as a three-step procedure. The two-step procedure is based on the administration of SA-antibody conjugate followed by radioactive biotin or biotinylated antibody is injected followed by radioactive SA[4,18]. In the three-step procedure, biotinylated antibody is injected followed by an excess of avidin/SA and then radioactive biotin is injected[4,19]. In both the methods, the radioactivity of tumor cells is increased[14,20]. Using pretargeting technique of avidin-biotin system, high T/NT ratios have been reported, not only in animal experiments, but also in clinical trials. The avidin-biotin system is also used to reduce the background radioactivity of directly labeled antibody as well as RIGS[9,21]. The success of the approach is due, in part, to the extremely high binding affinity of biotin to avidin/SA, which provides a high tumor targeting efficiency and a long retention at the tumor site. Further, the small size of biotin and its derivatives provides rapid renal clearance of radiolabeled biotin, eliminating much of the irradiation of nontarget tissues. In the present study, the two-step procedure applied consists of injection of radiolabeled SA after pretargeting with biotinylated antibody. The three-step procedure described in our study is as follows. First, tumor pretargeting was done by biotinylated anti-tumor McAbs. Second, when the uptake of McAbs in tumor cells reached its maximum within 24-48 h postinjection, circulating biotinylated McAbs were quickly (5-10 min) removed from the blood by the liver following the administration of avidin as a clearing agent. Finally, postlabeling of the tumor was done by a fast-clearing radioactive biotin and its derivatives. Since one molecule of McAb is capable of binding multiple molecules of biotin, the reaction of avidin with biotin and its derivatives is amplified. One molecule of avidin can bind to four molecules of biotin, and then further amplification of the reaction takes place. In other words, three-step pretargeting technique, like an amplifier, may result in an amplification of signals at the tumor target sites. Meanwhile, transchelating time of radioactive metal in vivo is shortened, and consequently normal tissue uptake of the antibody is reduced and antibody immunoreactivity is preserved unlike directly radiolabeled antibodies, which result in the loss of antibody immunoreactivity due to autoradiolysis and enzyme treatment.

Studies using 90Y-biotin have been successful[14,22]. Biotin has also been labeled with several chelated radionuclides for cancer therapy such as 99mTc, 188Re, 166Ho, and 211At[23-26]. ¹⁵³Sm is a radiolanthanide, which has not yet been widely used, but possesses nuclear characteristics suitable for RIT. It can be produced in reactors by enriched samarium (152Sm) through the (n, γ) reaction. This enables the production of ¹⁵³Sm at low cost. As far as we know, studies of labeling antibodies and biotin with ¹⁵³Sm are very few[27]. It has been recognized that the cation ¹⁵³Sm3+ has good chelating capabilities with polyaminopolycarboxylic acids, such as EDTA or DTPA. In our study, we chose DTPA as the intermediate chelating agent, which can be linked to the antibodies or SA via bicyclic anhydride (cDTPAa). The main purpose of our investigation was to establish the labeling method of McAb and SA as well as biotin with ¹⁵³Sm and to evaluate the pretargeting RII and RIT in nude mice bearing human colon carcinoma with SA-biotin system labeled with ¹⁵³Sm. In the three-step procedure, the tumor was clearly visualized at 4 h in γ-imaging and at the same time point tumor blood pool ratio was 5.76, which was significantly higher than that of control groups. In the two-step procedure, a significant accumulation of ¹⁵³Sm-SA in the tumor was observed only 24 h after injection. The tumor blood ratios at 4 and 24 h were 1.00 and 2.03, respectively. However, the higher radioactive accumulation was also observed in the liver, spleen, and kidney. This deposition may result from complex formation of biotinylated antibodies with radiolabeled SA in circulation. In addition, in molecule of SA there exists three-peptide amino acid sequences (Arg-Thy-Asp), which may bind to the surface of many types of cells[28]. The advantages of pretargeting technique lie in that it is safe and simple, biotinylation of antibody and other reagents are easily prepared. Since the clearance of radiolabeled biotin or SA from normal tissue is much more rapid than that of directly radiolabeled antibody because of its small molecular weight, background radioactivity levels are drastically reduced, and the high T/NT ratio can be reached shortly after injection of the radiolabel[3,15,29]. Our preliminary studies also showed that compared to directly labeled McAb with ¹⁵³Sm, multi-step pretargeting could efficiently decrease the blood background levels, elevate the T/NT ratio, shorten the imaging time and improve the quality of imaging. Earlier and better images of tumor can be obtained using these methods, especially the three-step procedure.

RIT has shown disappointing results in bulky, solid tumors, probably due to the low specific accretion of radiolabeled antibody in the tumor target as compared to normal tissues[30]. Tumors have intrinsic characteristics that unfavorably affect localization and intratumoral distribution of McAbs: heterogeneous expression of the antigen, relatively poor tumor vasculature, elevated interstitial pressure and tumor necrosis. Previous investigations have shown that tumor uptake and radiation doses are inversely correlated with tumor size[31,32]. Therefore, RIT may be a viable option, especially for small-volume and minimal residual tumors. ¹⁵³Sm emits β-rays with intermediate or low energy, which can effectively kill microtumors. In view of these considerations, our RIT experiment was devised to initiate on the 3rd d following tumor inoculation. The results showed that pretargeting RIT and ¹⁵³Sm-CEA McAb had a strong tumor inhibition effect. The tumor inhibitory rate was 80.67% and 78.44%, 5 wk after therapy. Histopathological evidence also indicated radioactive damage as necrosis of the tumor, which was not seen in the other controls. RIT and pretargeting RIT may be more effective for early stage carcinoma. The reason may be related to lower quantity of tumor cells in early stage, less heterogeneity in tumor antigen expression and greater sensitivity to radiation. In addition, there was, to certain degrees, tumor inhibitory effect in groups of ¹⁵³Sm-nmIgG and ¹⁵³Sm-DB2 with a tumor inhibitory rate of 29.78% and 23.99%, respectively, 5 wk after therapy. This may result from unspecific radiation effect. Because ¹⁵³Sm-nmIgG and ¹⁵³Sm-DB2 cannot bind specifically with anti-CEA McAb, effective localization in tumor cells in vivo would not take place, thus this unspecific radiation effect is limited.

In three-step pretargeting RIT, ¹⁵³Sm-DB2 at a single dose of 11.1 MBq/100 μL produced potent tumor-inhibiting effect, and no significant bone marrow toxicity was observed as evidenced by milder decrease of WBC counts. However, in ¹⁵³Sm-CEA McAb and ¹⁵³Sm-nmIgG groups, WBC counts significantly decreased, and WBC recovery was also slower than that in three-step pretargeting RIT, indicating that the hemopoietic function of bone marrow might be affected. When ¹⁵³Sm-CEA McAb and ¹⁵³Sm-nmIgG enter into blood circulation, its slower blood clearance may result in high background radioactivity levels and cause more radiation-related damage to normal tissues or organs. The rapid clearance of ¹⁵³Sm-DB2 from blood may drastically reduce background radioactivity levels. Consequently, radiation exposure to normal tissues especially bone marrow is decreased.

In conclusion, pretargeted RIT with ¹⁵³Sm-DB2 has higher anti-tumor efficacy but lower toxicity than ¹⁵³Sm-CEA McAb. It may serve as an adjuvant therapy for tumor, particularly for small tumor or micrometastatic disease and residual cancer cells in surgical tumor resection areas. Therefore, it is of good prospect in clinical application.