Specific imaging of VEGF-A expression with radiolabeled anti-VEGF monoclonal antibody.

Vascular endothelial growth factor-A (VEGF-A) is one of the most important angiogenic factors. Here, we studied in a nude mouse model whether the expression of VEGF-A in a tumor could be imaged with a radiolabeled anti-VEGF antibody. The humanized anti-VEGF-A antibody A.4.6.1. (bevacizumab), which is reactive with all VEGF-A isoforms, was radiolabeled with In-111 or with I-125. The accumulation of the radiolabeled antibodies in VEGF-A expressing tumors (LS174T) in nude mice was examined in biodistribution studies and by gamma camera imaging. The uptake of the In-111-bevacizumab in the tumor at 3 days p.i. was significantly higher than that of I-125-bevacizumab (19.4 +/- 7.0 %ID/g vs. 9.6 +/- 3.3 %ID/g, p = 0.04). Coinjection of an excess unlabeled antibody resulted in a significant decrease in radioactivity concentration in the tumor (<2.9 +/- 1.9 %ID/g, p < 0.005), indicating VEGF-mediated antibody uptake. Highest uptake in the tumor was observed at relatively low antibody protein doses (<3 microg) (20-25 %ID/g). VEGF-A-expressing tumors could be clearly visualized on planar scintigraphic images from 24-hr post injection onwards. In conclusion, VEGF-A expression in tumors can be visualized specifically with radiolabeled anti-VEGF-A-mAb.

Preparation and evaluation of glycosylated arginine-glycine-aspartate (RGD) derivatives for integrin targeting.

Arginine-glycine-aspartate (RGD) derivatives were prepared by a combination of solid-phase and solution-phase synthesis for selective targeting of alpha vbeta 3 integrin expressed in tumors. In order to evaluate the value of a triazole moiety as a proposed amide isostere, the side chain glycosylated cyclic RGD ( cRGD) peptides were synthesized with either a natural amide linkage or a triazole. Affinity of the cRGD constructs for the alpha vbeta 3 integrin was determined in a solid-phase competitive binding assay, showing strong similarity in binding affinity for each of the compounds under evaluation. Furthermore, the in vivo tumor targeting potential of glycosylated cRGD peptides, linked via amide or triazole, was investigated by determining the biodistribution of (125)I-labeled derivatives in mice with tumors expressing alpha vbeta 3. All of the cyclic RGD derivatives showed preferential uptake in the subcutaneous tumors, with the highest tumor-to-blood ratio measured for the triazole-linked glycosylated derivative. The results of the present study are a clear indication of the value of the triazole moiety as a suitable amide isostere in the development of glycosylated peptides as pharmaceuticals.

Effects of linker variation on the in vitro and in vivo characteristics of an 111In-labeled RGD peptide.

INTRODUCTION: Due to the selective expression of the alpha(v)beta3 integrin in tumors, radiolabeled arginine-glycine-aspartic acid (RGD) peptides are attractive candidates for tumor targeting. Minor modifications of these peptides could have a major impact on in vivo characteristics. In this study, we systematically investigated the effects of linker modification between two cyclic RGD sequences and DOTA (1,4,7,10-tetraazadodecane-N,N',N",N'''-tetraacetic acid) on the in vitro and in vivo characteristics of the tracer. METHODS: A dimeric RGD peptide was synthesized and conjugated either directly with DOTA or via different linkers: PEG4 (polyethylene glycol), glutamic acid or lysine. The RGD peptides were radiolabeled with 111In, and their in vitro and in vivo alpha(v)beta3-binding characteristics were determined. RESULTS: LogP values varied between -2.82+/-0.06 and -3.95+/-0.33. The IC50 values for DOTA-E-[c(RGDfK)]2, DOTA-PEG4-E-[c(RGDfK)]2, DOTA-E-E-[c(RGDfK)]2 and DOTA-K-E-[c(RGDfK)]2 were comparable. Two hours after injection, the tumor uptakes of the 111In-labeled compounds were not significantly different. The kidney accumulation of [111In]-DOTA-K-E-[c(RGDfK)]2 [4.05+/-0.20% of the injected dose per gram (ID/g)] was significantly higher as compared with that of [111In]-DOTA-E-[c(RGDfK)]2 (2.63+/-0.19% ID/g; P<.05) as well as that of [111In]-DOTA-E-E-[c(RGDfK)]2 (2.16+/-0.21% ID/g; P<.01). The liver uptake of [111In]-DOTA-E-E-[c(RGDfK)]2 (2.12+/-0.09% ID/g) was significantly higher as compared with that of [111In]-DOTA-E-[c(RGDfK)]2 (1.64+/-0.1% ID/g; P<.05) as well as that of [111In]-DOTA-K-E-[c(RGDfK)]2 (1.52+/-0.04% ID/g; P<.01). CONCLUSIONS: Linker variation did not affect affinity for alpha(v)beta3 and tumor uptake. Insertion of lysine caused enhanced kidney retention; that of glutamic acid also resulted in enhanced retention in the kidneys. PEG4 appeared to be the most suitable linker as compared with glutamic acid and lysine because it has the highest tumor-to-blood ratio and the lowest uptake in the kidney and liver.

Improved targeting of the alpha(v)beta (3) integrin by multimerisation of RGD peptides.

PURPOSE: The integrin alpha(v)beta(3) is expressed on sprouting endothelial cells and on various tumour cell types. Due to the restricted expression of alpha(v)beta(3) in tumours, alpha(v)beta(3) is considered a suitable receptor for tumour targeting. In this study the alpha(v)beta(3) binding characteristics of an (111)In-labelled monomeric, dimeric and tetrameric RGD analogue were compared. METHODS: A monomeric (E-c(RGDfK)), dimeric (E-[c(RGDfK)](2)), and tetrameric (E{E[c(RGDfK)](2)}(2)) RGD peptide were synthesised, conjugated with DOTA and radiolabelled with (111)In. In vitro alpha(v)beta(3) binding characteristics were determined in a competitive binding assay. In vivo alpha(v)beta(3) targeting characteristics of the compounds were assessed in mice with SK-RC-52 xenografts. RESULTS: The IC(50) values for DOTA-E-c(RGDfK), DOTA-E-[c(RGDfK)](2), and DOTA-E{E[c(RGDfK)](2)}(2)were 120 nM, 69.9 nM and 19.6 nM, respectively. At all time points, the tumour uptake of the dimer was significantly higher as compared to that of the monomer. At 8 h p.i., tumour uptake of the tetramer (7.40+/-1.12%ID/g) was significantly higher than that of the monomer (2.30+/-0.34%ID/g), p<0.001, and the dimer (5.17+/-1.22%ID/g), p<0.05. At 24 h p.i., the tumour uptake was significantly higher for the tetramer (6.82+/-1.41%ID/g) than for the dimer (4.22+/-0.96%ID/g), p<0.01, and the monomer (1.90+/-0.29%ID/g), p<0.001. CONCLUSION: Multimerisation of c(RGDfK) resulted in enhanced affinity for alpha(v)beta(3) as determined in vitro. Tumour uptake of a tetrameric RGD peptide was significantly higher than that of the monomeric and dimeric analogues, presumably owing to the enhanced statistical likelihood for rebinding to alpha(v)beta(3).

Synthesis and biological evaluation of potent alphavbeta3-integrin receptor antagonists.

INTRODUCTION: alpha(v)beta(3) Integrin is expressed in sprouting endothelial cells in growing tumors, whereas it is absent in quiescent blood vessels. In addition, various tumor cell types express alpha(v)beta(3) integrin. alpha(v)beta(3) Integrin, a transmembrane heterodimeric protein, binds to the arginine-glycine-aspartic acid (RGD) amino acid sequence of extracellular matrix proteins such as vitronectin and plays a pivotal role in invasion, proliferation and metastasis. Due to the selective expression of alpha(v)beta(3) integrin in tumors, radiolabeled RGD peptides and peptidomimetics are attractive candidates for tumor targeting. METHODS: A cyclic RGD peptide, a peptoid-peptide hybrid, an all-peptoid and a peptidomimetic compound were synthesized, conjugated with 1,4,7,10-tetraazadodecane-N,N',N'',N'''-tetraacetic acid (DOTA) and radiolabeled with (111)In. Their in vitro and in vivo alpha(v)beta(3)-binding characteristics were determined. RESULTS: IC(50) values were 236 nM for DOTA-E-c(RGDfK), 219 nM for DOTA-peptidomimetic, >10 mM for DOTA-all-peptoid and 9.25 mM for the peptoid-peptide hybrid DOTA-E-c(nRGDfK). (111)In-labeled compounds, except for [(111)In]DOTA-all-peptoid, showed specific uptake in human alpha(v)beta(3)-expressing tumors xenografted in athymic mice. Tumor uptake for [(111)In]DOTA-E-c(RGDfK) was 1.73+/-0.4% ID/g (2 h postinjection) and that of [(111)In]DOTA-peptidomimetic was 2.04+/-0.3% ID/g. Tumor uptake for the peptoid-peptide hybrid [(111)In]DOTA-E-c(nRGDfK) was markedly lower (0.45+/-0.07% ID/g). The all-peptoid [(111)In]DOTA-E-c(nRGnDnFnK) did not show specific uptake in tumors (0.11+/-0.04% ID/g). CONCLUSIONS: The peptidomimetic compound and the cyclic RGD peptide have a high affinity for alpha(v)beta(3) integrin, and these compounds have better tumor-targeting characteristics than the peptoid-peptide hybrid and the all-peptoid.

Pretargeting of carcinoembryonic antigen-expressing tumors with a biologically produced bispecific anticarcinoembryonic antigen x anti-indium-labeled diethylenetriaminepentaacetic acid antibody.

PURPOSE: The aim of these studies was to develop a pretargeting strategy for CEA-expressing cancers using biologically produced bispecific monoclonal antibodies (bsMAb). The bsMAbs used in this system have affinity for the carcinoembryonic antigen on the one hand, and for indium-labeled diethylenetriaminepentaacetic acid (DTPA), on the other. EXPERIMENTAL DESIGN: Stable quadroma clones producing bsMAb MN-14xDTIn-1 were isolated. LS174T tumor-bearing mice were injected with 1 to 100 microg of bsMAb followed by 1 to 60 ng of an (111)In-labeled bivalent peptide [Ac-Phe-Lys(DTPA)-Tyr-Lys(DTPA)-NH2]. Mice were killed at 24 hours postinjection and the biodistribution of the radiolabel was determined. The biodistribution of diDTPA labeled with four different radionuclides ((111)In, 99mTc, nonresidualizing 125I, and residualizing 125I) was determined at various time points postinjection following pretargeting of LS174T tumors with bsMAb MN-14xDTIn-1. RESULTS: Optimal tumor targeting was observed when tumors were pretargeted with 10 microg of bsMAb MN-14xDTIn-1 and when 6 ng of a radiolabeled peptide was given 72 hours later. The uptake of the four radiolabels in LS174T tumors at 4 hours postinjection was similar. However, at later time points, the (111)In-label and residualizing 125I-label were better retained in the tumor than the nonresidualizing 125I label. Although the absolute uptake in the tumor (in terms of percentage of injected dose per gram of tissue) was 5-fold lower than the uptake obtained with directly labeled MN-14, the pretargeting strategy revealed much higher tumor-to-blood ratios due to the rapid clearance of the radiolabel from the circulation as compared with (111)In-MN-14 (445 +/- 90 and 5.3 +/- 1.1, respectively, at 72 hours postinjection). CONCLUSIONS: Effective targeting of carcinoembryonic antigen-expressing tumors was achieved with a newly produced bispecific antibody. The (111)In-labeled L-amino acid peptide and 125I-D-amino acid peptide were better retained in the tumor than the 99mTc- and 125I-L-amino acid peptide. Very high tumor-to-blood ratios were obtained due to rapid background clearance.

Comparison of IgG and F(ab')2 fragments of bispecific anti-RCCxanti-DTIn-1 antibody for pretargeting purposes.

PURPOSE: An effective pretargeting strategy was developed for renal cell carcinoma (RCC) based on a biologically produced bispecific monoclonal antibody: anti-RCCxanti-DTPA(In) (bsMAb: G250xDTIn-1). Tumour uptake of a (111)In-labelled bivalent peptide after pretargeting with bsMAb G250xDTIn-1 was relatively high compared with that in other pretargeting systems using chemically coupled F(ab')(2) fragments. Here, we investigated the effect of the bsMAb form in the pretargeting strategy. METHODS: To determine the optimal interval between the administration of each of the bsMAb forms and the (111)In-labelled bivalent peptide, the biodistribution of the radioiodinated bsMAb forms was studied in athymic mice with subcutaneous SK-RC-1 RCC tumours. Since tumour targeting of the radiolabelled peptide depends on the bsMAb form and dose, a bsMAb dose escalation study was carried out for both bsMAb forms. Under optimised conditions, the biodistribution of the (111)In label in mice with pretargeted RCC was determined from 4 h up to 7 days p.i. RESULTS: The optimal interval between the two administrations was 72 h for the bsMAb IgG and 4 h for the bsMAb F(ab')(2). The optimal bsMAb dose for intact IgG was 67 pmol and the optimal bsMAb F(ab')(2) dose was 200 pmol. Targeting of the pretargeted RCC with 4 pmol (111)In-labelled bivalent peptide revealed high tumour uptake with both bsMAb forms. CONCLUSION: With the pretargeting strategy, using either bsMAb IgG or bsMAb F(ab')(2), very efficient peptide targeting of the tumour was obtained. Uptake and retention of the radiolabel in the tumour with the pretargeting approach are not affected by the bsMAb form used.

Pretargeting with bispecific anti-renal cell carcinoma x anti-DTPA(In) antibody in 3 RCC models.

UNLABELLED: We have developed an efficient pretargeting strategy for renal cell carcinoma (RCC) based on a biologically produced bispecific monoclonal antibody (bs-mAb). Tumor targeting with this 2-step pretargeting strategy in the NU-12 mouse RCC model was very efficient compared with other pretargeting strategies, possibly due to unique characteristics of the NU-12 tumor used in our studies. Here we describe the bs-mAb G250xDTIn-1 pretargeting strategy in 3 different RCC nude mouse models. METHODS: Three different human RCC xenografts in nude mice (NU-12, SK-RC-1, and SK-RC-52 tumors) were pretargeted with 100 pmol bs-mAb G250xDTIn-1. Three days after administration of the bs-mAb, mice were injected intravenously with 4 pmol 111In-labeled bivalent peptide, diDTPA-FKYK (DTPA is diethylenetriaminepentaacetic acid). At 1, 4, 24, 48, and 72 h after injection of the radiolabeled peptide, the biodistribution of the radiolabel was determined. The 3 RCC tumors were characterized in vivo and in vitro for G250 antigen expression, vessel density, vascular volume, and vascular permeability and by targeting with 111In-/125I-labeled cG250 mAb. RESULTS: Using the pretargeting strategy, relatively high uptake of the radiolabel was observed in all 3 tumor models (maximum uptake: SK-RC-1 [278 +/- 130 %ID/g (percentage injected dose per gram), 1 h after injection] > NU-12 [93 +/- 41 %ID/g, 72 h after injection] > SK-RC-52 [54 +/- 9 %ID/g, 4 h after injection]). Remarkably, uptake of the radiolabel in the tumor did not correlate with G250 antigen expression. The kinetics of the radiolabel in the tumor varied largely in the 3 RCC tumors: SK-RC-1 and SK-RC-52 tumors showed a washout of the 111In label from the tumor with time: only 30% of the radiolabel was retained in the tumor 3 d after injection, whereas the 111In label was fully retained in NU-12 tumors. Physiologic characteristics (vessel density, vascular volume, and vascular permeability) of the tumors may explain these differences. CONCLUSION: G250 antigen-expressing tumors can be pretargeted very efficiently with the bs-mAb G250xDTIn-1. There was no correlation between G250 antigen expression and uptake of the radiolabel in the tumor using the pretargeting strategy or with directly labeled mAbs. Therefore, these studies show that physiologic characteristics of the tumor, such as vascular permeability, play a significant role in pretargeting.

Biodistribution and therapeutic efficacy of (125/131)I-, (186)Re-, (88/90)Y-, or (177)Lu-labeled monoclonal antibody MN-14 to carcinoembryonic antigen in mice with small peritoneal metastases of colorectal origin.

UNLABELLED: Therapeutic efficacy in radioimmunotherapy depends, among other things, on the choice of the radionuclide. The aim of the present study was to determine the most suitable radionuclide for radioimmunotherapy with monoclonal antibody MN-14 to carcinoembryonic antigen in an experimental model of small peritoneal metastases of colorectal origin. METHODS: In nude mice with intraperitoneal LS174T tumors (diameter, 1-3 mm), the biodistributions of MN-14 labeled with (131)I ((131)I-MN-14), (186)Re-mercaptoacetyltriglycine ((186)Re-MN-14), and (88)Y-diethylenetriaminepentaacetic acid (DTPA) ((88)Y-MN-14) after intravenous and intraperitoneal administration were determined. Subsequently, the therapeutic efficacies of equally toxic activity doses of (131)I-MN-14 (9.25 MBq per mouse), (186)Re-MN-14 (9.25 MBq per mouse), (90)Y-MN-14 (3.15 MBq per mouse), and MN-14 labeled with (177)Lu-DTPA ((177)Lu-MN-14) (8.33 MBq per mouse) after intraperitoneal administration were determined. RESULTS: Each of the radioimmunoconjugates preferentially accumulated in tumor nodules, both after intravenous administration and after intraperitoneal administration. Values for clearance from blood were similar for all radioimmunoconjugates. The uptake of (88)Y-MN-14 in the liver and spleen was significantly higher than the uptake of (131)I-MN-14 or (186)Re-MN-14. Maximal uptake values (mean +/- SD) in tumors were 58 +/- 7 percentage injected dose per gram of tissue (%ID/g) for (131)I-MN-14 (24 h after administration), 83 +/- 19 %ID/g for (186)Re-MN-14 (72 h after administration), and 148 +/- 89 %ID/g for (88)Y-MN-14 (192 h after administration). Dosimetric analysis of the biodistribution data estimated that the radiation doses guided to the tumor by intraperitoneally administered (131)I-MN-14, (186)Re-MN-14, (90)Y-MN-14, and (177)Lu-MN-14 were 150, 100, 45, and 200 Gy, respectively. The median survival time of control mice, treated with unlabeled MN-14, was 42 d, whereas the median survival times of mice treated with (131)I-MN-14, (186)Re-MN-14, (90)Y-MN-14, and (177)Lu-MN-14 were 100 d (range, 58-142; P < 0.0001), 72 d (range, 46-84; P = 0.0002), 82 d (range, 46-142; P < 0.0001), and 136 d (range, 56-142; P < 0.0001), respectively. At the completion of the experiment (142 d after tumor cell inoculation), no residual disease was found in 8 of 9 long-term survivors ((131)I, n = 3; (90)Y, n = 1; and (177)Lu, n = 4). CONCLUSION: The uptake of (88)Y-MN-14 in small peritoneal LS174T xenografts was higher than the uptake of (131)I-MN-14 or (186)Re-MN-14. The present study indicates that (131)I and (177)Lu are the most suitable radionuclides for the radioimmunotherapy of small peritoneal metastases.

Experimental radioimmunotherapy of small peritoneal metastases of colorectal origin.

Radioimmunotherapy using radiolabeled monoclonal antibodies (MoAbs) directed against tumor-associated antigens might be an effective treatment modality for small volume disease. Our aim was to optimize an experimental model of radioimmunotherapy for small peritoneal metastases of colorectal origin using the anti-CEA MoAb MN-14. In nude mice with intraperitoneal (i.p.) LS174T tumors, a protein dose-escalation study was carried out to determine the maximal dose of radioiodinated MN-14 to be used in radioimmunotherapy. The biodistribution of radioiodinated MN-14 was determined after intravenous (i.v.) and i.p. administration. Finally, the therapeutic efficacy of escalating activity doses of (131)I-labeled MN-14 (62.5-500 microCi) was assessed and compared to that of unlabeled MN-14 or 500 microCi of (131)I-labeled irrelevant control antibody. At protein doses higher than 25 microg, uptake in tumor was reduced, presumably due to saturation of tumor antigen. During the first 24 hours i.p. administration led to higher tumor uptake and higher tumor:blood ratios than i.v. administration. Median survival of the control groups was 38 days (unlabeled MN-14) and 52 days ((131)I-labeled nonspecific antibody). Median survival of the groups treated with increasing activity doses of (131)I-labeled MN-14 was 42 days (62.5 microCi), 49 days (125 microCi), 63 days (250 microCi) and 101 days (500 microCi), respectively (p < 0.0001 compared to unlabeled MN-14). The present study shows that the anti-CEA-antibody MN-14 preferentially accumulates in i.p. LS174T tumor xenografts after both i.p. and i.v. administration. Intraperitoneal radioimmunotherapy using (131)I-labeled MN-14 delays significantly the outgrowth of peritoneal LS174T metastases, even at relatively low activity doses.