Preclinical Evaluation of Monoclonal Antibody 14C5 for Targeting Pancreatic Cancer

Abstract

The use of radiolabeled antibodies that are able to target primary tumors as well as metastatic tumor sites with minimal reactivity to normal tissues is a promising approach for treating pancreatic cancer. In this study, the integrin α_vß₅ is studied as a target for the diagnosis of and potential therapy for human pancreatic cancer by using the radiolabeled murine monoclonal antibody (mAb) 14C5. Biopsy specimens from human pancreatic tumors were examined for the expression of the integrin α_vß₅. The pancreatic tumor cell line Capan-1 was used to test the in vitro targeting potency of mAb 14C5 labeled with 125/131-iodine and 111-indium. Internalization, retention, and metabolism were investigated in cellular radioimmunoassays. Biodistribution and tumor-targeting characteristics were studied in Capan-1 xenografts. All tumor sections were positive for the integrin α_vß₅, with an extensive positive staining of the stroma. Saturation binding experiments showed high affinity with comparable K _ds. In vitro internalization experiments showed a longer intracellular retention of ¹¹¹In-p-benzyl isothiocyanate-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (p-SCN-Bz-DOTA)-14C5 in comparison to ¹²⁵I-14C5 and ¹¹¹In-p-isothiocyanatobenzyl diethylenetriaminepentaacetic acid (p-SCN-Bz-DTPA)-14C5. In vivo radioisotope tumor uptake was maximum at 48–72 hours, with the uptake of ¹¹¹In-p-SCN-Bz-DOTA-14C5 (35.84 ± 8.64 percentage of injected dose per g [%ID/g]) being 3.9- and 2.2-folds higher than ¹³¹I-14C5 (12.16 ± 1.03%ID/g) and ¹¹¹In-p-SCN-Bz-DTPA-14C5 (14.30 ± 3.76%ID/g), respectively. Planar gamma imaging with mAb 14C5 indicated clear localization of the pancreatic tumors versus minimal normal tissue uptake. mAb 14C5 is a promising new antibody for targeting the integrin α_vß₅ for the diagnosis of and potential therapy for pancreatic cancer.

Introduction

Despite recent advances in diagnostic modalities and therapeutic strategies, the prognosis of pancreatic cancer remains poor. With an incidence of 58,100 per year and a mortality rate almost identical to this cancer's incidence, it is the fifth leading cause of death in Europe. The 5-year survival rate for patients with this malignancy is less than 5%.¹ Although much research has been undertaken in the past 50 years, conventional treatment approaches, such as surgery, radiation, chemotherapy, or combinations of these, have little impact on the course of this aggressive neoplasm.² One of the reasons for the poor prognosis of pancreatic carcinoma is the difficulty of obtaining an early diagnosis. Metastatic spread, which would have already occurred in half of the patients at the time of diagnosis, is one of the major problems in the treatment of this aggressive malignancy.²

One potential approach is the use of radiolabeled antibodies that are able to target primary tumors as well as metastatic tumor sites with minimal affinity for normal tissues.³ Different monoclonal antibodies (mAb) have been used to target tumor antigens with high expression on malignant pancreatic cancer cells, including ferritin,⁴ HER-2/neu,⁵ MUC1,^6,7 α₆ ß₄,⁸ VEGF, and EGFR.⁹ Some of these mAbs have produced encouraging results, but none of them is yet commercially approved for pancreatic cancer therapy.

Recently, a promising new antibody targeting the integrin α_vß₅ was identifiedfor diagnosing human breast, colon, squamous-cell, and lung cancers.^10

–15 By FACScan analysis, the integrin α_vß₅ was shown to be overexpressed on the tumor surfaces of several cancer types, including pancreatic cancer.^13,15 Immunohistologic staining of human lung, colon, and breast tumor tissues with mAb 14C5 showed antigen expression predominantly in the stroma surrounding the tumor cells and on the stromal fibroblasts (squamous-cell carcinoma, 5 of 5; lung adenocarcinoma, 3 of 3; large cell carcinoma, 1 of 1; colon adenocarcinoma, 19 of 20) and was only localized on the tumor cells themselves in some cases (squamous-cell carcinoma, 2 of 5; lung adenocarcinoma, 1 of 3; large-cell carcinoma, 0 of 1; colon adenocarcinoma 10 of 20).¹⁵ Except for a weak reactivity within the myoepithelial cells in biopsies of breast tumor and in tubular cells of the kidney, the antibody was not reactive against normal tissues.^11,12,15

Integrins are not only key mediators in normal development, as recent evidence indicates that integrin-mediated tumor–stroma interactions play an active and crucial role in directing the malignant phenotype in different cancer types, including pancreatic cancer.¹⁶ A hallmark in pancreatic adenocarcinoma is the presence of desmoplasia, which is defined as proliferation of fibrotic tissue with an altered extracellular matrix (ECM) conductive to tumor growth and metastasis.¹⁷ This desmoplastic reaction is associated with an abnormal vasculature as well as abundant connective tissue in which matrix proteins, such as collagen, fibronectin, and vitronectin, interact with cell surface integrin receptors to provide survival signals for pancreatic stellate cells (PSCs) and adenocarcinoma cells.¹⁷ These PSCs or stromal fibroblasts express the integrin α_vß₅ ¹⁸ and play a pivotal role in the development of fibrosis in adenocarcinoma of the pancreas and chronic pancreatitis.^18,19 Targeting these PSCs by antifibrosis drugs has been suggested as a therapeutic strategy for pancreatic fibrosis.²⁰

In vitro experiments conducted by Grzesiak et al.¹⁶ revealed integrin α_vß₅ overexpression on all examined pancreatic tumor cells (13 of 13 for α_v and 10 of 10 for ß₅) and previous experiments demonstrated that mAb 14C5 showed a higher affinity for different cancer cells, including pancreatic cancer,¹³ compared with a commercially available anti-α_vß₅ mAb (P1F6).¹⁵ In vivo, integrin α_vß₅ expression on human carcinoma cells was correlated with a higher potential to metastasize²¹ and coinjection of PSCs with pancreatic tumor cells has led to a markedly increased growth rate in nude mice.²² Immunohistochemical studies revealed the presence of α_v on centroacinar cells of the normal pancreas,²³ and an increased heterogenous presence of α_v was shown in pancreatic carcinoma.²⁴

Even though pancreatic cancer differs markedly for normal pancreas through an altered ECM composition, activated PSCs, and changed distribution of integrins on the cell surface, the role of integrins, particularly integrin α_vß₅, in the development of pancreatic carcinoma needs further delineation.

Burvenich et al.¹³ showed that binding of ¹²⁵I-14C5 to its antigen resulted in the internalization of the complex in human lung and colon carcinoma cells. It is well-established that catabolism of iodine–tyrosine-labeled antibodies results in the generation of iodotyrosine within the lysosomes, which leaves the cell rapidly.²⁵ Conversely, indium chelator–labeled antibodies are metabolized to leave a chelator–amino acid fragment, which is typically not released from the cell and is trapped for further degradation.^26

–29 This effect can produce different residence times for two differently labeled forms of the same antibody. A longer retention time of a radioisotope within tumor cells is advantageous not only for imaging purposes but also for cancer RIT where a long duration of exposure is desired. Another benefit of specifically targeting tumors with a mAb coupled with 111-indium is the scouting potential before administration of therapeutic doses of ß-emitting radiometals such as 90-ytrium and 177-luthetium. Performing radioimmunoscintigraphy with trace-labeled mAbs as a scouting procedure before RIT confirms the tumor-targeting properties of the mAb, and quantification of the biodistribution allows estimation of dose delivery to tumors and normal tissues.³⁰

In this study, the potency of integrin α_vß₅ as a target for the diagnosis and potential treatment of pancreatic cancer with mAb 14C5 was tested. Cellular radioimmunoassays with human pancreatic carcinoma cells (Capan-1) were used to compare the retention of radioactivity, targeted to tumor cells using mAb 14C5, with iodine and 111-indium. Studies were performed to evaluate the radiolabeling efficiency of iodine and ¹¹¹In (with p-isothiocyanatobenzyl diethylenetriaminepentaacetic acid [p-SCN-Bz-DTPA] and p-benzyl isothiocyanate-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid [p-SCN-Bz-DOTA] as chelators) and their stability, in vitro binding capacity, and immunoreactivity. Comparative in vivo studies were performed with iodine-, indium-p-SCN-Bz-DTPA- and indium-p-SCN-Bz-DOTA-labeled 14C5 in Capan-1 tumor–bearing mice. At various timepoints, planar γ imaging scans were conducted with ¹²³I-14C5.

Materials and Methods

Monoclonal antibodies

All antibodies used are of murine origin. The mAb 14C5 (anti-integrin α_vß₅ immunoglobulin G1 [IgG1]) was produced and purified as described previously.^11,12 Control antibody MAB002 (IgG1 isotype control) was purchased from R&D Systems and anti-CD31 antibody (a blood-vessel marker) was purchased from Millipore. mAb LMH-3 and mAb F19 (antifibroblast activation protein [anti-FAP] mAb) were produced at the Biological Development Facility, Ludwig Institute for Cancer Research, Melbourne, Australia.

Cell lines

Capan-1, a human pancreatic cancer-cell line, was a kind gift of J&J Pharmaceutical Research and Development. Capan-1 cells were cultured in DMEM (Cambrex) containing 2 mM l-glutamine and 10% fetal bovine serum (FBS; Cambrex). Cells were cultured at 37°C in a 5% CO₂ humidified incubator and passaged with 0.05% trypsin–0.02% EDTA (Cambrex).

Immunohistochemistry

Biopsy specimens from human pancreatic tumors (n = 4) and normal pancreatic tissues (n = 3) (Austin Health, Ethics Committee number H2005/02169) were examined for the expression of the integrin α_vß₅. Sections (5 μm) from fresh, frozen tissues were cut and fixed in acetone for 1 minute at 4°C. For immunohistochemical detection, an avidin–biotin system (DakoCytomation) was applied using a 1 μg/mL dilution of mAb 14C5 as previously described.¹⁵ As a fibroblast marker, mAb F19³¹ was used at a concentration of 5 μg/mL. mAb LMH-3³² was included as an IgG1 isotype control, and anti-CD31 Ab was used as a blood-vessel marker. As a negative control, the primary antibody was omitted.

Radiolabeling

Preparation of iodine-labeled 14C5

Iodination of mAbs 14C5 and MAB002 was performed using the Iodo-Gen method as described previously.^10,13,14

Preparation of ¹¹¹In-labeled 14C5

The bifunctional chelators p-SCN-Bz-DTPA and p-SCN-Bz-DOTA were purchased from Macrocyclics. mAb 14C5 was conjugated with both chelators under strict metal-free conditions, using a modification of the procedure described by Perk et al.³³ The mAb was transferred in metal-free 0.1 M NaHCO₃ (pH 8 for DTPA, pH 9 for DOTA) via Centricon YM-30 (5500 rpm, 3 × 30 minutes) at a 15 mg/mL concentration. Conjugation of the chelator to the antibody was performed at a 50:1 concentration. DTPA conjugation was performed at room temperature for 10 minutes, and conjugation with DOTA was allowed to proceed overnight at 4°C. Buffer exchange into 0.1 M ammonium acetate buffer (pH 5.5) and removal of unbound chelator was performed using ultrafiltration with Centricon YM-30 (Centrifugal Filter Device YM 30; Millipore).

Indium [¹¹¹In] chloride (0.05 M HCl) was purchased from Tyco Healthcare NV. The pH of indium was adjusted to 5.5 by adding metal-free sodium acetate (pH 5.5). Typically, labeling reactions were performed using 50 μg of mAb with 3.7 MBq of radionuclide to obtain a specific activity of 74 MBq/mg of mAb. DTPA conjugates were incubated for 10 minutes at RT, whereas DOTA conjugates were incubated for 60 minutes at 45°C. The reaction was quenched by adding a 100-fold excess of EDTA. Removal of free ¹¹¹In was performed using ultrafiltration with Centricon YM-30 (5500 rpm, 3 × 30 minutes).

The molar ratios of p-SCN-Bz-DTPA and p-SCN-Bz-DOTA were determined following a method described by Meares et al.³⁴ In short, conjugates were labeled with a known excess of indium acetate spiked with ¹¹¹In. After labeling, the solution was challenged with EDTA and subsequently spotted for instant thin layer chromatography (ITLC). ITLC data were used to calculate the chelator-to-mAb molar ratio.

Quality control

All conjugates were analyzed by ITLC for radiochemical purity and by high-performance liquid chromatograph (HPLC) and sodium dodecyl sulfate (SDS)–polyacrylamide gel electrophoresis followed by phosphor imaging analysis for integrity. ITLC analysis of radiolabeled mAb was performed on silica gel-impregnated glass fiber sheets (Gelman Sciences). For iodine, citric acid (20 mM, pH 5.0) was used as an eluant. For ¹¹¹In-labeled antibodies, samples were incubated for 5 minutes in an EDTA solution (20 mM) and subsequently spotted for ITLC, using 0.9% NaCl as mobile phase.

The radioimmunoconjugates were monitored with HPLC using a Shodex KW 802.5 column (7.8 × 300 mm; Thomson Instrument Company). Elution was done with 0.1 M potassium phosphate buffer (pH 7.4) at a flow rate of 0.8 mL/minutes. Gel electrophoresis was performed on a Mini-Protean III cell system (BioRad Laboratories) using 10% SDS gels under nonreducing conditions and analyzed via phosphor imaging (Cyclone Storage Phosphor System; Perkin Elmer).

Immunoreactivity and in vitro stability

The immunoreactivity of the iodine- and ¹¹¹In-labeled mAb preparations was assessed using Capan-1 cells by the method of Lindmo et al.³⁵ Cells were diluted in PBS (0.5% BSA) at concentrations ranging from 1.6 × 10⁵ to 2.5 × 10⁶ cells/mL and 500 μL of these suspensions were incubated with a fixed amount of labeled mAb (250 μL, 39 ng/mL) for 2 hours at 4°C under agitation. To control the in vitro stability, ¹²⁵I-14C5, ¹¹¹In-p-SCN-Bz-DTPA-14C5, and ¹¹¹In-p-SCN-Bz-DOTA-14C5 were incubated at 4°C in PBS with 0.5% BSA or at 37°C in PBS:mouse serum (50:50). Samples were taken periodically for ITLC analysis up to 168 hours, using citric acid or ammonium acetate as an eluant.

Saturation-binding assay

The affinities of both conjugates were analyzed by a saturation-binding assay using Capan-1 cells. Specific activities of 37 MBq/mg protein were typically used. Twelve duplicate test samples containing increasing amounts of ¹²⁵I-14C5, ¹¹¹In-p-SCN-Bz-DTPA-14C5, ¹¹¹In-p-SCN-Bz-DOTA-14C5, or ¹²⁵I-MAB002 and 0.5 × 10⁶ Capan-1 cells in a total volume of 1 mL of cell medium were incubated at 4°C for 2 hours. The supernatant was removed by centrifugation (15 seconds, 8000 rpm, RT) and cells were washed twice with ice-cold PBS. Nonspecific binding was estimated in the presence of 167 nM (i.e., approximately 1000-fold excess of K _d value of mAb 14C5) unlabeled mAb. Nonlinear regression and K _d values were determined using GraphPad Prism Software 5.0.

Internalization experiments

The degree of internalization was investigated in a radioimmunoassay with Capan-1 cells as described previously.¹³ Petri dishes (10 cm²) with confluent Capan-1 cells were incubated with 37 kBq ¹²⁵I-14C5 and 37 kBq ¹¹¹In-p-SCN-Bz-DTPA-14C5 or ¹¹¹In-p-SCN-Bz-DOTA-14C5 (with a total of 0.2 μg mAb 14C5 per dish) at 4°C for 2 hours. Excess unbound mAb was removed and the cells were incubated for various time periods at 37°C to allow internalization or at 4°C to inhibit internalization. The labeled antibodies bound to the cell surface were stripped by washing the cells twice with ice-cold low-pH buffer (pH 2.25) for 1 minute. Cell surface-bound and intracellular activities were determined by measuring the activity of the acid washes and cell-associated radioactivity after treatment with low-pH medium. The % activity shown in the graphs represents the amount of radioactivity as a percentage of the total radioactivity available in the Petri dish at each timepoint (activity released in the medium + cell surface-bound activity + internalized activity). The radioactivity released in the supernatant and bound to the cell fraction was analyzed by passing the supernatant over a PD-10 column to determine the percentage of free isotopes and antibody-coupled radioactivity. The different fractions were counted with a γ counter (Cobra II; Canberra). The fractions containing antibody or antibody fragments were pooled, concentrated with YM-30, and analyzed with SDS–polyacrylamide gel electrophoresis and Cyclone phosphor imaging. For the analysis of the internalized radioactivity, cells were treated with a 2% RBS solution to disrupt the cell membranes, and after centrifugation, cell lysate was passed over a PD-10 column, as described previously.

Biodistribution and pharmacokinetic studies

Athymic mice (nu/nu; female, 5 weeks; Charles River) were used for in vivo biodistribution. Capan-1 cells (1 × 10⁶) were injected subcutaneously, and when tumors reached a size of 0.5–1.0 g, biodistribution studies were performed. All experiments were approved by the local ethics committee for animal experiments (ECD08/18).

For biodistribution studies, mice were injected in the tail vein with 0.074 MBq ¹³¹I-14C5 (37 MBq/mg) or 0.148 MBq ¹¹¹In-conjugated 14C5 (74 MBq/mg). Typically, groups of 3 mice were sacrificed at 1, 24, 48, 72, and 168 hours after injection. To show the specificity of mAb 14C5, a biodistribution with ¹³¹I-labeled nonspecific MAB002 was performed at preset timepoints (1, 24, 48, and 168 hours). Tumors and organs were removed immediately, blotted dry, and weighed. Tumor, organ, and blood activities were counted in a γ counter. Standards prepared from the injection solution were counted at each timepoint with the tissues and tumors. Tissue radioactivity concentrations are expressed as %ID/g of organ. Differences in blood clearance between the different isotopes were calculated from the blood samples. The biologic half lives were calculated and the data were analyzed by a two-phase exponential curve fit.

Statistical analysis

Experimental data were analyzed with GraphPad Prism Software. Data are expressed as mean ± standard deviation and evaluated for statistical significance with a nonparametric Mann-Whitney U test.

Results

Immunohistochemistry

Figure 1 shows the immunohistochemical investigation of human pancreatic carcinoma tissues and their surrounding stroma. All tumor sections (n = 4) were moderately to poorly differentiated ductal adenocarcinomas and all were positive for the integrin α_vß₅. An extensive positive staining of the stroma surrounding the tumor cells (4 of 4) was seen with no staining of tumor cells. A comparison with the fibroblast marker mAb F19 (Fig. 1B) and a blood-vessel staining antibody (anti-CD31; Fig. 1C) confirmed that 14C5 was not localized in the blood vessels but on the fibroblasts surrounding the tumor. No acute pancreatic inflammation was seen, but all sections showed areas of chronic pancreatitis. Intense staining of the pancreatic-cancer stroma was seen with minimal or no staining in the areas of chronic pancreatitis fibrous tissue. No staining was seen for the IgG1 isotype control (Fig. 1D) and the secondary antibody control (data not shown). Normal tissues of the pancreas (n = 3) were analyzed for expression of the antigen 14C5 and no specific staining with mAb 14C5 was detected (data not shown). In one section, a very weak focal staining of the pancreatic-duct cells and stroma was seen but considered nonspecific.

FIG. 1.

Immunohistochemistry on frozen tissue of human pancreatic carcinoma ( × 100). (A) 14C5 mAb (1 μg/mL), (B) anti-FAP F19 mAb (5 μg/mL), (C) anti-CD31 mAb (7 μL/mL), (D) nonspecific LMH-3 mAb (5 μg/mL). mAb, monoclonal antibody.

Radiolabeling, stability, and immunoreactivity

Premodification of mAb 14C5 resulted in 1.5 p-SCN-Bz-DTPA and 0.6 p-SCN-Bz-DOTA moieties per mAb. Subsequent labeling with ¹¹¹In resulted in an overall labeling yield of >82% for ¹¹¹In-p-SCN-Bz-DTPA and >60% for ¹¹¹In-p-SCN-Bz-DOTA. A radiochemical purity of at least 98% was noted for both products. Labeling of mAbs 14C5 and MAB002 with iodine resulted in an overall labeling yield of 90.4% ± 2.7% (n = 8) for 14C5 and 77.8% ± 3.5% (n = 3) for MAB002 and radiochemical purity was >98% for the purified products. The specific activities were 37 MBq/mg for ^125/131I-mAbs and 74 MBq/mg mAb for ¹¹¹In-14C5.

All radioimmunoconjugates were stable for 7 days at 4°C in 0.5% BSA. In serum at 37°C, ¹²⁵I- and ¹¹¹In-DOTA-14C5 were stable for at least 7 days (<5% of the radiolabel was released during this incubation), but a faster radiometal release was seen for ¹¹¹In-p-SCN-Bz-DTPA-14C5 (17% of radiolabel release at day 7). Radiolabeled mAb 14C5 was found to have an immunoreactivity of >78% for all immunoconjugates, even after storage for 48 hours in mouse serum at 37°C. At day 7, a decrease in immunoreactivity was seen for ¹¹¹In-p-SCN-Bz-DTPA-14C5 to 38%, whereas no decline was seen for iodine-14C5 and ¹¹¹In-p-SCN-Bz-DOTA-14C5.

Saturation-binding assay in Capan-1 cells

To determine the binding capacity of the radioimmunoconjugates, the dissociation constants (K _d) of ¹²⁵I-14C5, ¹¹¹In-p-SCN-Bz-DTPA-14C5, and ¹¹¹In-p-SCN-Bz-DOTA-14C5 were estimated for Capan-1 cells by a saturation-binding assay. ¹²⁵I-14C5 showed specific binding with a dissociation constant of 0.11 ± 0.01 nM. No major influence of p-SCN-Bz-DTPA or -DOTA conjugation was seen on the binding capacity of 14C5 with a K _d of 0.24 ± 0.02 and 0.11 ± 0.03 nM, respectively (n = 3). The saturation binding curves obtained were characteristic of high-affinity specific binding of an antibody to its antigen. No specific binding was seen for ¹²⁵I-MAB002.

Internalization experiments

The degree of internalization was investigated in a radioimmunoassay with Capan-1 cells. Cells were loaded with ¹²⁵I- and ¹¹¹In-labeled mAb 14C5 at 4°C for 2 hours. After removing unbound mAb 14C5 and adding fresh medium, the cells were incubated for various timepoints at 37°C to allow internalization or at 4°C to prevent internalization. Figure 2 shows the distribution of radioactivity in Capan-1 cells following incubation with ¹²⁵I-14C5 (Fig. 2A, n = 6), ¹¹¹In-p-SCN-Bz-DTPA-14C5 (Fig. 2B, n = 3), or ¹¹¹In-p-SCN-Bz-DOTA-14C5 (Fig. 2C, n = 3) at 37°C, corrected for the acid-resistant fraction observed at 4°C. At the start of the internalization assay (t = 0 minute) and at 4°C, the acid-resistant cell-associated fraction was 10%. Because the energy-dependent internalization process cannot occur at 4°C, the average 10% internalized fraction observed during incubation at 4°C was considered to be a result of handling of the cells during the internalization assay and not because of internalization of mAb 14C5. Therefore, the cell-associated fraction observed at 37°C was corrected for this cell-associated radioactivity observed at 4°C at each timepoint and the difference was considered the internalized fraction. A rapid internalization was seen for all radioimmunoconjugates, with a significantly faster and higher internalization for ¹¹¹In-p-SCN-Bz-DOTA-14C5. At 24 hours, ¹²⁵I-14C5 showed a significantly lower internalized fraction of 6.93% ± 0.88%, compared with 49.44% ± 0.75% for ¹¹¹In-p-SCN-Bz-DOTA-14C5 and 36.66% ± 1.42% for ¹¹¹In-p-SCN-Bz-DTPA-14C5 (p < 0.05, nonparametric Mann-Whitney U test). The activity released in the medium was higher for both ¹²⁵I-14C5 (64.40% ± 0.10%) and ¹¹¹In-p-SCN-Bz-DTPA-14C5 (40.83% ± 1.62%), compared with ¹¹¹In-p-SCN-Bz-DOTA-14C5 (19.13% ± 0.45%; p < 0.05, nonparametric Mann-Whitney U test). Analysis of the media revealed that the radioactivity of ¹¹¹In-p-SCN-Bz-DTPA-14C5 and ¹¹¹In-p-SCN-Bz-DOTA-14C5 was primarily present in a high–molecular-weight form, with no appreciable amounts of free ¹¹¹In, ¹¹¹In-p-SCN-Bz-DTPA, or ¹¹¹In-p-SCN-Bz-DOTA (<6%), whereas the released radioactivity derived from ¹²⁵I-14C5 was primarily present in the degraded form (>77%). Tumor cell lysates showed no free ¹²⁵I, consistent with the rapid release of free iodine from the lysosomes. In contrast, both free ¹¹¹In/¹¹¹In-p-SCN-Bz-DTPA/¹¹¹In-p-SCN-Bz-DOTA and intact ¹¹¹In-p-SCN-Bz-DTPA-14C5 and ¹¹¹In-p-SCN-Bz-DOTA-14C5 were observed in the tumor cell lysate.

FIG. 2.

Distribution of radioactivity in Capan-1 cells following incubation at 37°C with ¹²⁵I-14C5 (A, n = 6), ¹¹¹In-p-SCN-Bz-DTPA-14C5 (B, n = 3), or ¹¹¹In-p-SCN-Bz-DOTA-14C5 (C, n = 3). Bars: SD. p-SCN-Bz-DTPA, p-isothiocyanatobenzyl diethylenetriaminepentaacetic acid; p-SCN-Bz-DOTA, p-benzyl isothiocyanate-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid.

Biodistribution in Capan-1 tumor-bearing mice

For biodistribution studies, approximately 2 μg (37 MBq/mg) antibody per mouse was used. The biodistribution with nonspecific ¹³¹I-MAB002 in Capan-1–bearing mice showed a significantly lower uptake in tumor tissue at all timepoints (p < 0.05, nonparametric Mann-Whitney U test; Fig. 3A) and a significantly longer retention in the blood compared with ¹³¹I-14C5 (5.07 ± 0.84%ID/g and 2.24 ± 0.03%ID/g at 168 hours, respectively). At all timepoints, tumor uptake with ¹³¹I-14C5 was higher than that with ¹³¹I-MAB002, with the highest uptake of 12.16 ± 1.84%ID/g at 48 hours p.i. Tumor-to-blood ratios of ¹³¹I-14C5 exceeded 1 at 24 hours and remained high at all timepoints. At 48 hours, the time of highest tumor uptake, ratios of ¹³¹I-MAB002 were 6.6-fold lower than those of ¹³¹I-14C5. Tumor-to-blood ratios of ¹³¹I-14C5 were significantly higher compared with those of ¹³¹I-MAB002 for all timepoints (p < 0.05; Fig. 3B).

FIG. 3.

Antigen specificity of in vivo tumor binding. (A) Comparison of tumor uptake (expressed as percentage of injected dose per g[%ID/g]) of ¹³¹I-14C5 and ¹³¹I-MAB002 (nonspecific mouse IgG1) in Capan-1-bearing mice. (B) Comparison of tumor-to-blood ratios of ¹³¹I-14C and ¹³¹I-MAB002 (n = 3). Bars: standard deviation.

Table 1 summarizes the uptake of radiolabeled mAb 14C5 in the tumor and the important organs. The %ID/g of ¹³¹I-14C5 in Capan-1 xenografts peaked at 24–48 hours (11.22–12.16%ID/g), whereas tumor uptake for ¹¹¹In-conjugated 14C5 was highest at 48–72 hours after antibody injection (with 12.93–14.30%ID/g for p-SCN-Bz-DTPA and 24.68–35.84 for p-SCN-Bz-DOTA), with the uptake of ¹¹¹In-p-SCN-Bz-DOTA-14C5 being 2.2- and 3.9-fold greater than that of ¹¹¹In-p-SCN-Bz-DTPA-14C5 and ¹³¹I-14C5, respectively, and reaching statistical significance (p < 0.05, Mann-Whitney U nonparametric test). Tumor-to-blood ratios for ¹¹¹In-labeled-14C5 peaked at 11.54 ± 7.53 (p-SCN-Bz-DTPA) and 15.09 ± 4.16 (p-SCN-Bz-DOTA) at 168 hours.

Table 1.

Uptake of ¹³¹I-14C5, ¹¹¹In-p-SCN-Bz-DTPA-14C5, and ¹¹¹In-p-SCN-Bz-DOTA-14C5 in Capan-1 Tumor-Bearing Mice

Tissue	¹³¹I-14C5	¹¹¹In-p-SCN-Bz-DTPA-14C5	¹¹¹In-p-SCN-Bz-DOTA-14C5
Tumor
1 h	3.63 ± 0.50	2.26 ± 0.75	2.26 ± 0.52
24 h	11.22 ± 3.31	14.73 ± 3.22	20.62 ± 4.15
48 h	12.16 ± 1.03	12.93 ± 3.76	24.68 ± 5.44
72 h	8.45 ± 0.57	14.30 ± 3.76	35.84 ± 8.64
168 h	6.91 ± 1.84	9.71 ± 2.19	30.51 ± 12.71
Liver
1 h	9.89 ± 0.16	12.32 ± 1.79	12.48 ± 2.98
24 h	4.21 ± 1.56	13.66 ± 1.56	14.02 ± 1.42
48 h	3.28 ± 0.81	13.88 ± 1.74	11.30 ± 0.63
72 h	1.78 ± 0.75	8.85 ± 4.22	12.18 ± 4.39
168 h	1.62 ± 0.54	11.45 ± 3.40	11.36 ± 3.12
Spleen
1 h	5.52 ± 0.58	4.44 ± 0.43	6.26 ± 0.13
24 h	2.48 ± 0.70	3.86 ± 1.06	6.05 ± 0.77
48 h	2.41 ± 0.72	4.00 ± 0.53	7.58 ± 0.85
72 h	1.37 ± 0.59	3.63 ± 0.38	8.03 ± 1.84
168 h	0.87 ± 0.40	3.69 ± 1.21	5.35 ± 0.54
Kidneys
1 h	7.09 ± 0.41	5.16 ± 0.22	6.59 ± 0.76
24 h	3.74 ± 0.78	5.89 ± 0.69	4.63 ± 0.69
48 h	2.76 ± 0.30	5.70 ± 1.04	5.47 ± 0.34
72 h	1.73 ± 0.46	4.58 ± 0.73	6.39 ± 2.25
168 h	2.01 ± 0.78	4.13 ± 1.37	2.73 ± 0.35
Lungs
1 h	9.80 ± 1.81	5.14 ± 0.57	10.05 ± 4.13
24 h	5.72 ± 0.56	2.95 ± 0.69	3.15 ± 0.77
48 h	4.87 ± 0.56	2.27 ± 0.14	3.26 ± 0.32
72 h	4.22 ± 0.37	1.96 ± 0.41	2.21 ± 0.21
168 h	2.69 ± 0.68	1.44 ± 0.22	1.52 ± 0.45
Pancreas
1 h	1.66 ± 0.21	0.78 ± 0.22	1.04 ± 0.28
24 h	1.31 ± 0.26	0.85 ± 0.17	0.81 ± 0.12
48 h	0.89 ± 0.14	0.88 ± 0.11	0.83 ± 0.06
72 h	0.68 ± 0.19	0.92 ± 0.27	1.09 ± 0.47
168 h	0.54 ± 0.04	0.81 ± 0.31	0.51 ± 0.13
Blood
1 h	24.13 ± 1.37	19.39 ± 1.84	18.17 ± 1.81
24 h	10.79 ± 3.21	5.64 ± 1.32	6.80 ± 2.09
48 h	8.13 ± 2.01	1.99 ± 0.51	7.09 ± 0.93
72 h	6.92 ± 2.11	2.80 ± 1.19	5.95 ± 0.43
168 h	4.47 ± 0.67	1.05 ± 0.56	2.27 ± 1.38

Values are given in percent of injected dose per g (%ID/g); n = 3. Data are presented as mean ± standard deviation.

%ID/g, percentage of injected dose per g; p-SCN-Bz-DTPA, p-isothiocyanatobenzyl diethylenetriaminepentaacetic acid; p-SCN-Bz-DOTA, p-benzyl isothiocyanate-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid.

For iodine, other organs did not show specific uptake but showed a progressive washout of radioactivity, which mimicked the blood clearance. For both ¹¹¹In-p-SCN-Bz-DTPA-14C5 and ¹¹¹In-p-SCN-Bz-DOTA-14C5, the liver showed a high uptake of 8.85–13.88%ID/g and 11.30–14.02%ID/g, respectively. Kidney and spleen uptakes were also increased, but other organs showed no uptake of the indium-labeled 14C5. Bone uptake for both radiometals was <2.49%ID/g, uptake in muscle was <1%ID/g, and uptake in pancreas, intestines, and stomach was <2%ID/g. Uptake in the lungs was comparable for all three conjugates, with a high uptake at early timepoints, but with a declining uptake over time.

The half-life of each radioimmunoconjugate was estimated from the %ID/g of blood obtained over the timepoints sampled. Assuming an exponential two-phase decay in a two-compartmental model, the T _1/2α for ¹³¹I-, ¹¹¹In-p-SCN-Bz-DTPA-, and ¹¹¹In-p-SCN-Bz-DOTA-14C5 was 51.93, 55.38, and 24.94 minutes and mean T _1/2 ß was 2109, 868, and 1166 minutes, respectively.

To evaluate the influence of antibody concentration on liver, kidney, and spleen uptake, biodistributions with 50 and 100 μg (both 1.5 MBq) of ¹¹¹In conjugates were carried out (Table 2). No influence was seen for p-SCN-Bz-DTPA, but for p-SCN-Bz-DOTA significant differences were noted, with a decrease in the liver from 12.18 ± 4.40%ID/g (2 μg) to 3.80 ± 0.66%ID/g (100 μg) and in the spleen from 8.03 ± 1.84%ID/g (2 μg) to 4.50 ± 0.80%ID/g (100 μg), but with an increased uptake in the kidneys from 6.39 ± 2.25%ID/g to 13.60 ± 1.70%ID/g (p < 0.05, Mann-Whitney U nonparametric test).

Table 2.

Effect of Increased Amounts of ¹¹¹In-p-SCN-Bz-DTPA-14C5 and ¹¹¹In-p-SCN-Bz-DOTA-14C5 on the Uptake of Tracer in Pancreatic Tumor, Liver, Kidneys, and Spleen

	Uptake expressed as %ID/g (n = 3)
Dose (μg) at 72 h p.i.	Tumor	Liver	Kidneys	Spleen
¹¹¹In-p-SCN-Bz-DTPA-14C5
2	14.30 ± 3.76	8.85 ± 4.22	4.58 ± 0.73	3.63 ± 0.38
50	17.25 ± 5.78	10.42 ± 2.49	5.37 ± 1.15	4.43 ± 1.41
100	14.70 ± 3.83	10.00 ± 0.94	6.05 ± 1.10	3.55 ± 1.03
¹¹¹In-p-SCN-Bz-DOTA-14C5
2	35.84 ± 8.64	12.18 ± 4.40	6.39 ± 2.25	8.03 ± 1.84
50	27.77 ± 4.05	7.65 ± 1.32	6.92 ± 0.39	9.00 ± 1.50
100	32.16 ± 5.39	3.80 ± 0.66^*	13.60 ± 1.70^*	4.50 ± 0.80^*

%ID/g, n = 3; Data are presented as mean ± standard deviation.

Significant differences between uptake of 2 μg and 100 μg of ¹¹¹In-p-SCN-Bz-DOTA-14C5; p < 0.05 (Mann-Whitney U nonparametric test).

n.d., not done.

Planar γ camera imaging

Localization of radiolabeled mAb 14C5 in Swiss nu/nu mice bearing a Capan-1 tumor is presented in Figure 4, in which panel A shows ¹²³I-14C5, panel B shows ¹¹¹In-p-SCN-Bz-DTPA-14C5, and panel C shows ¹¹¹In-p-SCN-Bz-DOTA-14C5. For ¹²³I-14C5, no images were acquired after 48 hours because activity levels were too low for detection. For all radioimmunoconjugates, no tumor localization was evident at 3 hours after injection, with the image showing blood-pool activity and increased activity in the central portion of the scan corresponding to the cardiac and liver areas. At 24 hours, defined localization of mAb 14C5 to the Capan-1 tumor was seen. Although higher uptake in the liver area was seen for ¹¹¹In-conjugated 14C5 (Fig. 4B, C), consistent with biodistribution data, the Capan-1 tumor was clearly defined. Radioactivity was localized in the Capan-1 tumor, and via region of interest analysis, the tumor:contralateral-side ratios were determined. For all isotope conjugates, ratios exceeded 2 at 24 hours, with the highest ratio of 4.09:1 for ¹²³I-14C5 (48 hours), 3.77:1 for ¹¹¹In-p-SCN-Bz-DTPA-14C5, and 6.46:1 for¹¹¹In-p-SCN-Bz-DOTA-14C5 (both 96 hours).

FIG. 4.

Whole body planar images of 22.2 MBq ¹²³I-14C5 (A), ¹¹¹In-p-SCN-Bz-DTPA-14C5 (B), and ¹¹¹In-p-SCN-Bz-DOTA-14C5 (C). Images were taken at 3, 24, 48, 72, and 96 hours after injection of radiolabeled mAb 14C5. Animals were anesthetized by isoflurane, 5% for induction, 1%–2% rest. Thirty (30)–minute images were obtained with ¹²³I-14C5 and 15-minute images were obtained with ¹¹¹In-p-SCN-Bz-DTPA-14C5 and ¹¹¹In-p-SCN-Bz-DOTA-14C5. Tumors are indicated with arrowheads. n.d., not done.

Discussion

Pancreatic cancer has one of the poorest prognoses of all cancers, with no available curative treatment except surgical resection if possible.² 14C5 is a mAb that binds to the integrin α_vß,¹⁵ an antigen that is overexpressed on a variety of tumor cells, including pancreatic Capan-1 cells. This study investigated the use of mAb 14C5 to target radioisotopes specifically to pancreatic cancer. Labeling stability, in vitro binding capacity and internalization, and in vivo biodistribution of the mAb 14C5 were investigated, comparing iodine-, ¹¹¹In-p-SCN-Bz-DTPA-, and ¹¹¹In-p-SCN-Bz-DOTA-labeled mAb in a Capan-1 xenografted Swiss nu/nu mouse model.

Although the expression of integrin α_vß has been investigated extensively in breast-, colon-, and lung-cancer tissues,^11,12,15 its expression had not been examined in human pancreatic-cancer tissues. By immunohistochemistry, specific stromal uptake of mAb 14C5 was demonstrated in all tumor sections. An intensive staining was seen in the stroma surrounding the pancreatic tumor cells, but little Ag 14C5 was present on the tumor cells, similar to the data provided by Burvenich et al.¹⁵ Cancer-associated fibroblasts (CAFs) play an emerging role in the initiation and progression of cancer. These activated fibroblasts can enhance growth and metastasis of cancer cells by altering the ECM within the tumor stroma and changing the release and availability of growth factors.³⁶ The cell-surface serine protease known as FAP is a CAF-specific marker, and targeting FAP with sibrotuzumab in a phase I dose-escalation study showed specific uptake in the tumor sites without apparent side-effects.³⁷ Coexpression of the anti-FAP F19 antibody with mAb 14C5 in human pancreatic tumor sections shows that targeting the integrin α_vß on activated fibroblasts or PSCs with mAb 14C5 might serve as a new approach for cancer therapy. As described by Mahadevan and Von Hoff,¹⁷ activated fibroblasts could play an important role in the oncogenic process of pancreatic cancer, not only by altering the microenvironment of the tumor but also by altering the resistance to radiation and chemotherapy.³⁸ By not only targeting tumor cells but also CAFs, radioimmunotherapy with mAb 14C5 in pancreatic tumors could be a promising new approach for pancreatic cancer therapy.

The mAb 14C5 is a murine mAb that could potentially lead to immunogenic reactions when used in patients. Nowadays, chimerization and humanization techniques can be used to reduce immunogenicity^39,40 and even fully human antibodies can be generated.⁴¹ Efforts for genetically engineering mAb 14C5 to a chimeric and humanized mAb are currently undertaken.

mAb 14C5 could be efficiently labeled with iodine and ¹¹¹In in high yields and with high stability. In vitro experiments showed a small effect of conjugation with p-SCN-Bz-DTPA compared with p-SCN-Bz-DOTA on the affinity and immunoreactivity of mAb 14C5, with a twofold lower affinity for ¹¹¹In-p-SCN-Bz-DTPA-14C5 compared with ¹²⁵I- and ¹¹¹In-p-SCN-Bz-DOTA-14C5 and a decrease in immunoreactivity from 78% to 38% after 7 days at 37°C in mouse serum. Comparison with other mAbs used for targeting pancreatic carcinoma showed that mAb 14C5 has comparable affinity with hPAM4⁴² and a 2- to 20-fold higher affinity to pancreatic tumor cells compared with mAb AMB8LK (antiferritin), mAb DU-PAN 1, and mAb B20.4.1 and G6 (anti-VEGF).^3,43,44

Internalization of the 14C5–receptor complex was demonstrated in pancreatic tumor cells for all radioimmunoconjugates, but delayed retention times for ¹¹¹In-p-SCN-Bz-DTPA-14C5 and ¹¹¹In-p-SCN-Bz-DOTA-14C5 compared with ¹²⁵I-14C5 were shown, consistent with literature suggesting retention of ¹¹¹In-p-SCN-Bz-DTPA-lysine or ¹¹¹In-p-SCN-Bz-DOTA-lysine catabolites by tumor lysates.^25

–28 The prolonged intracellular retention of ¹¹¹In-labeled 14C5 leads to a long-lasting association between the radionuclide and the tumor cells, which would be beneficial not only for imaging but also for radioimmunotherapy of pancreatic cancer. Substituting 111-indium with the ß-emitter 90-ytrium or 177-lutetium could lead to an effective treatment of both bulky pancreatic tumors (⁹⁰Y with a ß⁻-energy range of 12.0 mm) as well as minimal residual disease (¹⁷⁷Lu with a ß⁻-energy range of 1.5 mm) with mAb 14C5. Nowadays, excellent results in imaging are also obtained with 124-iodine and 89-zirconium³³ and conjugation of these PET isotopes to mAb 14C5 could be promising in making a valuable PET tracer for pancreatic cancer.

In accordance with Press et al.,²⁹ the DTPA-conjugated antibody was more prone to passive dissociation. Because a higher number of p-SCN-Bz-DTPA molecules are coupled to 14C5 mAb compared with p-SCN-Bz-DOTA (1.5 for DTPA and 0.6 for DOTA), a critical lysine nearby the antigen-binding site could be occupied for the p-SCN-Bz-DTPA-conjugated 14C5 complex, leading to a decreased affinity and immunoreactivity. The lower stability of the ¹¹¹In-p-SCN-Bz-DTPA-14C5 mAb-Ag complex leads to a higher passive dissociation of ¹¹¹In-p-SCN-Bz-DTPA-14C5 and this consequently influences the internalization and trapping of the DTPA-conjugated mAb. The faster degradation of ¹²⁵I-14C5 and dissociation of ¹¹¹In-p-SCN-Bz-DTPA-14C5 could lead to a faster in vivo clearance and subsequently to a lower tumor uptake. Biodistributions with Capan-1 xenografts were conducted to compare the in vivo characteristics of the radioimmunoconjugates.

In the biodistribution, excellent and specific tumor uptake was observed for all mAb 14C5 conjugates compared with a nonspecific antibody (MAB002). A lower iodine-14C5 uptake compared with 111-indium-14C5 was seen in the Capan-1 xenografts, which has also been demonstrated in other studies,^45,46 and this, together with differences in tumor retention of the radiolabel, most likely represents dissimilarities in cellular processing as seen in the internalization study. The differences seen in the stability and in vitro experiments were also reflected in the in vivo results, with a significantly higher tumor uptake of ¹¹¹In-p-SCN-Bz-DOTA-14C5 compared with ¹³¹I- and ¹¹¹In-p-SCN-Bz-DTPA-14C5 (p < 0.05 at 48 and 72 hours, respectively). Sabbah et al.⁴ reported comparable tumor (12.8–23.6%ID/g) and blood (12.7–14.0%ID/g) values for an ¹¹¹In-p-SCN-Bz-DTPA-conjugated antiferritin mAb in Capan-1 tumor–bearing mice. However, tumor uptake for the p-SCN-Bz-DOTA-mAb was lower (11.2–14.1%ID/g), in contrast to the uptake of ¹¹¹In-p-SCN-Bz-DOTA-14C5 seen in the present experiments (24.7–35.8%ID/g). Also a significantly higher liver uptake (18.4–19.7%ID/g) was noted for the p-SCN-Bz-DOTA-conjugated antiferritin mAb compared with the p-SCN-Bz-DTPA-coupled mAb. Many different researchers^47

–50 have reported stable conjugations of mAbs with p-SCN-Bz-DTPA. However, it was not possible to establish this stability with p-SCN-Bz-DTPA-14C5.

In the present study, no differences in liver uptake were seen between both ¹¹¹In-conjugates in a low antibody concentration. Similar uptakes in liver and spleen have been published by several researchers,^7,48,49 and these results are consistent with known sequestration of radiometals in the reticuloendothelial organs. The higher uptake of ¹¹¹In-labeled 14C5 in the kidneys, compared with ¹³¹I-14C5, indicates differences in metabolism of radiometal conjugates as shown by different researchers.^18,45,46,51 Increasing the amount of 14C5 mAb reduced the liver and spleen uptake but elevated the kidney uptake for ¹¹¹In-p-SCN-Bz-DOTA-14C5, whereas no differences were seen for ¹¹¹In-p-SCN-Bz-DTPA-14C5. As postulated by Lub-de Hooge et al.,⁵² the lower stability of ¹¹¹In-p-SCN-Bz-DTPA-14C5 could cause transchelation to transferrin, leading to faster blood clearance and higher uptake in the liver. Knogler et al.⁵³ showed a decrease in liver uptake of p-SCN-Bz-DOTA-conjugated chCE7 antibody, with a decreasing ratio of chelator. The ¹¹¹In-p-SCN-Bz-DTPA-14C5 mixture has a higher number of chelator molecules compared with ¹¹¹In-p-SCN-Bz-DOTA-14C5, which could result in the high liver uptake at 100 μg with ¹¹¹In-p-SCN-Bz-DTPA-14C5. Preadministration of unlabeled mAb 14C5 did reduce the uptake of ¹¹¹In-p-SCN-Bz-DOTA-14C5 in the liver but not in the kidneys or spleen (data not shown). As shown by Morris et al.⁵⁴ and Pandit-Taskar et al.,⁵⁵ clearance of ¹¹¹In-mAb in serum and liver was dependent on antibody mass, with liver uptake being inversely proportional with mAb concentration. The reason for the increased uptake in the kidneys with increasing amounts of ¹¹¹In-p-SCN-Bz-DOTA-14C5 is unclear and should be further examined.

Uptake of radiolabeled 14C5 mAb in other organs reflects the blood flow and catabolization of the radiolabeled conjugates and is not a consequence of specific targeting attributable to the presence of 14C5 antigen expression.

Data from the biodistributions were confirmed in the planar γimaging, with ¹¹¹In-p-SCN-Bz-DOTA-14C5 showing the highest specific tumor uptake. Because of the longer half-life of 111-indium compared with 123-iodine, images could be taken even after 48 hours p.i., with the highest tumor-to-background ratio being 6.46:1 for ¹¹¹In-p-SCN-Bz-DOTA-14C5 at 168 hours p.i. The relatively high uptake of ¹¹¹In in the normal liver and the difficult anatomic location of the pancreatic lesions could hamper the diagnosis of pancreatic cancer. But, in patients, Pai-Scherf et al.⁵⁶ showed that even small liver metastasis gave a clear positive uptake against the ¹¹¹In-uptake of the normal liver. Future preclinical and clinical trials could elucidate the potential application of ¹¹¹In-conjugated 14C5 in the diagnosis of pancreatic adenocarcinoma. Nevertheless, radioimmunotherapy with p-SCN-Bz-DOTA-conjugated 14C5 could be very promising, considering the relative radioresistance of normal liver cells. Based on data from the present study, with a high and specific uptake of ¹¹¹In-p-SCN-Bz-DTPA-14C5 and ¹¹¹In-p-SCN-Bz-DOTA-14C5 in the pancreatic tumor and low uptake in normal tissues, ¹¹¹In-conjugated 14C5 and ¹¹¹In-p-SCN-Bz-DOTA-14C5 in particular might be an interesting mAb for scouting pancreatic tumors prior to RIT with ⁹⁰Y or ¹⁷⁷Lu. Because combinations of gemcitabine, the standard of care for advanced pancreatic cancer, with radionuclide therapy has shown very promising results,^57,58 and the possibility of therapy with gemcitabine and 14C5 mAb will be explored further.

Conclusions and Future Perspectives

This study shows that mAb 14C5 is a promising new antibody for targeting the integrin α_vß₅ in pancreatic cancer cells, not only in vitro but also in tumor-bearing xenografts. The superior affinity of mAb 14C5 to human pancreatic tumor cells was demonstrated in a cell-binding assay and in vivo biodistribution. An IgG1 isotype control antibody showed no specific targeting on human pancreatic tumor sections and pancreatic tumor cells in vitro as well as in vivo. Staining of human pancreatic tumor tissues showed intense staining of the tumor stroma and minimal to no staining of surrounding chronic inflammation areas. Radioimmunotherapy in which the CAFs are targeted with radiolabeled mAb 14C5 could be a promising new approach for therapy of pancreatic cancer, especially in combination with gemcitabine, a radiosensitizing chemotherapeutic targeting tumor cells.

Internalization experiments reveal a prolonged retention and association of ¹¹¹In to the tumor cells compared with ¹²⁵I-14C5, confirming reports from other investigators.^25

–29 Passive dissociation of ¹¹¹In-p-SCN-Bz-DTPA-14C5 as a result of lower stability of the mAb-Ag complex leads to a lower in vivo uptake in Capan-1 tumors compared with ¹¹¹In-p-SCN-Bz-DOTA-14C5. In the biodistributions, the superior internalization of ¹¹¹In-p-SCN-Bz-DOTA-14C5 is reflected in a significantly higher uptake of the radiometal in Capan-1 tumors compared with ¹³¹I-14C5 and ¹¹¹In-p-SCN-Bz-DTPA-14C5. The large liver uptake for the 111-indium-conjugated mAb 14C5 could hamper imaging of small primary pancreatic tumors. Therefore, it is recommended to use single photon emission computed tomography/computed tomography (SPECT/CT) scans in cases where tumors are located in the head to distinguish between hepatic uptake or tumor uptake and a prolonged clearance time after injection of ¹¹¹In-p-SCN-Bz-DOTA-14C5 could improve tumor-to-liver ratios. Changing the chelator to C-functionalized trans-cyclohexyl diethylenetriaminepentaacetic acid (CHX-A"-DTPA), a stable chelator for ¹¹¹In labeling as described by Tolmachev et al.,⁵⁹ could lower the hepatic background uptake, improving the imaging potential of ¹¹¹In-p-SCN-Bz-DOTA-14C5. For now, awaiting improved and more stabile chelator conjugations, mAb 14C5 and ¹³¹I-14C5 in particular are promising new tools in the detection of pancreatic cancer. Because of the low penetration of mAbs in solid tumors,⁶⁰ smaller IgG fragments such as F(ab′)₂ are promising tools in increasing the tumor penetration grade and hence radionuclide uptake. Burvenich et al.¹⁰ showed excellent tumor uptake of the bivalent F(ab′)₂ 14C5 fragment in a mouse A549 lung-tumor model, with improved tumor-to-background contrast images. The use of F(ab′)₂ 14C5 fragments in the diagnosis and treatment of pancreatic cancer should be explored.

For radioimmunotherapy, it is possible that targeting pancreatic cancer in patients with subacute or acute pancreatitis is hampered by the inflammation area frequently noted adjacent to pancreatic tumors, but further research has to be conducted to reveal the expression of α_vß₅ in these types of inflammation. As described by Sultana et al.,⁶¹ careful selection of patients prior to RIT with a pretherapy dose of mAb 14C5 could assist in the selection of patients with high enough tumor uptake before administration of therapeutic dosages. Supplementary immunologic research of pancreatic metastases could elucidate the possible application of mAb 14C5 in RIT in disseminated pancreatic adenocarcinoma.

To reduce radiotoxic damage to penetrating normal pancreatic glands or adjacent normal tissue, 14C5 variants conjugated with toxins or Auger electron emitters are interesting options. Because of the internalizing nature of mAb 14C5, nuclear localizing sequences could promote nuclear translocation and enhance the radiotoxicity of mAb 14C5 labeled with Auger electron emitters^62,63 targeting stromal fibroblast in the tumor and sparing surrounding normal tissue. Further investigation on the expression of the integrin α_vß₅ and its possible role as a marker for CAFs is necessary to understand fully the therapeutic and diagnostic potential of mAb 14C5.

Footnotes

Acknowledgments

The data and human normal and pancreatic tissues used in this project were provided by the Victorian Cancer Biobank with appropriate ethics approval. The Victorian Cancer Biobank is supported by the Victorian Government. The authors thank Dr. Peter Crowley, Director, and Dr. Siddhartha Deb, Senior Pathology Registrar, Department of Anatomical Pathology, Austin Health, for scoring the pancreatic tissues.

Disclosure Statement

No conflict of interest exists for any of the authors.

References

Ferlay

, Autier

, Boniol

et al. Estimates of the cancer incidence and mortality in Europe in 2006. Ann Oncol, 2007; 18:581.

, Xie

, Wolff

et al. Pancreatic cancer. Lancet, 2004; 363:1049.

, Song

, Li

et al. In vivo and in vitro inhibition of pancreatic cancer growth by targeted alpha therapy using ²¹³Bi-CHX.A′′-C595. Cancer Biol Ther, 2005; 4:848.

Sabbah

, Kadouche

, Ellison

et al. In vitro and in vivo comparison of DTPA- and DOTA-conjugated antiferritin monoclonal antibody for imaging and therapy of pancreatic cancer. Nucl Med Biol, 2007; 34:293.

Kimura

, Sawada

, Komatsu

et al. Antitumor effect of trastuzumab for pancreatic cancer with high HER-2 expression and enhancement of effect by combined therapy with gemcitabine. Clin Cancer Res, 2006; 12:4925.

Kigure

. Immunohistochemical study of the association between the progression of pancreatic ductal lesions and the expression of MUC1, MUC2, MUC5AC, and E-cadherin. Rinsho Byori, 2006; 54:447.

Gold

, Cardillo

, Goldenberg

et al. Localization of pancreatic cancer with radiolabelled monoclonal antibody PAM4. Crit Rev Oncol Hematol, 2001; 39:147.

Cruz-Monserrate

, O'Connor

. Integrin alpha 6 beta 4 promotes migration, invasion through Tiam1 upregulation, and subsequent Rac activation. Neoplasia, 2008; 10:408.

Longo

, Cacciamani

, Naso

et al. Pancreatic cancer: From molecular signature to target therapy. Crit Rev Oncol Hematol, 2008; 68:197.

10.

Burvenich

, Schoonooghe

, Blanckaert

et al. Biodistribution and planar gamma camera imaging of (123)I- and (131)I-labelled F(ab′)(2) and Fab fragments of monoclonal antibody 14C5 in nude mice bearing an A549 lung tumor. Nucl Med Biol, 2007; 34:257.

11.

De Potter

, Schelfhout

, De Smet

et al. A monoclonal antibody directed against a human cell membrane antigen prevents cell substrate adhesion and tumor invasion. Am J Pathol, 1994; 144:95.

12.

Coene

, Schelfhout

, De Ridder

et al. Generation of a monoclonal antibody directed against a human cell substrate adhesion molecule and the expression of the antigen in human tissues. Hybridoma, 1997; 16:77.

13.

Burvenich

, Schoonooghe

, Cornelissen

et al. In vitro and in vivo targeting properties of iodine-123- or iodine-131-labelled monoclonal antibody 14C5 in a non-small cell lung cancer and colon carcinoma model. Clin Cancer Res, 2005; 11:7288.

14.

Lahorte

, Burvenich

, Bacher

et al. Radiolabelling, biodistribution, and dosimetry of ¹²³I-mAb 14C5: A new mAb for radioimmunodetection of tumor growth and metastasis in vivo. J Nucl Med, 2004; 45:1065.

15.

Burvenich

, Schoonooghe

, Vervoort

et al. Monoclonal antibody 14C5 targets integrin α_vß₅ . Mol Cancer Ther, 2008; 7:3771.

16.

Grzesiak

, Ho

, Moossa

et al. The integrin-extracellular matrix axis in pancreatic cancer. Pancreas, 2007; 35:293.

17.

Mahadevan

, Von Hoff

. Tumor-stroma interactions in pancreatic ductal carcinoma. Mol Cancer Ther, 2007; 6:1186.

18.

Masamune

, Kikuta

, Watanabe

et al. Fibrinogen induces cytokine and collagen production in pancreatic stellate cells. Gut, 2009; 58:550.

19.

Omary

, Lugea

, Lowe

et al. The pancreatic stellate cell: A star on the rise in pancreatic diseases. J Clin Invest, 2007; 117:50.

20.

Jaster

. Molecular regulation of pancreatic stellate cell function. Mol Cancer, 2004; 3:26.

21.

Brooks

, Klemke

, Schön

et al. Insulin-like growth factor receptor cooperates with integrin α_vß₅ to promote tumor cell dissemination in vivo. J Clin Invest, 1997; 99:1390.

22.

Bachem

, Schunemann

, Ramadani

et al. Pancreatic carcinoma cells induce fibrosis by stimulating proliferation and matrix synthesis of stellate cells. Gastroenterology, 2005; 128:907.

23.

Rosendahl

, Neumann

, Chaloupka

et al. Expression and distribution of VLA receptors in the pancreas: An immunohistochemical study. Pancreas, 1993; 8:711.

24.

Gleason

, Adley

, Sambasiva

et al. Immunohistochemical detection of the ß4 integrin subunit in pancreatic adenocarcinoma. J Histochem Cytochem, 2005; 53:799.

25.

Duncan

, Welch

. Intracellular metabolism of indium-111-DTPA-labelled receptor targeted proteins. J Nucl Med, 1993; 34:1728.

26.

Smith-Jones

, Pandit-Taskar

, Cao

et al. Preclinical radioimmunotargeting of folate receptor alpha using the monoclonal antibody conjugate DOTA-MORAb-003. Nucl Med Biol, 2008; 35:343.

27.

Shih

, Thorpe

, Griffiths

et al. The processing and fate of antibodies and their radiolabels bound to the surface of tumour cells in vitro: A comparison of nine radiolabels. J Nucl Med, 1994; 35:899.

28.

Kyriakos

, Shih

, Ong

et al. The fate of antibodies bound to the surface of tumor cells in vitro. Cancer Res, 1992; 52:835.

29.

Press

, Shan

, Howell-Clark

et al. Comparative metabolism and retention of iodine-125, yttrium-90, and indium-111 radioimmunoconjugates by cancer cells. Cancer Res, 1996; 56:2123.

30.

Verel

, Visser

GWM

, van Dongen

. The promise of immuno-PET in radioimmunotherapy. J Nucl Med, 2005; 46:164S.

31.

Garin-Chesa

, Old

, Rettig

. Cell surface glycoprotein of reactive stromal fibroblasts as a potential antibody target in human epithelial cancers. Proc Natl Acad Sci USA, 1990; 87:7235.

32.

Liu

, Panousis

, Smyth

et al. Generation of anti-idiotype antibodies for application in clinical immunotherapy laboratory analysis. Hybrid Hybridomics, 2003; 22:219.

33.

Perk

, Visser

, Vosjan

et al. (89)Zr as a PET surrogate radioisotope for scouting biodistribution of the therapeutic radiometals (90)Y and (177)Lu in tumor-bearing nude mice after coupling to the internalizing antibody cetuximab. J Nucl Med, 2005; 46:1898.

34.

Meares

, McCall

, Reardan

et al. Conjugation of antibodies with bifunctional chelating agents: Isothiocyanate and bromoacetamide reagents, methods of analysis, and subsequent addition of metal ions. Anal Biochem, 1984; 142:68.

35.

Lindmo

, Boven

, Cuttitta

et al. Determination of the immunoreactive fraction of radiolabelled monoclonal antibodies by linear extrapolation to binding at infinite antigen excess. J Immunol Methods, 1984; 72:77.

36.

Kalluri

, Zeisberg

. Fibroblasts in cancer. Nat Rev Cancer, 2006; 6:392.

37.

Scott

, Wiseman

, Welt

et al. A phase I dose-escalation study of sibrotuzumab in patients with advanced or metastatic fibroblast activation protein-positive cancer. Clin Cancer Res, 2003; 9:1639.

38.

Neoptolemos

, Stocken

, Friess

et al. A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med, 2004; 350:1200.

39.

Koppe

, Postema

, Aarts

et al. Antibody-guided radiation therapy of cancer. Cancer Metastasis Rev, 2005; 24:539.

40.

Dübbel

. Handbook of Therapeutic Antibodies. I Technologies. Darmstadt: Wiley-VCH, 2007.

41.

Vaughan

, Williams

, Pritchard

et al. Human antibodies with sub-nanomolar affinities isolated from a large non-immunized phage display library. Nat Biotechnol, 1996; 14:309.

42.

Gold

, Goldenberg

, Karacay

et al. A novel bispecific, trivalent antibody construct for targeting pancreatic carcinoma. Cancer Res, 2008; 68:4819.

43.

Worlock

, Zalutsky

, Metzgar

. Radiolocalization of human pancreatic tumors in athymic mice by monoclonal antibody DU-PAN 1. Cancer Res, 1990; 50:7246.

44.

Liang

, Wu

, Peale

et al. Cross-species vascular endothelial growth factor (VEGF)–blocking antibodies completely inhibit the growth of human tumor xenografts and measure the contribution of stromal VEGF. J Biol Chem, 2006; 281:951.

45.

Clarke

, Lee

, Brechbiel

et al. In vivo biodistribution of humanized anti-Lewis Y monoclonal antibody (hu3S193) in MCF-7 xenografted BALB/c nude mice. Cancer Res, 2000; 60:4804.

46.

Sharkey

, Motta-Henessy

, Pawlyk

et al. Biodistribution and radiation dose estimates for yttrium- and iodine-labelled monoclonal antibody IgG and fragments in nude mice bearing human colonic tumour xenografts. Cancer Res, 1990; 50:2330.

47.

Michel

, Ochakovskaya

, Mattes

. Antibody localization to B-cell lymphoma xenografts in immunodeficient mice: Importance of using residualizing radiolabels. Clin Cancer Res, 2002; 8:2632.

48.

Behr

, Béhé

, Stabin

et al. High-linear energy transfer (LET) a versus low-LET b emitters in radioimmunotherapy of solid tumors: Therapeutic efficacy and dose-limiting toxicity of 213Bi- versus 90Y-labelled CO17-1A Fab′ fragments in a human colonic cancer model. Cancer Res, 1999; 59:2635.

49.

Brouwers

, Buijs

, Oosterwijk

et al. Targeting of metastatic renal cell carcinoma with the chimeric monoclonal antibody G250 labelled with ¹³¹I or ¹¹¹In: An intrapatient comparison. Clin Cancer Res, 2003; 9,suppl:3953S.

50.

Brouwers

, van Eerd

JEM

, Frielink

et al. Optimization of radioimmunotherapy of renal cell carcinoma: Labelling of monoclonal antibody cG250 with ¹³¹I, ⁹⁰Y, ¹⁷⁷Lu, or ¹⁸⁶Re. J Nucl Med, 2004; 45:327.

51.

DeNardo

, Motta-Henessy

, Pawlyk

et al. Pharmacokinetics of chimeric L6 conjugated to indium-111- and yttrium-90-DOTA-peptide in tumour-bearing mice. J Nucl Med, 1995; 36:829.

52.

Lub-de Hooge

, Kosterink

JGN

, Perik

et al. Preclinical characterisation of ¹¹¹In-DTPA-trastuzumab. Br J Pharmacol, 2004; 143:99.

53.

Knogler

, Grünberg

, Novak-Hofer

et al. Evaluation of ¹⁷⁷Lu-DOTA-labelled a glycosylated monoclonal anti-L1-CAM antibody chCE7: Influence of the number of chelators on the in vitro and in vivo properties. Nucl Med Biol, 2006; 33:883.

54.

Morris

, Divgi

, Pandit-Taskar

et al. Pilot trial of unlabeled and indium-111-labeled anti-prostate-specific membrane antigen antibody J591 for castrate metastatic prostate cancer. Clin Cancer Res, 2005; 11:7454.

55.

Pandit-Taskar

, O'Donoghue

, Morris

et al. Antibody mass escalation study in patients with castration-resistant prostate cancer using ¹¹¹In-J591: Lesion detectability and dosimetric projections for ⁹⁰Y radioimmunotherapy. J Nucl Med, 2008; 49:1066.

56.

Pai-Scherf

, Carrasquillo

, Paik

et al. Imaging and phase I study of ¹¹¹In- and ⁹⁰Y-labeled anti-Lewis^Y monoclonal antibody B3. Clin Cancer Res, 2000; 6:1720.

57.

Kang

, Saif

. Pharmacogenomics and pancreatic cancer treatment. Optimizing current therapy and individualizing future therapy. Pancreas, 2008; 9:251.

58.

Gold

, Modrak

, Schutsky

et al. Combined ⁹⁰Yttrium-DOTA-labelled PAM4 antibody radioimmunotherapy and gemcitabine radiosensitization for the treatment of a human pancreatic cancer xenograft. Int J Cancer, 2004; 109:618.

59.

Tolmachev

, Wållberg

, Andersson

et al. The influence of Bz-DOTA and CHX-A′′-DTPA on the biodistribution of ABD-fused anti-HER2 affibody molecules: Implications for (114m)In-mediated targeting therapy. Eur J Nucl Med Mol Imaging, 2009; 36:1460.

60.

Wong

JYC

. Basic immunology of antibody targeted radiotherapy. Int J Radiat Oncol, 2006; 66:S8.

61.

Sultana

, Shore

, Raraty

MGT

et al. Randomised phase I/II trial assessing the safety and efficacy of radiolabelled anti-carcinoembryonic antigen I131 KAb201 antibodies given intra-arterially or intravenously in patients with unresectable pancreatic adenocarcinoma. BMC Cancer, 2009; 9:66.

62.

Chen

, Wang

, Hope

et al. Nuclear localizing sequences promote nuclear translocation and enhance the radiotoxicity of the anti-CD33 monoclonal antibody HuM195 labelled with ¹¹¹In in human myeloid leukemia cells. J Nucl Med, 2006; 47:827.

63.

Constantini

, Chen

, Cai

et al. (111)In-labelled trastuzumab (Herceptin) modified with nuclear localization sequences (NLS): An Auger electron-emitting radiotherapeutic agent for HER2/neu-amplified breast cancer. J Nucl Med, 2007; 48:1357.

Preclinical Evaluation of Monoclonal Antibody 14C5 for Targeting Pancreatic Cancer

Abstract

Introduction

Materials and Methods

Monoclonal antibodies

Cell lines

Immunohistochemistry

Radiolabeling

Preparation of iodine-labeled 14C5

Preparation of 111In-labeled 14C5

Quality control

Immunoreactivity and in vitro stability

Saturation-binding assay

Internalization experiments

Biodistribution and pharmacokinetic studies

Statistical analysis

Results

Immunohistochemistry

Radiolabeling, stability, and immunoreactivity

Saturation-binding assay in Capan-1 cells

Internalization experiments

Biodistribution in Capan-1 tumor-bearing mice

Planar γ camera imaging

Discussion

Conclusions and Future Perspectives

Footnotes

Acknowledgments

Disclosure Statement

References

Preparation of ¹¹¹In-labeled 14C5