Clinical Cancer Research Vol. 12, 2191-2196, April 2006
© 2006 American Association for Cancer Research
Cancer Therapy: Preclinical |
Low and High Tenascin-Expressing Tumors Are Efficiently Targeted by ST2146 Monoclonal Antibody
Rita De Santis1,
Claudio Albertoni1,
Fiorella Petronzelli1,
Silvia Campo1,
Valeria D'Alessio1,
Antonio Rosi1,
Anna Maria Anastasi1,
Ragnar Lindstedt1,
Nadia Caroni1,
Brunilde Arseni1,
Pierino Chiodi1,
Antonio Verdoliva2,
Giovanni Cassani2,
Marco Chinol3,
Giovanni Paganelli3 and
Paolo Carminati1
Authors' Affiliations: 1 Sigma-Tau SpA, R&D, Rome, Italy; 2 Tecnogen SCpA, Località "La Fagianeria," Caserta, Italy; and 3 European Institute of Oncology, Milan, Italy
Requests for reprints: Rita De Santis, Immunology Department R&D, Sigma-Tau SpA, Via Pontina, Km 30.400, 00040 Pomezia, Rome, Italy. Phone: 39-06-9139-4283; Fax: 39-06-9139-3988; E-mail: rita.desantis{at}sigma-tau.it.
 |
Abstract
|
|---|
ST2146biot is a biotinylated anti-tenascin monoclonal antibody (mAb) to be used for Pretargeted Antibody Guided Radioimmunotherapy (PAGRIT) of solid tumors. In vivo biodistribution studies of 125I-labeled ST2146biot were done in nude mice transplanted with human HT-29 colon carcinoma and/or human U-118MG glioblastoma cells characterized for low and high tenascin expression, respectively. In vitro results show that ST2146 retains immunoreactivity upon biotinylation, in contrast to other anti-tenascin mAbs. In vivo biodistribution of ST2146 shows specific tumor accumulation up to 10 days after the i.v. injection, with no relevant differences between biotinylated and nonbiotinylated ST2146. A dose of 4 µg/mouse saturates the low tenascin-expressing human colon carcinoma HT-29, whereas the high tenascin-expressing human glioblastoma U-118MG seems to be saturated at a ST2146biot dose between 320 and 640 µg/mouse. The percentage of injected dose per gram of tumor ranges from 10% to 30%, corresponding to an amount of ST2146biot/g of tumor of 
400 ng/g and >200 µg/g for HT-29 and U-118MG, respectively. Tumor to normal organs uptake ratios are between 15 and 60, confirming high tumor selectivity of ST2146biot despite its cross-reactivity with the tenascin expressed at low level in the normal mouse organs. The ST2146biot localization data are substantially confirmed even when both low and high tenascin-expressing tumors are implanted in the same animal. To our knowledge, the absolute amount of ST2146biot, specifically localized in xenotransplanted human tumors, is the highest thus far described and supports the clinical use of this mAb in PAGRIT®.
The Pretargeted Antibody-Guided Radioimmunotherapy (PAGRIT) is based on i.v. sequential administration of a monoclonal antibody (mAb) biotinylated, avidin/streptavidin, and 90Y-labeled biotin. The main objective of pretargeting is to maximize the accumulation of the radioisotope at the tumor site while minimizing its distribution to nontarget organs. Therefore, the potency and selectivity of the mAb are essential prerequisites for optimal pretargeting. Proof of the safety and efficacy of PAGRIT has previously been shown in patients with high-grade glioma (1, 2) by using the anti-tenascin mAbs BC2 and BC4 (3, 4). These mAbs were found unsuitable for industrial development due to the high contamination by a nonfunctional light chain overproduced during high-density culture of the hybridomas in bioreactor. Second-generation antibodies with improved purity and potency have recently shown efficacy in preclinical models of tumor pretargeting, singularly or in combination (5, 6). The ST2146 mAb binds human tenascin at an epitope closely related to that of BC4. Compared with BC4, it showed similar selectivity but higher affinity and immunoreactivity. Biodistribution studies of biotinylated ST2146 (ST2146biot) and three other anti-tenascin mAbs showed the highest specific tumor localization of ST2146 in HT-29–grafted nude mice (5). For pretargeting applications, biotinylated mAbs are to be used at the optimal biotin/mAb ratio giving the highest avidin/streptavidin localization while preserving antibody immunoreactivity. In the present work, ST2146biot was analyzed in vitro for immunoreactivity and in vivo for the capacity to specifically localize to xenotransplanted human tumors expressing low or high level of tenascin. The biodistribution data of ST2146biot versus biotinylated normal mouse immunoglobulins (IgGbiot) showed high specific ST2146 tumor localization in both low and high tenascin-expressing models. Moreover, because ST2146 cross-reacts with mouse tenascin, the animal models adopted in the present study are relevant to the clinical PAGRIT perspective resembling the human condition of normal tissues expressing basal levels of tenascin and most tumors overexpressing tenascin in the stroma (7).
|
Materials and Methods
|
|---|
mAbs. BC4 anti-tenascin mAb (3, 4) was kindly provided by Dr. Luciano Zardi (Laboratory of Cell Biology, National Institute for Cancer Research, Genoa, Italy). ST1897 (original code 300-2) was obtained from Wistar Institute, Philadelphia, PA (8). ST2146 was produced by cultivation of cST2146 hybridoma cells in protein-free medium, in a 2-L perfusion bioreactor (MD2, B. Braun, Allentown, PA). Purification was done by three chromatographic steps, including MEP HyperCell (BioSepra, Fremont, CA), Q-Sepharose-XL, and SP-Sepharose-XL chromatography (Pharmacia, Milan, Italy). Normal mouse immunoglobulins were purchased from Sigma-Aldrich (St. Louis, MO).
Biotinylation and immunoreactivity/affinity evaluations. mAbs were biotinylated by reaction with 2X-AHBiotin-N-Hydroxysuccinimide ester (BioSPA, Milan, Italy) at the indicated biotin/mAb molar ratios. The biotinylation degree was determined by using the HABA test (9). For immunoreactivity evaluation, Immuno MAXISORP 96-well plates (Nunc, Rochester, NY) were coated at 4°C overnight with 100 µL/well of 5 or 0.5 µg/mL tenascin in PBS (pH 7.2). Tenascin was purified from the culture supernatant of the melanoma cell line SK-MEL-28 as previously described (10). The tenascin-coated plates were washed once with PBS and 0.1% Tween 20 (washing buffer) and blocked with 300 µL/well of PBS, 0.1% Tween 20, and 1% bovine serum albumin (blocking and diluting buffer) for 2 hours at room temperature. Plates were used immediately or dried and frozen at –20°C. The plates were incubated, in duplicate, with serial dilutions of biotinylated or nonbiotinylated anti-tenascin mAbs (100 µL/well) for 1.5 hours at 37°C. After three washings, the plates were incubated 1.5 hours at 37°C with 100 µL/well of alkaline phosphatase–conjugated anti-mouse IgG (Fc specific) antibody (Sigma-Aldrich) diluted 1:1,000 in blocking buffer. The plates were then washed four times, incubated with 200 µL/well of 4-nitrophenylphosphate (Sigma-Aldrich) for 30 minutes at 37°C, stopped with 100 µL/well of 3 mol/L NaOH, and read with an ELISA spectrophotometer (SEAC Sirio S, Florence, Italy) at 405 nm.
For affinity evaluations, a BIAcoreX instrument (Biosense, Milan, Italy) was used. Tenascin was immobilized on a CM5 sensorchip as previously described (6). The final immobilization response was 3,200 resonance units. Sensorgrams were generated by the injection of mAbs (at five concentrations with range of 3.9-500 nmol/L, in duplicate) at a flow rate of 30 µL/min, using 100 mmol/L NaOH as regeneration solution. The association and dissociation times were 60 and 120 seconds, respectively. Biosensor data were prepared, modeled, and fitted by means of BIAevaluation 3.1 software (Biosense). Data evaluation was done using a bivalent analyte model with simultaneous determination of association and dissociation constants.
Immunohistochemistry. Immunohistochemistry studies were done on cryosections and paraffin-embedded tissue slides (Super Biochips Laboratories, Seoul, Korea). Briefly, after deparaffination and hydration, the tissue sections were blocked with PBS and 1.5% normal horse serum (blocking solution). Slides were then incubated overnight at 4°C with ST2146 at 2 to 5 µg/mL in blocking solution. After three washings with PBS and incubation for 30 minutes with biotinylated goat anti-mouse antibodies, the slides were incubated for 30 minutes with an avidin-biotin-peroxidase complex. After three washings with PBS, diaminobenzidine substrate was added and the reaction stopped after 2 minutes by washing in tap water. Counterstaining was done with Mayer's hematoxylin for 10 seconds. Negative controls included slides incubated with a nonrelevant isotype-matched antibody or with the second antibody alone.
Tenascin titration. Tumors and normal organs from xenotransplanted mice were collected and kept at –80°C until protein extraction that was done as previously described (11, 12) with some modifications. After a brief washing with 20 mmol/L Tris, 150 mmol/L NaCl, 1 mmol/L EDTA (pH 7.4), each tissue was homogenized with 5 mL of extraction buffer [100 mmol/L CAPS, 0.5 mol/L NaCl (pH 11)]/g, in a PRO200 homogenizer (PBI, Milan, Italy) and mixed overnight at 4°C. The homogenate was then centrifuged at 16,000 x g for 30 minutes, and the pellet was reextracted with 3.3 mL/g of tissue of extraction buffer, mixing 5 hours at 4°C. The supernatants from the two extractions were brought to pH 7.5 with 2 mol/L NaH2PO4, and stored at –20°C until further analysis. A sandwich ELISA with ST2146 for capture and a secondary biotinylated anti-tenascin antibody, ST2077, was used for determination of tenascin. After an overnight coating with 5 µg/mL ST2146 in TBS [20 mmol/L Tris, 0.15 mol/L NaCl, 0.005% Thimerosal (pH 7.8)], plates were washed with TBS and 0.1% Triton X-100 (washing buffer) and then blocked with TBS, 0.1% Triton X-100, 0.5% casein (pH 7.8; blocking and diluent solutions) for 2 hours at room temperature. Samples of the protein extracts, diluted in diluent buffer, were added to the ST2146-coated plates and incubated for 2 hours at room temperature. After three washings, ST2077biot (biotinylated with the enhanced chemiluminescence protein biotinylation module kit from Amersham Biosciences, Uppsala, Sweden) was added to each well. After 1.5 hours of incubation, streptavidin-peroxidase and then tetramethylbenzidine substrate were used for detection. The plate was read at 450 nm. Tenascin from SKMEL28 was used as a standard. A 78% recovery was observed for tenascin added to protein lysates of normal organs.
Antibody radiolabelling and in vivo biodistribution. Biotinylated antibodies (5-20 µg)/100 µL of TBS (pH 7.4) were mixed with 50 to 200 µCi of Na125I according to the Iodo-Gen method (Pierce, Rockford, IL). Labeled antibodies were separated from free iodine by chromatography on a PD-10 column (Amersham Biosciences). Immunoreactivity of radiolabeled antibodies was checked by ELISA as previously described in plates coated with tenascin at 1.0 µg/mL. BALB/c nu/nu mice (Charles River, Lecco, Italy) underwent s.c. transplant of 5 x 106 HT-29 human colon carcinoma cells (DSMZ, Braunschweig, Germany) in 0.1 mL PBS or of 10 x 106 U-118MG human glioblastoma (American Type Culture Collection, Manassas, VA) in 0.1 to 0.2 mL sterile saline (pH 7.0), 2 to 3 weeks before treatment. Before antibody administration, mice were randomized into treatment groups (5-10 animals per group) and i.v. injected with different doses of 125I-radiolabeled biotinylated antibodies (1-4 x 106 cpm/animal). Normal mouse immunoglobulins (Sigma-Aldrich) were used as control. At fixed times, mice were sacrificed by cervical dislocation, and blood, spleen, kidney, liver, and tumor mass were collected. Each tissue was weighed and counted in a gamma-counter (Canberra Packard, Schwadorf, Austria). Data were expressed as the percentage of the injected dose/gram of tissue (%ID/g) or ng of antibody/g of tissue. Tumor versus normal tissue ratios were calculated by dividing ng of antibody/g of tumor by ng of antibody/g of blood, spleen, kidney, or liver. Localization indices were calculated by dividing tumor/normal tissue ratios to the ST2146biot groups by those of corresponding IgGbiot groups. In studies where an avidin chase was done, an avidin amount 100x the antibody dose was i.v. administered, and animals were sacrificed 24 hours later.
The care and husbandry of animals were in accordance with the European Directive 86/609 and Italian legislation.
Statistical analysis. Experiments were carried out in duplicate and repeated twice. The results were expressed as mean ± SE. Student's t test for unpaired data or one-way ANOVA followed by Student-Neuman-Keuls multiple comparison test was used for significance testing of the data, using a commercially available software, SigmaStat (Systat Software, Inc., Richmond, CA). A difference of P < 0.05 was considered statistically significant.
|
Results
|
|---|
Immunoreactivity of biotinylated ST2146. The immunoreactivity of biotinylated ST2146, BC4, and ST1897 anti-tenascin mAbs, exhibiting 1.7, 6.3, and 20.3 nmol/L affinity, respectively, was evaluated by ELISA. Results indicate that a low biotinylation level (2-3 biotins per mAb) slightly affects the immunoreactivity of ST2146, BC4, and ST1897 (range, 80-100%). At higher biotinylation degrees, up to 20 biotins per mAb, the immunoreactivity reduction is higher when the affinity of the nonbiotinylated antibody is lower. Particularly, the high-affinity ST2146 can be conjugated to up to 20 biotins per mAb, maintaining >50% immunoreactivity, whereas the lower-affinity anti-tenascin mAbs ST1897 and BC4 retain about 10% and 15% immunoreactivity, respectively (data not shown).
ST2146 cross-reactivity with mouse tenascin. In the perspective of investigating the biodistribution of ST2146 in nude mice xenografted with human tumors, the possible cross-reactivity of ST2146 with mouse tenascin was evaluated by immunohistochemistry. Figure 1, top
shows that ST2146 weakly stains the normal human intestine (A) but strongly stains a human breast tumor (B). Figure 1, bottom shows that ST2146 also stains the normal mouse intestine (C) and the LMM3 mouse breast carcinoma (D). Similar staining patterns are obtained with the commercial rat anti-mouse tenascin antibody ab6364 (Abcam, Milton Road, Cambridge, United Kingdom; data not shown). These data indicate that ST2146, generated against human tenascin, cross-reacts with mouse tenascin.
View larger version (128K):
[in this window]
[in a new window]
|
Fig. 1. ST2146 stains formalin-fixed and paraffin-embedded sections of human normal colon (A) and human breast carcinoma (B) as well as mouse normal colon (C) and mouse breast carcinoma (D). Isotype-matched control was negative (data not shown).
|
|
Characterization of low and high tenascin-expressing xenografts. To characterize the animal models to be used for evaluating the tumor pretargeting efficiency of biotinylated ST2146, xenotransplants of human colon carcinoma HT-29 or human glioblastoma U-118MG were done in nude mice. The tenascin content in xenotransplanted tumor masses was evaluated by immunohistochemistry and ELISA. Figure 2
shows that the immunohistochemical staining of HT-29 (A) seems weaker than that of U-118MG (B). In agreement with this observation, the tenascin values in excised tumor masses, reported in Table 1
, indicate that HT-29 and U-118MG xenotransplants contain 1 and 500 µg of tenascin/g of tumor, respectively (P < 0.001). The tenascin content of normal tissues, including blood, spleen, lung, brain, and colon, was <0.006 µg/g, whereas in normal heart and kidney, it was <0.14 µg/g. Therefore, HT-29 xenotransplant expresses 7 to 170 times the tenascin present in normal tissues, whereas the overexpression in U-118MG xenotransplant is 4.000 to 95.000 times.
View larger version (78K):
[in this window]
[in a new window]
|
Fig. 2. ST2146 staining of human colon carcinoma HT-29 (A) and human gliobalstoma U-118MG (B) cryosections from nude mice xenotransplants. Isotype-matched control was negative (data not shown).
|
|
Biodistribution of biotinylated ST2146 in low and high tenascin-expressing tumors. To evaluate the effect of biotinylation on the in vivo behavior of ST2146, a preliminary animal study was done in BALB/c nude mice implanted with the low tenascin-expressing HT-29 human colon carcinoma to compare the biodistribution of nonbiotinylated ST2146 with biotinylated ST2146. On the day of treatment, groups received a single i.v. administration of 125I-labeled ST2146 nonbiotinylated or biotinylated with 7.8 or 18.5 biotins per mAb. At days +2, +5, +7, and +10 after injection, the radioactivity in different organs and tumor was determined by gamma-counter and correlated to the total radioactivity injected. At day 2 after injection, the accumulation of either nonbiotinylated ST2146 or biotinylated ST2146 was about 10% of ID/g of tumor, 8% for blood and 2.5 to 3.5% for spleen, kidney, or liver. A time-dependent reduction of nonbiotinylated and biotinylated ST2146 was observed from day 2 up to day 10 in all nonspecific organs, whereas localization in the tumor was persistent. Ten days after injection, tumor localization of ST2146 and related biotinylated derivatives remained substantially unchanged, whereas in all nonspecific organs, the amount was significantly reduced, thus leading to tumor/blood ratio of >6 and tumor/spleen, kidney, or liver of >20 to 30 (data not shown).
In the attempt to find the optimal dose of biotinylated ST2146 for tumor pretargeting, biodistribution studies were done in nude mice transplanted with either HT-29 human colon carcinoma or U-118MG human glioblastoma cells. As shown in Fig. 3A
, ST2146biot and mouse IgGbiot are distributed, in a dose-dependent manner, in HT-29 tumor. At the dose of 4 µg/mouse, the amount of ST2146biot/g of tumor was 576 ng, whereas at doses of 2, 6, 8, and 10 µg/mouse it was 241, 439, 622, and 472, respectively. A dose of 4 µg/mouse seems to saturate HT-29 tumors, because at higher doses, the amount of antibody/g of tumor is not significantly increased, whereas the tumor/normal tissues ratios (Fig. 3B) as well as the localization indices (Fig. 3C) are lower, indicating nonspecific binding. The dose range of 2 to 10 µg/mouse used to identify the HT-29 saturating dose was not suitable for U-118MG, which was not saturated even when ST2146 was administered in a higher dose range of 10 to 80 µg (data not shown). Increased ST2146 doses were tested reaching 80 to 640 µg/mouse, and the results in Fig. 4A
show that ST2146biot accumulates in the U-118MG xenotransplants in a linear, dose-dependent manner, reaching about 200 µg/g of tumor. Tumor localization is highly specific as shown by the marginal localization of normal mouse IgGbiot and the low amount of ST2146 in nontarget organs, which yielded tumor/nontumor ratios between 45 and 93 in the dose range tested (Fig. 4B). The localization indices in Fig. 4C suggest that the saturating dose of ST2146biot, in this glioblastoma animal model, might be between 320 and 640 µg/mouse.
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 3. A, absolute amount of 125I-ST2146biot or 125I-IgGbiot at the indicated doses, localized in human colon carcinoma HT-29 xenotransplant. Columns, average ng of ST2146biot/g of tumor (ng/g) of five animals per group; bars, SE. *, 0.05 > P > 0.01; **, 0.009 > P > 0.001; ***, P < 0.001. B, tumor versus normal tissue ratios expressed as the ratio between ng of ST2146biot/g of tumor and ng of ST2146biot/g of blood, spleen, kidney, or liver. C, localization indices calculated by dividing tumor/normal tissue ratios to the ST2146biot groups by those of corresponding IgGbiot groups.
|
|
View larger version (36K):
[in this window]
[in a new window]
|
Fig. 4. A, absolute amount of 125I-ST2146biot or 125I-IgGbiot at the indicated doses, localized in human glioblastoma U-118MG xenotransplant. Columns, average ng of ST2146biot/g of tumor (ng/g) of five animals per group; bars, SE. *, 0.05 > P > 0.01; **, 0.009 > P > 0.001; ***, P < 0.001. B, tumor versus normal tissue ratios expressed as the ratio between ng of ST2146biot/g of tumor and ng of ST2146biot/g of blood, spleen, kidney, or liver. C, localization indices calculated by dividing tumor/normal tissue ratios to the ST2146biot groups by those of corresponding IgGbiot groups.
|
|
A further biodistribution study was done to evaluate ST2146 targeting efficiency of low and high antigen-expressing tumors when they coexist within the same host. Therefore, HT-29 and U-118MG were implanted in two sites of the same animal, and tumor as well as normal tissue ST2146 localization was evaluated as previously described. Data reported in Table 2
indicate that ST2146biot at a dose of 500 µg/mouse (dose range, 4-3,000 µg/mouse) effectively targets both HT-29 and U-118MG tumors, resulting in >5 and 100 µg/g of tumor, respectively. These localization data are associated with tumor/blood ratios, which are >4 and 100, respectively. The tumor versus other normal organ ratios were all superior to 1 and 30, respectively (data not shown). Overall results indicate that ST2146biot can efficiently and specifically target both low and high tenascin-expressing tumors even within the same host.
|
Discussion
|
|---|
Five of the eight mAbs registered thus far for cancer therapy are indicated for the treatment of hematologic malignancies, and the following three for carcinomas: Herceptin from Genentech (South San Francisco, CA) for metastatic breast cancer; Erbitux from ImClone (Branchburg, NJ) for metastatic colorectal cancer; Avastin from Genentech for metastatic colorectal cancer. The low number of antibodies registered for treatment of solid tumors reflects the difficulty in curing such tumors, which are less accessible and more heterogeneous than blood tumors. Herceptin, which targets HER-2/neu on cancer cells, showed the highest efficacy in clinical trials in patients with high level of HER-2 expression (<20% of all breast cancer patients), resulting in an overall response rate (ORR) of 17% when administered alone and in an improvement of 30% of the 14% ORR of Paclitaxel when combined (ORR 44%; Herceptin package insert from Food and Drug Administration web site). Erbitux, which is directed to the epidermal growth factor receptor, produced an ORR of about 14% as monotherapy and of about 25% when combined with Irinotecan (Erbitux package insert from Food and Drug Administration web site). Avastin, which recognizes the vascular endothelial growth factor, produced an improvement of 10% of the 35% ORR of chemotherapy when administered in combination (Avastin package insert from Food and Drug Administration web site). All three mAb therapeutic regimens registered an average improvement of time to progression between 2 and 10 months (package inserts from Food and Drug Administration web site). It is becoming evident that additional strategies are needed to exploit the therapeutic potential of mAbs. In a recent review, the prospective use of radioimmunotherapy is emphasized, based on the clinical success obtained with the two radiolabeled murine mAbs 90Y-Zevalin, (IDEC Pharmaceuticals Corp., San Diego, CA) and 131I-Bexxar, (GSK Research Triangle Park, NC) in the treatment of B-cell lymphomas (13). As for naked antibodies and chemotherapy, radioimmunotherapy was also less effective against solid tumors than against blood tumors. However, several preclinical studies provide guidelines for a prospective use of radioimmunotherapy in solid tumors, indicating that efficacy might be greatly improved in patients with minimal residual disease or small tumors (14). The use of antibody combinations might overcome the problems of low tumor antigen expression and high heterogeneity (15). A further evolution of radioimmunotherapy is based on pretargeting. The most clinically advanced pretargeting technique employs the avidin-biotin system. The first clinical experience with the avidin-biotin pretargeting system in cancer therapy began almost a decade ago at the European Institute of Oncology in Milan, and it has been collecting encouraging results (1, 2, 16–18). The present work describes the in vitro/in vivo binding features of biotinylated derivatives of the anti-tenascin mAb ST2146. To be employed for tumor pretargeting, biotinylated mAbs need to maintain the antigenic specificity and affinity of the nonbiotinylated ones and should distribute in vivo with similar selectivity. The in vivo biodistribution models, employed in the present work, were based on the human colon carcinoma HT-29 and human glioblastoma U-118MG xenografts in nude mice. The human grafted tumors were shown, by immunohistochemistry, to be stained by ST2146, which also stained the stromal component of a murine breast tumor as well as the normal mouse intestine, thus indicating that ST2146 recognizes both human and mouse tenascins. The intensity of ST2146 staining seemed higher with U-118MG compared with HT-29, correlating with a tenascin content about 500 times higher.
The biodistribution results indicate that adding biotins to ST2146 does not affect its in vivo immunoreactivity and confirm ST2146 efficiency in tumor targeting. The % ID/g of tumor was about 14 and 31 for low and high tenascin-expressing tumors, respectively, with T/NT ratios between 8 and 80. Because the relative tenascin content of HT-29 and U-118MG tumors is about 1:500, and the amount of ST2146/g of tumor is about 600 ng (4 µg/mouse) and 200 µg (640 µg/mouse), the ratio of antibody localization between the low and high tenascin-expressing tumors is at least 1:350, suggesting proportionality between antigen content and antibody binding. The distribution ratios calculated towards control IgG were also very high, resulting in localization indices ranging between 2.7 and 16.2 for HT-29 and 15.5 and 39.1 for U-118MG. To our knowledge, the present tumor localization data are the best described thus far in preclinical biodistribution models. Particularly, the murine parental mAb of Erbitux®, studied in nude mice transplanted with epidermal growth factor receptor–expressing human tumors, showed the best uptake 3 days after injection, of about 4 and 28 %ID/g in low and high antigen-expressing tumors, respectively (19). T/NT ratios were between 4.5 and 8.3, and localization indices ranged between 3.4 and 10.0. A recent article describes the biodistribution of 111In-labeled Herceptin® in human tumor-bearing mice (20). The most specific tumor uptake, in this case, was obtained 72 hours after injection with about 15 %ID/g of tumor and T/NT ratios between 1.6 and 6.5. Our ST2146 biodistribution data compare with those of the previously described 81C6 anti-tenascin mAb currently undergoing clinical trials (21). 81C6 showed about 20 %ID/g of tumor after 4 days; however, the values of 5 to 10 %ID/g of normal tissue indicate poorer T/NT ratios compared with current ST2146 data. It should be noted that none of the previously mentioned biodistribution studies included a chasing step to reduce nonspecific background, an additional advantage of the pretargeting method. Moreover, we report here an exceptionally high amount of ST2146/g of tumor compared with other previously described mAbs, reaching 200 µg/g of glioblastoma. This might be due to higher extracellular matrix protein content compared with cellular antigens, higher accessibility of tenascin compared with cellular antigens as a consequence of the association of tenascin with neovascularization, lack of internalization commonly observed with anti-cellular receptor mAbs, and lower variability in the frequency and expression level of tenascin compared with cellular target antigens. On the whole, PAGRIT, based on the use of biotinylated ST2146 alone or in combination with other antibodies, might represent a valid therapeutic opportunity for solid tumors.
|
Acknowledgments
|
|---|
We thank Dr. Carlo A. Nuzzolo for continuous advice and suggestions, Prof. Luigi G. Spagnoli for his contribution to immunohistochemistry, and Marlene Deutsch and Maria Grazia Martucci for language revision and article editing.
|
Footnotes
|
|---|
Grant support: Italian Ministry of University and Research.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 11/21/05;
revised 1/19/06;
accepted 1/26/06.
|
References
|
|---|
- Paganelli G, Grana C, Chinol M, et al. Antibody guided three-step therapy for high grade glioma with 90Y-biotin. Eur J Nucl Med 1999;26:348–57.[CrossRef][Medline]
- Grana C, Chinol M, Robertson C, et al. Pretargeted adjuvant radioimmunotherapy with yttrium-90-biotin in malignant glioma patients: a pilot study. Br J Cancer 2002;86:207–12.[CrossRef][Medline]
- Siri A, Carnemolla B, Saginati M, et al. Human tenascin: primary structure, pre-mRNA splicing patterns and localization of the epitope recognized by two monoclonal antibodies. Nucleic Acids Res 1991;19:525–31.[Abstract/Free Full Text]
- Balza E, Siri A, Ponassi M, et al. Production and characterization of monoclonal antibodies specific for different epitopes of human tenascin. FEBS Lett 1993;1:39–43.[Medline]
- De Santis R, Anastasi AM, D'Alessio V, et al. Novel antitenascin antibody with increased localisation for Pre-targeted Antibody-Guided RadioImmunoTherapy (PAGRIT®). Br J Cancer 2003;88:996–1003.[CrossRef][Medline]
- Petronzelli F, Pelliccia A, Anastasi AM, et al. Improved tumor targeting by combined use of two anti-tenascin antibodies. Clin Cancer Res 2005;11:7137–45s.[CrossRef]
- Natali PG, Nicotra MR, Bigotti A, et al. Comparative analysis of the expression of the extracellular matrix protein tenascin in normal human foetal, adult and tumor tissues. Int J Cancer 1991;47:811–6.[Medline]
- Herlyn M, Graeven U, Speicher D, et al. Guerry IV D Characterization of tenascin secreted by human melanoma cells. Cancer Res 1991;51:4853–8.[Abstract/Free Full Text]
- Green NM. A spectrophotometric assay for avidin and biotin based on binding dyes by avidin. Biochem J 1965;94:23–4c.
- Saginati M, Siri A, Balza E, et al. A simple procedure for tenascin purification. Eur J Biochem 1992;205:545–9.[Medline]
- Lightner VA, Slemp CA, Erickson HP. Localization and quantitation of hexabrachion (Tenascin) in skin, embryonic brain, tumors and plasma. Ann N Y Acad Sci 1990;580:260–75.[Medline]
- Ventimiglia JB, Wikstrand CJ, Ostrowski LE, et al. Tenascin expression in human glioma cell lines and normal tissues. J Neuroimmunol 1992;36:41–55.[CrossRef][Medline]
- Sharkey RM, Goldenberg DM. Perspectives on cancer therapy with radiolabeled monoclonal antibodies. J Nucl Med 2005;46:115–27s.
- Koppe MJ, Bleichrodt RP, Oyen WJG, et al. Radioimmunotherapy and colorectal cancer. Br J Surg 2005;92:264–76.[CrossRef][Medline]
- Paganelli G, De Santis R. Antibody-based cancer therapies: back to "polyclonals"? Eur J Nucl Med Mol Imaging 2004;31:1453–5.[CrossRef][Medline]
- Paganelli G, Bartolomei M, Ferrari M, et al. Pre-targeted locoregional radioimmunotherapy with 90Y-biotin in glioma patients: phase I study and preliminary therapeutic results. Cancer Biother Radiopharm 2001;16:227–35.[CrossRef][Medline]
- Grana C, Bartolomei M, Handkiewicz D, et al. Radioimmunotherapy in advanced ovarian cancer: is there a role for pre-targeting with 90Y-biotin? Gynecol Oncol 2004;93:691–8.[CrossRef][Medline]
- Boiardi A, Bartolomei M, Silvani A, et al. Intratumoral delivery of mitoxantrone in association with 90Y radioimmunotherapy (RIT) in recurrent glioblastoma. J Neuro-oncol 2005;72:125–31.[CrossRef][Medline]
- Goldenberg A, Masui H, Divgi C, et al. Imaging of human tumor xenografts with an Indium-111-labeled anti-epidermal growth factor receptor monoclonal antibody. J Natl Cancer Inst 1989;81:1616–25.[Abstract/Free Full Text]
- Lub-de Hooge MN, Kosterink JGW, Perik PJ, et al. Preclinical characterization of 111In-DTPA-trastuzumab. Br J Pharmacol 2004;143:99–106.[CrossRef][Medline]
- He X, Archer GE, Wikstrand CJ, et al. Generation and characterization of a mouse/human chimeric antibody directed against extracellular matrix protein tenascin. J Neuroimmunol 1994;52:127–37.[CrossRef][Medline]
Top
Abstract
Materials and Methods
