|Home | About | Journals | Submit | Contact Us | Français|
The monoclonal antibody, cetuximab, binds to epidermal growth-factor receptor and thus provides an opportunity to create both imaging and therapies that target this receptor. The potential of cetuximab as a radioimmunoconjugate, using the acyclic bifunctional chelator, CHX-A″-DTPA, was investigated. The pharmacokinetic behavior in the blood was determined in mice with and without tumors. Tumor targeting and scintigraphic imaging were evaluated in mice bearing xenografts of LS-174T (colorectal), SHAW (pancreatic), SKOV3 (ovarian), DU145 (prostate), and HT-29 (colorectal). Excellent tumor targeting was observed in each of the models with peak tumor uptakes of 59.8 ± 18.1, 22.5 ± 4.7, 33.3 ± 5.7, 18.2 ± 7.8, and 41.7 ± 10.8 injected dose per gram (%ID/g) at 48–72 hours, respectively. In contrast, the highest tumor %ID/g obtained in mice bearing melanoma (A375) xenografts was 6.3 ± 1.1 at 72 hours. The biodistribution of 111In-cetuximab was also evaluated in nontumor-bearing mice. The highest %ID/g was observed in the liver (9.3 ± 1.3 at 24 hours) and the salivary glands (8.1 ± 2.8 at 72 hours). Scintigraphy showed excellent tumor targeting at 24 hours. Blood pool was evident, as expected, but cleared over time. At 168 hours, the tumor was clearly discernible with negligible background.
Epidermal growth-factor receptor (EGFR) is a 170-kD polypeptide member of the erbB family of receptor tyrosine kinases that are crucial in signaling cell proliferation, differentiation, and survival. EGFR was the first of the erbB family to be described and characterized and has six known ligands with which it interacts.1 The receptor is expressed on normal cells, particularly those of epithelial origin, ranging in density from 40,000 to 100,000 receptors per cell, whereas in cancer cells, up to 2 × 106 receptors per cell have been reported.2 The receptor is overexpressed in cancers of the pancreas (30%–50%), colon (25%–77%), head and neck (80%–100%), lung (non-small-cell, 40%–80%), kidney (50%–90%), breast (14%–91%), prostate (40%–80%), ovary (35%–70%), bladder (31%–48%), and brain (glioma, 40%–63%).3–5 EGFR expression is associated with the inhibition of apoptosis, cell-cycle progression, angiogenesis, cell motility, and metastasis,1 resulting in a more aggressive malignant phenotype and poor prognosis.3–5
For these reasons, EGFR is an attractive candidate for targeted therapies. The strategies applied have included using small molecules that inhibit tyrosine kinases or monoclonal antibodies (mAbs) that bind to the receptor. Of the latter category, cetuximab (Erbitux) and panitumumab (Vectibix) have gained Food and Drug Administration (FDA) approval. Both of these mAbs bind to the extracellular domain of EGFR with high affinity and competitively inhibit ligand binding, resulting in the blockage of the activation of the tyrosine kinase.6–8
Cetuximab was approved in February 2004 under the FDA's accelerated approval program, which allows the FDA to approve products for cancer and other serious or life-threatening diseases based on early evidence of a product's effectiveness. The mAb is indicated for the treatment of patients with metastatic colorectal cancer whose tumors are positive for EGFR either in combination with irinotecan, or alone if patients cannot tolerate irinotecan. Cetuximab therapy has not been shown to extend patients' lives, but has been shown to shrink tumors in some patients and delay tumor growth, especially when used as a combination treatment.9,10
Cetuximab is a chimeric IgG1 that has activity, in vitro and in vivo, against a number of tumors through cell-cycle arrest at the G1 phase, induction of apoptosis, and antibody-dependent cell cytoxicity (ADCC).11 Antiangiogenic effects have also been ascribed to this mAb through the inhibition of vascular endothelial growth factor (VEGF) production.12 The mAb has also been shown to enhance the efficacy of conventional radiotherapy and several chemotherapeutics.
The potential of EGFR as a target for both imaging and therapy has been an active area of investigation.13–20 Cetuximab, in particular, has been the focus of a number of imaging studies for the purpose of monitoring disease,21 EGFR expression,21 patient selection18 or for performing dosimetry calculations for radioimmunotherapy trials.17,18
The studies described in this paper were designed to evaluate the in vitro and in vivo properties of radiolabeled cetuximab and to determine its potential for radioimmunodiagnostic and radioimmunotherapetic applications.
In vivo studies were conducted using human carcinoma cell lines of the colon (LS-174T and HT29), ovary (SKOV-3), pancreas (SHAW), prostate (DU145), epidermoid (A431), and a melanoma cell line (A375). LS-174T22 and A431 were grown in Dulbecco's minimum essential medium (DMEM), supplemented with 10 mM of glutamine. SKOV-3 cells were maintained in McCoy's 5a medium, SHAW and DU-145 in RPMI-1640, and A375 in DMEM, supplemented with 1 mM of sodium pyruvate and 10 μg/mL insulin. All media were also supplemented with 10% fetal bovine serum (FBS) and 1 mM of nonessential amino acids. Media and supplements were obtained from Quality Biologicals (Gaithersburg, MD), Invitrogen (Carlsbad, CA), or Lonza (Walkersville, MD).
EGFR expression of the cell lines was evaluated by standard flow-cytometric techniques.23 Briefly, cells were trypsinized, pelleted at 1500 × g for 10 minutes and resuspended in phosphate-buffered saline (PBS; pH 7.2) containing 1% bovine serum albumin and phosphate-buffered saline (BSA/PBS). The cells (1 × 106 cells in 100 μL) were added to 12 × 75 mm polypropylene tubes (Falcon Labware, Franklin Lakes, NJ) along with 1 μg of cetuximab (Erbitux; Amgen, Thousand Oaks, CA), or HuIgG (ICN Pharmaceuticals, Inc., Costa Mesa, CA). Following 1 hour of incubation at 4°C, the cells were washed three times with 3 mL of BSA/PBS, pelleting the cells at 1000 × g for 5 minutes and decanting the supernatant. Following the last wash, the cells were resuspended in 100 μL of BSA/PBS, containing 1 μg of Alexa Fluor 488 goat antihuman IgG (Invitrogen) and incubated for an additional 1 hour at 4°C. The cells were washed three times and analyzed by using a FACSCalibur (10,000 events collected) with CellQuest software (BD Biosciences, San Jose, CA).
The synthesis, characterization, and purification of the bifunctional ligand, CHX-A″-DTPA, used as the radiometal chelate, have been previously described.24 Conjugation of cetuximab with CHX-A″-DTPA was accomplished by using a modification of established methods.24 Briefly, 0.5 M of ethylenediaminetetraacetic acid (EDTA) and 0.05 M of sodium carbonate/bicarbonate (conjugation buffer) were added to cetuximab for a final concentration of 0.001 and 0.05 M, respectively. After vortexing, the solution was allowed to sit at ambient temperature for 30 minutes. The CHX-A″-DTPA was prepared at a concentration of 10 mg/mL in 0.05 M of conjugation buffer and added to the cetuximab solution dropwise while vortexing for a final molar excess of chelate to mAb of 10:1, 20:1, and 40:1. The reaction was incubated at 37°C for 4 hours. The excess unbound chelate was then removed by exhaustive dialysis against 0.15 M of NH4OAc (pH 7.0). The final concentration of cetuximab was quantified by the method of Lowry.25 The average number of CHX-A″-DTPA molecules linked to the mAb was determined by using a spectrophotometric assay based on the titration of ytrrium-Arsenazo(III) complex.26 Subsequent conjugations of CHX-A″-DTPA with cetuximab were conducted at 10:1.
Radiolabeling of CHX-A″-cetuximab (50 μg in 100 μL of 0.15 M NH4OAc buffer; pH 7.0) with 111In was performed by adding 0.5–1 mCi in 1–2 μL of 111InCl (in 0.05 M HCl) (PerkinElmer, Shelton, CT). The reaction was quenched after 30 minutes with 0.1 M of EDTA (3 μL) to scavenge free radiometal, and the radiolabeled product was purified by using a PD-10 desalting column (GE Healthcare, Piscataway, NJ). Integrity of the final product was evaluated by size-exclusion chromatography, using an analytic TSK-3000SW column (Tosoh Bioscience, Montgomeryville, PA) eluted at a flow rate of 0.5 mL/min.
Radioiodination of cetuximab with 125I was performed by the direct method by using Iodo-Gen (Pierce Chemical, Rockford, IL) as the oxidizing agent.27 Briefly, a solution of Na125I (10 μL, 674 μCi) in NaOHaq (0.05 M) was added to a solution of cetuximab (50 μg) in phosphate buffer (100 μL, 0.1 M; pH 7) contained in a test tube precoated with Iodo-Gen. After a 3-minute incubation at room temperature, the product was purified as above, using a PD-10 desalting column (GE Healthcare).
The immunoreactivity of the cetuximab-CHX-A″-DTPA conjugate was evaluated in a competition radioimmunoassay. Fifty (50) ng of EGFR (Sigma-Aldrich, St. Louis, MO) in 50 μL of PBS, containing Mg++ and Ca++, was added to each well of a 96-well plate. Following an overnight incubation at 4°C, wells were aspirated and 150 μL of PBS/BSA added to each well and allowed to sit for 1 additional hour at ambient temperature. The wells were aspirated and serial dilutions (0.01–500 ng in 50 μL of BSA/PBS) of unmodified cetuximab or cetuximab-CHX-A″-DTPA conjugate were added to the wells in triplicate, one set of wells received BSA/PBS without any competitor, along with 125I-cetuximab (~50,000 cpm in 50 μL of BSA/PBS). The wells were aspirated after a 4-hour incubation at 37°C and washed three times with BSA/PBS. The bound radioactivity was removed with 100 μL 0.2 M NaOH, adsorbed to cotton filters, placed in 12 × 75 mm tubes, and counted in a γ-scintillation counter (Wizard One, PerkinElmer, Shelton, CT). The percent inhibition was calculated by using the buffer control and plotted. HuM195, a mAb that reacts with human CD33, served as a negative control.
The immunoreactivity of the 111In-cetuximab was assessed in a radioimmunoassay, as detailed previously, using methanol-fixed cells.23,28 Serial dilutions of 111In-cetuximab (~200,000–12,500 cpm in 50 μL of BSA/PBS) were added to 12 × 75 mm test tubes containing A431 cells (1 × 106/50 μL PBS/BSA). Following a 2-hour incubation at 37°C, the cells were washed, pelleted, and counted in a γ-scintillation counter. The percentage of binding was calculated for each dilution and averaged. The specificity of the radiolabeled cetuximab was confirmed by incubating one set of cells with radiolabeled cetuximab and 10 μg of unlabeled cetuximab.
All in vivo studies were performed by using 4–6-week-old female athymic (nu/nu) mice (Charles River Laboratories, Wilmington, MA). All animal protocols were approved by the National Cancer Institute Animal Care and Use Committee.
Mice were injected subcutaneously (s.c.) in the right rear leg with either 2 × 106 LS-174T cells or 4 × 106 Shaw, SKOV-3, DU145, HT-29, or A375 cells in 200 μL of a 20% mix of Matrigel (BD Biosciences) in media. Mice were utilized in studies when the tumor xenografts maximal diameter measured 0.4–0.6 cm. Mice (n = 5) were injected intravenously (i.v.) with 111In-CHX-A″-cetuximab (~7.5 μCi on 0.6 μg) and sacrificed by exsanguination at time points from 24 to 168 hours. The blood, tumor, and major organs were collected, wet-weighed, and counted in a scintillation counter. The percent injected dose per gram (%ID/g) and standard deviation were calculated.
The blood pharmacokinetics were performed with both nontumor-bearing (n = 5) and mice bearing s.c. LS-174T tumors (n = 5). Following an i.v. injection of the 111In-CHX-A″-cetuximab (200 μL) blood samples were collected at various time points through the tail vein in heparinized capillary tubes (10 μL; Drummond Scientific, Broomall, PA). The blood was transferred to a cotton filter, placed in a 12 × 75 mm tube, and the radioactivity measured in a γ-scintillation counter. The percent injected dose per milliliter of blood was calculated for each of the samples; the averages and standard deviations were calculated and plotted. The T½α and T½β were calculated by using Sigmaplot 2001, version 7.101 (SPSS Inc., San Jose, CA).
Radioimmunoscintigraphy was performed with mice bearing s.c. LS-174T Shaw, HT29, DU-145, SKOV-3, and A375 tumor-bearing mice to further validate tumor targeting with 111In-CHX-A″-cetuximab. Tumor-bearing mice (n = 4–5) were given i.v. injections of 111In-CHX-A″-cetuximab (~80–100 μCi on 6–7.5 μg) in 200 μL of PBS. The mice were chemically restrained with 2.5% isoflurane (Abbott Laboratories, North Chicago, IL) delivered in O2, using a Model 100 vaporizer (SurgiVet, Waukesha, WI) at a flow rate of ~1.0 L/min. Images (100,000 counts) were acquired at 24, 48, 72, 96, and 168 hours with a large field of view (LFOV) gamma camera equipped with a pinhole collimator, using a 20% window centered on both photopeaks (173 and 247 KeV).29
The human tumor cell lines DU-145 (prostate), LS174T (colon), HT-29 (colon), SKOV-3 (ovary), SHAW (pancreas), and A375 (melanoma) were screened for EGFR expression, using flow-cytometric techniques. The human epidermoid carcinoma cell line, A431, known to express high levels of EGFR (2.6 × 106 receptors per cell),30 was included for comparison. The analyses (Fig. 1) indicated that EGFR is expressed on 92.3%–99.8% of the cells. The mean fluorescence intensity (MFI) varied to a greater degree, with the DU-145 having the highest (855.6) and the lowest MFI (48.6) observed with the A375. This is compared to 95.4% of A431 cells being positive for EGFR expression, which had a MFI of 1480.5.
To determine conditions for conjugating cetuximab with CHX-A″-DTPA, reactions were performed at a molar excess of CHX-A″-DTPA to cetuximab of 10:1, 20:1, and 40:1. These reactions resulted in a final product with a chelate-protein ratio of 1.8, 2.7, and 4.3, respectively. Radiolabeling of each of these cetuximab-CHX-A″-DTPA preparations resulted in a percent incorporation of 72, 75, and 77, with specific activities of 14.5, 16.4, and 15.8 mCi/mg, respectively. A subsequent conjugation was performed at a 10:1 molar excess of CHX-A″-DTPA to cetuximab, and the final chelate-protein ratio obtained was 2.3.
Modification of cetuximab with this number of chelates did not affect the immunoreactivity of the mAb. When the cetuximab-CHX-A″ was evaluated in a competition radioimmunoassay (Fig. 2), it was found that 2.9 ng of unmodified cetuximab was required to obtain 50% inhibition, while for cetuximab-CHX-A″ it was 3.0 ng.
Radiolabeling of cetuximab-CHX-A″ with 111In was efficient (81.9% ± 13.5%) with an average specific activity of 13.6 ± 1.4 mCi/mg. Size-exclusion high-performance liquid chromatography analyses indicated that the radioactivity was associated with a retention time consistent with intact IgG with <2% (each) associated with higher and lower molecular weight species (data not shown). When the radioimmunoconjugate (RIC) was incubated with fixed A431 cells for 2 hours, 70.1 ± 13.5% of the radioactivity was bound. When 10 μg of unlabeled cetuximab was included in the assay, only 4.5% ± 0.4% of the radioactivity was cell bound, demonstrating that binding was specific.
Athymic mice bearing s.c. LS-174T xenografts (n = 5) were injected (i.v.) with 111In-cetuximab to establish and define tumor targeting and normal organ distribution of the RIC. As detailed in Table 1, the 111In-CHX-A″-cetuximab demonstrated excellent tumor targeting with a tumor %ID/g of 25.7 ± 8.4 at 24 hours. The tumor %ID/g peaks at 59.8 ± 18.1 at 72 hours, and even though this uptake decreases, the level remained high at the end of the study, with a %ID/g of 26.4 ± 6.7 at 168 hours. Of the normal organs, the highest %ID/g was observed in the blood (12.9 ± 1.7) at 24 hours, which then decreased to 1.4 ± 1.4 by 168 hours. The liver demonstrated the next highest %ID/g, with 10.0 ± 0.9 noted at 48 hours, which then steadily decreased to 4.4 ± 1.1 by 168 hours. The spleen attained a %ID/g of 7.8 ± 1.1 at 48 hours. Other normal tissues that peaked at 48 hours included the small and large intestine, the ovaries, uterus, bladder, and femur. The %ID/g of the kidneys, lung, heart, and stomach attained their highest value at 72 hours, all of which were <6.
The normal organ distribution of 111In-CHX-A″-DTPA-cetuximab was also evaluated in nontumor-bearing mice. Mice were injected with ~7.5 μCi of the RIC and euthanized at 24, 48, 72, 96, 120, and 168 hours for a comprehensive organ harvest. As observed with the mice bearing the LS-174T tumors, the highest %ID/g (12.5 ± 1.5) was obtained in the blood at 24 hours (Table 2). The difference is that in the nontumor-bearing mice, the %ID/g values were consistently lower throughout the study, with the exception of the 168-hour time point, at which time a value of 4.1 ± 1.7 was obtained. The liver resulted in the second highest %ID/g (9.3 ± 1.3), also at the 24-hour time point. Among the remainder of the normal organs that were collected, with the exception of the salivary glands with a %ID/g of 8.1 ± 2.8 at 72 hours, the %ID/g were below 6. In addition, all of the tissue %ID/g, with the exception of the uterus and skin, were highest at 24 hours and decreased over the course of the 168-hour study period.
Pharmacokinetic studies were then conducted to determine the clearance rates of the 111In-CHX-A″-DTPA-cetuximab from the blood compartment in both tumor and nontumor-bearing mice (Fig. 3). The i.v. injection resulted in an initial %ID/mL in the blood of 33.1 ± 7.1 in the mice bearing s.c. LS-174T xenografts, whereas it was 31.6 ± 3.3 in the group of mice without tumors. In the former group, the %ID/mL attained a value of 10.9 ± 3.6 at 24 hours and by 168 hours, it was 3.9 ± 2.2. In the nontumor-bearing group, a greater amount of the 111In-CHX-A″-DTPA-cetuximab appeared to remain circulating in the blood. The %ID/mL was 13.1 ± 1.0 at 24 hours, and at 168 hours, it was 6.0 ± 0.6. This observation was confirmed when the calculations based on a biphasic clearance were performed. The T½α and T½β for the tumor-bearing mice were 1.0 and 51.9, respectively; corresponding values for the nontumor-bearing mice were 3.5 and 115.6.
The tumor targeting of 111In-CHX-A″-DTPA-cetuximab was assessed further in several other tumor models to establish the potential of radio-labeled cetuximab for therapeutic and diagnostic applications. As outlined for the previous biodistribution studies, athymic mice bearing tumor xenografts of DU-145, HT-29, SHAW, or SKOV-3 were injected with ~7.5 μCi of 111In-CHX-A″-DTPA-cetuximab and then euthanized at 24, 48, 72, 96, and 168 hours (n = 4–5 mice per time point). Excellent tumor targeting (Table 3) was observed in each of these tumor xenografts, with the highest tumor %ID/g (41.7 ± 14.6) observed with the HT-29 at 48 hours, and was still at 33.0 ± 10.3 at 168 hours. The peak tumor %ID/g for the DU-145, SHAW, and SKOV-3 tumors occurred at 72 hours, with values of 18.2 ± 3.7, 22.5 ± 4.7, and 33.3 ± 5. The highest blood %ID/g was found in the mice bearing the HT-29 tumors; 17.5 ± 2.8 at 24 hours, which then steadily declined to 5.6 ± 3.4 at the end of the 168-hour study. The liver also had a high %ID/g with a value of 17.0 ± 5.0 at 48 hours. In comparison to the LS-174T tumor-bearing mice, the %ID/g of the rest of the normal organs was similar. In the biodistribution study with the mice bearing HT-29 tumor xenografts, the spleen resulted in a %ID/g of 7.9 ± 2.3 and the lungs were 8.3 ± 1.4 at 48 hours. The %ID/g of the kidneys, heart, and femur was <6 at 24 hours; by 168 hours, the values were lower, ranging from 1.6 to 3.4.
In each of the other tumor models, the highest uptake, in normal tissues, of 111In-CHX-A″-DTPA-cetuximab was observed in the liver; the %ID/g was 6.2 ± 2.6 for DU-145 at 24 hours, 10.5 ± 2.5 for SHAW at 24 hours, and 11.0 ± 2.3 for SKOV-3 at 48 hours. In all of these models, the liver %ID/g decreased to <5 at 168 hours. The %ID/g of the remaining normal tissues were found to be <5 throughout the study.
The distribution of 111In-CHX-A″-DTPA-cetuximab was also evaluated in mice with s.c. melanoma (A375) xenografts, a tumor with low, or no, expression of EGFR. When compared to the other tumor models, low tumor targeting was observed with the A375 model. The peak tumor %ID/g was 6.3 ± 1.1, which was obtained at 72 hours, which then decreased to 2.6 ± 1.4 by 168 hours. A close comparison of A375 and LS-174T tumor models reveals that the LS-174T tumor attains a %ID/g that is 8.3-fold higher than A375. The %ID/g of the normal organs is on par with the other tumor models and, in fact, the tumor is not the highest value. At 24 hours, the liver has the highest %ID/g (8.6 ± 1.1) in the A375 model.
Imaging of each of the tumor models was also performed by using scinitigraphy. Mice bearing each of the tumor xenografts already discussed were injected i.v. with 80–100 μCi of 111In-CHX-A″-DTPA-cetuximab and then imaged at 24, 48, 72, 96, and 168 hours. Shown in Figure 4, the LS-174T xenograft is clearly visualized at 24 hours. There is an increase in the intensity of the tumor in the subsequent images, along with a decrease of activity in the blood pool (heart, lungs, and liver). In contrast, only blood pool activity was observed in the nontumor-bearing mice. Tumor targeting is also clearly visualized in the other tumor models, with the exception of the A375 xenograft. As illustrated by the images taken at 72 hours (Fig. 5), excellent images of the tumors were obtained and the activity in the blood pool was low. The lack of tumor uptake of the 111In-CHX-A″-DTPA-cetuximab in the A375 model is consistent with the low tumor %ID/g obtained in the biodistribution study.
Cetuximab binds to the EGFR, which is expressed in a wide variety of tumors, with high affinity.3–6 In contrast to Herceptin® therapy, where it is established that tumor responsiveness correlates with higher HER2 expression, no such correlation has been made between EGFR expression and tumor responsiveness to cetuximab therapy.31
Interestingly, the investigators reported that the HT-29 carcinoma cell line was one of the cell lines that was insensitive to cetuximab therapy. Yet, in the studies reported in this paper, HT-29 xenografts were found to have the highest uptake of 111In-labeled cetuximab. For patients with nonreponsive disease or disease refractory to cetuximab therapy, radioimmunotherapy would provide a viable alternative treatment.
The flow-cytometric analysis on EGFR expression of the cell lines versus the in vivo tumor-targeting studies also seems contradictory. A ranking of the cell lines from high to low EGFR expression would be A431, DU-145, SKOV-3, HT-29, Shaw, LS-174T, and A375. The ranking of those used in the in vivo tumor targeting would be LS-174T, HT-29, SKOV-3, Shaw, DU-145, and A735. The LS-174T cell line was not overly impressive, with a mean fluorescence intensity of 72.2, albeit 97.6% of the cells were found to express EGFR, yet excellent tumor targeting was obtained. These results would imply that flow-cytometric analysis of receptor expression might not be valid or actually predictive of tumor targeting; a similar phenomenon was observed with HER2 expression.32 The requisite for the in vivo milieu and spatial conformation to induce antigen expression is not a novel concept.33 This apparent dichotomy attests to empirically evaluating the potential of a target and targeting vehicle, such as cetuximab, in an intact biologic system.
Conjugation of the mAb with the CHX-A″-DTPA chelate with a 10-fold molar excess of the chelate resulted in a chelate-protein ratio of 2.3; the modification did not affect the immunoreactivity of the cetuximab with its cognate antigen, as determined by a competition radioimmunoassay. The chelate-protein ratio is within the range for what has been published for the modification of cetuximab with other chelates.18,20 Further, when radiolabeled with 111In, excellent binding was obtained when the RIC was incubated with A431 cells, which express high levels of EGFR. Specificity of the 111In-CHX-A″-cetuximab was maintained and could be demonstrated by inhibiting binding with the addition of excess unlabeled cetuximab.
The current literature on radiolabeled cetuximab is focused on utilizing cetuximab for the imaging of tumor lesions. One purpose for this is to provide a means of rationally selecting patients for cetuximab monotherapy. Other purposes include 1) monitoring of patients during therapy to assess efficacy, 2) to provide a method that would allow physicians to optimize dosages of cetuximab, alone or in combination with other therapeutics, and 3) to perform dosimetry calculations prior to cetuximab radioimmunotherapy. In these studies five different tumors were clearly visible within 24 hours of the i.v. injection of 111In-CHX-A″-DTPA-cetuximab. The blood pool was evident, but not excessive, and cleared considerably over time. Over the course of the week, as the radioactivity cleared from the blood pool, the tumor xenografts remain prominent. Interestingly, two other reports have been published in which cetuximab was radiolabeled with 111In and utilized to image tumor xenografts.14,34 In one, the investigators sought to reduce liver uptake of 111In-cetuximab by modifying the mAb with polyethylene glycol.14 In that report, A431 tumors were poorly visualized at 24 hours and were no longer evident at 48 hours, whereas the blood-pool activity (i.e., the liver) was very high. They reported a liver %ID/g of 46.9 ± 2.5 at 48 hours, which was 4.6-fold higher than what we report from our study. Since similar amounts of 111In-cetuximab (up to 100 μci) were injected, the only apparent differences were in the tumor xenografts and the choice of chelate used for the RIC. The study utilized the dianhydride of DTPA to produce the chelated conjugate, which is known to release 111In. This would account for the high liver uptake of radioactivity.35 The values that were calculated for the clearance of the RIC (T½α of 0.4 hours and T½β of 9.11 hours) also indicates that there was a lack of stability.36 In the second study, the potential of optical dyes for imaging tumor lesions was explored. As part of the study, cetuximab-dye conjugates were compared to single-positron emission computed tomography (SPECT) images, using 111In-labeled cetuximab.34 In this instance, the same chelate, CHX-A″-DTPA, was used for radiolabeling cetuximab.34 Similar to the data reported herein, visualization of tumor xenografts was obtained at 24 hours and was still very evident at 4 days. The activity in the blood pool in the SPECT images did not appear to decrease over time. One of the differences between that study and this one was that the investigators injected 280 μCi on 50 μg of cetuximab per mouse, whereas the studies in this paper employed an injected dose of 80–100 μCi on ~6–7.5 μg of cetuximab. Thus, the 111In-CHX-A″-cetuximab in that prior report may not have been clearing as rapidly as it should due to saturation.
Cetuximab, radiolabeled with 64Cu and 89Zr, has also been investigated as an imaging agent for PET for the purposes of quantitating EGFR expression and to identify patients for cetuximab therapy.18,20,37 In the studies with 89Zr, comparisons were made between 89Zr-labeled cetuximab and 88Y and 177Lu conjugates to assess if 89Zr would serve as a surrogate for the 88Y and 177Lu, which would then permit functions such as monitoring patients or dosimetric calculations.18 These comparisons were accomplished with in vivo biodistribution studies using mice bearing A431 xenografts. Again, one is comparing different radionuclides and chelates to accomplish the radiolabeling; however, the tumor-targeting data reported herein with the 111In-CHX-A″-DTPA, are similar to this published data.18 Differences are evident in the blood clearance and in normal organ uptake, such as the liver. Higher liver uptake was observed in mice injected with cetuximab radiolabeled with 88Y using DOTA, 89Zr with N-sucDF, 177Lu with SCN-DTPA, and 88Y with SCN-DTPA. In this particular study, the only biodistribution that showed comparable normal tissue uptake to the studies reported in this paper were when the mice that were injected i.v. with cetuximab labeled with 177Lu, using DOTA as the chelating agent. A study was just published that demonstrated tumor targeting with micro-PET with 64Cu-cetuximab, using the DOTA chelate.37 Tumor (A431) xenografts were clearly visualized at 24 hours. Biodistribution studies also demonstrated tumor uptake, with a %ID/g of 18.49 at 24 hours. Unfortunately, accretion of radioactivity in the liver was also evident in the imaging and biodistribution experiments. The researchers admit that DOTA is not the ideal chelate for Cu radioisotopes. In fact, they were able to demonstrate the presence of metabolites in the liver. There are chelates more appropriate for Cu, such as cross-bridged macrocycles; however, the harsh conditions required to complex Cu with the chelate preclude its use with biomolecules such as proteins. These studies, once again, attest to the requirement for thoughtful selection of the chelate for preparation and study of a RIC.
The pharmacokinetic data described in this paper revealed differences in the bi-exponential clearance of the 111In-CHX-A″-DTPA cetuximab when given i.v. to tumor and nontumor-bearing mice. In the group of mice without a tumor burden, the T½α and T½β were calculated to be 3.5 and 115.6 hours, respectively. These values are very similar to plasma clearance rates reported for other chimeric mAb.38,39 In the group of mice bearing LS-174T xenografts, the clearance was much faster with a T½α of 1.0 hours and T½β of 51.9 hours. This T½β value is consistent with the values reported for both radiolabeled cetuximab36 and unlabeled cetuximab.40 The tumor burden of the mice is most likely acting as a “sink” for the cetuximab and thus facilitating its rapid disappearance from the blood compartment. The much shorter T½β time of the tumor-bearing mice may be explained by the mAb having a high affinity for EGFR, and that EGFR is an internalizing receptor depleting the mAb from circulation.
The biodistribution data presented in this report demonstrates the broad potential of cetuximab has a vehicle for delivering therapeutic levels of a radionuclide. The mAb, which is FDA approved for the treatment of colorectal cancer, not only targeted two human colorectal tumor xenografts, but also clearly targeted prostate, pancreatic, and ovarian tumors with significant efficiency.
In summary, cetuximab is readily modified with the bifunctional acyclic chelate, CHX-A″-DPTA. When labeled with 111In, its immunoreactivity is retained. The studies detailed above demonstrate the flexibility of cetuximab as a potentially useful RIC for both imaging and therapeutic applications for the treatment and management of a spectrum of cancers. Studies are ongoing to further expand and define its role in both of these arenas.
This research was supported by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research (Bethesda, MD).