DOI: 10.1148/radiol.2462070229
(Radiology 2008;246:497-507.)
© RSNA, 2008
Metastatic Human Colonic Carcinoma: Molecular Imaging with Pretargeted SPECT and PET in a Mouse Model1
Robert M. Sharkey, PhD,
Habibe Karacay, PhD,
Shankar Vallabhajosula, PhD,
William J. McBride, PhD,
Edmund A. Rossi, PhD,
Chien-Hsing Chang, PhD,
Stanley J. Goldsmith, MD, and
David M. Goldenberg, ScD, MD
1 From the Center for Molecular Medicine and Immunology, Garden State Cancer Center, 520 Belleville Ave, Belleville, NJ 07109 (R.M.S., H.K., D.M.G.); Citigroup Biomedical Imaging Center and Department of Radiology, Weill Medical College of Cornell University, New York, NY (S.V., S.J.G.); Immunomedics, Morris Plains, NJ (W.J.M., C.H.C.); and IBC Pharmaceuticals, Morris Plains, NJ (E.A.R., C.H.C.). Received February 2, 2007; revision requested April 5; revision received April 20; accepted May 9; final version accepted July 19. Supported in part by grant CDG-06-103 from the New Jersey Department of the Treasury.
Address correspondence to R.M.S. (e-mail: rmsharkey{at}gscancer.org).
 |
ABSTRACT
|
|---|
Purpose: To prospectively determine if a bispecific monoclonal antibody (MoAb) pretargeting method with a radiolabeled hapten peptide can depict small (<0.3 mm in diameter) microdisseminated human colon cancer colonies in the lungs of nude mice.
Materials and Methods: Animal studies were approved in advance by animal care and use committees. Animals injected intravenously with a human colon cancer cell line to establish microdisseminated colonies in the lungs were pretargeted with TF2—a recombinant, humanized, anti–carcinoembryonic antigen (CEA) and anti–histamine-succinyl-glycine (HSG) bispecific MoAb; 21 hours later, a radiolabeled HSG peptide was given. Imaging and necropsy data for tumor-bearing animals given the anti-CEA bispecific MoAb (n = 38, all studies) were compared with those of animals given fluorine 18 (18F) fluorodeoxyglucose (FDG) (n = 15, all studies), peptide alone (n = 20, all studies), or an irrelevant anti-CD22 bispecific MoAb (n = 12, all studies). Uptake of these agents in the lungs of non–tumor-bearing animals enabled assessment of specificity (n = 15, 4, and 6 for TF2 pretarget, hapten peptide alone, and 18F-FDG, respectively).
Results: TF2-pretargeting helped localize tumors in the lungs within 1.5 hours of the radiolabeled HSG peptide injection, while the peptide alone, irrelevant bispecific MoAb pretargeted peptide, and 18F-FDG failed. Necropsy data indicated that the signal in tumor-bearing lungs was five times higher than in blood within 1.5 hours, increasing to 50 times higher by 24 hours. Peptide uptake in tumor-bearing lungs pretargeted with TF2 was nine times higher than in non–tumor-bearing lungs, while it was only 1.5-fold higher with 18F-FDG or the peptide alone. Micro-positron emission tomographic (PET) images showed discrete uptake in individual metastatic tumor colonies; autoradiographic data demonstrated selective targeting within the lungs, including metastases less than 0.3 mm in diameter.
Conclusion: Bispecific antibody pretargeting is highly specific for imaging micrometastatic disease and may thus provide a complementary method to 18F-FDG at clinical examination.
© RSNA, 2008
|
INTRODUCTION
|
|---|
Even in cancers in which fluorine 18 (18F) fluorodeoxyglucose (FDG) positron emission tomography (PET) imaging already provides a high sensitivity and specificity for detection (1–4), there are opportunities for new procedures that could enhance specificity, thereby providing greater confidence that the FDG targeting is indicative of malignancy (5–9). The specificity of antibodies for tumor-associated antigens could provide these improvements in molecular imaging. Although most of the previously approved antibody-based imaging products are no longer in clinical use because they were not as effective as 18F-FDG in their respective indications (10,11), recent developments suggest that a new generation of such agents may become available. By means of molecular engineering, a variety of antibody constructs that have more desirable targeting properties than the earlier products, including humanized antibodies that have reduced immunogenicity, have been developed (12,13).
A two-step pretargeting approach (Fig 1), which includes humanized recombinant bispecific monoclonal antibodies (MoAbs) in concert with a small radiolabeled peptide, has been promising (15–17). Tumor uptake and targeting ratios have far exceeded those achieved with directly radiolabeled Fab' fragments, and, even in comparison with 18F-FDG, this pretargeting method, combined with an iodine 124 (124I)-labeled peptide, improved imaging of a well-established, subcutaneously grown, human colonic tumor xenograft (18). This pretargeting method and 18F-FDG were further studied to determine their ability to help localize micrometastatic disease by using another colon cancer model (19). In this model, the GW-39 human colon cancer cell line (20) that is serially propagated in nude mice can be injected intravenously, which results in microdisseminated tumor colonies growing in the lungs. The purpose of our study, therefore, was to prospectively determine if a bispecific MoAb pretargeting method with a radiolabeled hapten peptide can depict small (<0.3 mm in diameter), microdisseminated human colon cancer colonies in the lungs of nude mice.
View larger version (37K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1: Schematic representation of bispecific MoAb (bsMAb) pretargeting method. Bispecific MoAb is administered, and, over time, the antitumor binding arm will localize in the tumor. After allowing sufficient time for bispecific MoAb to clear from the blood to a prescribed level, radiolabeled hapten peptide is administered. Within minutes, radiolabeled hapten peptide leaves the bloodstream and enters the extravascular space, where it can bind to the antihapten portion of the bispecific MoAb on the tumor. In this pretargeting procedure, the peptide contains two histamine-succinyl-glycine (HSG) moieties that help stabilize binding of the radiolabeled peptide in the tumor by cross linking adjacent bispecific MoAbs, a process known as affinity enhancement system (14). The majority of the radiolabeled hapten peptide is quickly filtered by the kidneys, where it is then voided from the body in urine.
|
|
|
MATERIALS AND METHODS
|
|---|
Bispecific MoAbs and Radiolabeled Peptides
TF2, a recombinant 157-kD humanized bispecific MoAb with divalent specificity to carcinoembryonic antigen (CEA) (based on the humanized MoAb MN-14, labetuzumab) (21) and with monovalent specificity to the hapten HSG (based on the MoAb 679) (22), and TF6, a recombinant 157-kD humanized bispecific MoAb with divalent specificity to CD22 (based on the humanized MoAb LL2, epratuzumab) (23) and with monovalent binding to HSG, were produced and purified by using a "dock-and-lock" procedure (16). The purified bispecific MoAbs were shown to be reactive with both CEA and CD22, by virtue of their sequential binding to anti-idiotype antibodies derived against the parental antibodies, and reactive with an HSG-containing peptide (16). These agents were provided by Immunomedics (Morris Plains, NJ) and IBC Pharmaceuticals (Morris Plains, NJ). Four authors (D.M.G., W.J.M., E.A.R., C.H.C.) have financial interest in Immunomedics or IBC Pharmaceuticals. All other authors had control of the data and information submitted for publication.
IMP-245 is a divalent HSG peptide used for radiolabeling with technetium 99m (99mTc) (17,22). IMP-288 and IMP-325 are divalent HSG peptides used for radiolabeling with indium 111 (111In) and 124I or 131I, respectively (18), and were provided by Immunomedics. We purchased 99mTc pertechnetate from a local Mallinckrodt radiopharmacy (Pinebrook, NJ), 111In chloride (111InCl3) and sodium 131I (Na131I) from PerkinElmer (Boston, Mass), and sodium 124I (Na124I) from IBA Molecular (Sterling, Va). The peptides were labeled with a radioisotope on the day of the animal injection by using freshly delivered radioactivity at the two institutions (H.K. and W.J.M. at Center for Molecular Medicine and Immunology [CMMI] and S.V. at Citigroup Biomedical Imaging Center [CBIC]). The products had less than 3% unbound radionuclide at the time of their injection and, by using size-exclusion high-performance liquid chromatography, 100% of the radioactivity bound to the bispecific MoAbs. Fluorine 18 FDG was purchased from a local radiopharmacy for studies performed at CMMI or was made in-house for micro-PET imaging studies performed at CBIC.
Animal Studies
All studies were conducted with prior approval of each institution's animal care and use committee. CMMI was responsible for the acquisition of the animals, implantation of tumors, all necropsy and gamma imaging data, histologic assessment of tumor involvement in the lungs, and autoradiographic targeting analysis (R.M.S., H.K., and D.M.G.). Tumor-bearing animals were shipped to CBIC, where micro-PET imaging studies were performed. Female athymic nude mice (athymic NCr-nu/nu; Taconic Labs, Germantown, NY; NCI colony) were examined without implantation of tumor or were injected intravenously and/or subcutaneously with 30 µL of a 10% suspension of the GW-39 human colon cancer cell line (19,20). Animals were injected intravenously with GW-39 at CMMI and then were shipped to CBIC for micro-PET imaging studies. Animals given an intravenous injection of GW-39 develop tumor nodules within the lungs and die of extensive pulmonary involvement within 6–8 weeks, with no evidence of spread to other regions of the body (19).
One to 3 weeks after tumor inoculation, animals were given intravenous injections of the test articles (ie, bispecific MoAb, radiolabeled peptide, or 18F-FDG). For the pretargeted animals, the radiolabeled peptide was administered 21–24 hours after the bispecific MoAb was injected and was always administered at a bispecific MoAb-to-peptide molar ratio of 10:1, as described previously (24). The specific dosing information for each study is provided in the text, tables, or figures. Pretargeting with the TF2 anti-CEA bispecific MoAb and its associated 99mTc-, 131I-, 124I-, or 111In-labeled peptide was compared with results from animals given 18F-FDG, the radiolabeled peptide alone, or TF6, an irrelevant anti-CD22 bispecific MoAb that was pretargeted in an identical manner as the specific TF2 anti-CEA bispecific MoAb. All animals given a radioiodinated agent received Lugol iodine solution in their drinking water 2 days before the injection to reduce thyroid uptake. All animals receiving 18F-FDG were fasted starting the night before (approximately 18 hours) the intended injection, but water was given ad libitum.
At defined intervals after the radiolabeled peptide injection, groups of animals were imaged and/or necropsy was performed. For studies involving 99mTc imaging performed at CMMI (under the supervision of H.K. and R.M.S.), planar images were obtained by placing a group of three to six anesthetized animals prone on a low-energy, high-resolution collimator for 99mTc fitted to a single photon emission computed tomographic (SPECT) camera (Solus; ADAC Laboratories, Milpitas, Calif). Images were acquired for 15–20 minutes (typically collecting 30 000 to 70 000 counts). Unless otherwise indicated, images were adjusted by an author (R.M.S.) without background reduction and with the signal intensity increased until the activity in the pixels of a given tissue (except the urinary bladder) was maximized.
Micro-PET imaging studies were performed at CBIC (S.V.) by using a camera (Focus TM 220; CTI Concorde Microsystems, Knoxville, Tenn) that had a bore size of 22 cm and a transverse field of view of 7.6 cm. The resolution at the center of the field of view was less than 1.3 mm. Three-dimensional histograms of emission data were generated by using dead-time correction. Images were reconstructed by using a two-dimensional ordered-subset expectation maximization algorithm with no attenuation and scatter correction and were reviewed by authors (S.V., S.J.G.). These image files were then transferred to CMMI, where selection of sections and final adjustments to the image signal intensity were performed by an author (R.M.S.).
Micro-PET imaging studies were performed in concert with tissue distribution studies performed at CMMI (under the supervision of H.K. and R.M.S.). Thirty-six nude mice were implanted intravenously with the GW-39 cell line. Fifteen animals were shipped to CBIC for micro-PET imaging, while the remaining 21 were retained at CMMI for biodistribution assessment by using tissue counting (see below). Eleven days after implantation, eight of the animals at CBIC were given an intravenous injection of a bispecific MoAb (TF2, n = 4; TF6, n = 4) and then, 21 hours later, they received 124I-IMP-325 (0.216 mCi [7.99 MBq]). Another four animals received 18F-FDG (0.81 mCi [29.97 MBq]), and three animals received 124I-IMP-325 alone (no pretargeting). For 18F-FDG, images were obtained at 1.5 hours and again at 7 hours after injection; for the pretargeted peptide, images were first obtained at 1.5–2.0 hours and then again at 5 and 21 hours after injection. Animals given the radiolabeled peptide alone were imaged at 1.5 and 21 hours after injection. Anesthetized animals were placed prone on the imaging table in pairs, with count acquisition for 20 minutes (S.V.). At the conclusion of the final imaging session, animals were euthanized and the lungs were placed in formalin for histologic evaluation of tumor involvement, which was determined at CMMI.
Timed to correspond with the micro-PET studies, the additional 21 animals retained at CMMI received identical doses of the pretargeting agents (TF2, n = 9; TF6, n = 4), and then, 21 hours later, they received the peptide that was radiolabeled with 131I rather than 124I. Five animals received 131I-peptide alone, and three animals were given 18F-FDG. Necropsy was performed in most of the animals at 1.5 hours, with additional animals receiving 131I-peptide alone (n = 2) or pretargeted with TF2 (n = 4) also examined at 24 hours after 131I-peptide was given.
For autoradiographic studies (performed at CMMI), 3 weeks after intravenous tumor inoculation, animals were given TF2 (n = 4) or TF6 (n = 4), and 21 hours later they received 111In peptide. Three hours later, necropsy was performed, and the lungs were placed in cold formalin for 2 hours. The lungs were then embedded in paraffin, and the next day, slides containing two 5-µm slices were wrapped in plastic wrap and placed on x-ray film with an intensifier screen (H.K.). The film was exposed at –80°C for 14 days. The slides were then removed, and slices were stained with hematoxylin-eosin. Microscopic identification of the tumor colonies was confirmed (R.M.S. and D.M.G.; each with more than 30 years of experience). Separate photographs of the autoradiographs and hematoxylin-eosin–stained slices were taken at the same magnification by using a digital camera fitted with a macrozoom lens (R.M.S.). In addition to the individual photographs, the two images were overlaid to illustrate the autoradiographic uptake in concert with the tumor nodules seen on the hematoxylin-eosin slices.
Data Collection and Statistical Analysis
CMMI was responsible for all necropsy data obtained (under the supervision of H.K. and R.M.S.). At necropsy, animals were first weighed, then anesthetized and bled intracardially, and then euthanized by means of cervical dislocation. The tissues were weighed and placed in glass vials in their entirety, and then radioactivity measurements were made by using a gamma scintillation counter with windows set for the individual radionuclide administered (99mTc, 80–200 keV; 131I, 200–450 keV; 18F, 400–800 keV; 111In, 100–450 keV; and 124I, 400–800 keV). A dilution of the injected radiolabeled product was included with each set of tissues; this was used to calculate the total injected activity. The counts per minute per gram of tissue were divided by the total injected activity and expressed as the percentage of injected dose per gram. In addition, a ratio of the activity per gram in the tumor (subcutaneous tumor or tumor-bearing lungs) to that in nontumor tissue was determined. Mean ± standard deviation was determined for each set of animals, with the number of animals in each set shown in the tables or text.
All gastrointestinal tissues were weighed and counted with their contents. Because tumor colonies within the lungs were not visible at gross examination, direct quantitation of uptake in the tumor masses could not be performed. Instead, the tumor-bearing lungs were counted in toto, and uptake in the lungs was compared with uptake measured in a separate group of age- and weight-matched non–tumor-bearing animals. In addition, mean ± standard deviation weight of the lungs from all non–tumor-bearing animals (n = 23) was calculated and used as a baseline against which the average weight of the tumor-bearing lungs taken from animals that received a given radiotracer could be compared. The animals included in the first set of studies, in addition to having received the intravenous injection of GW-39, were also given a subcutaneous injection of GW-39, which allowed direct measurement of tumor uptake for each of the radiotracers included in this study.
The lungs from all animals were placed in formalin after radioactivity determination and were embedded in paraffin, and two 5-µm slices were cut and stained with hematoxylin-eosin. Tumor colonies in each slice were counted by using low-power microscopy (R.M.S.). Photographs of representative areas of tumor-bearing lung specimens were obtained at the same magnification as that of a hemocytometer, from which the size of tumor nodules was determined (R.M.S.).
The comparison of primary interest was the uptake of the radiotracer in tumor-bearing lungs versus that in non–tumor-bearing lungs to assess whether there was a significantly enhanced signal intensity related to tumor content in the lungs. Other comparisons between the uptake of one radiotracer and that of another in tumor (either subcutaneous tumor or tumor-bearing lungs), between the weights of tumor-bearing lungs and the weights of non–tumor-bearing lungs, and between animal body weights were also made. Statistical analysis was performed (R.M.S.) at CMMI and was reviewed by a consulting statistician. A two-sample two-tailed t test assuming equal variances as provided with software (Excel 2003; Microsoft, Redmond, Wash) was used. P 
.05 was considered to indicate a significant difference.
|
RESULTS
|
|---|
Bispecific MoAb Pretargeting with SPECT
Uptake of the TF2-pretargeted 99mTc-peptide in the subcutaneous tumor 1.5 hours after injection was about 20 times higher than that of the peptide alone and was significantly higher than that in the animals given 18F-FDG (P = .041) (Table 1). The distribution of the radiotracer in normal tissues of the tumor- and non–tumor-bearing mice was not appreciably different. However, there was substantially less 99mTc-peptide, alone or pretargeted, in all tissues except the kidneys, as compared with uptake in mice given 18F-FDG, particularly the brain, heart, and femur (eg, bone marrow) (Table 1).
View this table:
[in this window]
[in a new window]
|
Table 1. Biodistribution Data of 99mTc-IMP-245 Peptide Alone and Pretargeted with TF2 or 18F-FDG in Tumor- and Non–Tumor-bearing Nude Mice
|
|
Each slice from the 25 tumor-bearing animals included in this study revealed multiple (ranging from 80 to 120) tumor nodules, all measuring less than 1 mm in diameter (Fig 2). Collectively, the lungs taken from all non–tumor-bearing animals included in this study weighed 0.16 g ± 0.02 (n = 23). The weight of the lungs taken from the tumor-bearing animals that received TF2-pretargeted 99mTc-peptide or 18F-FDG at 1.5 hours was not significantly higher than that of the non–tumor-bearing lungs (Table 1), but the weight of lungs in the other groups of animals was approximately 50 mg higher (P < .05), an indication of the relatively low tumor content in the lungs.
View larger version (51K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2: Histologic appearance of micrometastatic GW-39 (T) (arrows) in the lungs. Left: Representative sample of lungs removed from a mouse in the first set of studies, approximately 3 weeks after implantation. Center: Representative sample of lungs removed from a mouse given pretargeted 99mTc-peptide 1 week after implantation. Right: Magnified portion of boxed area in center image shows size of one of the larger lesions in the field of view, which measured approximately 0.2 mm in diameter. (Hematoxylin-eosin stain.)
|
|
In the TF2-pretargeted animals, the tumor-infiltrated lungs at 1.5 hour after injection showed 3.32% injected dose per gram ± 1.23 compared with only 0.37% injected dose per gram ± 0.03 in the tumor-free lungs (nine-fold increase) (Table 1). The ratios of activity in the tumor-bearing lungs to that in the blood, liver, spleen, femur, and kidneys in the tumor-bearing animals given TF2-pretargeted 99mTc-peptide were 11.6 ± 2.9, 12.7 ± 3.8, 17.6 ± 5.3, 21.6 ± 10.8, and 0.9 ± 0.2, respectively, at 1.5 hours (n = 5). Uptake in the tumor-bearing lungs for the TF2-pretargeted peptide was persistent during 24 hours (2.49% injected dose per gram ± 0.21 and 2.26% injected dose per gram ± 0.53 at 3 and 24 hours, respectively), while activity in the normal tissues continued to decrease. For example, by 24 hours, there was 2.5-, 56-, and 23-fold more activity in the lungs than in the kidneys, blood, and liver, respectively. Activity in the lungs of the TF2-pretargeted, non–tumor-bearing animals also decreased over time, with only 0.06% injected dose per gram ± 0.01 at 24 hours, or approximately 38-fold less than that seen in the pretargeted tumor-bearing lungs.
While the uptake in the tumor-bearing lungs of the animals given the peptide alone or 18F-FDG was significantly higher than in the tumor-free lungs (P = .031 and .005, respectively), the average uptake was only about 1.5-fold higher in the lungs with metastases. The ratios of activity in the tumor-infiltrated lungs to that in the blood, liver, spleen, femur, and kidneys were 5.6 ± 0.4, 2.8 ± 0.3, 1.2 ± 0.2, 1.4 ± 0.5, and 2.2 ± 0.1, respectively, in animals given 18F-FDG (n = 5) but were not appreciably different than those found with 18F-FDG in the non–tumor-bearing animals (4.4 ± 0.7, 2.2 ± 0.1, 1.1 ± 0.3, 0.8 ± 0.1, and 1.9 ± 0.3, respectively) (n = 4).
Animals injected intravenously with GW-39 were evaluated just 1 week after tumor inoculation. Evaluation of the hematoxylin-eosin–stained lung slices indicated a similar number of tumor nodules (approximately 80–120 in a 5-µm slice), but all colonies were smaller, being no more than 0.25 mm in diameter (Fig 2). At 1.5 hours after injection of the pretargeted 99mTc-peptide, uptake in the tumor-bearing lungs exceeded that in most normal tissues by at least approximately fivefold (Table 2). By 24 hours, the ratios of activity in the lungs with metastatic disease versus that in the blood, liver, spleen, femur, and kidney were 48.1 ± 7.5, 17.0 ± 1.4, 40.6 ± 4.0, 64.1 ± 20.4, and 1.5 ± 0.1, respectively (n = 3), while these same ratios averaged no higher than 1.8:1 for the peptide alone at this time. The ratio of activity in the tumor-bearing lungs to activity in the other normal tissues for the animals given 18F-FDG was similar to that observed in the previous group of animals, but lung uptake was no higher than the uptake seen in two reference tumor-free animals included in this group of animals. Although the ratio of 18F-FDG activity in the tumor-infiltrated lungs to that in blood was approximately 6:1, a similar ratio was seen in the tumor-free animals, which suggests a lack of specificity for 18F-FDG uptake in the tumor-bearing lungs. As noted earlier, there was substantially more 18F-FDG activity in many of the other normal tissues than was observed in the pretargeted animals.
View this table:
[in this window]
[in a new window]
|
Table 2. Biodistribution Data of Pretargeted Peptide, Peptide Alone, or 18F-FDG in Nude Mice 1 Week after Implantation of GW-39
|
|
Whole-body planar images (Fig 3) showed a diffuse uptake in the chest in all animals with lung metastases that were given the pretargeted 99mTc-peptide compared with that with the peptide alone at 3 and 24 hours. At 3 hours, kidneys were prominent, as well as some lower abdominal activity, most likely in the intestine, but by 24 hours, only lung uptake in pretargeted animals, along with residual renal activity, was seen.
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3: Images (whole-body planar images overlaid on top of photographs of the mice for anatomic orientation) from SPECT camera of targeting of micrometastatic GW-39 tumors in lungs with TF2-pretargeted 99mTc-peptide or 99mTc-peptide alone. Top: Images obtained 3 hours after injection of 99mTc-peptide in three mice pretargeted with TF2 (315 µg [2 nmol]) administered 21 hours before 99mTc-peptide (0.2 nmol; approximately 300 µCi [11.1 MBq]) and three mice given the same dose of the peptide alone. Bottom: Same three mice imaged at 24 hours. Magenta arrows = plane of lungs, white arrows = uptake that is likely passage of 99mTc in intestine, and green arrows = position of kidneys. UB = urinary bladder.
|
|
Bispecific MoAb Pretargeting with PET
Histologic findings in the lungs removed from the animals undergoing the micro-PET study at CBIC or the tissue distribution study at CMMI showed that the number and size of the tumor nodules were similar in both groups, with no single lesion exceeding 0.3 mm in diameter.
The necropsy data of the animals given the 131I-peptide (Table 3) showed an approximately 15-fold higher concentration in the lungs of the animals that received the specific TF2 bispecific MoAb than in the lungs of those that received the peptide alone at 1.5 hours (1.79 ± 0.84 vs 0.12 ± 0.02, respectively). At 24 hours, the uptake in lungs persisted in the TF2-pretargeted animals, while the concentration of the peptide alone decreased 10-fold. Animals given the irrelevant TF6 bispecific MoAb had approximately threefold higher lung uptake than those given the peptide alone at 1.5 hours, but this uptake was still at least fivefold lower than that achieved with the specific TF2 bispecific MoAb. The average 18F-FDG accretion in the tumor-bearing lungs at 1.5 hours exceeded that in the TF2-pretargeted animals, but again there was a higher level of 18F-FDG activity in the normal tissues, such that the ratios of activity in the lungs (containing the human colon tumor) to that in normal tissue in the TF2-pretargeted animals far exceeded the same ratios in the animals that received 18F-FDG.
Results from micro-PET imaging with TF2-pretargeted 124I-peptide showed clear evidence of enhanced uptake in the chests of the four animals, while neither the animals injected with peptide alone nor the TF6 (control)-pretargeted animals showed accretion in the lungs. In addition to uptake in the kidneys, gastric uptake was observed at 1.5 and 5 hours in animals given the 124I-labeled peptide, but by 21 hours, this activity subsided considerably, while the signal in the lungs remained intense (Fig 4).
View larger version (95K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4: Micro-PET images of pretargeting of micrometastatic GW-39 tumors in lungs with 124I-peptide. Paired mice were pretargeted with specific TF2 anti-CEA bispecific MoAb (A, C, E, and F) or with irrelevant TF6 anti-CD22 bispecific MoAb (B, D, G, and H). Mice were given 315 µg (2.0 nmol) of TF2 or TF6, followed 21 or 24 hours later with 220–240 µCi (8.14–8.88 MBq) (approximately 0.2 nmol) of the 124I-peptide. Transverse sections through the chest (A–D) were adjusted to the same scale and signal intensity and reveal an appreciably higher concentration of activity in the chest with TF2 bispecific MoAb (A, C) than with TF6 bispecific MoAb (B, D). Selective uptake was apparent within just 1.5 hours. The 21-hour images in TF2-pretargeted mice show uptake in individual tumor colonies (arrows in C). E–I, Coronal sections provide an overview of uptake in the body, showing specific uptake in the lungs (Lg) (with corresponding arrows) in TF2-pretargeted mice and not in TF6-pretargeted mice. Coronal views were obtained in a posterior plane cut so that the kidneys (K) (with corresponding arrows) and a portion of the stomach (S) (with corresponding arrows) could be seen. Stomach uptake is due to dehalogenation, while renal uptake is due to clearance of the activity by urinary excretion. Activity found in the urinary bladder (UB) in all animals at 1.5 hours is not seen in these more posterior coronal sections. Gastric and urinary bladder activity was reduced at later imaging. Yellow arrows = activity that likely represents contamination of the skin from excreted activity. Coronal section (I) in one of the mice given the 124I-peptide alone.
|
|
Regardless of the cross section viewed in the four animals given 18F-FDG, no tumor targeting was observed, whether examined at 1.5 hours (Fig 5, representative images of one pair of animals) or at 7 hours (not shown). Visualization was impeded in part by the high concentration of radioactivity in the bone marrow of the ribs, scapulae, and spine surrounding the lungs. This uptake overwhelmed the images because the intensity was increased in an attempt to reveal targeting within the chest cavity. Substantial heart wall activity in two of the four animals added to the difficulty of viewing the lungs, even on the transverse views. High uptake of 18F-FDG in the brain also was observed on the sagittal views.
View larger version (62K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a: Fluorine 18 FDG (810 µCi [29.9 MBq]) micro-PET images in animals bearing GW-39 micrometastatic colon cancer in the lungs. (a, b) Coronal sections (a more anterior than b). (c) Individual sagittal sections. (d, e) Transverse sections through different lung planes. Uptake in bone marrow is most prominent feature (white arrows). Activity in the heart wall (red arrow) was seen in one of the two animals (animal 1). Uptake in the brain (blue arrow) was apparent, particularly in the sagittal views. No accretion in the lungs could be appreciated in any of the views. An1 = animal 1, An2 = animal 2, UB = urinary bladder.
|
|
View larger version (67K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b: Fluorine 18 FDG (810 µCi [29.9 MBq]) micro-PET images in animals bearing GW-39 micrometastatic colon cancer in the lungs. (a, b) Coronal sections (a more anterior than b). (c) Individual sagittal sections. (d, e) Transverse sections through different lung planes. Uptake in bone marrow is most prominent feature (white arrows). Activity in the heart wall (red arrow) was seen in one of the two animals (animal 1). Uptake in the brain (blue arrow) was apparent, particularly in the sagittal views. No accretion in the lungs could be appreciated in any of the views. An1 = animal 1, An2 = animal 2, UB = urinary bladder.
|
|
View larger version (56K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5c: Fluorine 18 FDG (810 µCi [29.9 MBq]) micro-PET images in animals bearing GW-39 micrometastatic colon cancer in the lungs. (a, b) Coronal sections (a more anterior than b). (c) Individual sagittal sections. (d, e) Transverse sections through different lung planes. Uptake in bone marrow is most prominent feature (white arrows). Activity in the heart wall (red arrow) was seen in one of the two animals (animal 1). Uptake in the brain (blue arrow) was apparent, particularly in the sagittal views. No accretion in the lungs could be appreciated in any of the views. An1 = animal 1, An2 = animal 2, UB = urinary bladder.
|
|
View larger version (32K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5d: Fluorine 18 FDG (810 µCi [29.9 MBq]) micro-PET images in animals bearing GW-39 micrometastatic colon cancer in the lungs. (a, b) Coronal sections (a more anterior than b). (c) Individual sagittal sections. (d, e) Transverse sections through different lung planes. Uptake in bone marrow is most prominent feature (white arrows). Activity in the heart wall (red arrow) was seen in one of the two animals (animal 1). Uptake in the brain (blue arrow) was apparent, particularly in the sagittal views. No accretion in the lungs could be appreciated in any of the views. An1 = animal 1, An2 = animal 2, UB = urinary bladder.
|
|
View larger version (40K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5e: Fluorine 18 FDG (810 µCi [29.9 MBq]) micro-PET images in animals bearing GW-39 micrometastatic colon cancer in the lungs. (a, b) Coronal sections (a more anterior than b). (c) Individual sagittal sections. (d, e) Transverse sections through different lung planes. Uptake in bone marrow is most prominent feature (white arrows). Activity in the heart wall (red arrow) was seen in one of the two animals (animal 1). Uptake in the brain (blue arrow) was apparent, particularly in the sagittal views. No accretion in the lungs could be appreciated in any of the views. An1 = animal 1, An2 = animal 2, UB = urinary bladder.
|
|
Autoradiography
Micro-PET images of the TF2-pretargeted animals were highly suggestive that the individual tumor colonies could be disclosed, but because it was impossible to compare the image with anatomic findings, autoradiography was performed to determine if the uptake was selectively localized in tumor. Whole-body planar images obtained at CMMI of the TF2-pretargeted animals 3 hours after injection of 111In-peptide again displayed specific targeting of the lungs as compared with findings in animals pretargeted with the irrelevant TF6 anti-CD22 bispecific MoAb (not shown). Necropsy data indicated that there was only 0.34% injected dose per gram ± 0.15 (n = 4) in the lungs of the TF6-pretargeted animals, compared with 3.2% injected dose per gram ± 1.0 (n = 4) in the lungs of the TF2-pretargeted animals. Slices of lungs taken from the TF6-pretargeted mice failed to show any signal intensity on the x-ray film after 14 days of exposure (not shown). The x-ray film disclosed radioactivity surrounding many of the corresponding visible tumor colonies seen on the hematoxylin-eosin–stained slices in animals pretargeted with TF2 (Fig 6).
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6: Autoradiographic data of lungs taken from two representative animals implanted 3 weeks earlier with GW-39 and pretargeted with TF2 anti-CEA bispecific MoAb (120 µg [0.76 nmol]) followed 21 hours later with 0.076 nmol (0.045 mCi [1.665 MBq]) of 111In-labeled peptide. Necropsy was performed 3 hours later; lungs were fixed in formalin for 2 hours and embedded in paraffin overnight. A, B, Hematoxylin-eosin–stained 5-µm slices show location of several lung nodules (arrows) that were about 0.3–0.8 mm in diameter. Other, smaller nodules were seen at microscopic examination (not shown). C, D, Respective autoradiographs after 14-day exposure on x-ray film. E, F, Stained slices overlaid with autoradiographs show the activity surrounding visible and microscopic metastatic colon cancer in the lungs. There was a small number of lesions that did not show uptake. Thus, uptake in the lungs was specifically associated with tumor nodules and was not a generalized, diffuse uptake.
|
|
|
DISCUSSION
|
|---|
An anti-CEA Fab' (arcitumomab) labeled with 99mTc was introduced as a clinical imaging agent for metastatic colon cancer on the basis of its ability to help stage disease in patients with liver metastases for salvage resection (11,25). These directly radiolabeled antibody fragments, however, did not provide the image resolution needed for broad clinical acceptance that was achieved with 18F-FDG PET for this and many other cancer types. Results of animal studies (26) with an 18F-labeled anti-CEA diabody have been reported recently, but this product had low yields (<2%), with tumor uptake of only 2.7% at 1 hour and tumor-to-nontumor ratios no better than 1:1, requiring 6 hours for tumor-to-blood ratios to improve to 6.8:1. Our prior experience in the same xenograft model, as well as the work presented in our study, has shown higher tumor uptake and tumor-to-blood ratios of 5:1 or greater at 1 hour with 18F-FDG (18). Thus, these 18F-labeled anti-CEA diabodies did not have a level of targeting that could rival 18F-FDG. In this context, the anti-CEA pretargeting imaging modality may constitute a specific and sensitive molecular probe for CEA-expressing malignancies. Indeed, this prototype human cancer antibody provides a proof of principle for the use of this pretargeting strategy with both SPECT and PET radionuclides, so that this technology now can be applied to other antibodies against immunologic and biochemical markers having specificity for cancer.
A signal amplification has been shown with SPECT (99mTc) (17) and PET (124I) (18) in established subcutaneous CEA-expressing human colon cancer xenografts with this pretargeting method and has now been shown in a lung micrometastatic model of colon cancer. These data indicate that pretargeting provides a more selective tumor uptake (tumor-bearing lungs vs non–tumor-bearing lungs) and a better contrast ratio (tumor-bearing lungs vs normal tissues) when compared with 18F-FDG uptake. Although the lungs are not the primary metastatic site for colorectal cancer, this and a number of other CEA-producing cancers can and do spread to the lungs, which makes this model relevant. However, the most important finding is that this pretargeting method amplified the signal to such a degree that tumor nodules smaller than the threshold of detection with even micro-PET (27) appeared as distinct foci. This amplification was a result of the method's ability to enhance radiotracer uptake in the target tissue while simultaneously allowing rapid and thorough removal of the radiotracer from normal tissues, thus achieving a high signal-to-noise ratio. Indeed, the HSG-peptide was designed for its versatility in facilitating conjugation with many SPECT and PET diagnostic radionuclides, as well as therapeutic isotopes (28). The TF2 bispecific MoAb construct used herein is also unique because it represents a new platform technology to customize the production of multivalent, multitargeting proteins, including antibodies, by using covalent tethering of Fab and single-chain constructs of diverse antibodies (16). Novel constructs have also been prepared for a streptavidin-radiolabeled biotin pretargeting approach that can provide a similar signal intensification advantage as the bispecific MoAb-based pretargeting approach, but streptavidin's immunogenicity would likely limit repeated imaging studies (29).
Fluorine 18 FDG PET is currently used for the detection of colon cancer with a sensitivity and specificity that have been reported to be as high a 90% (4); however, the sensitivity can be lower in the more mucinous cancer types, and other factors affect specificity (1,9). A limitation of our study was that visualization of tumor within the lungs with 18F-FDG was hampered in part by the elevated uptake in the bone marrow of mice. However, even a transverse section through the chest could not reveal any evidence of tumor involvement in the lungs with 18F-FDG, while uptake could be easily seen with pretargeting. While there undoubtedly will be differences in the uptake of a CEA-pretargeted radiolabeled peptide on the basis of CEA expression and physiologic factors in colorectal cancer, as there will also be differences in 18F-FDG uptake, this pretargeting technique has potential not only to provide a highly sensitive method for localizing colorectal cancer but also to enhance the specificity of cancer targeting in situations in which 18F-FDG is known to be problematic (eg, inflammation or infection).
Although we have shown a capability to localize tumors in a pretargeting setting within a similar time after the radiotracer injection as that of 18F-FDG, the pretargeting procedure requires patients to be given the bispecific MoAb 1 day or more in advance of their receiving the radiotracer. In practice, the bispecific MoAb could be given by the patient's oncologist, who would then coordinate the scheduling of the radiotracer study with the nuclear physician. A limitation is that the rapid filtration of the radiotracer into the urinary bladder could obscure uptake in the surrounding region, but one of the advantages of using a longer-lived radionuclide, such as 124I, is that images could be obtained several hours or even as long as 1 day later, when the majority of the activity would have been eliminated from the bladder. Gastric uptake that represents release of radioiodide from the 131I- or 124I-labeled peptide, because this same peptide radiolabeled with 111In showed no gastric activity (18), occurs in the early images and could interfere with interpretation in this region. Other positron-emitting radionuclides, such as gallium 68, copper 64, or 18F, may be preferred for this reason, but because gastric uptake is largely eliminated in the later images, the longer half-life of 124I may compensate for this. Another limitation is that the higher photon energy of 124I does degrade image resolution to some degree, which might have an impact on the sensitivity afforded by this procedure, but the very high contrast ratios and persistent uptake in the tumor obtained with this method might offset this limitation.
In conclusion, our studies introduce multivalent, bispecific antibodies for signal amplification of pretargeted molecular imaging that can be used with SPECT and PET to improve imaging of metastatic CEA-expressing human colonic cancers. This may be complementary to 18F-FDG imaging because of its antibody-based specificity.
|
ADVANCES IN KNOWLEDGE
|
|---|
- A new two-step molecular imaging procedure that utilizes a technetium 99m– or iodine 124–labeled peptide pretargeted with an anti–carcinoembryonic antigen (CEA) bispecific antibody is highly specific and more sensitive for detection of small (<0.3 mm in diameter) microdisseminated human colon tumors in the lungs of nude mice than fluorine 18 (18F) fluorodeoxyglucose (FDG).
- Autoradiographic data provide correlative evidence of the highly selective uptake of the pretargeted radiolabeled peptide in the metastatic tumor colonies.
|
IMPLICATION FOR PATIENT CARE
|
|---|
- Bispecific antibody pretargeting with tumor-associated antibodies introduces a potentially more specific method of tumor imaging than 18F-FDG, which could provide a complementary method of molecular imaging for cancers expressing the target antigen, such as in colorectal, pancreatic, and lung carcinomas expressing CEA.
|
ACKNOWLEDGMENTS
|
|---|
We thank Louis Osorio, BS, Lenka Muskova, BS, Dion Yeldell, BS, Tom Jackson, NPT, and Marisol Hernandez, BS, for their technical assistance, and Zhiliang Ying, MD (Columbia University, New York, NY) for statistical review. At CBIC, we thank Michael Synan, Klaus Hamacher, and Paresh Kothari for their technical assistance with micro-PET studies.
|
FOOTNOTES
|
|---|
Abbreviations: CBIC = Citigroup Biomedical Imaging Center • CEA = carcinoembryonic antigen • CMMI = Center for Molecular Medicine and Immunology • FDG = fluorodeoxyglucose • HSG = histamine-succinyl-glycine • MoAb = monoclonal antibody
Guarantors of integrity of entire study, R.M.S., H.K.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, R.M.S., H.K., S.V., E.A.R., C.H.C., S.J.G.; experimental studies, R.M.S., H.K., S.V., W.J.M., E.A.R., C.H.C., S.J.G.; statistical analysis, R.M.S.; and manuscript editing, all authors
|
References
|
|---|
- Abouzied MM, Crawford ES, Nabi HA. 18F-FDG imaging: pitfalls and artifacts. J Nucl Med Technol 2005;33:145–155. [Abstract/Free Full Text]
- Kostakoglu L, Hardoff R, Mirtcheva R, Goldsmith SJ. PET-CT fusion imaging in differentiating physiologic from pathologic FDG uptake. RadioGraphics 2004;24:1411–1431. [Abstract/Free Full Text]
- Rohren EM, Turkington TG, Coleman RE. Clinical applications of PET in oncology. Radiology 2004;231:305–332. [Abstract/Free Full Text]
- Esteves FP, Schuster DM, Halkar RK. Gastrointestinal tract malignancies and positron emission tomography: an overview. Semin Nucl Med 2006;36:169–181. [CrossRef][Medline]
- Atri M. New technologies and directed agents for applications of cancer imaging. J Clin Oncol 2006;24:3299–3308. [Abstract/Free Full Text]
- Kelloff GJ, Hoffman JM, Johnson B, et al. Progress and promise of FDG-PET imaging for cancer patient management and oncologic drug development. Clin Cancer Res 2005;11:2785–2808. [Abstract/Free Full Text]
- Kelloff GJ, Krohn KA, Larson SM, et al. The progress and promise of molecular imaging probes in oncologic drug development. Clin Cancer Res 2005;11:7967–7985. [Abstract/Free Full Text]
- Yang DJ, Kim EE, Inoue T. Targeted molecular imaging in oncology. Ann Nucl Med 2006;20:1–11. [Medline]
- Subhas N, Patel PV, Pannu HK, Jacene HA, Fishman EK, Wahl RL. Imaging of pelvic malignancies with in-line FDG PET-CT: case examples and common pitfalls of FDG PET. RadioGraphics 2005;25:1031–1043. [Abstract/Free Full Text]
- Goldenberg DM, Larson SM. Radioimmunodetection in cancer identification. J Nucl Med 1992;33:803–814. [Free Full Text]
- Goldenberg DM, Juweid M, Dunn RM, Sharkey RM. Cancer imaging with radiolabeled antibodies: new advances with technetium-99m-labeled monoclonal antibody Fab' fragments, especially CEA-Scan and prospects for therapy. J Nucl Med Technol 1997;25:18–23. [Abstract]
- Kenanova V, Wu AM. Tailoring antibodies for radionuclide delivery. Expert Opin Drug Deliv 2006;3:53–70. [CrossRef][Medline]
- Olafsen T, Kenanova VE, Sundaresan G, et al. Optimizing radiolabeled engineered anti-p185HER2 antibody fragments for in vivo imaging. Cancer Res 2005;65:5907–5916. [Abstract/Free Full Text]
- Le Doussal JM, Martin M, Gautherot E, Delaage M, Barbet J. In vitro and in vivo targeting of radiolabeled monovalent and divalent haptens with dual specificity monoclonal antibody conjugates: enhanced divalent hapten affinity for cell-bound antibody conjugate. J Nucl Med 1989;30:1358–1366. [Abstract/Free Full Text]
- Rossi EA, Chang CH, Losman MJ, et al. Pretargeting of carcinoembryonic antigen-expressing cancers with a trivalent bispecific fusion protein produced in myeloma cells. Clin Cancer Res 2005;11:7122s–7129s. [Abstract/Free Full Text]
- Rossi EA, Goldenberg DM, Cardillo TM, McBride WJ, Sharkey RM, Chang CH. Stably tethered multifunctional structures of defined composition made by the dock and lock method for use in cancer targeting. Proc Natl Acad Sci U S A 2006;103:6841–6846. [Abstract/Free Full Text]
- Sharkey RM, Cardillo TM, Rossi EA, et al. Signal amplification in molecular imaging by pretargeting a multivalent, bispecific antibody. Nat Med 2005;11:1250–1255. [CrossRef][Medline]
- McBride WJ, Zanzonico P, Sharkey RM, et al. Bispecific antibody pretargeting PET (ImmunoPET) with an 124I-labeled hapten-peptide. J Nucl Med 2006;47:1678–1688. [Abstract/Free Full Text]
- Sharkey RM, Weadock KS, Natale A, et al. Successful radioimmunotherapy for lung metastasis of human colonic cancer in nude mice. J Natl Cancer Inst 1991;83:627–632. [Abstract/Free Full Text]
- Goldenberg DM, Hansen HJ. Carcinoembryonic antigen present in human colonic neoplasms serially propagated in hamsters. Science 1972;175:1117–1118. [Abstract/Free Full Text]
- Sharkey RM, Juweid M, Shevitz J, et al. Evaluation of a complementarity-determining region-grafted (humanized) anti-carcinoembryonic antigen monoclonal antibody in preclinical and clinical studies. Cancer Res 1995;55(23 suppl):5935s–5945s. [Abstract/Free Full Text]
- Sharkey RM, McBride WJ, Karacay H, et al. A universal pretargeting system for cancer detection and therapy using bispecific antibody. Cancer Res 2003;63:354–363. [Abstract/Free Full Text]
- Leung SO, Goldenberg DM, Dion AS, et al. Construction and characterization of a humanized, internalizing, B-cell (CD22)-specific, leukemia/lymphoma antibody, LL2. Mol Immunol 1995;32:1413–1427. [CrossRef][Medline]
- Sharkey RM, Karacay H, Richel H, et al. Optimizing bispecific antibody pretargeting for use in radioimmunotherapy. Clin Cancer Res 2003;9:3897S–3913S. [Abstract/Free Full Text]
- Hughes K, Pinsky CM, Petrelli NJ, et al. Use of carcinoembryonic antigen radioimmunodetection and computed tomography for predicting the resectability of recurrent colorectal cancer. Ann Surg 1997;226:621–631. [CrossRef][Medline]
- Cai W, Olafsen T, Zhang X, et al. PET imaging of colorectal cancer in xenograft-bearing mice by use of an 18F-labeled T84.66 anti-carcinoembryonic antigen diabody. J Nucl Med 2007;48:304–310. [Abstract/Free Full Text]
- Tai YC, Ruangma A, Rowland D, et al. Performance evaluation of the microPET focus: a third-generation microPET scanner dedicated to animal imaging. J Nucl Med 2005;46:455–463. [Abstract/Free Full Text]
- Sharkey RM, Goldenberg DM. Advances in radioimmunotherapy in the age of molecular engineering and pretargeting. Cancer Invest 2006;24:82–97. [CrossRef][Medline]
- Goldenberg DM, Sharkey RM, Paganelli G, Barbet J, Chatal JF. Antibody pretargeting advances cancer radioimmunodetection and radioimmunotherapy. J Clin Oncol 2006;24:823–834. [Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
W. J. McBride, R. M. Sharkey, H. Karacay, C. A. D'Souza, E. A. Rossi, P. Laverman, C.-H. Chang, O. C. Boerman, and D. M. Goldenberg
A Novel Method of 18F Radiolabeling for PET
J. Nucl. Med.,
June 1, 2009;
50(6):
991 - 998.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. M. Goldenberg, E. A. Rossi, R. M. Sharkey, W. J. McBride, and C.-H. Chang
Multifunctional Antibodies by the Dock-and-Lock Method for Improved Cancer Imaging and Therapy by Pretargeting
J. Nucl. Med.,
January 1, 2008;
49(1):
158 - 163.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
