125I-Labeled trivalent, bispecific monoclonal antibody construct TF10 that targets mucin-1 and is reactive against a histamine-succinyl-glycine hapten IMP-288

Most individuals suffering from pancreatic adenocarcinoma (PAC) do not survive for more than 1 year after diagnosis, and <1% of these patients live beyond 5 years (1). Although surgical resection of the cancer is a possible intervention for this disease, only 10%–25% of the patients are considered suitable for this treatment because, by the time that the neoplasm is detected, the malignancy has metastasized to other organs and the tumor load in the patient is too high to warrant surgery (2). Patients with nonresectable PAC are treated either with gemcitabine or radiotherapy or a combination of the two; however, these treatments are not curative because they only prolong survival and improve the quality of life of the patient (2). Early detection of this invasive cancer would facilitate proper staging of the disease so that a suitable treatment regimen can be initiated to possibly improve patient prognosis (3). In this regard, the monoclonal antibody (mAb) PAM4, which specifically targets mucin 1 (MUC1), a glycoprotein that is overexpressed only in PAC tumors, was developed, radiolabeled with I or In, and shown with scintigraphy to detect neoplastic tumors in patients with pancreatic malignancies (4). However, intact radiolabeled antibodies (5) are known to have incomplete tumor penetration due to their large size (~150 kDa), and they are of limited utility to visualize cancerous lesions with different imaging modalities because they have a long blood circulating half-life and usually generate a low signal/noise (S/N) ratio (5). To amplify the signal obtained from an imaging agent so that it can be used to detect malignant tumors noninvasively, investigators have developed and evaluated a variety of strategies in preclinical studies with animals. One such strategy involves the pretargeting of cancer lesions with a suitable mAb (or its derivative), allowing some time for the pretargeting Ab to bind to the cancerous tissue and clear from blood circulation; the animals subsequently are injected with an appropriate radiolabeled small molecular weight ligand that binds to the pretargeting mAb or its derivative. This technique has been shown to generate higher S/N ratios during imaging compared to ratios obtained with a directly labeled mAb alone (6-8). Use of the pretargeting technique with multivalent (i.e., containing more than one antigen binding site), multispecific (i.e., can bind more than a single type of antigen) Abs for the imaging and therapy of cancer has been discussed in detail elsewhere (9, 10). Cardillo et al. developed bsPAM4 (or bsmAb), a divalent, bispecific F(ab’) mAb, by cross-linking a PAM4 Fab’ fragment to a murine anti–In-diethylenetriamine pentaacetic acid (In-DTPA) mAb Fab’ fragment and used the unlabeled bsPAM4 to pretarget human CaPan-1 cell xenograft PAC tumors in nude mice (4). The animals were then exposed to a radiolabeled peptide hapten that bound to the In-DTPA binding mAb Fab’; imaging showed that this technique generated higher S/N ratios compared to the directly labeled bsmAb alone. From this study, the investigators concluded that the peptide was suitable for use with scintigraphy to target and visualize human xenograft PAC tumors in nude mice (4). In an effort to further improve the S/N ratio that can be obtained with the pretargeting technique, Gold et al. generated a recombinant trivalent (i.e., three binding sites) bsmAb, designated TF10, and evaluated its use with scintigraphy for the visualization of xenograft PAC tumors in mice (1). The TF10 bsmAb is divalent for binding to MUC1 and monovalent for binding to a histamine-succinyl-glycine (HSG) motif containing hapten (DOTA-d-Tyr-d-Lys(HSG)-d-Glu-d-Lys(HSG)-NH (IMP-288, HSG hapten)), and was generated by linking two PAM4 Fab fragments to an anti-HSG hapten mAb Fab fragment as described elsewhere (1). The biodistribution of I-labeled TF10 ([I]-TF10) was studied in mice bearing PAC tumors and is discussed in this chapter. The biodistribution of In-labeled IMP-288 ([In]-IMP-288) and the visualization of PAC tumors in mice with this labeled hapten is discussed in a separate chapter of MICAD (www.micad.nih.gov) (11).