TF2 is a recombinant bispecific antibody composed of:
Anti-carcinoembryonic antigen (CEA) Fab fragments: Derived from humanized antibodies (hMN-14 or labetuzumab) for tumor targeting .
Anti-histamine-succinyl-glycine (HSG) Fab fragment: Enables binding to radiolabeled hapten peptides like IMP288 .
Dock-and-Lock (DNL) assembly: Creates a trivalent, bispecific structure (~157 kDa) with high binding avidity .
Pretargeting: TF2 is administered first, binding to CEA-expressing tumors.
Radiolabeled peptide delivery: A secondary agent (e.g., ⁶⁸Ga-IMP288) binds to TF2’s HSG site, enabling precise tumor imaging or therapy .
Immuno-PET: ⁶⁸Ga-IMP288 with TF2 achieved 88% sensitivity and 100% specificity in detecting CEA+ tumors, outperforming FDG-PET .
Pancreatic Cancer: TF10 (PAM4×HRPO variant) enabled early detection with high tumor-to-normal tissue contrast .
¹⁷⁷Lu-IMP288: Delivered targeted radiation to CEA+ tumors with minimal off-target toxicity .
Dose Optimization: 75 mg TF2 + 25 µg IMP288 achieved therapeutic efficacy in colorectal cancer .
TF2 is a bispecific antibody created using the Dock-and-Lock protein engineering platform technology. Unlike conventional antibodies that attach to only one receptor, TF2 contains an additional binding site that recognizes a radioisotope-carrying peptide while maintaining its primary binding to carcinoembryonic antigen (CEA or CEACAM5). This dual-binding capability allows for innovative pretargeting approaches where the delivery of radiation can be separated from the tumor-targeting antibody infusion .
Anti-TF antibodies like Nemod-TF1 and Nemod-TF2 target the Thomsen-Friedenreich (TF) antigen (CD176, Galβ1-3GalNAc), a carbohydrate structure O-glycosidically linked to threonines and serines of membrane glycoproteins. This antigen is masked in normal adult tissues by extended carbohydrate chains and sialic acid but becomes exposed during malignant transformation. The TF antigen has been implicated in endothelial adhesion and tumor invasion, suggesting its potential as a marker of cancer cell aggressiveness .
Bispecific antibodies (BsAbs) contain two distinct binding sites directed at different antigens or different epitopes on the same antigen. In the case of TF2, one binding domain recognizes tumor-associated CEA while the second binding domain targets a radioisotope-carrying peptide. This molecular design enables a pretargeting approach where the antibody first localizes to the tumor, followed by administration of a small radioisotope-labeled peptide that binds to the second domain of the already-localized antibody. This strategy improves tumor-to-background signal ratios for imaging and potentially enhances therapeutic efficacy .
In a phase II clinical trial (NCT02587247), pretargeted immunoPET with TF2 and 68Ga-labeled IMP288 demonstrated superior diagnostic performance compared to conventional FDG-PET. Based on lesion analysis, the pretargeting approach showed sensitivity, specificity, positive predictive value, and negative predictive value of 88%, 100%, 100%, and 67%, respectively, compared to 76%, 67%, 87%, and 33% for FDG-PET. This indicates that TF2-based pretargeting offers significantly enhanced tumor detection capabilities, particularly for smaller tumors where sensitivity reached 44% compared to 0% in control groups .
Research has demonstrated the efficacy of TF2 antibodies in targeting various cancer types, particularly colorectal cancer, where TF2-based immunoPET with 68Ga-labeled peptides has shown high specificity and sensitivity. Additionally, promising results have been observed in pancreatic adenocarcinoma models using TF10 (a bispecific antibody targeting MUC1), suggesting broader applications across multiple solid tumor types. The TF antigen itself has been studied in lung, breast, and liver carcinomas, indicating potential applications for TF-targeting antibodies in these cancer types as well .
| Antibody | Positive Expression Rate | Association with Clinical Parameters |
|---|---|---|
| Nemod-TF1 | ~40% of breast cancers | Inversely correlated with Ki67; Predicts shortened survival in triple-negative BC |
| Nemod-TF2 | ~40% of breast cancers | Higher in ER+ (46.0%) and PR+ (46.9%) cancers; Positively correlated with TA-MUC1 |
The pretargeting approach with TF2 involves a two-step protocol: first, administration of the TF2 bispecific antibody that binds to the tumor target (e.g., CEA); second, after allowing sufficient time for tumor localization and clearance of unbound antibody from circulation, administration of a radioisotope-labeled peptide (typically gallium-68 labeled) that binds to the second specificity of TF2. Key methodological considerations include optimizing the interval between antibody and peptide administration (typically several hours), selecting appropriate peptide radioisotope (68Ga provides good imaging properties), and determining optimal dosing to maximize tumor-to-background ratios. This approach has demonstrated superior sensitivity (67% vs. 31% in control groups) for tumor detection, particularly for smaller tumors (<200 mg) .
When designing experiments with TF2 antibodies, appropriate controls should include: (1) conventional monoclonal antibodies targeting the same antigen to compare efficacy; (2) non-specific antibodies of the same isotype to control for non-specific binding; (3) in pretargeting experiments, a control group receiving directly radiolabeled targeting antibodies without the pretargeting approach; and (4) tissue samples known to be positive or negative for the target antigen. For example, in pretargeted immunoPET studies, comparing TF2 pretargeting with direct labeling approaches demonstrated superior tumor-to-background ratios with the pretargeting strategy, providing a critical methodological validation .
Immunogenicity remains a significant challenge in bispecific antibody development, including TF2. Advanced strategies to mitigate this include: (1) humanization of antibody sequences; (2) removal of potential T-cell epitopes through protein engineering; (3) implementation of deimmunization algorithms during antibody design; and (4) careful monitoring of anti-drug antibody responses in experimental models. Interestingly, the immunogenicity of the TF antigen itself may actually be beneficial in cancer therapy contexts, as it has been hypothesized that TF expression might enhance anti-tumor immune responses by presenting as a "non-self" epitope that facilitates immune recognition of cancer cells .
The heterogeneous expression of targets like CEA or the TF antigen presents a significant challenge for antibody-based cancer targeting. Advanced strategies to address this include: (1) targeting multiple cancer-associated antigens simultaneously using cocktails of bispecific antibodies; (2) focusing on antigens enriched on cancer stem cells, such as the TF antigen which has been found to co-express with cancer-initiating cell markers like CD44 and CD133; (3) utilizing pretargeting approaches that amplify signal from low-expressing regions; and (4) developing imaging analysis algorithms that can detect and quantify heterogeneous expression patterns. Studies have demonstrated that TF is co-expressed with CD44 or CD133 in various carcinomas, suggesting its potential utility in targeting cancer-initiating cell populations despite heterogeneous expression .
Distinguishing specific from non-specific binding is crucial for accurate interpretation of results with TF2 antibodies. Advanced approaches include: (1) competitive binding assays using excess unlabeled antibody; (2) tissue clearing techniques combined with 3D imaging to improve signal-to-noise ratios; (3) dual-labeling approaches to confirm co-localization with known markers; and (4) validation using multiple antibody clones targeting different epitopes of the same antigen. For TF antigen detection, using both Nemod-TF1 and Nemod-TF2 antibodies provides complementary information due to their different fine specificities and affinities, despite their strong positive correlation (p<0.001) in tissue staining .
The combination of TF-targeting strategies with immune checkpoint inhibitors represents a promising frontier in cancer immunotherapy research. The TF antigen's nature as an "oncofetal" carbohydrate epitope that appears foreign to the immune system makes it potentially synergistic with checkpoint inhibition. Future research directions should explore: (1) whether TF-specific antibodies can enhance tumor immunogenicity and improve responses to checkpoint inhibitors; (2) development of bispecific antibodies targeting both TF and immune checkpoint molecules; (3) investigation of TF expression as a biomarker for checkpoint inhibitor response; and (4) evaluation of combination therapies in preclinical models. Research suggests that TF expression may increase immunogenicity of tumor cells, which could theoretically enhance the efficacy of checkpoint inhibition strategies .
Advanced glycobiology approaches could significantly enhance TF-targeting antibody development through: (1) detailed glycoproteomic analysis to identify specific carrier proteins of the TF antigen in different cancer types; (2) engineering antibodies that recognize cancer-specific TF-glycopeptide epitopes rather than just the TF glycan; (3) exploiting differential glycosylation patterns between normal and malignant tissues; and (4) developing glycosyltransferase inhibitors that could potentially increase TF exposure on cancer cells. Research has already identified that CD44 serves as a carrier molecule for the TF antigen not only in colorectal carcinomas but also in lung, breast, and liver carcinomas, suggesting this as a more general phenomenon that could be exploited for enhanced targeting specificity .
TF2 antibodies hold significant promise for theranostic applications due to their bispecific nature and pretargeting capabilities. Future research directions include: (1) development of companion diagnostic and therapeutic pairs using the same antibody platform; (2) integration of TF2 pretargeting with therapeutic radioisotopes beyond imaging applications; (3) exploration of TF2 variants capable of delivering payloads beyond radioisotopes, such as photosensitizers or cytotoxic drugs; and (4) clinical trials evaluating the safety and efficacy of theranostic approaches. The demonstrated efficacy of TF2 in pretargeted immunoPET, with its superior tumor-to-background ratios and enhanced detection of small tumors, provides a strong foundation for expanding into therapeutic applications using the same molecular platform .