Safety and Efficacy of Bispecific Antibodies, ADCs, and Combination Therapies

Efficacy of bispecific antibodies (bsAbs) in solid tumours is limited by (i) lack of accessibility of the tumour site for immune effector cells, (ii) lack of sufficiently tumour-specific target antigens, and (iii) lack of costimulatory “signal 2” that enables thorough and long-lasting T cell activation. We plan to overcome these limitations by a combination of functionally interrelated bsAbs that target two different antigens expressed on both tumour cells and the tumour microenvironment/tumour vasculature and stimulate CD3 and CD28 on T cells.

IMV-M is a bispecific antibody that co-targets MUC16 and death receptor 5 (DR5) to induce tumour-selective apoptosis. MUC16 is overexpressed in non-small cell lung cancer, ovarian cancer, and pancreatic adenocarcinoma, while its expression in normal tissues is minimal. DR5 is broadly expressed across these tumour types. IMV-M has demonstrated potent antitumour activity in xenograft models, favorable safety in non-human primates, and is CMC-ready.

Since the beginning of immunotherapy, the limits for higher efficiency have been pushed. This shift did not come without tradeoffs for safety and immunogenicity. This presentation showcases several examples of drug enhancing approaches that also can have a profound impact on immunogenicity. Understanding the underlying mechanisms can help to avoid immunogenic transformation and lead to the development of better drugs for patients.

Bispecific antibody has already evolved as a well-validated therapeutic modality with multiple regulatory approvals mostly in oncology setting over the past few years. Multispecific antibody targeting more epitopes on the same or different targets has recently emerged as a novel modality with unique advantages to afford better potency via binding to additional epitopes, broader target space by engaging more challenging receptor complexes and higher specificity from restricting activity to only overlapping target expression pattern. FGF21 is a master coordinator for fatty acid homeostasis between liver and adipose tissue. Developing FGF21 mimetics has been one of a few therapeutic strategies currently being tested to treat MASH (metabolic dysfunction-associated steatohepatitis) clinically. Harnessing structure-aided design, we have engineered a multispecific AltibodyTM agonist mimicking FGF21 ligand function by activating FGFR1c only in the presence of its coreceptor KLB. Format-based stepwise optimization further enhances its agonist activity. Our study illuminates that a combination of factors, such as IgG subclass, linker length and building block, can contribute to the improved agonism, highlighting the challenges and complexity of engineering multispecific agonist antibody. Our FGFR1c x KLB multispecific Altibody offers an attractive alternative strategy to the current Fc-FGF21 fusion tested in clinic with superior PK, lower immunogenicity and similar efficacy in preclinical settings.

Allogeneic CAR T cells can overcome limitations associated with autologous cancer therapies, providing immediate access to standardised, affordable batches of CAR T with improved efficacy. We systematically interrogated genes providing resistance to allogeneic rejection using in vivo genome-wide CRISPR KO in T cells. We identified Fas and B2m and demonstrated using base editing in human CAR T cells that TCR/FAS-inactivation outperforms TCR/B2M-deletion in the resistance to both T and NK cell-mediated allo-rejection.

Broad application of T cell engagers to patients with solid tumors is limited by the availability of antigens that discriminate malignant from non-malignant tissues. We show that T cell engagers can be readily built for classic solid tumor targets with a lab-in-the-loop approach to AI/ML-guided antibody design, efficiently clearing tumors without established on-target toxicities. We provide examples of avidity-gated and Boolean logic-gated T cell engagers in diverse malignancies.

Vγ9Vδ2-T cells constitute a relatively homogeneous population of pro-inflammatory immune effector cells. This presentation will focus on the preclinical and early clinical development of bispecific Vγ9Vδ2-T cell engagers as a novel approach for cancer immunotherapy.

IL-2 is a clinically validated immunotherapy. However, toxicity limits its use and despite years of research, the optimal approach to deliver IL-2 has yet to be achieved. AZD6750 applies cis-guiding to deliver a modified IL-2 mutein preferentially to CD8+ T cells. This approach potentially limits the side effects of Aldesleukin, typically attributed to the broad effects of IL-2R signalling, which should result in a safer, more effective IL-2.

The blood-brain barrier (BBB) restricts the effective delivery of protein therapeutics to the central nervous system (CNS). To address this, we have developed an engineered Fc BBB transport vehicle (TV) that binds a highly expressed brain endothelial cell target and exploits receptor-mediated transcytosis to facilitate biotherapeutic delivery to the CNS. TVs were engineered to bind the apical domain of the human transferring receptor (hTfR) using directed evolution and without the use of amino acid insertions, deletions, or unnatural appendages. The TV platform is a highly modular CNS delivery platform that readily accommodates diverse antibody architectures, including bispecific formats, protein fusions and oligonucleotides.

In this work, we describe the development of a new TV platform that enables transport of ASOs into the brain, called “Oligonucleotide Transport Vehicle” (OTV). Systemically delivered OTV drives significant, cumulative, and sustained knockdown of the target gene expression across multiple CNS regions and all major cell types. When compared with other clinically relevant ASO delivery routes, such as a high affinity TfR antibody or direct ASO delivery to the CSF, systemic OTV delivery also results in more uniform ASO biodistribution and target knockdown. Our data supports systemically delivered OTV as a potential therapeutic oligonucleotide-delivery platform for neurological disorders.

This talk explores the development of 177Lu-AKIR001, a novel radio-antibody drug conjugate (RADC) for targeted cancer therapy. By integrating antibody engineering with radionuclide technology, this approach combines precise tumour targeting with localised radiation delivery. Preclinical findings demonstrate strong efficacy and safety, supporting the ongoing clinical trial. The presentation will discuss the design, challenges, and translational potential of RADCs, bridging protein engineering and radiopharmaceutical innovation.

Rheumatoid arthritis is a chronic autoimmune disease marked by joint inflammation and destruction. Citrullinated proteins, produced by PAD2 and PAD4 enzymes, drive autoantibody generation and joint damage. AZD1163, a novel bispecific monoclonal antibody, potently inhibits human PAD2 and PAD4. It shows high-affinity binding, preserves neutrophil function, avoids cytokine release, and is well tolerated. Favourable human pharmacokinetics support AZD1163’s continued clinical development.