The 2025 PEGS Boston Summit (The Essential Protein & Antibody Engineering Summit) was held in Boston, MA on May 12-16. Bringing together a global community of experts, innovators, and leaders in the field, the Summit offered cutting-edge insights into the latest advances in drug development, protein and antibody engineering, immunotherapy, radiotherapy, and AI/ML-driven biologics research. With over 300 presentations, breakout sessions, and interactive seminars, attendees had unparalleled opportunities to network, learn, and engage in meaningful collaborations.
Engineering
Oncology
Multispecifics
Immunotherapy
Expression
Immunogenicity
Emerging therapeutics
Machine learning
→ Biointron’s Highlighted Points:
1. Engineering Antibodies
A modular approach for cytosolic penetration of tumor cells based on bispecific antibodies containing a masked cytosol-penetrating Fab on one arm and a tumor-targeting scFv linked via an endosomal cleavable linker on the other arm: Such TME-dependent as well as TAA-specific cytosol-penetrating antibodies have the potential to serve as a platform to deliver macromolecular cargoes for addressing intracellular targets in tumor cells. (Technische Universität Darmstadt)
Mammalian display to secretion switchable libraries for antibody preselection and high-throughput functional screening. Libraries secreting antibodies are a perfect match for microfluidics-assisted high-throughput function first screens. The versatility and fruitful options arising from combining these emerging technologies were discussed. (Merck Healthcare KGaA)
An engineered tumor-targeting peptide (PIP) can selectively bind several tumor-associated integrins, offering broad applicability across solid tumors. The development of multispecific PIP-drug conjugates were discussed. (TwoStep Therapeutics)
2. Antibodies for Cancer Therapy
Advancing TCRm Antibody Therapy: Tumor cells present specific peptides on HLA-I molecules, which can be targeted by TCRm antibodies to address otherwise undruggable proteins. However, the limited peptide copies and the specificity challenges of targeting these epitopes pose therapeutic hurdles. (MD Anderson Cancer Center)
Despite the curative potential of cancer immunotherapy, most patients do not benefit from treatment. Glyco-immune checkpoints—interactions of cancer glycans with inhibitory glycan-binding receptors called lectins—have emerged as prominent mechanisms of resistance to existing immunotherapies. Antibody-lectin chimeras are a biologic framework for glyco-immune checkpoint blockade that is now moving toward the clinic. (Massachusetts Institute of Technology)
Immunogenicity risk assessment and mitigation is desirable to aid successful engineering of protein therapeutics including antibodies. Multiple in silico and ex vivo assays suggest that some mutations (knob-into-hole and Fab) used to facilitate in vivo assembly of bispecific IgG represent a low risk for immunogenicity. Common light chains may reduce the immunogenicity risk of bispecifics. (Genentech)
A significant effort has improved antibody drug conjugate (ADC) linker, payload, and conjugation strategies, but less emphasis has been placed on engineering the protein. Strategies for antibody engineering include the use of multivalent and/or biparatopic antibodies, epitope selection with antibody/ADC combinations, and Fc-engineering to improve the therapeutic window. (University of Michigan)
3. Multispecifics
Recent research evaluates trispecific T cell engagers targeting two tumor-associated antigens in multiple myeloma/lymphoma. Successful T cell-engaging antibodies target antigens selectively expressed on tumors with minimal/no expression in other tissues and eliminate malignant cells resulting in long-term clinical benefit. Several bispecific T cell engagers have been approved in hematological malignancies. (Johnson & Johnson)
Boosting memory T cells with co-stimulatory bispecifics targeting 4-1BB or CD28 have emerged in oncology. T cell bispecifics (TCBs) are aiming for chemo-free treatments, with emerging clinical data showing compelling safety and efficacy profiles, evidence of costimulatory action, and enhanced memory T cell responses, paving the way for novel and innovative combination therapies. (Roche)
Engineered T cells secreting TCE antibodies (STAb): T lymphocyte redirection has been addressed by infusion of bispecific T cell engagers (TCEs) and adoptive transfer of CAR-T cells. A novel strategy based on in situ secretion of TCEs by engineered T cells, termed STAb-T, aims to combine active tumor trafficking of T cells with polyclonal T cell recruitment by locally secreted TCEs. STAb-T cells have demonstrated potent antitumor activity in mouse models. (Spanish National Cancer Research Centre (CNIO))
4. Advances in Immunotherapy
Immune checkpoint blockade (ICB) and particularly the combination of multiple checkpoint inhibitors, while responsible for improved rates of survival in multiple malignancies, has suffered from recent clinical trial failures. Research into combining multiple ICB molecules and the importance of Fc/FcgR interactions for antibody therapeutics are needed for the development of next-generation checkpoint inhibitors. (Vir Biotechnology)
TCR-T cell therapy is a promising approach with high specificity and sensitivity to low-abundance antigens, targeting both surface and intracellular proteins. In contrast, mAbs and CAR T cells only target surface antigens. TCR-based bispecific T cell engagers (BiTEs) may be a solution to the challenges in manufacturing, cost, and accessibility. (BlueSphere Bio)
Tumor-target heterogeneity presents a major obstacle for immunotherapies, often resulting in incomplete tumor elimination and treatment resistance. A novel class of T cell-redirecting multispecific antibodies addresses this problem with a single-molecule approach that converts target-negative cells into target-positive cells by transforming endogenous tumor self-defense mechanisms into targetable vulnerabilities. (Seattle Childrens Research Institute)
5. Protein Expression
A typical human gene has a markedly different structure than transgenes used to manufacture proteins. By reintroducing introns back into the cDNA sequence, a mammalian transgene can be transformed. An AI model of "genetic syntax" can be used to learn how to combine coding and non-coding DNA to improve protein expression. (ExpressionEdits Ltd.)
Mammalian cell cultures undergo a variety of stresses under bioprocess conditions, including nutrient limitations, waste product accumulation, and increased osmolarity. In response to these stress conditions, cells alter their gene expression, tRNA prevalence, and tRNA synthetase enzymes—and exhibit epitranscriptomic modifications to their tRNAs. By analyzing these changes, we can recode both the gene sequence for our therapeutic protein and transcription factors to improve productivity. (University at Albany)
Protein medicines have revolutionized our ability to prevent and treat human diseases. However, the World Health Organization estimates that at least 30% of the world’s population still lacks access to essential medicines. A portable, decentralized synthesis platform is being created to make medicines when and where they are needed. (Stanford University)
6. Emerging Therapeutics
Neonatal Fc receptor is a popular target for treatment of autoimmune disorders due to its role in maintaining IgG levels. Fc-ABDEG and albumin-binding VHH were combined to develop next-generation FcRn blockers with improved IgG clearance. Step-wise engineering was applied to optimize position and number of VHHs, their affinity to albumin, and the linker connecting it to Fc-ABDEG. (Argenx)
KnotBodies: lon channels are an important target class which are underserved by biologics. Small cysteine-rich peptides with ion-channel modulating activity can be inserted into antibody CDR loops while retaining their function. The resulting molecules modulate ion channel activity while benefitting from the optimal drug-like properties of antibodies. (Maxion Therapeutics)
Development of oral VHHs via simultaneous optimization of the affinity and protease stability. This oral VHH matches the efficacy of an injected antibody in a murine colitis model and exerts a strong pharmacodynamic effect in non-human primates. With high potency, gut stability, and favorable developability, oral VHHs offer a promising approach for IBD. (Genentech)
7. Machine Learning Approaches
Advancing Humanization through Computational Design: Besides low immunogenicity, immune-system-derived antibodies have favorable in vivo properties like long half-life and low selfreactivity. Designing nanobodies indistinguishable from human ones is therefore important. Deep learning strategies for designing and humanizing antibodies and nanobodies indistinguishable from immunesystem-derived ones, using tools like AbNatiV. (University of Cambridge)
Generative AI can be used to accelerate prediction and design in biomedicine and sustainability. Development of new machine learning methods that can exploit the enormous natural sequence diversity and sequence DNA at scale. (Harvard Medical School)
OASis, a granular antibody humanness score, and Sapiens, a language model for antibody humanization, are integrated into the open-source BioPhi web platform. Similar principles may be applied to de novo designed proteins, but there are challenges stemming from computational complexity and fundamental biological constraints. (MSD)
Thank you to everyone who joined us at PEGS! We had a fantastic time connecting with you and sharing how we can help you achieve antibody development. Our expert team would be happy to answer any follow-up questions. Feel free to email us at info@biointron.com or visit our website at www.biointron.com.
