Back in June this year, I went to Amsterdam to give a talk at “Antibody Engineering & Therapeutics Europe 2024”. I had a great time at the conference, and it presented many opportunities to gain some insights into research that is directly relevant to me, as well as research to broaden my horizons a little beyond the CDR loops. While I would love to go through all the fantastic talks, I’m opting to give some takeaways on only a subset:

Conditional Activity

• Maarten Merkx (Eindhoven University of Technology) gave a fascinating talk on designing caps to bind the CDR loops of antibodies and conjugating oligonucleotides onto these caps to allow them to dimerise. He showed that it is possible to design these oligonucleotides so that the linker between the two “plugs” is pH-sensitive, thereby leading to the activation of the antibody in the low-pH tumour microenvironment to reduce off-target effects [1].
• Dario Neri (ETH Zurich) gave an outline of a number of antibody formats, but one I found particularly interesting was an antibody-drug conjugate with a linker that would be targeted by tumour proteases, allowing the use of non-internalising antibodies to deliver cytotoxic payloads [2].

Antibodies beyond the Fv region

• Thom Sharp (Leiden University) presented his work on IgM-mediated complement activation. Using in situ cryo-ET, he shows that the hexagonal dome structures that are adopted when these antibodies bind antigens on a membrane surface allow for the deposition of complement system components on the membrane, leading to opsonisation [3].
• Mark Cragg (University of Southampton) presented research from his group on the importance of the hinge region of IgG antibodies. Using experimental and computational approaches, his team were able to show that differences in the disulphide linkages in the hinges between IgG1 and IgG2 isotypes result in vastly different conformational profiles and effectively change the downstream immune response of an anti-CD40 antibody [4].

Computational Antibody Design

• Laila Sakhnini (Novo Nordisk) gave a talk on tackling the challenge of non-specific binding prediction. She used language model embeddings and traditional machine learning approaches to accurately predict whether a sequence will result in non-specific binding. Interestingly, her results suggest that most of the predictive power of her models comes from the VH domain.
• Rahmad Akbar (University of Oslo) presented on using synthetic data to allow testing of new machine learning methodologies without running into the issue of data set size. He outlined how suites like Absolut! [5] would allow us to be able to test the expressivity of our models.

[1] Engelen W, Zhu K, Subedi N, Idili A, Ricci F, Tel J, Merkx M. Programmable Bivalent Peptide-DNA Locks for pH-Based Control of Antibody Activity. ACS Cent Sci. 2020 Jan 22;6(1):22-31. doi: 10.1021/acscentsci.9b00964. Epub 2019 Dec 23. PMID: 31989023; PMCID: PMC6978833.

[2] Dal Corso A, Cazzamalli S, Gébleux R, Mattarella M, Neri D. Protease-Cleavable Linkers Modulate the Anticancer Activity of Noninternalizing Antibody-Drug Conjugates. Bioconjug Chem. 2017 Jul 19;28(7):1826-1833. doi: 10.1021/acs.bioconjchem.7b00304. Epub 2017 Jul 6. PMID: 28662334; PMCID: PMC5521252.

[3] Sharp TH, Boyle AL, Diebolder CA, Kros A, Koster AJ, Gros P. Insights into IgM-mediated complement activation based on in situ structures of IgM-C1-C4b. Proc Natl Acad Sci U S A. 2019 Jun 11;116(24):11900-11905. doi: 10.1073/pnas.1901841116. Epub 2019 May 30. PMID: 31147461; PMCID: PMC6575175.

[4] Orr CM, Fisher H, Yu X, Chan CH, Gao Y, Duriez PJ, Booth SG, Elliott I, Inzhelevskaya T, Mockridge I, Penfold CA, Wagner A, Glennie MJ, White AL, Essex JW, Pearson AR, Cragg MS, Tews I. Hinge disulfides in human IgG2 CD40 antibodies modulate receptor signaling by regulation of conformation and flexibility. Sci Immunol. 2022 Jul 15;7(73):eabm3723. doi: 10.1126/sciimmunol.abm3723. Epub 2022 Jul 8. PMID: 35857577.

[5] Robert, P.A., Akbar, R., Frank, R. et al. Unconstrained generation of synthetic antibody–antigen structures to guide machine learning methodology for antibody specificity prediction. Nat Comput Sci 2, 845–865 (2022). https://doi.org/10.1038/s43588-022-00372-4