The Antibody Society

Approval of Jazz Pharma’s zanidatamab sets up blockbuster fight in oncolocy. ADCs have become a hot topic over the last year or two but anti-HER2 ADCs for the treatment of various cancers have been on the market for over a decade now. Kadcyla from Genentech, an ADC version of the anti HER2 trastuzumab, was first approved in 2013. For many years this dominated the ADC space for HER2 positive tumors but recently AstraZeneca and Daiichi Sankyo’s Enhertu, an alternative trastuzumab ADC first approved in 2019, has become the biggest selling ADC for solid tumors (approx. $3B annual sales). Enhertu has been shown to out-perform Kadcyla in patients with metastatic breast cancer. Now a new drug enters the fray with the very recent approval of zanidatamab (Ziihera) from Jazz pharmaceuticals. Zanidatamab is a biparatopic antibody that targets two different epitopes on HER2 and is produced using an asymmetric Fc with a Fab on one side and scFv on the other.?The Fc contains ‘next generation’ knobs-into-holes mutations from Zymeworks. So far as I can tell the scFv is trastuzumab and the Fab is a slightly modified version of pertuzumab, two anti-HER2 antibodies originally from Genentech that have been approved for many years. Compared to the combination of these two antibodies, the bispecific leads to unique HER2 clustering and more effectively interferes with signaling of the cancer driver. But, rather than going head to head with those traditional mAbs Jazz is initially setting its sights on patients with biliary tract cancer which will put it head to head with Enhertu. It would then make sense to aim for the much larger breast cancer market, where Enhertu is accepted as a leading therapy and is heading towards becoming a front-line treatment. Will the potential toxicity of the bispecific antibody enable it to win out over the ADC? Or does ADC potency beat the HER2 biparatopic approach? Time will tell. My son recently got into the Godzilla film franchise, so this battle of ADC versus bispecific reminded me of the radioactive monster and its fight with the multi-headed monster King Ghidorah. I couldn’t find a good picture of that so I went with Godzilla versus Kong instead. ----- I'm Ian, I post about antibody engineering, recombinant proteins and my journey to bootstrap Gamma Proteins into a leading supplier of Fc receptors. If you like my content please reshare with your network and follow me to see more.

?? ?????????????????? ??????????????????????????'?? ???????? ???? ???????? ???????????????????????? ?????????????? ???????????????????????? ???????????????? ?? The development of ?????????????????????? ???????????????????? ???????????????????? demands a deep understanding of their ?????????????????????????? ???????????????? and their impact on immune effector functions. This study presents a ???????????????????? ?????? ???????????????????? ???????????????? to evaluate these effects using ???????????????????????? ???????? ???????????????? and in vitro glycoengineering—a step forward in aligning therapeutic protein design with QbD principles. ???? ?? ?????? ????????????????: ? ???? ???????????????????????? significantly influences ????????????????-?????????????????? ???????????????? ???????????????????????? (????????), with just a 1% decrease leading to >25% increase in ADCC. ?? ? ???????????????????????? ???????????????????? dramatically enhance FcγRIIIa binding and ADCC. ?? ? ?????????????????????????????? boosts complement activation via increased C1q binding ??? but minimally impacts ADCC in fucosylated species. ? ???????? ?????????????? ?????????????? show elevated FcγRIIIa binding ?? but reduce complement activation ??. ? ?????????????????????? generally lowers FcγR and C1q binding, underscoring the complexity of its role. ?? Through rigorous ?????????????????????? ????????????????, this research provides actionable tools to predict the relationship between glycan structures and therapeutic outcomes—enabling precise control over mAb quality and efficacy. ?? ?? ?????? ??????????????????: ? Glycosylation is a ???????????????? ?????????????? ?????????????????? in therapeutic mAb design. ?? ? Afucosylation is a primary driver of ADCC enhancement. ?? ? High mannose and galactose glycans show distinct roles in FcγR and complement activation. ?? ? Multivariate approaches are essential for optimizing therapeutic performance under QbD frameworks. ?? #Biopharma #MonoclonalAntibodies #Therapeutics #Glycosylation #ADCC #QualityByDesign Lek Pharmaceuticals University of Ljubljana, Biotechnical Faculty Tamara Cviji?, Matej Horvat, Jakob Plahutnik, Ana Golob and Jaka Maru?i?

EMA invites all professionals involved in marketing authorisation applications for new medicines to a Q&A session. Join Francesca Day (nee Mazzei), EMA’s Head of Therapeutic Areas, and Alessandro Faia, EMA press officer, as they delve into a critical topic: “The Predictability of Marketing Authorisation Application Submissions.” ?? Date: Thurs, 5 December 2024? ? Time: 12:30 to 13:15 Amsterdam time (CET)? ? ?? Did you know? In 2023, only 35% of marketing authorisation applications were submitted on their agreed dates. These delays impact long-term planning and approval timelines for much-needed medicines across the EU. This session will:? ? Outline the impact of these delays on EMA and the European network ? ? Discuss the challenges companies face in meeting submission deadlines? ? Highlight EMA’s initiatives to improve efficiency in assessment and approval processes ? Share tools and recommendations to help companies better plan their submissions ?? Join us live on LinkedIn! Ask your questions during the session via the comment section. You can also share your questions in advance in the comment section of this post.

The diversity of AI de novo protein design tools has as many choices now as there are favour of jams ??????. While having my tea and toast (no marmalade today) I read the Nabla Bio team’s recent technical report. Now this is the JAM I like! ??The combination of their #JAM (Joint Atomic Modeling) with experimental validation for four (epitope, in vitro function, affinity and developability) of the nine attributes of a high quality antibody #CandidateDrug is impressive. In addition, the scaffold that enables antigen presentation of multi-pass transmembrane extra cellular loops will be super impactful. ? This is a real step forward in demonstrating an integrated and digital AI enabled antibody discovery platform. ?? Great to see them include a novel target X as a critical challenge in R&D is choose the right target. ??Technical ??: https://lnkd.in/eSB6aRn3 #antibodies #biotechnology #drugdiscovery #structureprediction #AI #ML #biologics #GPCRs #human #chat

The race for de novo antibody design - did Nabla Bio just crush the opposition? The AI based antibody design and discovery space has become white hot. A few months ago I made two posts on this topic. The consensus was that we’ve made huge strides in mini-proteins but still have a long way to go for antibodies. Only 2 months on, have we just hit judgement day? It’s difficult to pick out the real progress in AI discovery. The hype is so loud it drowns out the real achievements. This isn’t helped by the clamour of some to claim the ultimate achievement of designing an antibody from scratch (de novo) with drug-like properties. A year ago Absci made outlandish claims to be the first to have achieved this. In reality what they did is take an existing antibody, took out CDR3 and redesigned a new CDR. Out of some 400k designs about 1% bound the antigen. This isn’t de novo design. You would likely achieve the same thing with random sequences. To claim “zero shot” or “de novo” design for efforts like this is ludicrous. Sadly the field is littered with similar claims, often with a lack of wet-lab work to backup them up. As a non-AI/ML specialist I find it hard to understand many of the papers and work out fact from fiction. The best approaches so far seem to have focussed on VHHs and produced binders with modest μM affinities. Last Friday, in a very understated manner, Nabla Bio put out a blog on their website with a link to a technical report. Usually I would avoid commenting on a non-peer reviewed report but this is fascinating. They’ve developed a generative model for antibody design called JAM. They demonstrate true de novo design on 9 highly distinct antigens, including 1 for which no PDB structure exists. 2 of the antigens are multi-pass membrane proteins which typically pose a major challenge for conventional mAb discovery. They prepared small libraries of approximately 10k AI designed sequences, screened for binding and reported binding rate. They then analysed a sub-set to confirm affinities and various developability parameters. They typically obtained 10s of binders with an average hit rate of 0.35%. For each target the best binder had an affinity of double digit nM. They even followed this approach for a classical VH+VL antibody using scFvs! They then developed two methods that can improve hit rates and affinities further. In one example achieving pM binders. Report and more details in the first comment. To me this seems like a giant step forward! There are some incredibly well funded companies fighting to lead this market. Nabla Bio only came out of stealth mode this year with $26M series A funding. Peanuts compared to many. Kudos to them! Hasta la vista, baby. I’ll be back. ----- I'm Ian, I post about antibody engineering, recombinant proteins and my journey to bootstrap Gamma Proteins into a leading supplier of Fc receptors. If you like my content please reshare with your network and follow me to see more.