Chris Behrenbruch

Immunogenicity assessments - why do we bother? As an industry we spend millions of $ measuring the pre-clinical immunogenicity of our biologics. We do this because none of us want our project stopped in the clinic due to a high anti-drug antibody (ADA) response. But to deimmunize or triage those with high immunogenicity you need to be able to reliably predict ADA in an in vitro or in silico assay. Andrew Buchanan, FRSC recently made an excellent post on this topic. For most late stage clinical antibodies the ADA data is publicly available but it is buried in datasheets and reports. There is no single freely available source of the truth when it comes to ADA. I can well imagine the countless days it took Andrew to gather this data because Geoff Hale and I went through a similar exercise years ago. We extracted ADA data for 80 mAbs and hoped to show some correlation between in silico prediction and clinical ADA. We had seen others report impressive results on much smaller datasets. We were sorely disappointed with the random scatter plot we obtained. There was no correlation between prediction and reality. But is this a surprise? At times immunogenicity feels like a biological random number generator. It's dependent on sequence, structure, aggregation, dose, route of injection, health of the patient and many other factors. Most people go off the ADA value stated in the info sheet provided with a drug. But dig deeper and for something like adalimumab (Humira) you find hundreds of clinical ADA readouts with values ranging from 0 to 100%. And then we have golimumab, another anti-TNF mAb but it stands out as the only one with low immunogenicity, reported at 2.8%. But years after this measurement the scientists at Janssen decided to re-run the samples in an updated assay. The ADA response jumped 8-fold. Good on them for doing this, many others wouldn't have! For clinical ADA measurements there is no consistency in the assay setup, sensitivity or reporting. The more of this data you look at the more you question why you are even bothering! The assumption is also that ADA is a project killer, but that's not always the case. Humira is the poster child for ADA but has total sales of +$230B. People are hyping up AI as the solution to immunogenicity prediction. Surely this is a nonsense? AI needs accurate, complete and consistent data. Real world ADA data is far from this! Almost everyone I speak to treats pre-clinical immunogenicity predictions with a pinch of salt but we all continue to do them to give management, investors or advisors a warm fuzzy feeling inside that their asset is squeaky clean. Problem is, the first real experiment isn't until phase 1. ----- 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.

712

105 Comments

It’s truly a momentous day for Tryp Therapeutics as the talented clinical team at Swinburne University of Technology, begin Patient Recruitment for the study of TRP-8803, in concert with psychotherapy, in the treatment of Binge Eating Disorder (BED). With 94% of BED sufferers also living with at least one other neuropsychiatric condition, reducing and controlling binge eating episodes is only one of the outcomes the team will be measuring. Co-morbid Depression, Anxiety, OCD or PTSD are common conditions that place an additional burden right at the feet of a patient diagnosed with BED. Tryp Therapeutics expects to deliver topline results from this study during Q425. #BED #TRP8803

65

11 Comments

A key lesson from mAbs in the 90's was not to build too much capacity. Did we learn our lesson for CGT? Andy Lewin joined us recently on our new podcast 𝗟𝗲𝗻𝘁𝗶 𝘄𝗶𝘁𝗵 𝗙𝗿𝗶𝗲𝗻𝗱𝘀 where he recalls a lot of discussion in the late 90's about the demand for manufacturing capacity in mAb bioproduction. At that time there was a lot of demand for the production of clinical material, with the expectation that mAb candidates would flourish. Developers built a lot of manufacturing capacity to meet this demand. While many mAbs did become highly successful and could fill this capacity. Many others did not. This led a lot of developers to either just shut down this additional capacity, or sell off whole sites to CDMO's. 𝗗𝗶𝗱 𝘁𝗵𝗶𝘀 𝗲𝘅𝗽𝗲𝗿𝗶𝗲𝗻𝗰𝗲 𝘁𝗲𝗮𝗰𝗵 𝘂𝘀 𝗮𝗻𝘆𝘁𝗵𝗶𝗻𝗴 𝗳𝗼𝗿 𝘃𝗶𝗿𝗮𝗹 𝘃𝗲𝗰𝘁𝗼𝗿 𝗺𝗮𝗻𝗳𝘂𝗮𝗰𝘁𝘂𝗿𝗶𝗻𝗴? Fast forward circa 2018 and the dilemma is the same. Huge demand for viral vector manufacturing, and not enough capacity. Many developers were wary, and rather than build they relied on CDMOs. 𝗦𝗼 𝘆𝗲𝘀, 𝗱𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗿𝘀 𝗱𝗶𝗱 𝗹𝗲𝗮𝗿𝗻 𝘁𝗵𝗲 𝗹𝗲𝘀𝘀𝗼𝗻 (𝗺𝗼𝘀𝘁𝗹𝘆𝗅). Unfortunately, the CDMOs didn't... they built a lot of manufacturing capability modeled on highly optimistic demand assumptions. That CDMOs like Exothera are closing up shop should not surprise us - we have seen it before. 25 years ago! Last time around CDMOs picked up the pieces of developers unwanted mAb capacity. Do you think anyone will pick up unwanted vector capacity now? Let me know in the comments below. P.S. if you want to see the full podcast, I will put the links in the comments.

59

13 Comments

“My meeting with LiKang's Chairman, Mr Zhuang Zhong, in Nanjing last Friday confirms that Intellego's products are currently being marketed and are in use with several larger hospitals in China while LiKang continues to prepare the broader market launch to coincide with new Chinese disinfection regulations which are expected to come into force in 2026 and to require the use of products such as Intellego's. LiKang is actively following the regulatory process and preparing the market. We reaffirmed our mutual commitment to our collaboration and our belief in its full potential“, says Intellego chairman, Greg Batcheller. Usage of our dosimeters in China is increasing and we look forward to the national regulations, expected in 2026, when 30 000+ hospitals MUST use products like Intellego’s (that these regulations were coming has been known for some time.) Due to this and based on previous cases, its now likley that its just a matter of time before EU, the US and other markets implement the same regulations.

161

2 Comments

Cross-border biotech deals often fail to deliver expected value. Over 15 years, we’ve closed 90+ deals on 5 continents despite being a boutique team. Here’s the framework that makes it work: Cross-border biotech deals often fail due to regulatory hurdles, pricing issues, and mismatched priorities. Here's how we avoid that fate: 1. Understand each market deeply Every region has its own commercial and regulatory frameworks, IP norms, and cultural context. We invest in learning that upfront because misalignment, not valuation, kills most deals. For instance, in China, most drugs are sold through hospital systems. If you don't know the distributors with hospital access, your product won't move. From volume-based pricing to the National Reimbursement Drug List (NRDL), every layer matters. Missing these details kills value before the deal is even discussed. 2. Surface hidden value others miss Our team has launched, manufactured, developed, and commercialized products in key global markets. That on-the-ground experience allows us to unearth hidden value. In Brazil, we generated additional gross margin by optimizing the global supply chain of our client's product. That gave the greenlight for the deal to proceed. In Japan, we understand the government's approach to drug loss and the policies addressing it. That lets us create favorable deal environments for Phase 2 companies. 1 in 3 cross-border deals only move forward after we reframe how value is understood on both sides. 3. Build strong trust-based personal relationships Teams that trust each other are more likely to collaborate, find solutions, and push internal alignment. Over 15 years, we've built relationships with decision-makers in every key market. That means we can speak directly to the people who matter. A quick yes mobilizes the org. A quick no saves months of useless conversations. 4. Once interest is real, move with precision Cross-border deals stall when teams lose focus. We act as deal leads: removing blockers, aligning terms, and setting the pace. Deals with clear cadence get done. 5. Keep teams lean and clear We run small, senior teams (3–4 core people) across all mandates. Big teams slow things down. Small teams close. In cross-border deals, clarity of purpose and focused execution beats headcount every time. Some quick metrics: • 90+ deals closed globally • Cross-border close rate >70% • Median close time: 6–9 months • Over 50% repeat clients Execution, not scale, is the differentiator. You don't need a big bank to close global biotech deals. You need clarity, follow-through, and credibility across borders. We've built that over hundreds of negotiations, through trust, not volume. If you're exploring global partnerships & want a partner who knows the path and the pitfalls, let's talk. At Kybora.com, we help leaders navigate cross border deals with clarity. Follow me for more on biotech M&A, strategy, global deals, and market shifts.

168

24 Comments

TCG Labs Soleil has created more than 10 single-asset biotechs, which receive early-stage clinical development support from a centralized R&D team. Now, the Bay Area firm has doubled its capital pool to $800 million and opened an office in Shanghai to run even more efficiently. One of its portfolio companies, Juri Biosciences, licensed a drug from a Shanghai-based biotech earlier this year. “The science, there’s really no border,” CEO Jin-Long Chen said in an interview with Endpoints News. “It’s all about efficiency,” chief operating officer Lucinda Quan said in a joint interview. “Having boots on the ground there, keeping an ear to the innovation that’s happening there. But also we use a number of CROs globally, including in China, so having boots on the ground to manage those relationships, to cut down on the time zone differences is just all about adding to our efficiency.” #TCGLabsSoleil #thecolumngroup #biotechvc #biotechventurecapital #shanghaibiotech #chinabiotech #bayareabiotech https://lnkd.in/g45QpwpR

104

2 Comments