Antibodies 2025: Venerable & Naked Or Complex & Multi-Specific?

By Matthew Pillar, Editor, Bioprocess Online

Antibodies are the biotherapeutic bulwark, the steadfast stalwart in not just oncology, but infectious and chronic inflammatory diseases as well. Despite the impression created by a vocal minority of ATMP developers, traditional monoclonal antibodies account for more than 65 percent of biologics market share. The next largest segment, at 15 percent, isn’t cell or gene therapies or even vaccines—it’s the growing legion of antibody-based recombinant proteins—bi-specific antibodies, antibody fragments, radio-immunotherapies, nanobodies, and antibody-drug conjugates built on the back of monoclonal antibody research and discovery.

Rather than give the venerable antibody short-shrift in the face of new modalities deemed higher priority for prognostication, we’re leaning into the antibody’s deeply-set roots and looking ahead to what’s in store for antibody-based drug development in 2025 and beyond. Here, AVEO Oncology, an LG Chem Company CEO Michael Bailey, Uniquity Bio CEO Brian Lortie, AP Biosciences CEO Jeng Her, Ph.D., and AP Biosciences VP of Antibody Discovery JJ You, Ph.D. weigh in on the trends to watch in antibody manufacturing capacity, artificial intelligence, regulatory trends, and more.

ANTIBODY MANUFACTURING CAPACITY GETS RATTLED

Michael Bailey is an antibody-based biotech builder. He’s shepherded a number of therapies from conception to commercialization, many of them antibodies, and that’s exactly what he intends to do in his current capacity as CEO at AVEO Oncology. The company’s got a late-stage mAb asset for HPV-HNSCC and AML, and a couple of Phase 1/2 mAb candidates for Cachexia/oncology and solid tumors. His biggest concern about their progress isn’t rooted in the outsourced manufacturing environment. “A lot of manufacturing is happening in China, and that’s creating some challenges for companies as the BioSecure Act takes shape,” he says. “You’ve seen some acquisitions of clinical and commercial biologic manufacturing facilities by some of the major companies who have leading antibodies and need to secure a lot of volume. For the small biotechs and startups in the space, that creates a logistical situation that forces us to plan way, way ahead or risk waiting in line.”  

When he was at Imclone Systems, the company took a big risk when it built a mAb manufacturing plant prior to a positive Phase 3 study. That gamble ultimately paid off with the approval of Erbitux (cetuximab), but it looked a bit crazy at a time when virtually every company in that class was using contract manufacturers. Now, with Chinese CDMOs at risk of being boxed out of an already restricted U.S. contract manufacturing ecosystem, the politically-influenced laws of supply and demand are poised to meter the advance of antibody innovation.

The potential fallout from this scenario is uniquely threatening to the U.S. and Western countries. Veteran antibody developer Jeng Her, Ph.D. is the man who co-invented a technology that converts non-human antibodies into engineered human-like antibodies with sequences close to the germline to minimize immunogenicity. He’s now the CEO at the Taiwanese biotech AP Biosciences, which is developing a host of bi-specific antibodies in programs ranging from pre-clinical to late- phase 3 trials. Dr. Her’s tone, when asked about the outsourced manufacturing environment in Taiwan, is palpably relaxed compared to that of the several biotech builders, specifically those in CMC, who have expressed off-the-record anxiety about investor and shareholder perception of their dealings with Chinese CDMOs. Asked about the challenge of aligning with an outsourced manufacturing partner in Taiwan, Dr. Her points to the comparably mundane. “Our considerations are cost, timelines, and communication,” he says. “We expect to run into some unforeseen challenges during process development, so communication is pretty important. I believe our local CDMOs have proven to be more flexible than the large, global CDMOs.”

Back in the States, AVEO Oncology found manufacturing capacity relief – or at least, insurance – when it was acquired by LG Chem in January 2023. LG Chem boasts state-of-the-art antibody manufacturing capabilities, and while AVEO hasn’t transferred its late-stage projects there, Bailey sees that capacity as a leg-up as he charts the course for the company’s early-stage candidates. “This is supply-and-demand. You can’t just ignore the BioSecure Act and say, ‘this isn’t gonna happen.’ You have to prepare. And because manufacturing lead times are so long, that preparation has to be in anticipation of this act going through and becoming a standard we have to work under. I can’t imagine there’s a company working with a Chinese manufacturer right now who’s not devising multiple plans to manage through this.” That, he says, might look like taking on the initiative and risk that comes with building their own manufacturing plants, or, as in AVEO’s case, taking advantage of a strategic acquisition situation.

ANTIBODIES OFFER THE AI USE CASE

Bailey is also ready to take advantage of AI, and if the antibody’s service is the tip of the spear in AI-enabled drug discovery and design is any indication, there should be plenty of opportunity to do so. We’ll see more use of the technology in those and more applications in 2025, as the cottage industry of techbio firms based on it continues to grow.

“AI is primarily an efficiency enabler, and that efficiency isn’t just about being able to more rapidly and more accurately predict targets or binding sites in drug development,” he says. “It’s also about patient selection, accessing patient data that we can translate into a more efficient clinical study design that's got a higher probability of success.” He sees AI as a key to improving the odds of the therapies produced by an industry that’s become accustomed to a 95% failure rate.

Like AP Biosciences, Uniquity Bio is developing bi-specific antibodies, in addition to a monoclonal lead candidate. Uniquity CEO Brian Lortie is also bullish on AI’s impact on antibody development, and more specifically, antibody engineering. “We’re taking antibodies that are already quite precise and safe, and then engineering them to be even more specific, by partnering with people who think of nothing else every day,” he says, and the momentum for companies to keep doing so is not slowing down. “The way the incentive structure appears to be set up post-IRA is re-catalyzing the development and commercialization of biologics, and artificial intelligence is playing a key role in the exploitation of that opportunity,” he says.

But that’s not Lortie’s only hope for AI in biotech. His company is one that’s intent on mining those tens of thousands of failed drugs that constitute the 95% of failures. “I’m interested in how AI can help a company like mine build a pipeline of two or three thoughtful and high-value additions to our core program – programs that have not advanced for the wrong reasons, or that have been strategically parked for some reason,” he says. “Consider the throughput, the efficiency of going into all of these databases, and in an hour coming up with a report of 500 programs with interesting data, good safety, but maybe were developed for the wrong thing or parked for the wrong reason.”

Still, Bailey says landing on sophisticated use cases for AI will take time. “There are some quick wins for efficiency, and we’re leveraging that for antibody design. Even if that’s not a core competency, you can purchase it as a service,” he says. “Search out those opportunities, but be realistic in your expectations. I was at an AI conference, where the prevailing sentiment was that all these biotech CEOs are ready to throw money at AI. If that’s true, we’re probably wasting much of this spend it at this point, because the industry at large hasn’t become mature in its ability to use it. Hopefully that will change as we become more sophisticated.”  

IS THE REGULATORY PATH MORE EFFICIENT?

As biologics go, the antibody’s regulatory path is well-trodden, and Bailey says the FDA in particular has been working in earnest on innovation and efficiency in the regulatory process. “They’ve streamlined clinical trial designs and requirements using surrogate endpoints, created accelerated approval opportunities, and made some progress on harmonization with global authorities,” he says. That’s an important consideration for global efforts like AVEO’s, with incremental costs as they apply to individual regulatory authorities.

Bailey also points to the FDA OCE (Oncology Center of Excellence) Project Optimus initiative, focused on reforming the dose optimization and dose selection paradigm in oncology. “I think that’s a really important step forward. So many of these drugs have just incredibly high levels of toxicities,” he says.

AP Biosciences’ Dr. Her notes that the FDA’s guidance on bi-specific antibody development, last updated in 2021, is far less definitive than that on monoclonal development. He suggests engaging with a reputable consultant, leveraging as much communication with the Agency as possible prior to a pre-IND meeting, and leaning into the regulatory expertise of your CDMO as this guidance—and presumably forthcoming guidance on multi-specifics, plays out. 

Looking ahead to 2025, Bailey says there’s plenty of room to build on the Agency’s clinical trial diversity initiatives. “As an industry, we haven’t done a great job here. AI could be a useful tool, but it’s only as valuable as the data that’s being put in, and we have limited data on minorities,” he says. “The diversity initiatives the FDA are taking on seek to change that, and will enable us to then use that data for future decision making and tailor therapeutics to different populations.”

MULTI-SPECIFIC ADVANCES DON’T TARNISH THE ANTIBODY’S SHINE

Multi-specific antibodies like those AP Biosciences and Uniquity Bio are working on add a degree of process development and manufacturing complexity. They require a combination of two different, unrelated protein sequences in a single molecule, and if achieving that conjugation isn’t challenging enough, they’re required to, as Dr. Her puts it, “be well-behaved” in terms of protein expression, stability, and aggregation. Target selection and mechanism of action are also create more challenges than their naked antibody counterparts. “When you put these two sequences together, you’re looking for not just an additive effect, but a synergistic one,” he says. That’s not to mention the risk of potentially doubling toxic side effects.

JJ You, Ph.D. is VP of Antibody Discovery at AP Biosciences. To address the mechanism of action challenge, he says the company takes advantage of IGG, which lends itself to multi-specific pairings. The stabilization and aggregation challenges, he says, require some good old-fashioned optimization work to solve. “We use the rational design [methods to ensure the correct pairing of heavy and light chains to produce bi-specific molecules with optimal manufacturability properties] to determine what kind of residue will contribute to aggregation, or what kind of a sequence will contribute to antibody cleavage,” he says. “Before the project enters into cell line development, we will try to optimize the antibody sequence. This not only addresses the stability and degradation risk, it improves productivity and helps us control production costs.”

At AVEO, Bailey doesn’t disagree with the reality that some bi-specifics have shown therapeutic advantages over single-target antibodies, but his agreement comes with a caveat. “They’re showing advantages because they're hitting multiple targets. I see opportunity with single-target antibodies in combinations.” That opportunity, he says, is rooted in an inherent safety and tolerability advantage. To manage and optimize the patient experience and outcome, you have to be able to modulate the dose, which gets tricky with multi-specifics. “The challenge that some of these more complex, multi-targeted antibody approaches present is that you have a bit of a one-size-fits-all scenario. If you have to down-dose, you’re going to have to down-dose both targets, even if you’re experiencing toxicity with just one component of the multi-specific,” he says. In a combination approach, on the other hand, you can modulate both of the molecules independently.

AP Biosciences addresses this challenge with a conditional T cell activation mechanism, by which T cells are only activated when the antibody binds to both the tumor-associated antigen and CD137, a feature that Dr. You says significantly reduces the risk of cytokine release syndrome. Its development platform, he adds, is designed for modularity of target binding sites for rapid target replacement.

Bailey, who worked for Genentech during the early days of antibody development, remains bullish on the naked antibody’s future. He also refuses to discount multi-specifics simply because they’re harder to develop, manufacture, and control. “Is there competition between ADC developers and multi-specific developers and mAb developers? Sure,” he says. “But what’s to stop us from developing an ADC, or a bi-specific, to hit more comprehensive targets as disease evolves and changes over time? Or to combine our mabs with bispecfics or ADCs to hit an even broader set of targets?” These are  salient points. The antibody is the foundation of them all. And the real opponent in this contest isn’t the biotech across the street – it’s the cancer we’re all working to outsmart.