Known in New: Applying Validated Biology at Scale

Advances in artificial intelligence and machine learning have dramatically improved our ability to generate biological insights and identify promising drug candidates. And yet, rather than becoming meaningfully faster or cheaper, drug development has become increasingly slower and more expensive over time, as summarized by Eroom’s Law. This is fundamental to how Formation Bio views the development bottleneck – while clinical development has become more expensive, the number of promising therapeutic hypotheses and candidates continues to grow. This means many potentially valuable drugs may never be developed, and the number of attractive programs bottlenecked by clinical development will likely increase.

These dynamics compound in ways that make the bottleneck even harder to overcome. For example, recent analyses of pharmaceutical pipelines have highlighted another phenomenon: once a mechanism demonstrates clinical success, development activity often clusters around it, producing a “herding effect.” This occurs when multiple companies pursue similar targets within the same disease indications, amplifying competitive dynamics. This concentration is evident across the development ecosystem, with a 2025 L.E.K. analysis showing that over 200 programs target the GLP-1 receptor, and hundreds of drugs target the PD-1/PD-L1 pathway in oncology. More broadly, the same analysis showed that 38 targets, roughly 2% of active R&D targets, are each associated with 50 or more drugs and collectively account for approximately 25% of all drug–target pairs in the pipeline. Ultimately, this raises the question: can developers benefit from validated biology without simply joining an increasingly crowded field? One strategy that aims to address this is what Formation Bio calls “Known in New”.

What is Known in New?

Known in New is one of the key strategies informing how we build our pipeline. In contrast to traditional drug repurposing, which typically directs an existing molecule to a new indication (often as a salvage or rescue due to limitations or failures with the primary indication), Known in New is characterized by applying a well-understood (known) mechanism of action that has been successfully validated in humans, to a new indication of high unmet need via a New Chemical Entity (NCE). This approach is compelling for several reasons, especially in how it redistributes development risk:

• It absorbs less fundamental biological risk (e.g., on-mechanism safety) because the mechanism has already been interrogated in another disease context

• It reduces the competitive risk of pursuing established indications where there is already significant development activity

• Because it relies on new chemical matter, it avoids some of the intellectual property limitations that often accompany repurposing strategies

• It takes targeted translational biology risk, expanding the potential applications of a validated mechanism

The strategy looks for opportunities where validated biology may apply in disease areas that have been comparatively underexplored. “Known in New” is one way to pursue those opportunities, connecting existing biological insight with diseases that may benefit from it. While the term “Known in New” is not yet widely used, the underlying strategy of applying validated biological mechanisms to new disease contexts has precedent across the biotech and pharma industries. Programs such as remibrutinib (BTK inhibitor) in chronic spontaneous urticaria, ocrelizumab (CD20 monoclonal antibody) in multiple sclerosis, and etrasimod/ozanimod (S1P modulators) in ulcerative colitis illustrate how well-characterized pathways can be translated into new disease contexts when supported by biological rationale.

“Known in New” is clearly a compelling strategy for redistributing development risk, but the underlying challenge is in execution. Identifying that a validated mechanism might apply in a new disease context is just a starting point. What’s more difficult is doing this rigorously, which involves exhaustively following every relevant evidence chain across human genetics, published literature, pathway biology, clinical precedent, and competitive landscape, across a broad universe of potential mechanism-indication pairs.

Doing this for a single asset is already a substantial undertaking, but across the universe of known mechanisms, it becomes nearly impossible without the right infrastructure. Historically, this has been an artisanal process, dependent on a team of experts going deep across all chains of relevant evidence. This is why, for most traditional drug developers, this kind of cross-indication biological mapping happens opportunistically, driven by the intuition of individual scientists or ad-hoc observations rather than a scalable, data-driven process. The search space is often too large, and the required breadth of expertise, spanning translational biology, clinical development, competitive intelligence, and real-world evidence, are often too wide for a single team to cover comprehensively or exhaustively (i.e., the total number of target indication pairs is very large, and hard to do manually at scale). The result is that “Known in New” opportunities are routinely missed or identified too late. This is where Formation Bio's model creates a unique advantage. Our ability to combine deep scientific and clinical expertise with proprietary data infrastructure and AI-enabled analysis means we can evaluate mechanisms across disease areas systematically, surface non-obvious connections between validated biology and unmet need, and move quickly when we find them.

Known in New in Practice

Our recent transaction with Lynk Pharmaceuticals exemplifies Known in New in practice. In December 2025, Formation Bio licensed global rights (excluding Greater China) to LNK01006, a next generation central nervous system (CNS)-penetrant highly selective TYK2 inhibitor, which will be developed within our Bleecker Bio subsidiary.

TYK2 is a well-characterized signaling node within the JAK-STAT pathway and has already demonstrated clinical relevance in immune-mediated disease through approved therapies targeting the same mechanism (e.g., BMS’ Sotyktu in psoriasis). Rather than pursuing another entrant in the same indications where TYK2 inhibition has already been validated (or is currently being validated), the opportunity with BLKR201 (formerly LNK01006) lies in exploring new disease contexts where the underlying biology may also play a role.

This brings us back to the framework behind the development strategy. There are many early stage development programs where the underlying science and data package are compelling. However, the proposed lead indication may present commercial or competitive challenges, such as a crowded field or limited differentiation, that make it difficult for downstream partners to see the competitive potential of an asset. The “Known in New” approach starts from a validated biological mechanism, then asks a different question: what other disease contexts, comparatively underexplored, could this mechanism meaningfully address? In the case of BLKR201, we evaluated how the combination of biology and the compound’s profile could unlock new therapeutic opportunities in TYK2, assessing both the strength of the biological rationale and the competitive landscape. This ultimately led to a new opportunity for Formation Bio for TYK2 inhibition in a disease area with high unmet need.

Formation Bio's approach to drug development also gives us a distinct advantage when it comes to applying “Known in New”, particularly in how we use real-world data and AI to better understand disease progression and inform development strategy. For example, we’ve analyzed longitudinal patient data to compare progression in primary efficacy endpoint(s) across treated and matched control populations, using methods such as propensity matching to account for confounding factors. In some cases, this allows us to explore, in silico, whether drugs with related or adjacent mechanisms may meaningfully alter disease trajectory, helping to generate and prioritize new therapeutic hypotheses. While these analyses require careful interpretation and do not fully isolate causal effects, they provide an additional layer of evidence to inform both indication selection and development strategy. Over time, we see an opportunity to scale this approach across diseases, combining data science and clinical insight to systematically identify where validated biology may translate into new therapeutic opportunities.

Known in New at Scale

This is also ultimately how we think about applying Known in New at scale. The strategy works best when it is driven by a systematic process for identifying where validated biology and unmet medical need intersect. That means combining translational expertise, clinical development insight, and structured data analysis to evaluate mechanisms across disease areas and prioritize the opportunities most likely to translate in human trials. Some examples include:

• Competitive landscape analysis and benchmarking: Our indication landscaping tool is built on a framework that collects and structures comprehensive competitive data (efficacy and safety metrics, patient populations, and dosing regimens), enabling us to systematically map the safety-efficacy tradeoffs across competitive therapeutic landscapes, identify where best-in-class drugs set the clinical benchmark, and surface overlooked opportunities for development.

• Clinical development plan benchmarking and development: Forge, our in-house clinical trial design tool, is built on a foundation of structured, third-party historical clinical trial data, which AI agents leverage to automate research and analysis, integrate findings across sources, and produce novel study designs grounded in precedent, regulatory guidelines, and internal documents.

• Agentic trial outcome and probability of technical/regulatory success prediction: We’ve built a multi-agent forecasting model to help predict an asset’s probability of success in the clinic – defined as an asset reaching its primary endpoint. The system leverages individually-scoped agents to gather evidence across drug safety, efficacy, trial design, and precedent to generate a probability of trial success score, along with confidence intervals and evidence-backed rationale.

• Causal biology linked to disease progression: We integrate human genetics, perturbation data, and multi-omics datasets to establish causal links between biological pathways and disease outcomes. This can enable us to prioritize targets and mechanisms with strong human validation, map their relevance across disease contexts, and better predict whether modulating a given pathway is likely to translate into meaningful clinical benefit.

As biological knowledge continues to expand, the number of potential connections between mechanisms and diseases will only grow. The challenge, and the opportunity, is building the capabilities to identify those connections consistently and early, engage the right experts, move the most promising hypotheses into development, and execute them with the right clinical strategy.

What’s Next

Looking ahead, we’re continuing to expand how we identify and evaluate Known in New opportunities at scale. To date, much of this work has relied on evidence such as genetic insights, published literature, and existing clinical data to help map validated mechanisms to new disease contexts. The next phase is building on that foundation by integrating additional datasets, ranging from causal biology to real-world clinical evidence, to improve how we surface and prioritize these opportunities.

Our goal is to make this process increasingly structured and repeatable at scale, combining human expertise with richer datasets to generate unique insights to guide our asset acquisition and development decisions. As this work evolves, we expect Known in New to remain a key strategy in our pipeline, informing how we bring new medicines to patients faster and more efficiently.