Beyond the First Readout

The moment the question changes

Clinical validation for a new target class does not happen at IND filing. It does not happen when Phase 1 safety data clears the way forward. It happens at the first randomized, placebo-controlled Phase 2 POC trial that shows an effect size large enough to matter for patients in the real world. Before that moment, biology is largely hypothesis—promising, perhaps even compelling, but unproven in the setting that counts. After it, the question flips entirely.

You stop asking whether the pathway works. You start asking what the winning drug on this pathway needs to look like to still be winning commercially a decade from now.

For Fibonacci, that is the moment the best-in-class floodgates open.

We use "best-in-class" as a design constraint, not a marketing adjective: the future-proof TPP representing the minimum bar across efficacy, safety, and convenience needed to become the drug physicians reach for, patients stay on, and payers keep on formulary. A strong Phase 2 POC accomplishes two things at once. It de-risks the mechanism, stepping up PoS in ways investors and BD teams can underwrite. It also exposes the validator's liabilities (PK constraints, tolerability gaps, ROA friction, dosing inconvenience), which define the design space for what comes next.

Some of the highest-signal readouts look messy on first glance. Great Phase 2 efficacy paired with a toxicity of known provenance gives you a blueprint: preserve the efficacy, engineer out the toxicity. Great efficacy and safety combined with obvious convenience friction tells you that when outcomes are close, convenience becomes destiny (oral versus injection, Q12W versus Q2W, lower burden wins). And sometimes a drug "works" clinically without "winning" commercially; the validator proves the pathway, then reveals its ceiling. That ceiling becomes the specification for best-in-class.

Two live examples illustrate where this pattern is already unfolding: Na_V1.8 and the resurrection of non-opioid pain after Vertex's VX-548 Phase 2 POC, and chronic urticaria and the emergence of mast cell depletion via KIT inhibition after Celldex's barzolovolimab Phase 2 POC. We'll also show how Fibonacci's medicine engine (our Retrieval Augmented Reasoning knowledge graph and clinical refinery) treats that Phase 2 moment as a starting gun and value creation signal rather than an endpoint.

From mechanism graveyards to gold rushes

Across drug targets and disease areas, the same script repeats. It begins with plausibility: a mechanism looks compelling, often backed by human genetics and elegant preclinical biology. Early programs proceed with optimism, then fail—on efficacy, on safety, on execution. Capital exits. The target acquires a reputation. It becomes "cursed," a graveyard where good ideas and good money went to die. Only a handful of teams keep going, often the ones with the longest time horizons or the stubbornest conviction.

Then the first randomized Phase 2 POC lands. A well-powered, placebo-controlled trial finally demonstrates that the pathway can move disease in humans. Mechanism-level risk collapses in a single readout. Investors and BD teams reprice the opportunity. The floodgates open.

What happens next is subtler but equally important: the best-in-class design space crystallizes. The validator is rarely perfect, and its weaknesses (tolerability gaps, PK constraints, dosing inconvenience) become a specification for follow-ons. Over time, a differentiated later entrant can take category leadership and expand the treated market in ways the pioneer never could.

The commercial record on this is blunt. In therapeutic classes where follow-on entrants offer only incremental improvement, the first-in-class product often retains more than 60% market share while later entrants struggle to break 20%. But when a later entrant is genuinely differentiated, order of entry matters far less. "Differentiated" tends to be concrete rather than abstract: a meaningful step up on at least one adoption-driving dimension like efficacy, safety, route of administration, or dose frequency. When that step up is real, late entrants can displace pioneers and take majority share even if they arrive years later.

Na_V1.8 and mast cell depletion are both in the early innings of this dynamic right now.

Case study 1: Na_V1.8, when pain stopped being a graveyard

For a long time, pain was where good ideas went to die. The world needs a non-opioid, non-addictive way to control serious pain, and the opioid epidemic made the cost of "effective but addictive" painfully clear. A credible non-opioid option would be a release valve for a healthcare system still struggling with addiction and diversion.

On paper, Na_V1.8 is straightforward: a peripheral pain signaling channel with strong human genetic validation. In practice, the industry's earlier sodium channel wave trained investors to flinch. The history was littered with selectivity traps, noisy clinical endpoints, placebo response, and repeated high-profile failures. The target acquired a reputation.

Then Vertex kept going. On August 3, 2023, they announced positive Phase 2 POC results for VX-548 in acute pain. That kind of readout changes the ambient belief in an entire target class, and capital follows conviction: suddenly the mechanism was validated, the space was investable again, and the best-in-class race was on.

You could see it in capital formation almost immediately. Latigo launched with a $135M Series A to develop non-opioid pain medicines, explicitly positioning around best-in-class. Latigo later closed a $150M Series B, supporting highly selective Na_V1.8 inhibitors already in clinical development. SiteOne raised a $100M Series C to advance selective ion channel modulators. And then Lilly announced it would acquire SiteOne outright, with shareholders eligible to receive up to $1.0B in upfront and milestone payments.

Vertex ultimately converted the acute pain story into regulatory success: on January 30, 2025, the FDA approved Journavx (suzetrigine) for moderate to severe acute pain in adults. But Phase 2 is not just an upside unlock. It also reveals ceilings.

As Vertex's suzetrigine story moved from acute pain into chronic indications, the market's enthusiasm oscillated. Billions can appear quickly when a "graveyard" becomes investable, and billions can disappear when first-generation profiles fail to translate into larger indications or longer treatment durations. The whiplash follows naturally from the underlying logic: once the biology is validated, the game becomes simple to state and hard to execute. Who can build the TPP that wins?

For Na_V1.8, the best-in-class wedge is crisp. Faster onset matters for patients in acute distress. Chronic use needs to feel boring: predictable efficacy, clean safety, minimal monitoring, the kind of profile physicians prescribe reflexively. And removing the metabolic liabilities that constrain dosing would let you drive stronger efficacy at lower exposures, with more margin for chronic use and combinations.

When "pain is impossible" turns into "this is an engineering problem," the floodgates open.

Case study 2: Barzolovolimab and the mast cell depletion design space

Urticaria is not rare in the lived experience sense. It is relentless: itch, wheals, swelling, sleep loss, steroid cycles, and a quality-of-life tax that compounds over years. Mechanistically, KIT is a master survival signal for stem cells and mast cells. If you interrupt KIT signaling strongly enough, you can deplete mast cells and, in principle, turn off the immune cascade in a way that can feel like a functional cure.

The problem pattern is familiar. KIT is not mast cell exclusive. Historically, KIT inhibition came packaged with on-target collateral toxicity that made it a non-starter for chronic immunology indications where patients might stay on therapy for years.

On November 6, 2023, Celldex announced positive topline results from a Phase 2 study of barzolovolimab in chronic spontaneous urticaria. Within a day, Celldex announced pricing of a public offering at $27.00 per share, a vivid example of how quickly markets reprice a target class when Phase 2 de-risks the biology. The stock moved sharply upward after the readout and later ran to roughly $50 per share as the efficacy story matured.

Then longer follow-up and chronic tolerability questions, including 52-week safety event scrutiny, forced a major repricing. The safety signal became the signal to engineer around. Validators do this: they make the biology real and the tradeoffs explicit, and once the tradeoffs are explicit, the best-in-class specification becomes legible.

The follow-on wave is now organizing itself along two axes. The first is precision mast cell depletion through biologic engineering. If the limitation is "KIT is everywhere," the best-in-class response is selectivity, which explains the interest in bispecific "AND gate" strategies: pair KIT targeting with a mast cell biased marker, preserve potency, narrow where the drug acts.

The second axis is oral KIT inhibition, using PK as the safety feature and ROA as the adoption feature. Instead of committing patients to long-acting depletion kinetics, orals let you tune the exposure window (dose, half-life, on-off kinetics) with the goal of durability of control and a safer chronic footprint. Oral therapy is also a real adoption lever; when outcomes are close, patients and physicians often prefer simpler routines. Third Harmonic, for example, has described Phase 1 data for an oral KIT program with a half-life of about 40 hours enabling once-daily dosing.

Put it together and it is the same Phase 2 logic as Na_V1.8: Phase 2 validates transformative efficacy, the liabilities define the blueprint, and the second wave engineers best-in-class around safety and real-world convenience.

The market is already moving around these concepts. Santa Ana Bio launched with $168M to advance precision medicines for inflammatory diseases, including a bispecific approach designed to focus activity on mast cells. Alys launched with $100M to build an immuno-dermatology pipeline including mast-cell-selective bispecific strategies. And Sanofi agreed to acquire Blueprint Medicines for more than $9B, citing pipeline assets that include BLU808, a highly selective oral wild-type KIT inhibitor with potential in immunology.

The commercial scoreboard: best-in-class usually wins

"Best-in-class wins" gets invoked like an aspiration, but it only holds when the differentiation is real, and "real" has specific dimensions that vary by therapeutic class. When follow-on entrants offer only modest improvement, the first-in-class product often retains outsized market share while later entrants struggle to break out. When a later entrant is genuinely differentiated, order of entry matters far less.

Differentiation can come from efficacy (Tagrisso in EGFR-mutant lung cancer), safety (Eliquis against warfarin), route of administration (Firazyr enabling self-administered subcutaneous dosing for hereditary angioedema attacks), or dose frequency (Fasenra's transition from monthly to every-eight-weeks maintenance). Differentiated entrants can also expand the market itself: advanced therapy penetration in psoriasis more than tripled over roughly a decade as mechanisms improved and dosing became more livable.

But which dimensions matter varies by therapeutic area. Our analysis of 55 oncology antibodies found that dosing frequency has almost no relationship with commercial success in that setting. Efficacy is king in oncology; convenience is queen. In chronic immunology, the calculus flips: when efficacy is table stakes, convenience and safety become the battleground. Knowing which dimensions actually drive adoption in a given space is half the work of defining a winning TPP.

Phase 2 POC matters because it's the first moment you can credibly define the performance thresholds a winning TPP must hit, then make a rigorous call on whether you can build (or access) a drug that clears them.

Inside the engine: encoding Phase 2 into a knowledge graph that reasons

We built Fibonacci's engine to make this pattern repeatable. Instead of treating Phase 2 POCs as anecdotes, we encode them as structured objects inside an ontology-backed Retrieval Augmented Reasoning knowledge graph, capturing the biology (targets, pathways, genetic validation, mechanism of action), the clinical context (indication, endpoints, inclusion and exclusion criteria, comparator, SoC), the liabilities (adverse events with mechanistic provenance, PK constraints, route and dosing friction), and the competitive context (what else is in development, which differentiation dimensions historically drive adoption in similar classes).

The questions we ask are not "find similar text." They are decision questions: show me targets where the first randomized Phase 2 POC de-risks the mechanism in a large chronic indication and the validator has clear liabilities in PK, tolerability, or convenience. For a given target, what would a best-in-class TPP need to look like to win share and expand the treated population? Which modalities can realistically deliver that TPP?

Operationally, we do not start at Phase 2. We continuously monitor Phase 1 safety and early efficacy readouts so we can form TPP hypotheses, map the design space, and build diligence paths before the pivotal POC lands. When the floodgates open, we are already running at full speed. Phase 2 is the trigger; Phase 1 is the early warning system.

From floodgate to portfolio: choosing where we win

Our approach is deliberate, and it is staged. As we launch and grow, we will prioritize licensing, partnering, and acquiring assets that match our medicine engine's winning TPP and already sit inside Phase 2 validated design spaces. That is how we move quickly while the market is still regrading the mechanism.

At the same time, our ambition is not static. Over a ten-year horizon, the goal is to build the winning assets with the winning TPP ourselves, then execute development at full speed. Fibonacci's medicine engine helps us filter for the right opportunities: Is the Phase 2 signal strong enough to matter against today's and tomorrow's competition? Does the validator expose a crisp best-in-class wedge in efficacy, safety, or convenience? Can we define performance thresholds that are ambitious and achievable, then build a development plan around them?

When we find those opportunities, the clinical refinery turns pattern recognition into execution: trial designs aligned to the TPP, dosing strategies built to maximize real differentiation, and combination approaches that raise the efficacy ceiling without raising the safety floor.

Building future-proof medicines at the Phase 2 edge

The industry loves to debate first-in-class versus best-in-class, but the more useful question is: when does a mechanism become real enough that you can design the medicine patients will still be using, and still buying, ten years from now?

Our answer is unromantic and practical: the first randomized, placebo-controlled Phase 2 POC in a new target class. Before that moment, biology remains hypothetical. After it, the best-in-class design space appears and you can credibly define a TPP, stress test it against commercial history, and decide whether you can build something that actually deserves to exist.

Na_V1.8 and mast cell depletion are current examples, but the pattern generalizes across therapeutic areas and decades of industry history. Phase 2 turns biology into a design space, and the winners translate that design space into a best-in-class TPP, then execute.