A Systematic Methodology for the Identification, Evaluation, and Acquisition of Phase 1 Clinical Assets

The Twin Drivers of Failure

The biotech market is currently bifurcated. On one side, late-stage assets command astronomical premiums. On the other, early-stage science—specifically Phase 1 assets—is trading at a discount, often abandoned in what the industry lazily calls the "Valley of Death."

At Fibonacci, we view this not just as a valley, but as an opportunity for Value Engineering.

The industry assumes that the 90% failure rate of early assets is due entirely to bad biology. Our thesis is that failure is driven by two distinct vectors:

1. Undifferentiated Assets: Too many drugs enter the clinic with marginal efficacy or safety profiles that are indistinguishable from standard of care.
2. Execution Failure: Suboptimal trial designs, unscalable CMC processes, and regulatory stumbling blocks.

These are engineering problems. And engineering problems can be solved.

This protocol outlines our systematic methodology for using our Medicine Engine to identify the highest-probability clinical assets and then deploying our AI-native infrastructure to bridge the execution gap efficiently and scalably.

1. Operational Alpha: Beyond Simple Arbitrage

Large pharma faces a patent cliff; they need volume. Small biotech faces a capital cliff; they need runway. This structural imbalance creates a buyer's market for Phase 1 assets—drugs that have cleared the initial safety hurdle (First-in-Human) but haven't yet reached the expensive "Proof of Concept" stage.

By acquiring at Phase 1, we secure three strategic advantages that late-stage buyers miss:

• Valuation Entry Point: We pay for the option on efficacy, not the certainty. This allows us to acquire multiple "shots on goal" for the price of one Phase 3 asset.
• Operational Control: We don't inherit a flawed Phase 2 trial design. We build the pivotal trial from the ground up using our own AI-optimized protocols, ensuring the data we generate is regulatory-grade.
• The "Fallen Angel" Premium: Many valid assets stall simply because their parent company ran out of cash. These are "distressed assets" financially, but "prime assets" scientifically. Our diligence is tuned to spot the difference.

2. Building the Dragnet: Systematic Sourcing

The "inbound model" of business development—waiting for bankers to send pitch decks—is broken. In a global ecosystem with 5,000+ active early-stage trials, passive sourcing ensures you only see what others have already passed on.

We treat sourcing as a data engineering problem. We don't "network"; we index. Our Sourcing Engine ingests the global clinical landscape daily, filtering for signal across three distinct layers.

2.1 The Intelligence Stack

No single source of truth exists in pharma. The "truth" is a triangulation of scientific promise, operational reality, and commercial viability. We architect our stack to collide these datasets:

• The Predictive Engine (Scientific Signal): We don't rely on third-party scores. Our proprietary models ingest raw patent feeds, preclinical publications, and mechanism-of-action data to predict clinical success. We filter for biological validity first—if the pathway isn't sound, the asset doesn't exist to us.
• The "Fallen Angel" Scraper (Operational Reality): We built custom daemons that monitor global clinical registries for negative space. We look for the absence of data: Phase 1 trials that marked "Completed" 8 months ago with no follow-up. These data gaps signal capital constraints, not scientific failure.
• The Market Simulator (Commercial Filter): Most science projects fail the spreadsheet test. We don't wait for a consultant's report. Our agents run real-time epidemiological models against every potential target. If the market size can't support the R&D cost, the code kills the deal instantly.

2.2 Algorithmic Daemons

We don't browse. We set daemons. Our system monitors for specific state changes that signal a "distressed opportunity":

1. The "Zombie Trial" Alert: Monitors for status changes to "Active, not recruiting" with no updates for >180 days. This is the signature of a biotech maximizing runway by pausing development.
2. The Abstract Scraper: Before a press release hits the wire, the data hits the conference abstract submission system. We scrape ASCO, AACR, and JP Morgan for keywords like "Preliminary Safety" and "Dose Escalation" to catch data before it is presented.
3. The Patent Watch: PCT filings precede clinical data by 12-18 months. A new composition-of-matter filing is the earliest possible "ping" of a new asset coming online.

2.3 The Human Element: Academic Hubs & RESI

While our bots scrape the public web, our scouts map the networks of academia. However, we recognize a distinct boundary: Academia is the engine of early-stage ideation, not clinical development. University Technology Transfer Offices (TTOs) are excellent sources for novel biology and preclinical concepts, but rarely offer the structured clinical data we require for Phase 1 acquisition.

Therefore, we flip the conference strategy. We skip the crowded main floor of BIO International. Instead, we focus on RESI (Redefining Early Stage Investments). This is where the seed-stage, pre-Series A founders live—the ones who have moved beyond the lab bench and need an operational partner to survive the clinic.

3. Translational Validity: Decoding the Signal

Once the dragnet identifies a target, the question shifts from "Is it available?" to "Is it real?" For Phase 1 assets, the challenge is Translational Validity: discerning whether a signal in a dozen patients is a biological truth or a statistical fluke.

3.1 The Architecture of Failure: 3+3 vs. Bayesian

The industry standard "3+3" dose-escalation design is a relic of the chemotherapy era. It is computationally inefficient, statistically conservative, and ethically dubious (it treats patients as disposable sensors).

We view the uncritical use of a 3+3 design as a competence signal. We acknowledge that regulatory agencies (often the FDA) frequently mandate these conservative designs for safety reasons. However, we look for evidence that the sponsor attempted to advocate for smarter designs like Bayesian Optimal Interval (BOIN) or TITE-CRM. If a sponsor accepted a sub-optimal design without leveraging modern statistical arguments, it signals operational passivity.

3.2 Safety is a Signal, Not a Checkbox

"Safe" is a relative term. In modern oncology, undefined toxicity is the enemy. We do not adhere to arbitrary rules, but we look for a clearly defined Therapeutic Window. If the toxic dose and the effective dose overlap, there is no drug.

3.3 The Holy Trinity: Exposure, Engagement, Effect

We do not buy "hope." We do not buy "encouraging trends." We buy the PK/PD Trinity. If any one of these three pillars is missing, the asset is a science project, not a drug.

1. Exposure (PK): Is the molecule actually getting into the blood? If the plasma concentration doesn't match the preclinical model, the model is wrong.
2. Target Engagement (PD): Is it hitting the receptor? We require biomarker evidence (e.g., receptor occupancy or phosphorylation reduction) in humans.
3. Downstream Effect: Is the biology responding? We need to see a functional consequence—tumor shrinkage, cytokine reduction, or cell death.

Differentiation is the final filter. If the safety profile doesn't support a "best-in-class" claim, we walk away. Being the 5th best PD-1 inhibitor is not a business model.

4. The Silent Killer: CMC & Scalability

Chemistry, Manufacturing, and Controls (CMC) is where deals go to die. In Phase 1, you can make a drug in a beaker. In Phase 2, you need to make it in a tank. The gap between those two realities is the Scalability Gap.

4.1 The Metric-Ton Reality Check

Our diligence agents audit the synthetic route for industrial viability, not just chemical elegance. We ask three binary questions:

• Is the route scalable? If the synthesis requires cryogenic temperatures (-78°C) or hazardous reagents that are banned in commercial facilities, the process is a dead end.
• Is the formulation patient-ready? Phase 1 trials often use "powder-in-bottle" solutions. We demand a roadmap to a commercial formulation. What is the pill burden? Are registration batches planned? If these questions aren't answered, the asset isn't ready for late-stage development.
• Is the impurity profile stable? Scaling up often introduces new impurities. If the Phase 2 material looks different from the Phase 1 material, the FDA will demand new toxicology studies. That’s a "Clinical Hold" waiting to happen.

4.2 Process Amnesia

The greatest risk in early-stage acquisitions isn't IP; it's tacit knowledge. The "recipe" for a drug often lives in the head of a single scientist at a boutique CDMO, not in the official documentation.

We treat Tech Transfer as a forensic exercise. We don't just buy the IP; we buy the "Tech Transfer Package"—every batch record, every raw data file, and every deviation report. If the documentation is incomplete, we assume the process is broken.

5. The Regulatory Moat: IP & IND Forensics

Technical success is irrelevant if the asset cannot be owned or approved. We treat IP and Regulatory diligence as a forensic audit. We are looking for the skeletons in the closet.

5.1 Crown Jewel IP Only

We only value Composition of Matter patents. Patents on "methods of use" or "formulations" are weak shields that generic competitors will pierce in court. We require a filed patent with a runway of at least 12-14 years post-launch.

Equally critical is Freedom to Operate (FTO). It’s not enough to own your molecule; you must prove you aren't trespassing on someone else's foundational IP (a common trap in ADC and CRISPR assets).

5.2 The Regulatory Audit

The Investigational New Drug (IND) application is the asset's regulatory passport, but the audit goes deeper. We look for the "Regulatory Audit Trail." Our agents scan the correspondence logs for "Partial Clinical Holds"—the regulatory equivalent of a check engine light. We also mine the FDA Pre-IND and End-of-Phase 1 meeting minutes. Did the agency flag the animal models? Did they demand cardiac safety monitoring that will kill commercial uptake? We find out before we write the check.

6. Pricing the Option: Valuation & Reality

Valuing a Phase 1 asset isn't guesswork; it's a high-stakes exercise in probability management. Our agents and proprietary know-how allow us to systematically select assets with a lower inherent risk and a significantly higher probability of success. This enables us to capture disproportionately more value than the status quo, pricing the option, not the promise. Our models account for the brutal attrition rates while capturing the asymmetric upside of a true blockbuster.

6.1 The rNPV Engine: Quantifying the Future

The risk-adjusted Net Present Value (rNPV) model is the core of our valuation engine. Unlike simplistic discounted cash flows, rNPV explicitly adjusts future cash flows by the probability of success (PoS) at each development stage.

Our PoS benchmarks are continuously updated with real-world data (Phase 1 to Approval historically at 6.7% - 12%), and our discount rates (10-14%) reflect the true cost of capital for high-risk assets. We don't inflate probabilities; we model them with ruthless accuracy.

6.2 Real Options: The Value of Deferral

For early-stage assets, the greatest value isn't in a guaranteed outcome, but in the option to proceed. Our Real Options Analysis (ROA) treats a Phase 1 asset as a financial call option. This is how we structure asset deals: We license the asset with the explicit intention to develop it, but with predefined options to terminate the agreement if the data—after key go/no-go decision points—does not meet the necessary required bar. This minimizes capital at risk and maximizes strategic flexibility.

Crucially, we recognize that the Phase 1 to Phase 2 Proof-of-Concept (POC) is the single biggest value inflection point in drug development. Our entire engine is designed to systematically de-risk and accelerate assets through this critical transition, enabling small teams to achieve scale and capture this outsized value.

7. Engineering the Deal: Risk Sharing & Incentives

A well-structured deal isn't just a financial transaction; it's a blueprint for operational partnership. While big pharma often approaches these deals with large upfront cash payments, our advantage as an agile, AI-native biotech lies in engineering frameworks where both parties are incentivized for success. For Phase 1 assets, the primary goal is Risk Sharing.

7.1 Blueprinting for Alignment

• The Option Deal: This is our preferred mechanism. We pay an upfront fee to fund the critical "killer experiment" (Phase 1b or early Phase 2a). Upon positive data readout, we hold the unilateral option to acquire the asset at a pre-negotiated price. This limits our downside and de-risks the commitment.
• Equity Components: Strategic equity investments align long-term incentives. We become partners in the platform, not just purchasers of a single asset.
• Milestone Engineering: Milestones are triggers for value release. We tie them to operational data points: Phase 2 start, NDA filing, first sale. These are not arbitrary dates but hard-won validations of technical progress.

7.2 First Principles from Recent Deployments

We analyze the market's most sophisticated transactions to extract universal principles for optimal deal architecture:

7.3 Optimized Royalty Structures

Royalty rates (typically single to low double digits for Phase 1 assets) are calibrated against IP strength and competitive intensity. We also track the rise of "synthetic royalties"—a mechanism where investors buy future revenue streams to fund current development, providing non-dilutive capital and increasing leverage for sellers.

8. PMI: The Crucible of Value

The deal closing isn't the finish line; it's the starting gun. We view Post-Merger Integration (PMI) as a high-velocity data migration event. Pragmatically, we often find that the people at these companies—the licensors—can be significant bottlenecks due to inertia or resistance to change. If the asset isn't fully ingested into the Fibonacci Engine within 14 days, and these human elements aren't aligned, the acquisition is failing.

8.1 The 100-Day Execution Script

We don't use generic checklists. We deploy a specialized integration squad with a singular mandate: Transfer and Accelerate.

1. Regulatory Transfer (Day 1-14): The "Change of Sponsor" letter must be filed with the FDA immediately. Until the IND is in our name, we are flying blind.
2. Data Ingestion (Day 15-45): We migrate the entire clinical database (SAS files, raw imaging data) from the legacy CRO to our centralized data lake. Incompatible standards are the enemy of speed; we enforce standardization ruthlessly.
3. Site Audit & Optimization (Day 30-60): We audit every active clinical site. High-performing sites are doubled down on; laggards are cut. We don't tolerate "zombie sites" that dilute our resource focus.
4. Operational Harmony (Day 60-90): We balance the target's "agile" culture with our "industrial" rigor. We don't impose bureaucracy; we install infrastructure. The goal is to give their scientists better tools, not more meetings.

Conclusion

The acquisition of Phase 1 clinical assets is a high-stakes endeavor that requires a synthesis of scientific acumen, financial discipline, and strategic foresight. By moving away from ad-hoc "shopping" and adopting the systematic methodology outlined in this report—leveraging advanced intelligence stacks, conducting rigorous translational diligence, and structuring risk-shared deals—companies can effectively bridge the "Valley of Death."

In 2025, the winners will not be those who pay the most, but those who see the most clearly—using data to peel back the layers of uncertainty inherent in early-stage drug development. This framework provides the lens for that clarity.