Apotek Signal: Quantifying Clinical Trial Success Where Asset Value Is Decided

Retrospective Probability of Success (PoS) answers “what usually happens across programs.” Apotek Signal answers “what is likely to happen for this asset.”

Key Takeaways

• Apotek Signal quantifies clinical risk at the asset level: It estimates the probability that a specific trio of drug–target–indication will succeed at a given clinical phase.

  
  

• Phase-aware by design: Separate models for clinical trial phase I, II, and III reflect how decision criteria change across development.

  
  

• Built for real decisions, not averages: Predictions incorporate both drug developability and trial design, moving beyond historical phase benchmarks.

  
  

• Directly usable for BD and portfolio governance: Outputs support rNPV models, indication-expansion assessments, and licensing discussions.

  
  

• Focused where risk and capital density are highest: Currently optimized for oncology small-molecule programs with clear translational hypotheses.

Why We Build Apotek Signal, and How It’s Used

Making clinical risk explicit before it is priced in

In practice, most clinical development decisions are made under uncertainty long before definitive data are available. Yet, in licensing negotiations and internal portfolio reviews, drug assets are still commonly valued using risk-adjusted Net Present Value (rNPV) models that require an explicit assumption about the probability that a drug pipeline will advance through the next clinical milestone. In many organizations, those probabilities are based on historical phase averages, therapeutic-area benchmarks, or expert consensus.

 

Multiple large-scale studies have shown that only ~10% of drug candidates entering Phase I ultimately reach regulatory approval, with oncology programs exhibiting even higher attrition [1–3]. As a result, probability assumptions become one of the most sensitive—and most debated—inputs in asset valuation and portfolio decision-making.

Apotek Signal is built to address this gap by reframing the core question:

Clinical success probability should be estimated for a specific drug, acting on a specific target, in a specific indication, under a specific study design—not inferred from averages.

This shift from static benchmarks to asset-specific risk estimation is increasingly emphasized in modern pharma strategy and portfolio management, particularly as organizations move toward more data-driven decision frameworks [4, 5].

Data Foundation Behind Apotek Signal

Apotek Signal integrates pre-clinical developability signals with structured representations of clinical study design, reflecting how real-world decisions balance molecule quality with execution strategy. This design aligns with broad industry consensus: clinical success is rarely a property of the molecule alone, but an interaction between pharmacology, patient enrollment, and trial design [2, 6].

Inference Perspective: How Teams Use It

Apotek Signal is designed for early and mid-development decision points, not only retrospective analysis.

To generate a prediction, teams provide:

·       a Drug–Target–Indication definition, and

·       a minimal clinical study outline.

This enables use before trials are executed, when uncertainty reduction has the highest impact on capital allocation and strategic direction, a point frequently highlighted in pharma portfolio strategy analyses [7].

How Apotek Signal Compares Existing Approaches in Oncology

The Industry Baseline

 In oncology, success probabilities are still commonly estimated using:

• historical phase transition rates published in industry reports,

• benchmark databases, and

• expert committee review.

While these provide useful base rates, they are not asset-specific and often fail to reflect how trial design choices, patient stratification, and translational fit influence outcomes. Longitudinal studies show that success rates vary meaningfully across time, phase, and development strategy [1, 3].

Apotek Signal’s Positioning

 Apotek Signal is designed to complement—and improve upon—existing approaches by being:

• Asset-specific, not average-based

• Trial design-aware, not molecule-only

• Directly usable in valuation workflows (rNPV / PTRS)

• Decision-oriented, augmenting expert judgment rather than replacing it

The goal is not to produce a single “answer,” but to give teams a shared quantitative starting point for discussion—especially in high-stakes portfolio and partnering decisions.

Practical Use Cases Enabled by Apotek Signal

• Indication expansion for existing oncology assets, where base-rate assumptions are particularly misleading.

• Portfolio prioritization across clinical phases, using consistent probability logic rather than qualitative rank [4].

• Licensing and due diligence support, anchoring rNPV assumptions to explicit PoS estimates [8, 9].

• Early study design stress-testing, a use case increasingly emphasized in portfolio optimization frameworks.

These are the moments where uncertainty is high, data are incomplete, and decisions are hardest to reverse.

About Apotek

Apotek harnesses groundbreaking Causal AI and GenAI technologies to build an integrated platform that transforms multi-omics data into actionable insights — enabling novel target discovery, biosimulation, and generative drug design. By uncovering cause-and-effect relationships rather than spurious correlations, our approach enhances biological interpretability, improves model robustness, and accelerates decision-making across the drug discovery pipeline. 👉 More Apotek Solutions

References

1. BIO / Informa Pharma Intelligence. Clinical Development Success Rates and Contributing Factors. https://www.bio.org/clinical-development-success-rates-and-contributing-factors-2011-2020

2. Hay M. et al. Clinical development success rates for investigational drugs. Nature Reviews Drug Discovery (2014). https://www.nature.com/articles/nbt.2786

3. Wong C.H., Siah K.W., Lo A.W. Estimation of clinical trial success rates and related parameters. Biostatistics (2019). https://academic.oup.com/biostatistics/article/20/2/273/4817524

4. ZS Associates. Mispriced risk: Finding undervalued biopharma business development opportunities. https://www.zs.com/insights/build-smarter-pharma-portfolio-strategy-using-ai-ml-quantify-risk

5. EY. How AI Can Support Drug Development Portfolio Decisions. https://www.ey.com/en_us/insights/life-sciences/how-ai-can-support-drug-development-portfolio-decisions

6. RAPS. Estimating the Probability of Regulatory Registration Success. https://www.raps.org/news-and-articles/news-articles/estimating-the-probability-of-regulatory-success

7. McKinsey & Company. How Biopharmaceutical Leaders Optimize Their Portfolio Strategies. https://www.mckinsey.com/industries/life-sciences/our-insights/how-biopharmaceutical-leaders-optimize-their-portfolio-strategies

8. Alacrita Consulting. Valuing Pharmaceutical Assets: When to Use NPV vs rNPV. https://www.alacrita.com/whitepapers/valuing-pharmaceutical-assets-when-to-use-npv-vs-rnpv

9. WIPO. Valuation in Biotechnology and Pharmaceuticals. https://www.wipo.int/web-publications/intellectual-property-valuation-in-biotechnology-and-pharmaceuticals/en/index.html