Published on 24/12/2025
Establishing Go/No-Go Criteria for Phase 2 Trials: Making the Right Decision to Advance to Phase 3
Introduction
One of the most critical junctures in drug development occurs at the end of a Phase 2 trial—deciding whether to advance a candidate to Phase 3. This decision, often referred to as a “Go/No-Go” decision, involves a thorough evaluation of safety, efficacy, pharmacokinetics, pharmacodynamics, and strategic considerations. A well-defined decision framework helps ensure objectivity, minimizes bias, and avoids the costly mistake of proceeding with a product that lacks sufficient evidence of benefit. This tutorial outlines the key components, methodologies, and best practices for defining and executing Go/No-Go decisions in Phase 2 clinical trials.
Why Go/No-Go Criteria Matter
- Data-Driven Decisions: Prevent emotional or financial bias from influencing progression
- Risk Mitigation: Avoids Phase 3 failures due to weak efficacy or unacceptable safety signals
- Resource Optimization: Allocates time and capital only to the most promising programs
- Regulatory Readiness: Facilitates productive end-of-Phase 2 meetings with agencies
Key Decision Inputs for Go/No-Go Evaluation
- Primary Efficacy Signal: Magnitude and consistency of effect vs. control or baseline
- Safety Profile: Frequency and severity of adverse events; tolerability profile
- Dose-Response Relationship: Clarity of optimal dose selection
- Pharmacokinetics/Pharmacodynamics: Evidence of target engagement and therapeutic
Quantitative Go/No-Go Thresholds
To promote objectivity, many teams predefine specific thresholds. Examples include:
- Response rate: ≥ 30% vs. ≤ 10% in historical control
- Effect size: Minimum of 0.5 standard deviations on a key endpoint
- Safety limits: Grade 3 or higher AE rates below 15%
- PK parameters: AUC within predefined therapeutic range
Bayesian and Predictive Approaches
- Use of Bayesian probability models to predict Phase 3 success based on Phase 2 data
- Examples: Probability that response rate ≥ 30% exceeds 85% confidence threshold
- Simulations can estimate likelihood of success under various Phase 3 assumptions
Multifactorial Decision-Making Frameworks
Many sponsors use structured frameworks to combine multiple data sources:
1. TPP-Based Scoring
Score trial results against a Target Product Profile (TPP) covering efficacy, safety, dosing, and formulation.
2. Decision Matrices
| Criteria | Weight | Score | Weighted Score |
|---|---|---|---|
| Efficacy (ORR) | 40% | 4/5 | 1.6 |
| Safety (Grade ≥3 AEs) | 30% | 5/5 | 1.5 |
| PK/PD | 15% | 3/5 | 0.45 |
| Market Potential | 15% | 4/5 | 0.6 |
| Total Weighted Score | 4.15 / 5 | ||
Operationalizing the Go/No-Go Process
- Define go/no-go criteria during protocol design or before database lock
- Conduct internal decision-making meetings including clinical, regulatory, commercial, and biostatistics
- Document decisions and rationale in official development plans
- Align decisions with regulatory interactions (e.g., End-of-Phase 2 meeting)
Common Pitfalls to Avoid
- Moving goalposts: Changing criteria post hoc introduces bias
- Over-reliance on underpowered findings: Misinterpreting trends as proof
- Lack of cross-functional input: Decisions made in silos miss strategic factors
- Ignoring patient or market needs: Efficacy alone is not enough
Regulatory and Strategic Considerations
- Discuss go/no-go criteria at the End-of-Phase 2 meeting with the FDA or EMA
- Align expectations across internal teams, investors, and partners
- Use decision outputs to update the Target Product Profile (TPP)
Conclusion
Making a robust Go/No-Go decision after a Phase 2 trial is as much a strategic milestone as it is a scientific one. Clear, quantitative, and pre-specified criteria—supported by a structured review process—can help avoid costly mistakes and ensure only the most promising candidates advance. By combining statistical rigor, cross-functional input, and market awareness, sponsors can increase the odds of success in Phase 3 and beyond.