How Model-Based and Model-Assisted Designs Are Advancing Precision Medicine
Advancing Precision Medicine Through Model Based and Model Assisted Design Approaches
The selection of an appropriate Phase 1 trial design can significantly impact early drug development outcomes. While the 3+3 design was the standard for many years, contemporary approaches offer enhanced methods with a higher likelihood of identifying the true maximum tolerated dose and allow for the collection of information that can efficiently lead into dose optimization all while maintaining patient safety.
Modern sponsors can choose from multiple options, ranging from streamlined designs to mathematically rigorous methodologies. Selecting an effective dose-finding approach requires a thoughtful collaboration with all stakeholders, scientific, operational, and business partners to ensure careful consideration of patient safety, speed, and data quality.
This guide examines the capabilities and limitations of current clinical trial designs to help inform the selection process for your program.
BOIN Trial Design is A Simple Yet Statistically Rigorous Option for Phase 1 Trials
BOIN (Bayesian Optimal Interval) design bridges traditional and modern approaches. It provides a transparent framework that maintains statistical integrity while remaining accessible to implementation teams.
| Strengths | Limitations |
|
|
Optimal use: Phase 1 oncology trials requiring balanced statistical strength and operational efficiency, particularly with sites new to adaptive designs.
i3+3 and Backfill Designs Can Enhance Safety in Traditional Dose Escalation
The i3+3 design updates conventional 3+3 methodology. Its backfill variant (Bi3+3) enables lower dose assessment—particularly important for targeted therapy development.
| Strengths | Limitations |
|
|
Optimal use: Programs prioritizing safety considerations or regulatory compliance, and organizations transitioning from traditional designs.
mTPI-2 Design in Phase 1 Trials Strikes a Balance Between Simplicity and Power
The mTPI-2 (Modified Toxicity Probability Interval) design represents a middle ground between rule-based and model-based approaches. It has gained acceptance in immuno-oncology trials where dose-effect relationships often defy linear patterns.
| Strengths | Limitations |
|
|
Optimal use: Programs seeking enhanced statistical rigor while avoiding full model-based complexity.
CRM Design for Dose-Finding Can Deliver Precision, Speed, and Statistical Rigor
The CRM (Continual Reassessment Method) delivers precise dose recommendations, proving particularly valuable for novel mechanisms and complex safety profiles.
| Strengths | Limitations |
|
|
Optimal use: Programs where precise dose-finding is essential, especially for new drug classes or non-linear dose-response relationships.
BLRM in Phase 1 Trials Uses Historical Data to Guide Dose Escalation
The BLRM (Bayesian Logistic Regression Model) excels when deployed with an over-dose control method like EWOC (Escalation with over-dose control).BLRM has the flexibility for the use in trials evaluating multiple drugs.
| Strengths | Limitations |
|
|
Optimal use: Programs with substantial prior data, combination therapy studies, or work in well-documented therapeutic areas.
Key Factors and Framework for How to Choose the Right Phase 1 Trial Design
When choosing a Phase 1 clinical trial design, consider these important factors before making your final decision:
- Available statistical expertise throughout the trial duration
- Pre-clinical findings and extent of prior compound behavior knowledge
- Implementation complexity tolerance
- Regulatory strategy requirements
The most effective design aligns with your specific circumstances and operational capabilities rather than pure statistical sophistication.
The Future of Phase 1 Clinical Trial Designs Account for Personalization, AI, and Statistical Innovation
As medicine advances toward greater personalization and therapeutic sophistication, selecting the right trial design becomes increasingly critical. Precision for Medicine offers deep expertise in tailoring approaches to align with each program's unique needs. Our experienced statisticians can provide guidance on strategy for navigating regulatory, operational, and business needs. Partnering with Precision for Medicine provides you with access to thought partners useful for the efficient and economic scientific research to reach the best outcomes for treated patients and business entities.
Through our full-service and Functional Service Provider (FSP) biostatistics capabilities, we ensure that every trial benefits from precise statistical modeling, robust data interpretation, and seamless integration of design methodologies.
Partner with Precision to leverage our multidisciplinary team dedicated to the success of your trial and program.
References
1. Li N, Zhou X, Yan D. Phase I clinical trial designs in oncology: A systematic literature review from 2020 to 2022. J Clin Transl Sci. 2024;8(1):e134. Published 2024 Sep 24. doi:10.1017/cts.2024.599.
