ABOUT US
Overview
Our Leadership Team
Strategic Consulting
OUR SERVICES
CAREERS
Life at Maxis
Openings
OTHER
INSIGHTS
MAKE AN APPOINTMENT
FAQ
CONTACT
Policies & Terms
ABOUT US
Overview
Our Leadership Team
Strategic Consulting
OUR SERVICES
CAREERS
Life at Maxis
Openings
OTHER
INSIGHTS
MAKE AN APPOINTMENT
FAQ
CONTACT
Policies & Terms

Adaptive Protocols in Early Trials: Boosting Success with AI

  • 2 MIN READ
RWE
Clinical trials remain the backbone of medical research, yet traditional trial designs characterized by fixed sample sizes and limited treatment arms often struggle to address the increasingly complex questions researchers face today. Adaptive clinical trial designs (ADs) offer a powerful alternative, allowing pre-planned modifications in response to accumulating data. In 2019, the FDA issued final guidance encouraging the use of adaptive designs, outlining recommendations for their development, application, and transparent reporting1. Compared to conventional methods, adaptive designs offer several advantages: they improve statistical efficiency by better controlling error rates and increasing power; they address ethical concerns by allocating more patients to better-performing treatments and enabling early trial termination when safety issues or lack of benefit arise; and they enhance understanding of treatment effects by enriching trials with participants who are more likely to respond2.

What Are Adaptive Designs? Flexibility and Efficiency in Action

Adaptive designs introduce flexibility into clinical trials by permitting planned changes based on interim data. This flexibility can improve trial quality and shorten or reduce trial duration. However, successful implementation requires careful planning to overcome practical challenges such as budget constraints, communication hurdles, and operational complexities3. Learning from past experiences and sharing insights through detailed reporting is essential for advancing adaptive methodologies. While adaptive designs are not suitable for all studies, especially those with long-term endpoints, they should be a key tool in the clinical investigator’s arsenal. These designs help guard against incorrect initial assumptions but are no substitute for poor trial planning3. Though adaptive designs carry some risk of bias, this can be minimized through meticulous preparation, transparency, oversight by skilled Independent Data Monitoring Committees (IDMCs), and blinding trial personnel to ongoing adaptations. Accurate and comprehensive reporting of design details and outcomes is critical and is now codified in an adaptive design appendix to the CONSORT guidelines.3

Key Adaptive Design Strategies

Adaptive trials enable a range of modifications to improve efficiency and ethical standards:

    • • Group Sequential Designs: These incorporate scheduled interim analyses that allow early stopping for efficacy or futility based on predefined criteria4,5.
      • Adaptive Randomization: Techniques like minimization adjust treatment assignment probabilities to favor more successful therapies as the trial progresses 6, balancing patient benefit with the need for robust statistical inference to maintain trial integrity and power1.
      • Sample Size Re-estimation: This allows adjustment of enrollment targets to maintain adequate statistical power, reducing the risk of underpowered studies7.
      • Adding or Dropping Treatment Arms: Trials can dynamically introduce or remove treatment groups based on interim results, optimizing resource use8.
      • Enrichment Strategies: By focusing on subpopulations likely to benefit, these designs increase the precision of treatment effect estimates9.
      • Seamless Phase II/III Designs: These combine phases into a continuous trial, using phase II data to inform phase III decisions without stopping the study10.
      • Adaptive Dose-Finding: Dose levels are adjusted in response to participant reactions, maximizing safety and efficacy11.

By allowing these modifications, adaptive designs address ethical concerns by minimizing patient exposure to ineffective treatments and improving trial success rates.

Artificial Intelligence: A Game Changer for Adaptive Trials

AI is transforming adaptive trials by enabling advanced data analytics, predictive modeling, and real-time decision-making. The explosion of complex clinical, molecular, and imaging data requires powerful tools to extract actionable insights quickly and accurately8.

AI enhances adaptive protocols by:

    • • Improving patient recruitment and selection through analysis of clinical, genetic, and lifestyle data, identifying eligible and suitable participants more efficiently.
      • Optimizing dosing and treatment allocation using machine learning models that predict efficacy and safety outcomes.
      • Supporting dynamic randomization to favor better-performing treatment arms.
      • Simulating trial outcomes and design scenarios to refine protocols and reduce failure rates.
      • Predicting participant dropout risks, enabling targeted retention strategies to maintain trial integrity.
Picture1

By automating and accelerating these processes, AI makes adaptive trials more responsive, efficient, and patient-centric.

Enhancing Trial Design and Execution with AI-Driven Adaptive Protocols

Traditional clinical trials often suffer from limited generalizability and rigid protocols. AI-powered adaptive designs overcome these limitations by enabling:

    • • Stratified cohort identification that selects patient subgroups based on multiple variables such as genetics, lifestyle, and environment.
      • Biomarker-driven trial designs that improve precision by targeting patients most likely to respond.
      • Multi-modal treatment optimization that integrates diverse data sources for better decision-making.

AI models also predict clinical trial success probabilities, regulatory approval likelihood, and patient outcomes, helping sponsors make informed decisions early. In oncology, AI-driven in-silico trials simulate patient cohorts and treatment responses, reducing costly trial failures and accelerating progress.

A practical illustration of these advances can be seen in Maxis Clinical Sciences’ recent use case in ulcerative colitis drug development, where the integration of Real-World Evidence (RWE) with adaptive trial design and AI-driven analytics enabled dynamic patient selection and treatment arm adjustments. This approach accelerated the study timeline by up to two years compared to traditional designs, while maintaining regulatory compliance and scientific rigor. By leveraging Bayesian statistical methods and machine learning algorithms, Maxis Clinical Sciences demonstrated how combining adaptive designs with AI and RWE can transform early clinical trials and improve success rates 12. The flexibility of Bayesian methods, their ability to formally incorporate prior knowledge, and their emphasis on probabilistic inference offer significant advantages for dynamic trial designs. These strengths can translate into smaller required sample sizes and more efficient, data-driven decision-making compared to traditional frequentist approaches13.

Challenges in Integrating AI with Adaptive Designs

Despite their promise, AI-enhanced adaptive trials face several challenges:

    • • Data quality and availability: AI requires large, well-curated, and labeled datasets to perform reliably.
      • Ethical and regulatory compliance: Transparency (“explainability”) of AI decisions and compliance with evolving guidelines are essential.
      • Operational complexity: Implementing AI-driven adaptive protocols demands collaboration across disciplines and robust IT infrastructure.
      • Bias and validation: AI models must be tested across diverse populations to avoid bias and ensure broad applicability.

Addressing these challenges is critical to fully unlocking AI’s potential in clinical research.

Practical Applications: Improving Patient Selection, Randomization, and Treatment Arms

    • • Patient Selection: AI analyzes electronic health records (EHRs), genomics, imaging, and lifestyle data to identify eligible patients and predict disease progression. Tools like natural language processing (NLP) automate data extraction, improving recruitment speed and accuracy.
      • Randomization: Beyond traditional methods, AI enables response-adaptive randomization, dynamically adjusting treatment allocations based on ongoing results. AI can also create synthetic control arms using historical data, reducing the need for placebo groups and improving patient enrollment.
      • Treatment Arm Adaptation: Adaptive dose-finding and drop-the-loser strategies focus resources on the most promising therapies. AI-powered platform trials allow adding or removing treatments in real time, enhancing trial relevance and efficiency.

Conclusion

Adaptive protocols empowered by AI represent a paradigm shift in early-phase clinical trials. By combining flexibility with intelligent data analysis, these designs enable faster, more ethical, and patient-focused studies. As regulatory frameworks evolve and AI technologies mature, AI-driven adaptive trials are poised to become the standard, accelerating drug development and improving patient outcomes worldwide.

By embracing this innovative approach, clinical researchers can overcome traditional trial limitations and pioneer a future where treatments reach patients more efficiently and precisely than ever before.

References

  1. Center for Drug Evaluation and Research. Adaptive Design Clinical Trials for Drugs and Biologics Guidance for Industry [Internet]. U.S. Food and Drug Administration. 2020 [cited 2025 Jul 11]. Available from: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/adaptive-design-clinical-trials-drugs-and-biologics-guidance-industry
  2. Center for Drug Evaluation and Research. Complex Innovative Trial Designs [Internet]. U.S. Food and Drug Administration. 2024. Available from: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/interacting-fda-complex-innovative-trial-designs-drugs-and-biological-products
  3. Pallmann P, Bedding AW, Choodari-Oskooei B, Dimairo M, Flight L, Hampson LV, et al. Adaptive designs in clinical trials: why use them, and how to run and report them. BMC Medicine. 2018 Feb 28;16(1).
  4. Perkins GD, Quinn T, Deakin CD, Nolan JP, Lall R, Slowther AM, et al. Pre-hospital Assessment of the Role of Adrenaline: Measuring the Effectiveness of Drug administration In Cardiac arrest (PARAMEDIC-2): Trial protocol. Resuscitation. 2016 Nov;108(108):75–81.
  5. Burnett T, Jennison C. Adaptive enrichment trials: What are the benefits? Statistics in Medicine. 2020 Nov 26;40(3):690–711.
  6. Taves DR. The use of minimization in clinical trials. Contemporary Clinical Trials. 2010 Mar;31(2):180–4.
  7. Santis FD. Power Priors and Their Use in Clinical Trials. The American Statistician [Internet]. 2006 Feb 1;60(May):122–9. Available from: https://www.researchgate.net/publication/4741256_Power_Priors_and_Their_Use_in_Clinical_Trials
  8. Adaptive platform trials: definition, design, conduct and reporting considerations. Nature Reviews Drug Discovery. 2019 Aug 28;18(10):797–807.
  9. Thall PF. Adaptive Enrichment Designs in Clinical Trials. Annual Review of Statistics and Its Application. 2021 Mar 7;8(1):393–411.
  10. Cuffe RL, Lawrence D, Stone A, Vandemeulebroecke M. When is a seamless study desirable? Case studies from different pharmaceutical sponsors. 2014 Jul 1;13(4):229–37.
  11. Iasonos A, O’Quigley J. Adaptive Dose-Finding Studies: A Review of Model-Guided Phase I Clinical Trials. Journal of Clinical Oncology. 2014 Aug 10;32(23):2505–11.
  12. Use Case | Adaptive Designs For Clinical Trials With RWE [Internet]. Maxis Clinical Sciences. 2024 [cited 2025 Jul 11]. Available from: https://maxisclinical.com/insights/use-case/adaptive-designs-for-clinical-trials-rwe-approach/
  13. Berry, D. A. (2006). Bayesian clinical trials. Nature Reviews Drug Discovery, 5(1), 27–36. DOI: 10.1038/nrd1927

SHARE

Other Blogs