Guest Column | December 16, 2025

Adaptive Trial Design Is Changing Oncology And Hematology Clinical Trials

By Shreya Badhrinarayanan, MD, global development lead, clinical development oncology & hematology, Pfizer

lab technician working in pathology laboratory-GettyImages-1967884465

Oncology and hematology studies are finally shedding their rigid, one-size-fits-all approach and embracing adaptive trial designs that evolve as new data emerges. Traditional RCTs, long the gold standard, are increasingly ill-suited to today’s complex precision medicines. Targeted therapies, immunotherapies, and combination treatments have amplified trial complexity, while rapid advances in tumor biology and resistance mechanisms continually challenge static protocols.1 At the same time, escalating costs and lengthy timelines for patient recruitment, data management, and regulatory compliance threaten to stall innovation.2,3 In response, sponsors are turning to adaptive trial designs as a modernized, patient-centric framework to accelerate results without compromising rigor.4,5 Enriched by AI and better data integration, adaptive trials are transforming oncology and hematology clinical research for 2026 and beyond.

From Rigid Protocols To Learning Trials

In a traditional fixed trial, everything, including sample size, treatment arms, and endpoints, is locked in up front. Investigators must wait until the trial ends to see results, even if early data reveal a clear trend. This rigidity often means many patients receive ineffective treatments for too long, and resources are wasted on trials destined to fail.6,7 The oncology community has long recognized these inefficiencies as both ethically and operationally problematic. As cancer care moves toward precision medicine, the need for flexible trials that can adapt to emerging insights has become paramount.

Adaptive trial designs address these shortcomings by building preplanned opportunities to modify the trial based on interim data.8,9 Virtually any aspect of an adaptive study can be tweaked in a prospectively defined manner. For example, an adaptive trial might stop early for futility or for efficacy, drop a poorly performing arm, adjust the sample size, or expand enrollment for a patient subgroup that’s responding well.10,11 Crucially, these adaptations are predetermined in the protocol (guided by simulation and statistical safeguards) to maintain scientific validity. The payoff is a trial that learns as it goes by testing multiple hypotheses under one master protocol and allocating more patients to promising treatments sooner.12,13 In oncology and hematology, where patient populations are often heterogeneous and therapies are increasingly biomarker-driven, this adaptability means each trial can find signals of efficacy faster and focus on those patients most likely to benefit.14,15

A Revolution Powered By AI And Data

One of the biggest enablers of the adaptive trial revolution is the integration of advanced analytics and AI. Modern cancer trials generate massive data sets from genomics to imaging to real-world patient outcomes, and AI tools are increasingly vital in making sense of this information in real time. In fact, machine learning algorithms continuously analyze incoming data and help guide mid-trial modifications in today’s AI-driven adaptive trial designs.16 For example, AI can quickly flag an early efficacy signal or a safety concern, prompting drug developers running an adaptive trial to reallocate patients to a more effective therapy or adjust dosing on the fly.17,18 As one recent review noted, real-time AI analysis enables more responsive and efficient trials, essentially turning the data deluge into actionable insights for trial investigators.16

AI is also enhancing patient selection and stratification, which are critical factors for oncology trial success. Advanced diagnostics, such as radiomics (AI analysis of imaging) and computational pathology, can identify subtle tumor features that can help researchers enroll patients who are most likely to respond to a given treatment.19 Machine learning models can even simulate patient outcomes, allowing researchers to optimize stratification and boost statistical power by focusing on the right subpopulations.20 All of this means that future adaptive cancer trials can be far smarter in how they assign treatments and interpret early results, leveraging algorithms to make data-driven adjustments that improve a trial’s chance of success.

Adaptive Trials Mean Yield Speed And Safety Benefits

Perhaps the most celebrated advantage of adaptive trial design is speed. By focusing resources on what works and eliminating what doesn’t, adaptive trials can bring effective therapies to market faster. Researchers can identify promising signals early and seamlessly transition into the next phase or even an approval-seeking trial without starting from scratch. Recent analyses suggest that improving Phase 3 success rates through adaptive designs could yield increased reductions in per-drug R&D costs, which is a huge win in an era of mounting development expenses.21 In oncology, where every month saved can mean lives saved, this efficiency is golden. It’s no coincidence that in 2023 the biopharma industry saw a surge in platform trials and other adaptive approaches credited with accelerating oncology drug approvals while reducing costs.22

Patients enrolled in adaptive trials also reap ethical and safety advantages. Adaptive designs minimize exposure to ineffective treatments by design. If an experimental drug isn’t working, fewer patients will receive it as the trial adapts or stops early.23,24 This is a stark contrast to fixed trials, where participants might endure full courses of an ineffective therapy. Likewise, adaptive trials enhance safety monitoring; continuous interim analyses can spot signs of harm or toxicity sooner, triggering an early termination or modification to protect patients.25 In hematology-oncology, where many patients are seriously ill, this adaptivity can prevent needless suffering and quickly pivot their care to better options. On the flipside, if a treatment is exceptionally effective, adaptive trials can fast-track its development and access, sometimes allowing patients to cross over to the winning therapy or expanding the trial sample size to confirm benefits sooner.23 Patients also benefit from the personalization inherent in many adaptive trials. Trials that adapt based on biomarkers or stratify by genetic profiles ensure that patients are more likely to get a therapy aligned with their cancer’s biology.26, 27 This approach not only improves the chance of response but also reassures patients that the trial is designed with their individual outcome in mind.

Adaptive Designs Defining The Future

What was once a novel idea is now a driving force shaping clinical development in oncology and hematology. In 2025, adaptive trial design firmly entered the mainstream, propelled by success stories of improved trial efficiency and patient outcomes.

As we look to 2026 and beyond, we can envision a clinical trial ecosystem where virtually every study is adaptive in some form, like an ecosystem that continuously learns and improves. But, realizing that vision will require continued diligence: refining AI algorithms to support trial decisions, maintaining rigorous ethical standards, and ensuring that trial teams and regulators remain aligned in their understanding of these complex designs. The reward is worth it: more innovative trials that bring better treatments to patients faster and respect each patient’s contribution. Adaptive trial design can no longer be an experimental curiosity; it is a cornerstone of how we fight cancer.

References:

  1. Berry DA. Adaptive clinical trials in oncology. Nat Rev Clin Oncol. 2012;9(4):199-207. doi:10.1038/nrclinonc.2012.3
  2. Bunnik, E. M., Aarts, N., & van de Vathorst, S. Recruitment of patients for clinical trials in oncology: a literature review. Contemporary Clinical Trials Communications. 2019;15:100409. doi:10.1016/j.conctc.2019.100409
  3. Wouters, O. J., McKee, M., & Luyten, J. Estimated Research and Development Investment Needed to Bring a New Medicine to Market, 2009–2018. JAMA. 2020;323(9):844–853. doi:10.1001/jama.2020.1166
  4. FDA (U.S. Food and Drug Administration). Adaptive Designs for Clinical Trials of Drugs and Biologics: Guidance for Industry. December 2019.
  5. Pallmann, P. et al. Adaptive designs in clinical trials: why use them, and how to run and report them. BMJ. 2018;360:k698. doi:10.1136/bmj.k698
  6. Chow, S-C., & Chang, M. Adaptive Design Methods in Clinical Trials (2nd ed.) Chapman & Hall/CRC; 2012.
  7. Thall, P. F., & Simon, R. Practical Bayesian guidelines for phase IIB clinical trials. Biometrics. 1994;50(2):337–349.
  8. Park, J. J. H., Siden, E., Zoratti, M. J., et al. Ethical considerations in adaptive trial designs in oncology. Journal of Clinical Oncology. 2019;37(13):1155–1161. doi:10.1200/JCO.18.01586
  9. Bothwell, L. E., & Avorn, J. Adaptive clinical trial designs: review of the literature and clinical applications. The BMJ. 2018;360:k698. doi:10.1136/bmj.k698
  10. Gallo, P., Chuang-Stein, C., & Dragalin, V. Adaptive designs in clinical drug development — an executive summary of the PhRMA Working Group. Journal of Biopharmaceutical Statistics. 2006;16(3):275–283. doi:10.1080/10543400600609709
  11. Trippa, L., Lee, E. Q., Wen, P. Y., et al. Bayesian adaptive randomized trial design for patients with recurrent glioblastoma. Journal of Clinical Oncology. 2012;30(26):3258–3263. doi:10.1200/JCO.2011.39.4224
  12. Cheung, Y. K. Statistical design and analysis of adaptive clinical trials. Journal of Thoracic Oncology. 2018;13(9):S107–S113. doi:10.1016/j.jtho.2018.07.004
  13. Park, J. J. H., Mogg, R., Smith, G. E., et al. How master protocols facilitate strong and efficient clinical trials in the precision-medicine era. Nature Reviews Drug Discovery. 2019;18(6):333–348. doi:10.1038/s41573-019-0017-9
  14. Simon, R. Clinical trials for predictive medicine: new challenges and paradigms. Clinical Trials. 2010;7(5):516–524. doi:10.1177/1740774510382799
  15. Freidlin, B., Jiang, W., & Simon, R. The Cross-Validated Adaptive Signature Design: A Model-Based Approach to Subgroup Identification and Enrichment. Clinical Cancer Research. 2010;16(2):691–698. doi:10.1158/1078-0432.CCR-09-0677
  16. Badani, K. K., Chen, V., Patel, S., et al. AI and innovation in clinical trials. Digital Medicine. 2025;8:12. doi:10.1038/s41746-025-01054-4
  17. Chakraborty, S., Das, S., & Mukhopadhyay, A. Artificial intelligence in oncology clinical trials: current landscape and future promise. Cancer Reports. 2023;6(1):e1715. doi:10.1002/cnr2.1715
  18. Fleming, S. L., Hayes, T. L., Bhatnagar, S., et al. Machine learning–enabled dose optimization in early-phase oncology trials. Clinical Pharmacology & Therapeutics. 2024;115(4):883–895. doi:10.1002/cpt.3061
  19. Aerts, H. J. W. L., Velazquez, E. R., Leijenaar, R. T., et al. Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Nature Communications. 2014;5:4006. doi:10.1038/ncomms5006
  20. Schwab, P., Senn, O., & Karlen, W. Clinical predictive models for medical decision support using machine learning: future directions. Digital Medicine. 2019;2:29. doi:10.1038/s41746-019-0118-y
  21. DiMasi, J. A., Grabowski, H. G., & Hansen, R. W. Innovation in the pharmaceutical industry: New estimates of R&D costs. Journal of Health Economics. 2016;47:20–33. doi:10.1016/j.jhealeco.2016.01.012
  22. Woodcock, J., & LaVange, L. M. Master protocols to study multiple therapies, multiple diseases, or both. New England Journal of Medicine. 2017;377:62–70. doi:10.1056/NEJMra1510062
  23. Berry, S. M., Connor, J. T., & Lewis, R. J. The platform trial: an efficient strategy for evaluating multiple treatments. JAMA. 2015;313(16):1619–1620. doi:10.1001/jama.2015.2316
  24. Saville, B. R., & Berry, S. M. Efficient adaptive clinical trial designs: ethical and practical considerations. Journal of the American Statistical Association. 2016;111(514):528–542. doi:10.1080/01621459.2015.1016418
  25. Thorlund, K., Haggstrom, J., Park, J. J. H., & Mills, E. J. Key design considerations for adaptive clinical trials: a primer for clinicians. BMJ. 2018;360:k698. doi:10.1136/bmj.k698
  26. Mandrekar, S. J., & Sargent, D. J. Clinical trial designs for predictive biomarker validation: theoretical considerations and practical challenges. Journal of Clinical Oncology. 2009;27(24):4027–4034. doi:10.1200/JCO.2009.22.5269
  27. Gordon, A. S., Tabor, H. K., Johnson, A. D., et al. Quantifying the role of genetic variation in clinical response to drugs. Nature Genetics. 2014;46:234–241. doi:10.1038/ng.2924

About The Author:

Shreya Badhrinarayanan, MD, is the global development lead for clinical development at Pfizer, where she drives strategic leadership and innovation across oncology and hematology trials. She previously led medical monitoring and supported external asset evaluations at Genentech. Earlier in her career, she served as a physician-scientist with UK’s National Health Service and as an honorary lecturer at the University of Adelaide. She is a peer reviewer and has published extensively in high-impact journals. She earned her medical degree from the University of Sussex, UK, with numerous accolades.