Adaptive clinical trials explained: how a study can change while it is running

The idea that a clinical trial can change while it is running surprises many people who are considering participation. A traditional study locks every rule in place before the first participant enrolls, and those rules stay fixed until the last data point is collected. An adaptive clinical trial takes a different approach. It builds specific decision points into the study plan in advance, and the research team is allowed to make certain pre-approved changes based on what the accumulating information shows.

For a participant, that raises reasonable questions. What can change? What cannot change? Is the study still safe if it looks different halfway through? This article explains how adaptive clinical trials work, what stays the same for participants enrolled in them, and what to ask a research site team before joining one.

What an adaptive clinical trial is

The U.S. Food and Drug Administration (FDA) defines an adaptive clinical trial as a study design that allows for prospectively planned modifications to one or more parts of the study, based on data gathered from participants already enrolled. The key word is prospectively. Every possible change is written into the study protocol (the master document that describes how the study will be run) before the first participant enrolls. Nothing is improvised during the study.

A useful way to picture it is a recipe with pre-approved checkpoints. A traditional recipe is followed exactly as written from start to finish. A recipe with pre-approved checkpoints includes a small number of specific "if this happens, do that" instructions that were agreed on before cooking started. The cook does not invent new steps. The cook only follows steps that were planned in advance.

Adaptive clinical trials are governed by strict regulatory guidance. The FDA finalized its guidance on adaptive designs for drugs and biologics in November 2019, and a separate guidance for medical device studies has been in place since 2016. International guidelines from the International Council for Harmonisation (ICH), a body that sets shared standards for medical research across major regulatory regions, add further requirements around statistical rigor and study integrity.

For a plain-language introduction to how clinical trials work in general, see Clinical Trials Explained: Simple Guide for Beginners.

How adaptive clinical trials differ from traditional clinical trials

A traditional clinical trial fixes its design at the start. The number of participants, the doses being tested, the ratio of participants assigned to each group, and the total duration are set in advance and held constant. Researchers analyze the results only after the study is complete.

An adaptive clinical trial keeps most of those same protections but adds one important feature: pre-planned interim analyses. An interim analysis is a scheduled review of the study data while the study is still running. These reviews happen at points that were chosen before the study began, and they trigger only the specific changes that the protocol allows.

The trade-off is straightforward. Adaptive designs can be more efficient, meaning they may answer a research question with fewer participants or in less time. They can also reduce the number of participants exposed to a study intervention that is not working. In exchange, they require more up-front planning, specialized statistics, and stronger independent oversight than a traditional design.

For further background on how clinical research is structured, see How Clinical Trials Advance Medicine and Change Lives.

The kinds of pre-planned changes that can happen during an adaptive trial

The changes allowed in an adaptive clinical trial come from a defined menu, not from open-ended freedom. The FDA guidance names the main types.

Sample size re-estimation allows researchers to increase or decrease the number of participants based on how variable the early results are. If the early data suggest more participants are needed to reach a clear answer, the study can expand within pre-agreed limits.

Group sequential design allows the study to stop early at a planned interim look if the evidence is already conclusive, either for benefit or for a lack of benefit (a situation called futility).

Dropping or adding study arms allows the research team to remove a group that is clearly not working, or to add a new investigational product to a study designed to test several candidates over time. This is common in platform trials, which are built to keep running and rotate candidates in and out.

Response-adaptive randomization changes the probability of being assigned to a particular group as the study progresses, based on how each group is performing. New participants may have a higher chance of being assigned to a group that appears to be doing better.

Adaptive enrichment narrows enrollment toward a subgroup of participants, for example people with a specific biomarker (a measurable biological indicator), who appear more likely to benefit.

Seamless phase 2/3 designs combine what would normally be two separate studies into one continuous protocol, saving time between phases.

To understand how randomization and control groups work in any clinical study, see Placebos and Controls: What It Means in Your Study.

What stays the same for participants when a study adapts

The design of an adaptive clinical trial can shift within its pre-approved rules. The core protections for the people enrolled do not.

Informed consent remains an ongoing process throughout any clinical trial, adaptive or not. When a change to the study protocol affects a participant’s rights, safety, or reasons for continuing, an Institutional Review Board (an independent committee that reviews research to protect participants) determines whether participants must be re-consented. Re-consent means the participant is told about the change, given a revised consent document, and asked to sign again before continuing.

Safety monitoring continues without interruption. Adverse events, meaning any medical problems that occur during the study, whether related to the study intervention or not, are tracked and reported in the same way they would be in a fixed study.

Participant rights do not change. The right to ask questions, to receive clear answers, to know what the study involves, and to withdraw at any time for any reason without penalty applies in every clinical trial. Withdrawing from an adaptive clinical trial does not affect a participant’s access to standard medical care.

The independence of the research site team also remains constant. Final eligibility determination, informed consent, study walk-through, and enrollment decisions are handled by the site team.

Real examples of adaptive clinical trials

Several well-known clinical trials have used adaptive designs to reach answers faster than a traditional study would have.

The RECOVERY trial, launched in the United Kingdom in March 2020, was designed as an adaptive platform study to test multiple candidates for hospitalized COVID-19 patients. It enrolled tens of thousands of participants across nearly 200 hospital sites. Its most widely reported finding was that the low-cost steroid dexamethasone reduced deaths in the sickest hospitalized patients, a result that reached the public within months of the study opening.

The I-SPY 2 trial, running since 2010, uses an adaptive platform design to test new therapies for high-risk breast cancer. Several investigational products have "graduated" from the study, meaning the data suggest a high probability of success in a later confirmatory study.

The HEALEY ALS Platform Trial, launched in 2020, applies an adaptive design to amyotrophic lateral sclerosis (ALS), a progressive neurological disease. Multiple investigational products are tested in parallel against a shared placebo group, and new candidates can be added over time. It is one of the most closely watched applications of adaptive design in a rare disease.

The Ebola PALM trial in the Democratic Republic of the Congo, conducted between 2018 and 2020, used an adaptive design that allowed the independent monitoring board to stop the study early once the data crossed a pre-specified threshold showing that certain candidates were more effective than others.

For a broader view of research in rare diseases, see Hope in Research: How Clinical Trials Are Transforming Rare Disease Treatment.

How safety is monitored when a study can change

Every adaptive clinical trial has an independent oversight group, commonly called a Data Safety Monitoring Board or Independent Data Monitoring Committee. This board is made up of clinicians, statisticians, and other experts who are not part of the study team and have no financial stake in the outcome.

The board reviews the accumulating data at the pre-planned interim points. It checks whether the study should continue as planned, be modified according to the pre-approved rules, or stop early for safety, benefit, or futility. In an adaptive clinical trial the board’s role expands, because it is the only group that sees the unblinded comparative data during the study. This separation, sometimes called a firewall, protects the study team from information that could unconsciously bias how the remaining participants are cared for or how the data are collected.

The pre-specified stopping rules are a safeguard, not a warning sign. A study that stops early under its stopping rules is doing exactly what the design intended. It usually means the answer became clear sooner than expected.

For a deeper look at the protections built into every clinical trial, see Clinical Research Basics: What Every Trial Participant Should Understand Before Enrolling.

Questions to ask the research site team about an adaptive trial

A participant considering an adaptive clinical trial can bring a short list of specific questions to the research site team. Useful questions include:

• Is this an adaptive clinical trial, and what kinds of pre-planned changes are allowed?
• Could my assigned group, my dose, or the total study duration change while I am enrolled?
• What are the specific reasons the study could stop early?
• If a change is made to the protocol, how will I be told, and will I need to sign an updated consent form?
• Who is on the independent monitoring board, and how often do they review the data?
• What happens to my study visits and follow-up care if the study ends earlier than planned?
• If I choose to withdraw, how does that affect my usual medical care?

The research site team is the correct source for all of these answers. They hold the protocol, the consent documents, and the site-specific procedures that apply to a participant’s care during the study.

For a broader list of pre-enrollment questions, see Top Questions to Ask Before Joining a Clinical Study.

Finding an adaptive clinical trial that may be right for you

Adaptive clinical trials are becoming more common across many areas of medicine, from oncology and rare disease to infectious disease and neurology. For someone considering trial participation, the design of a study is one factor among several. Location, eligibility criteria, study intervention, visit schedule, and personal health situation also matter.

DecenTrialz uses AI-assisted participant matching and registered nurse-led pre-screening to help people share their information, find studies that may fit, and receive a referral to a research site. The research site team handles study walk-through, final eligibility determination, informed consent, and enrollment decisions.

To explore whether an adaptive clinical trial or any other type of study may be a match, begin the matching process at DecenTrialz.