The one-trial era: what the FDA’s shift signals for the future of drug development

The landscape of drug development is undergoing a significant transformation as the FDA begins to embrace a more flexible approach to evidence generation. Traditionally, two pivotal trials were required to establish the safety and efficacy of new therapies, but recent changes signal that, in some cases, a single pivotal trial may suffice. This shift reflects a broader evolution in how regulators and sponsors evaluate evidence, emphasizing the integration of clinical data with predictive modeling and mechanistic insights. Masoud Jamei discusses what this shift signals about the future of drug development.

Author: Masoud Jamei, Senior Vice President, Certara Research & Development (PA, USA)

For decades, drug development has followed a familiar path: two adequate, well-controlled pivotal trials to ensure both safety and efficacy. That expectation has gradually shaped how sponsors plan clinical programs, manage risk and ultimately make the case required for regulatory approval.

However, that long-standing rule is now beginning to shift. The US Food and Drug Administration (FDA; MD, USA) recently signaled that, in certain cases, a single pivotal trial may be sufficient to support approval. While that change may sound procedural on the surface, it reflects a broader evolution in how regulators and sponsors think about evidence generation. What we’ll see is that the focus will be less on repeating similar studies, and more on if the overall evidence is strong enough to support a regulatory decision.

In practice, sponsors pursuing a single pivotal trial will still need to demonstrate a robust and convincing understanding of a therapy’s safety, efficacy and clinical behavior. The difference: this confidence may increasingly come from the totality of evidence surrounding a drug, rather than simply repeating similar trials. Approaches like model-integrated evidence (MIE) including physiologically-based pharmacokinetic (PBPK) modeling, are starting to play a critical role in helping sponsors build supporting pivotal evidence and better understand how therapies perform in clinical settings.

From replication to integrated evidence

The reason the FDA required two pivotal trials initially served an important purpose: replication. A requirement that confirmed if a treatment effect is real and reproducible. While that principle still matters, recent advances have evolved the way that is achieved.

Today, regulators look more closely at how different types of evidence intersect, from pharmacology data to predictive modeling outputs to biomarkers. When these elements are aligned, they can reinforce each other and provide a clearer picture of how a therapy is expected to perform in a pivotal trial.

As a result, the way development programs are designed is changing as well. Rather than relying on repetition alone through multiple trials, sponsors are building evidence frameworks that combine clinical data with predictive insights from the very beginning. Adaptive trial designs, including Bayesian statistical approaches – along with the growing use of model-informed drug development (MIDD) and MIE – are part of that shift, connecting findings across studies.

Ultimately, the goal is not simply to reduce the number of trials. It is to ensure that the evidence generated across a development program provides regulators with a comprehensive and scientifically coherent picture of how a therapy works.

Strengthening evidence through model-informed drug development

MIDD is among the tools supporting this transition toward building stronger evidence. At its core, MIDD integrates data from diverse sources – from preclinical findings to clinical trial data – into quantitative models to simulate how drugs behave in the human body.

That means answering questions that would otherwise require additional studies. For instance, how might a therapy perform across different patient populations? What happens when it is taken alongside other medications? How do changes in dosing affect exposure? These are all areas where modeling approaches can add clarity without requiring new trials, helping fill evidence gaps that traditional trials alone may not address.

PBPK modeling is one of the more established approaches in this area. By simulating how a drug is absorbed, distributed, metabolized and eliminated, PBPK models can extend insights well beyond what has been directly observed through clinical studies. When those models are grounded in clinical data, they can provide regulators with greater confidence in how a drug will behave in real-world scenarios that may not have been examined in trials.

In the one-trial era, such tools will become increasingly valuable for building the scientific context that helps regulators interpret and trust the results of a single trial.

Practical examples of model-integrated evidence

In practice, this shift toward integrated evidence is already playing out across development programs, where modeling is being used to complement clinical data and fill gaps that would typically require additional studies. Some recently FDA approved drug labels (e.g. Ibrutinib) do not distinguish between drug-drug interaction evidence generated from clinical studies or PBPK simulations, both equally informing drug labels.

One example is asciminib, a treatment for chronic myeloid leukemia (CML), where researchers used PBPK modeling to evaluate how the drug behaves across a range of clinical scenarios. Rather than running a series of additional individual studies, mechanistic modeling was leveraged to bridge between dosing regimens, evaluate potential drug-drug interactions, and better understand how physiological factors might influence exposure. The FDA ultimately accepted these predictions in place of over ten dedicated clinical pharmacology studies.

Similar approaches are being used more broadly across therapeutic areas. In another case, PBPK modeling was used to demonstrate bioequivalence, eliminating the need to re-run the original, costly clinical study. More broadly, these methods have supported regulatory decisions for over 125 novel drugs, helping answer critical questions around dosing, interactions and variability without requiring additional trials.

Together, these examples showcase how model-integrated evidence approaches can meaningfully expand the information available to regulators. By providing predictive confidence and mechanistic understanding, modeling can strengthen the overall development package that surrounds a therapy.

This type of evidence generation becomes especially relevant in the context of single-trial approvals. If sponsors rely on one pivotal study, the surrounding scientific evidence must help demonstrate that the therapy’s clinical behavior is well understood across populations and clinical scenarios.

Designing development programs for the one-trial era

As regulators become more open to single pivotal trials, sponsors will need to think differently about how they build evidence. That often means designing trials that capture more insights initially, whether through adaptive approaches, more targeted patient selection, or advanced statistical frameworks. At the same time, modeling and simulation approaches provide insights that strengthen the interpretation of clinical results.

Sponsors must also consider how evidence generated in a single pivotal trial will be interpreted across global regulatory agencies. While the FDA’s evolving stance is influential, many development programs ultimately need to satisfy regulators in multiple regions. Bringing clinical data together with predictive modeling can help create a clearer, more complete picture of how a therapy is expected to perform – especially important when that evidence needs to hold up across different regulatory frameworks.

While modeling and simulation are not a replacement for clinical research, they can help round out evidence and ensure regulators have the information they need to evaluate a therapy with full confidence.

Implications for the future of drug development

The FDA’s updated approach points to a broader shift in drug development. It’s no longer just about shortening timelines, but to effectively use data (including Real-World Evidence) that already exists to generate stronger evidence.

For sponsors, success will depend on how effectively they can integrate different sources of evidence. Strong clinical data will always be essential, but it will increasingly be complemented by modeling, mechanistic insights, and more innovative trial designs.

Rather than lowering the bar, the one-trial era reflects a change in how evidence is generated and interpreted. By integrating clinical data with predictive modeling, drug developers can gain a more holistic understanding of how a therapy will behave. Sponsors that take this approach will be better positioned to move efficiently and bring new therapies to patients, while still meeting the level of rigor required for regulatory confidence.

Find more content on the FDA’s updates here.

The opinions expressed in this interview are those of the interviewee and do not necessarily reflect the views of Bioanalysis Zone or Taylor & Francis Group.