2026 Archives

FDA’s New Default: One Pivotal Trial for Drug Approval

A Paradigm Shift Sparking Optimism and Questions

In February 2026, the U.S. Food and Drug Administration (FDA) announced a landmark policy change that one adequate and well‑controlled pivotal trial, supplemented by confirmatory evidence, will now serve as the default basis for drug approval. This decisive shift — articulated by FDA Commissioner Marty Makary and CBER director Vinay Prasad in The New England Journal of Medicine — effectively ends a decades‑long “two‑trial dogma” and reframes the evidentiary foundation of U.S. drug regulation.

“Going forward, the FDA’s default position is that one adequate and well-controlled study, combined with confirmatory evidence, will serve as the basis of marketing authorization of novel products. The FDA will carefully examine all aspects of study design with particular focus on controls, end points, effect size, and statistical protocols.”1

It is important to remember that it has always been possible to obtain a marketing authorization on the basis of a single adequate and well-controlled study in combination with confirmative evidence, but typically this approach was mainly applied in breakthrough program designation, accelerated approval, and priority review pathways.

Why the FDA is moving away from two trials

Makary and Prasad argue that requiring two trials made sense when biology was poorly understood and therapeutics were often blunt instruments rather than targeted molecular tools. In today’s world, duplicative trials may be unnecessarily costly, slow, and redundant.

The original argument for two clinical trials is a statistical one: If a substance does not have any efficacy, then the chances of showing an effect in two studies are much lower than showing it in only one study. The article quantifies this chance as 0.06% instead of 2.5%, assuming that the test is performed at the typically applied one-sided 2.5% level (but that calculation assumes that the two studies are independent of each other, which is not necessarily the case).

The more important argument is that modern drug development provides much more clarity on a precise mechanism of action, assessed by biomarkers as well as a variety of endpoints, thus supporting statistical with biologic inference.

They emphasize several points:

Modern science provides multiple layers of corroboration

Mechanistic data, class‑effect consistency, real‑world evidence, and surrogate endpoints can complement a single pivotal study.

Two trials don’t guarantee correctness

Even under the two‑trial regime, the FDA has approved drugs later found ineffective or unsafe — not because of too few trials, but because trial design quality matters more than quantity.

Lowering trial count may reduce costs and time

One pivotal trial can cost $30–150M and takes years to complete. Reducing this burden may spur innovation and could reduce price‑justification arguments tied to Research & Development investment.

Focus on trial design and analysis

The article clearly articulates the importance of various aspects of trial design to support the credibility of trial results, including the use of a contemporary control group, pre-specification of a hypothesis, choice of a primary endpoint, statistical power, randomization, and blinding. These are key statistical aspects documented in the ICH E9 guidance on Statistical Principles for Clinical Trials, and as such they have been underlying drug development for almost 30 years. What is new and encouraging is that the article specifically states that these can also be provided by a Bayesian framework, referencing the recently published draft FDA guidance on this topic, and described by Cytel’s Savina Jaeger..

Unclear implications for global drug development

For most companies, drug development is a global business, and as such it’s unclear whether this change in FDA policy will affect the expectations from regulatory authorities in other regions and countries. Will they follow suit or maintain their current expectations? Cytel’s Strategic Consulting group will be monitoring this closely as this will have a fundamental impact on designing trials for global approvals.

It is also uncommon for the FDA to announce a major change in policy through a publication, so we will also monitor FDA’s official channels for further announcements on this topic in the future.

Final takeaway: A defining regulatory moment

The FDA’s new one‑trial default represents a significant policy shift in U.S. drug regulation. It aligns with trends in precision medicine, leverages mechanistic and statistical advances, and may unlock faster access and lower development burdens. Yet it also raises profound questions about evidence standards, risk tolerance, and the balance between speed and certainty. Most importantly, though, it reinforces the importance of solid statistical principles underlying credible drug development, with a clear statement that both Bayesian and frequentist approaches can provide them.

Enhancing Pharmacovigilance: Leveraging Generative AI to Transform Patient Safety Narratives

Rethinking clinical documentation with generative AI

Generative artificial intelligence (AI) is rapidly reshaping the landscape of clinical documentation. Traditionally, writing patient safety narratives (PSNs) for Clinical Study Reports (CSRs) has required hours of manual data extraction and synthesis — a time-consuming process that slows pharmacovigilance workflows.

New advances in large language models (LLMs), such as Google Gemini, are demonstrating how AI can generate coherent, accurate narratives from structured clinical data. By doing so, these models promise to improve both speed and consistency while maintaining compliance with International Council for Harmonization (ICH) standards.

Study overview: Automating PSNs with a RAG framework

In our recent study, we explored how a retrieval-augmented generation (RAG) system could automate PSN drafting for semaglutide-related adverse events. The system merged structured case data with adaptive AI prompting techniques — specifically, Automatic Prompt Engineering (APE) — to optimize the factual accuracy of the generated narratives.

Using an ICH E3–aligned template, the model generated PSNs across four key sections:

Patient Demographics and Study Information
Relevant Medical History
Adverse Event (AE) Details
Laboratory and Diagnostic Findings

Thirty published case reports were analyzed to assess how well the AI performed in extracting, contextualizing, and summarizing information.

Measuring quality and efficiency

Each AI-generated narrative was evaluated by clinical documentation experts on a 1–10 scale across multiple criteria — including completeness, clarity, and accuracy. Evaluation metrics focused on core demographic details, drug administration data, adverse event description, and diagnostic relevance.

The average processing time per case was approximately 10 seconds, compared to the several hours typically required for manual PSN drafting. This represents a remarkable productivity gain for pharmacovigilance teams.

Key results

The AI-generated narratives achieved an average composite score of 7.5/10 for narrative quality.

Highest-performing areas included:
- Accuracy of AE/SAE Identification (9.8/10)
- Relevance of Key Findings (9.8/10)
- Disease/Treatment Context Accuracy (9.4/10)
- Extraction of Prior Medications (9.0/10)

These results underscore the model’s strength in synthesizing clinical information into concise, ICH-compliant summaries.

However, the patient demographics section scored lower (6.4–7.0), mainly due to missing temporal details or incomplete demographic data. These gaps reflected the model’s sensitivity to inconsistencies in source reports — a known challenge in real-world data processing.

Discussion: The balance between automation and oversight

Our findings reveal that integrating generative AI within a structured RAG framework can significantly accelerate PSN drafting without compromising clinical accuracy. The approach supports a hybrid workflow in which AI handles repetitive data synthesis, while human reviewers focus on interpretation, validation, and scientific review.

Still, the study also highlights that expert oversight remains essential. Variability across cases — especially when data formats or terminology differ — underscores the importance of human supervision to ensure contextual completeness and regulatory compliance.

The road ahead

Future research will refine prompt design through adaptive APE techniques to improve temporal and contextual accuracy. Expanding the framework across multiple therapeutic areas and languages will be key to scaling adoption in global regulatory environments.

By combining AI-driven generation with expert validation, pharmacovigilance teams can achieve the best of both worlds: faster, more accurate, and more standardized safety documentation.

Key takeaways

AI tools — when integrated with structured RAG systems — hold enormous promise for the future of pharmacovigilance. They can dramatically reduce drafting time, enhance consistency, and allow safety experts to focus where it matters most: interpreting data and protecting patients.