FDA's Real-Time Clinical Trials: The Shift From Data Submission to Signal Readiness

On April 28, 2026, the FDA did more than announce two real-time clinical trial proof-of-concept studies. It raised a harder question for clinical development leaders: can the industry produce regulator-ready safety and efficacy signals before database lock — without sacrificing the rigor, governance, and statistical discipline that make clinical trials credible?

That is the real significance of real-time clinical trials.

The announcement itself is deliberately limited. FDA identified two oncology proof-of-concept trials, AstraZeneca's TRAVERSE study in treatment-naïve mantle cell lymphoma and Amgen's STREAM-SCLC study in limited-stage small cell lung carcinoma, and issued a Request for Information for a broader pilot program. The operational ambition, however, is larger than the pilot. FDA is testing whether clinical development can move from episodic data submission toward governed, auditable, near-real-time signal exchange, with Paradigm Health publicly identified as a platform partner supporting real-time signal sharing in the proof-of-concept collaboration.

For those of us working at the intersection of risk-based quality management, ICH E6(R3), ICH M11, AI-enabled workflows, and clinical development operating models, RTCT does not feel like a sudden departure. It feels like the next pressure test of whether clinical trial modernization is real or still mostly language.

Real-time clinical trials should not be interpreted as a shortcut around evidence generation. They are better understood as a shift in visibility. The question is no longer only whether a sponsor can collect, clean, analyze, and submit data at the end of a study. The question is whether the sponsor, CRO, technology partners, and regulator can operate from pre-specified signals that are traceable, auditable, interpretable, and governed while the study is still running.

The Safety Question: Real-Time Visibility Is Not the Same as Premature Action

It is fair to ask whether real-time review introduces new safety risk. The concern is understandable. If more people can see partial data earlier, could that lead to premature interpretation, unnecessary intervention, or pressure to make decisions before the evidence is mature?

That is the right question, but I would frame the risk differently.

Today's clinical trial system has safety latency built into it. A signal may emerge at a site, move through source documentation, monitoring, EDC entry, query resolution, safety review, medical monitoring, and periodic reporting before it becomes visible in a broader regulatory context. That process is familiar, but it is not always fast.

Real-time clinical trials have the potential to shorten that latency. That is not inherently a safety cost. Properly governed, it can be a safety gain.

What RTCT does not change is the safety architecture around a clinical trial. The investigator, medical monitor, sponsor pharmacovigilance function, IRB, DSMB or DMC where applicable, and established escalation pathways remain central. The regulator is not replacing bedside medical judgment or sponsor safety oversight. The intended model is visibility into pre-agreed signals and endpoints within a structured framework.

The real risk is not real-time visibility itself. The risk is acting too quickly on incomplete or poorly contextualized information. That is why pre-specified decision rules, statistical discipline, audit trails, role clarity, and governance are essential. RTCT can improve safety oversight only if the speed of signal visibility is matched by rigor in interpretation.

Streaming the Data, Not Changing the Evidentiary Standard

A second concern is whether RTCT changes what it means to run a clinical trial. If FDA can see accumulating signals from individual patients, has the regulator moved toward decision-making based on n=1?

The answer should be no.

Clinical trials are still designed to evaluate safety and efficacy across groups of patients using pre-specified endpoints, eligibility criteria, statistical methods, and analysis plans. RTCT does not change the evidentiary standard required to interpret whether a therapy is safe, effective, approvable, or developmentally viable.

The better way to think about RTCT is this: it changes when regulators can see accumulating information, not the scientific standard needed to interpret that information.

That distinction matters. Individual-level data can raise a signal. Aggregated data, pre-specified analyses, medical judgment, and statistical context determine what that signal means. A serious adverse event, an unexpected pattern of toxicity, or a dramatic response in an early-phase oncology study may influence dose escalation, cohort expansion, biomarker strategy, or regulatory dialogue. That is not new. What is new is the possibility that those signals can enter a structured regulatory review pathway more rapidly.

The statistical logic of clinical development remains intact. Adequately powered studies, pre-specified endpoints, estimands, statistical analysis plans, and complete datasets still matter. RTCT does not make weak evidence stronger. It makes relevant signals visible earlier.

Adding a Layer, Not Replacing Standards

For sponsors thinking about submission strategy, one point deserves clarity: RTCT does not replace existing regulatory data and submission standards.

Structured clinical trial datasets, validated analyses, audit trails, regulatory submissions, and the complete evidence package remain essential. CDISC SDTM and ADaM, eCTD, ICH E6(R3), and ICH M11 remain part of the broader foundation for regulatory-quality clinical development.

What appears likely is an additional technical and governance layer for real-time signal sharing. That layer may include:

• Common data models and API specifications between sponsor, platform, CRO, and FDA systems
• Pre-agreed safety and efficacy signal definitions, with threshold rules for escalation
• Data validation expectations, access controls, and role-based visibility
• Audit trails and data provenance across the signal pipeline
• Procedures for documenting review, interpretation, and action on preliminary or accumulating data

In other words, RTCT does not eliminate the need for traditional submission standards. It adds a new question: how should preliminary or accumulating signals be transmitted, interpreted, documented, and governed before the final locked dataset is available?

That is one of the most important issues the pilot should clarify. The field will need a common understanding of what constitutes a meaningful signal, how preliminary data should be handled, how over-interpretation should be avoided, and how real-time review fits into formal regulatory decision-making.

This is also where ICH M11 becomes relevant. Structured protocols are not just cleaner documents. They create the foundation for more consistent downstream interpretation of objectives, endpoints, schedules, eligibility criteria, statistical concepts, and operational requirements. If real-time signal sharing is to scale, protocol structure, data architecture, and quality governance cannot remain disconnected workstreams.

The Workload Question: More Upfront, Less Downstream

For clinical development organizations, RTCT will not be operationally neutral. It will shift work earlier in the study lifecycle.

The upfront work should not be underestimated. Clinical development organizations would need to define which signals are being shared, map those signals to source data and EDC data, validate the data pipeline, establish governance for review and response, and ensure that audit trails, cybersecurity, privacy, and documentation standards are in place.

For organizations that have already invested in risk-based quality management, centralized monitoring, key risk indicators, quality tolerance limits, integrated data review, and cross-functional safety governance, some of that infrastructure may already exist. For organizations still operating under more traditional monitoring models, RTCT would represent a more significant operational change.

Where the model may reduce burden is later in the development cycle.

Better real-time visibility could reduce last-minute data reconciliation, delayed signal detection, avoidable protocol amendments, repeated regulatory information requests, and long gaps between study completion and regulatory discussion. It may also reduce the need to reconstruct decision logic months after the fact, because signal definitions, thresholds, and review pathways would have to be defined prospectively.

The important point is that RTCT does not simply add a dashboard to an existing trial. It changes the operating model. The trial must be designed from the beginning to support signal readiness.

Escalation, Not Surveillance

A natural worry is whether FDA can monitor dozens or hundreds of clinical trials in real time. No agency could reasonably operate that way, and that should not be the model.

The better model is escalation, not surveillance.

Real-time data streams should not require reviewers to watch dashboards continuously. Instead, pre-defined signals and thresholds should trigger human attention when review may be needed. That is similar in principle to how RBQM is intended to work at the sponsor level. The system should be designed so that experts are alerted to meaningful issues, not overwhelmed by continuous noise.

The technical challenge is not whether humans can watch more screens. The challenge is whether the field can design signal thresholds that are sensitive enough to detect meaningful issues, specific enough to avoid excessive noise, and transparent enough to support regulatory trust.

AI-assisted triage will likely play a role, particularly in identifying patterns, prioritizing review, and distinguishing noise from meaningful signals. Human judgment, however, remains essential. Real-time clinical trial data creates obligations around cybersecurity, audit trails, secure data environments, data provenance, reviewer training, and disciplined interpretation of pre-lock data.

That is why the pilot approach is appropriate. Starting with a limited number of trials, defined signals, and controlled technical pathways allows FDA, sponsors, CROs, and technology partners to learn what works before the model is scaled more broadly.

Why This Matters for Clinical Development Organizations

The practical implications for clinical development organizations are significant.

1. Data infrastructure becomes a strategic capability

Real-time signal sharing assumes that study data can be captured, validated, mapped, and interpreted continuously. Organizations that depend on retrospective EDC locks, manual data review cycles, disconnected vendor systems, and fragmented monitoring processes are not well positioned for this model.

The infrastructure that enables RTCT is also the infrastructure that enables credible RBQM, AI-enabled operations, structured protocol execution, and faster regulatory dialogue. These are converging requirements, not separate modernization initiatives.

2. Signal pre-specification becomes a regulatory discipline

RTCT requires more than deciding what data to collect. It requires defining what signals matter, how they will be measured, what thresholds will trigger review, how missing or incomplete data will be handled, and who is responsible for interpretation.

That discipline draws on clinical development, biostatistics, data management, pharmacovigilance, medical monitoring, regulatory affairs, quality, and technology. It cannot be assembled at the last minute.

3. Sponsor-partner integration becomes more important

Traditional outsourcing models may struggle in a real-time environment. If a delivery partner is executing tasks on a periodic deliverable schedule while the sponsor, platform provider, and regulator are operating from continuous signal visibility, governance gaps will emerge quickly.

RTCT assumes a more integrated partnership model. Data, safety, quality, medical, statistical, and regulatory teams must operate from common definitions and shared escalation pathways. Sponsors will need to evaluate which partners can operate with integrated governance, modern data infrastructure, and a strategic orientation — rather than primarily as transactional service providers.

4. This is not only a large-pharma story

The first proof-of-concept studies involve large pharmaceutical companies, but the broader question is not limited to large pharma. Emerging pharma and biotech companies may actually have an opportunity if they build the right operating model early.

Emerging pharma and biotech companies often have fewer legacy systems to unwind. With the right CRO, technology platform, and regulatory strategy, they may be able to design signal-ready trials more efficiently than larger organizations burdened by complex internal infrastructure.

The equity issue, however, is real. If RTCT depends on expensive bespoke platforms, it could widen the gap between large and small sponsors. A scalable model should include proportional requirements, CRO-enabled implementation pathways, and technology solutions that are accessible to emerging sponsors.

A Practical Readiness Test for Clinical Development Leaders

For clinical development teams, the immediate question is not whether every trial should become a real-time clinical trial. Most should not, at least not in the near term.

The better question is whether the organization has the capabilities that RTCT presumes.

Use the following questions as a practical internal readiness test.

These five questions are intended as a quick gut check. They translate into a more formal governance and operating-model self-assessment in the companion scorecard below.

RTCT Is a Capability Roadmap, Not an Immediate Mandate

The April 2026 announcement does not require every sponsor to redesign every clinical trial. Most programs will continue to run under conventional models in the near term. The cost-benefit calculation will not support real-time approaches for every indication, phase, endpoint, or sponsor.

But the direction of travel is clear.

Clinical development is moving toward more structured protocols, more continuous data visibility, more AI-enabled review, more risk-based oversight, and more integrated sponsor-CRO operating models. RTCT sits at the intersection of all these trends.

That is why the announcement matters beyond the two proof-of-concept trials.

The organizations that are best positioned for this future will not simply be those with the most technology. They will be the organizations that can connect protocol design, endpoint strategy, data standards, safety governance, statistical discipline, operational execution, and regulatory dialogue into one coherent operating model.

The risk for the industry is not that real-time clinical trials move too fast. The risk is that the technology moves faster than the governance needed to use it responsibly.

Closing

Real-time clinical trials are not a departure from the scientific method. They are an acceleration of the feedback loop that clinical development has been building toward through RBQM, structured protocols, ICH E6(R3), ICH M11, and data-driven oversight.

The pilot approach is the right one. The real test is whether standards, access, governance, and operational discipline mature as quickly as the technology.

For clinical development leaders, the immediate action is not to chase the latest pilot. It is to ask a more practical question: if a regulator asked tomorrow for governed, auditable, near-real-time visibility into pre-specified trial signals, would our organization be ready?

For many organizations, the honest answer will be: not yet.

That gap — between regulatory ambition and operational readiness — is not a failing. It is the work.

That is exactly why this conversation matters now.

Further reading on kushdhody.com

• The RTCT Readiness Scorecard (May 2026)
• ICH M11 & E6(R3) Protocol Readiness Scorecard (April 2026)
• Implementing ICH M11: A CRO Leadership Perspective on Protocol Standardization and AI Integration (March 2026)
• AI-Enabled Clinical Research Operations: From Concept to Implementation (June 2025)
• The Evolving CRO-Sponsor Partnership Model: Strategic Value Beyond Outsourcing (July 2024)
• Risk-Based Quality Management in a Post-Pandemic CRO Environment (September 2023)

Selected sources

• FDA — "FDA Announces Major Steps to Implement Real-Time Clinical Trials" (April 2026)
• Federal Register — AI-Enabled Optimization of Early-Phase Clinical Trials Pilot Program (Docket FDA-2026-N-4390, April 2026)
• Paradigm Health — FDA collaboration on regulatory review of clinical trial data (April 2026)
• ICH — E6(R3) Good Clinical Practice Step 4 final guideline (PDF, Jan 2025)
• ICH — M11 Clinical Electronic Structured Harmonised Protocol Step 4 guideline (PDF, Nov 2025)
• ICH — M11 Clinical Electronic Structured Harmonised Protocol Step 4 Technical Specification (PDF, Nov 2025)