Closer to the Patient: How Imaging Data Is Redefining Regulatory Evidence

Introduction 

As drug development moves faster, regulators are asking a simple but increasingly important question: does this evidence actually reflect what’s happening in patients? 
 
Claims and EHR data still provide important context, but they’re often shaped by how care is documented and billed, rather than by what’s happening in the patient. Imaging data is different. It captures disease directly. 
 
As guidance from the U.S. Food and Drug Administration continues to evolve, two themes are becoming clear: regulators want evidence that is closer to the patient—and they want pathways that allow high-quality evidence to move faster. Imaging data sits at the intersection of both. It offers clinical truth, traceability, and increasingly, the opportunity to generate earlier, decision-grade signals without slowing development. 

This shift is especially important in therapeutic areas where disease progression is complex and meaningful outcomes can take years to observe. 

 

1. Why Imaging Endpoints Are Gaining Momentum 

Imaging endpoints address many of the challenges regulators face when evaluating treatment effect: 

• They show disease directly — visualizing and measuring pathology rather than inferring change from secondary signals 

• They capture change over time — making it possible to detect treatment response earlier and more sensitively 

• They can support surrogate endpoints — when imaging measures are validated and shown to reflect meaningful clinical benefit 
  

Where this becomes truly meaningful is at the therapeutic-area level: 

Oncology 
In oncology, imaging often provides the earliest indication of treatment response. Changes in tumor size or activity can often be seen on scans well before differences in survival can be reliably measured, making imaging a long-standing input into regulatory decision-making, particularly in accelerated approval settings.  
 
Cardiology 
In cardiology, imaging helps researchers see how the heart is responding to treatment as it happens. Changes in heart structure or function often appear long before symptoms worsen or clinical events occur.  
 
Neurology 
In neurology, the value of imaging is often even more pronounced. Diseases like Alzheimer’s, multiple sclerosis, and Parkinson’s progress slowly and vary widely from patient to patient, making functional outcomes difficult to measure in the near term. Imaging can reveal structural or pathological changes in the brain long before symptoms noticeably change, offering an earlier window into disease progression and treatment effect.  

Across these areas, imaging doesn’t just add data —
it changes the timing and confidence of regulatory decision-making. 

 
2. What the FDA is Signaling  

Rather than introducing a single new rule, the FDA has been sending consistent signals about how imaging fits into regulatory evidence: 

• Strong imaging practices still matter 
  When imaging is used to support an endpoint, the FDA expects clear standards around how images are collected, reviewed, and managed. 

• Imaging can support surrogate endpoints 
  Radiographic and imaging-based measures may be used as surrogate or intermediate endpoints when they are shown to reflect meaningful clinical benefit. 

• Early, credible signals enable faster pathways 
  Imaging often provides early insight into whether a treatment is working—supporting accelerated approval when follow-up evidence is planned. 

• High-quality real-world data is easier to review 
  Recent FDA actions have made it easier to evaluate de-identified real-world datasets, signaling openness to non-traditional data sources when their origin and relevance are clear. 

• AI increases the focus on data trust 
  As AI is applied to imaging, regulators are placing greater emphasis on understanding where the data comes from, how models are validated, and how outputs can be explained. 
  

Taken together, the message is clear: imaging is welcome —
but only when the evidence is defensible. 

 

3. Imaging vs. Claims and EHR: Proximity to Clinical Truth 

Claims and EHR data offer valuable longitudinal context, but they weren’t designed to serve as endpoints. Claims are built for reimbursement, not disease measurement. EHRs reflect real-world care, but are shaped by local practices, incomplete documentation, and variability across systems. 

Both introduce distance between the data and what’s actually happening in the patient. 

Imaging, by contrast, captures what is happening in the patient — tumor growth, lesion burden, structural change. That proximity to the clinical source matters when regulators are evaluating whether an endpoint truly reflects treatment effect, and whether it can be trusted in surrogate or accelerated approval discussions. 

 

4. The Role of AI  

AI has dramatically expanded what’s possible with imaging, from quantitative feature extraction to pattern recognition and longitudinal disease modeling across populations. 

Importantly, recent FDA guidance suggests a growing openness to AI-enabled tools that accelerate interpretation, summarize findings, or support clinical decision-making—provided clinicians can understand and trust how conclusions are reached. 

But AI doesn’t fix weak data. It magnifies it. 

Regulatory frameworks around AI consistently emphasize validation, transparency, and traceability. If imaging data lacks provenance or consistency, AI-derived outputs become harder — not easier — to defend. High-fidelity imaging isn’t a nice-to-have for AI; it’s the foundation. 
 

5. How Avandra Enables Regulatory-Grade Imaging Evidence 

For imaging to stand up in regulatory conversations, sponsors need more than access — they need confidence in the data itself. 

Avandra is built to support that confidence. By enabling source-level imaging access directly from health systems, Avandra preserves provenance and minimizes unnecessary transformation. A federated model keeps data where it originates, improving traceability and auditability. And expert clinical and scientific curation ensures imaging datasets are understood in context, not just at scale. 

 

Closing Thought 

The FDA isn’t requiring imaging, but its direction is clear: evidence that is objective, traceable, and rooted in real patient biology is easier to trust—and easier to move forward. 

In therapeutic areas like oncology, neurology, and cardiology, imaging often delivers the earliest insight into treatment effect. For sponsors pursuing faster development and modern endpoints, imaging is becoming a cornerstone of regulatory decision-making.