The Rare Disease Challenge
Let me start with a fundamental problem that has plagued rare disease research for decades: finding enough patients. If you’re studying diabetes, you won’t have any issues finding enough patients for a clinical trial. But in rare disease? You might have to search all over the world to find just a few patients and may never find enough for a robust trial. This reality has historically made traditional randomized placebo-controlled trials incredibly difficult, sometimes impossible, to conduct in rare diseases.
This is where RWD becomes a game-changer.
What RWD Actually Means
There’s often confusion about what we mean by RWD, so let me clarify: it’s medical data collected during the routine provision of care, not as part of a specific investigation or clinical trial. This includes electronic health record data (EHRs) generated at doctor visits, insurance claims generated for reimbursement, and registries maintained by public health agencies. This data is being captured every second of every day, all over the world, and we’re now learning how to harness it to accelerate medical breakthroughs and generate reliable evidence.
Transforming the Drug Development Lifecycle
What excites me most is how RWD can support rare disease research across the entire drug development lifecycle. Early in development, it helps us understand the burden of illness and treatment patterns for patients we’re trying to help. This is particularly crucial in rare disease, where gathering even basic information about patient experiences has historically been enormously challenging.
RWD also revolutionizes clinical trial design. We can use it to optimize inclusion and exclusion criteria, identify important clinical outcomes to measure, find which medical centers actually have the patients we’re looking for, and even understand whether adverse events during trials are expected or unexpected within the specific patient population. I’ve seen trials pause unnecessarily because investigators couldn’t quickly determine if a few cases of anemia were within normal background rate of the disease or a cause for concern. RWD can answer that question in days instead of weeks or months.
RWD also enables external control arms (ECAs) for single-arm trials, which are very common in rare disease. Instead of requiring a placebo group, we can compare treatment outcomes to what patients typically experience in their natural history. This approach is not only more ethical in serious rare diseases; it’s often the only feasible path forward.
Real Examples, Real Impact
In TriNetX’s recent CSO Perspectives: Research Impact Report — Unlocking Rare Disease Insights, we highlighted several case studies that demonstrate this potential. Work on paroxysmal nocturnal hemoglobinuria (PNH) included studies with data on 150 and 1,200 patients from the TriNetX global network of over 250 healthcare organizations. These are possibly the largest studies in PNH ever conducted. With TriNetX, researchers were able to accomplish in months what would have taken years in the past.
Similarly, a study on eosinophilic granulomatosis with polyangiitis (EGPA) utilizing TriNetX data examined treatment patterns and patient outcomes in a population where no single clinician sees enough patients to have a comprehensive view. These insights are invaluable for identifying care gaps and designing interventions that address real patient needs.
The Team Sport Approach
I want to be clear: working with RWD is not for the faint of heart. The data are complex, and they weren’t collected for research purposes. You need a team that includes RWD experts who understand why data were collected, epidemiologists who understand research methods, biostatisticians, and clinicians who can explain clinical documentation practices and standards of care within the target patient population.
RWD is sometimes scary because there are data quality issues is all RWD data sources – there will be someone weighing 10,000 pounds or a temperature of 1,000 degrees or a woman with prostate cancer. Although these data quality issues are in all datasets, they are expected and exceedingly rare. But with the right expertise, you learn to identify and handle these anomalies appropriately and distinguish between expected errors and true data quality issues. The key is understanding the data as they are collected and matching your research question to the right data source and methodology for the intended use.
The Regulatory Frontier
One area where I believe we need to push harder as a community is regulatory acceptance of RWD for approval decisions. Between 2017 and 2024, regulatory agencies worldwide have issued about 50 guidance documents on the use of RWD, demonstrating clear interest. Yet the actual acceptance of RWD in regulatory submissions remains limited.
I understand the hesitation. Regulators are accustomed to clinical trial data that’s meticulously documented and collected specifically for research. RWD comes with uncertainty, and that’s scary when you’re making decisions that affect patient safety and treatment decisions. But we’ve been using RWD for over 50 years. We understand the methodology and the limitations. For rare diseases, especially serious life-threatening conditions, the benefit-risk calculus should support broader acceptance.
The AI Revolution Ahead
Looking forward, I’m particularly excited about the potential of artificial intelligence (AI) to unlock information in clinical notes. While we’ve become quite good at using structured data like diagnoses and lab results, vast amounts of valuable information remain trapped in free-text clinical notes. AI will help us extract and validate these insights, potentially helping us identify rare disease patients more quickly and accelerate the entire research cycle.
However, we must be careful. Extracting information from notes is relatively easy; knowing whether to trust that information is much harder. We’re still years away from having truly trustworthy, analysis-ready facts extracted from clinical notes at scale.
The Path Forward
The RWD community is still relatively small but growing. Expanding this community (i.e., bringing more stakeholders into the fold who understand and trust these methods) will speed up research and ultimately improve patient care. For rare disease patients who have waited too long for treatments, this acceleration could be life-changing.
Ready to dive deeper? Listen to the full RARECast podcast episode for my conversation with Danny Levine, complete with insights into how RWD is reshaping rare disease research, including detailed discussions of methodology, regulatory considerations, and future innovations.
Want to see RWD in action? Download TriNetX’s CSO Perspectives: Research Impact Report — Unlocking Rare Disease Insights to explore detailed case studies and learn how this approach is generating new knowledge and accelerating drug development for patients who need it most.
About Jeffrey Brown, PhD
With more than 25 years of experience in research and consulting, Jeff is an internationally recognized expert in the use of RWD to support the evidentiary needs of regulatory agencies and medical product sponsors and an expert in the assessment of data quality of RWD resources.