Rare diseases collectively affect hundreds of millions of people worldwide, yet for life sciences organizations, they remain among the most complex areas for research and drug development. At the center of this challenge is a persistent gap: the inability to translate fragmented clinical and patient data into usable insights.
During a recent webinar, leaders across patient advocacy, translational research, and clinical informatics explored how improving data quality and interoperability can fundamentally reshape rare disease diagnosis – and, critically, accelerate therapeutic innovation.
Click below to watch the full webinar on-demand or keep scrolling for six key takeaways.
The diagnostic odyssey is a data problem with R&D consequences
For many patients, the path to diagnosis spans years, directly impacting clinical trial readiness, cohort identification, and time-to-market for new therapies.
Charlene Son Rigby, CEO of Global Genes, shared a firsthand account of this challenge through her daughter’s journey:
This type of fragmented journey results in missed opportunities to identify patients earlier, delaying enrollment in trials and limiting the availability of high-quality longitudinal profiles.
Even when a genetic cause is identified, variability in disease presentation complicates classification.
As Rigby said, “Juno’s presentation was not typical… and so, this kind of heterogeneous presentation for Juno extended the period of time for her diagnosis.”
Why rare disease data remains underutilized
Despite the rapid growth of healthcare data, rare diseases remain difficult to “see” within clinical systems. The issue is not volume – but structure.
Melissa Haendel, PhD, FACMI, Director of Precision Health and Translational Informatics at UNC Chapel Hill, highlighted a core limitation:
Efforts like the Mondo disease ontology aim to address this by standardizing disease definitions and linking disparate datasets. For R&D teams, this creates a more reliable foundation for cohort identification, real-world evidence generation, and target validation.
AI in rare disease depends on precision
AI is increasingly central to rare disease research – from identifying undiagnosed patients to accelerating drug discovery. However, its effectiveness depends entirely on the quality and structure of underlying data.
As Amol Bhalla, MD, M.SCI, MHSA, MBA, Chief Informaticist at IMO Health, noted:
For life sciences applications, this underscores a critical requirement: AI must be grounded in standardized, clinically meaningful data models.
When supported by robust terminologies and knowledge graphs, AI can:
- Improve rare disease patient identification within large datasets
- Enhance clinical trial matching and recruitment
- Capture nuanced clinical concepts for precise disease tracking
- Accelerate drug repurposing with NLP-derived insights
Without those guardrails, AI is more likely to produce unreliable or misleading results.
Real-world data as a strategic asset in rare disease research
Patient registries and real-world data (RWD) are becoming essential components of rare disease research strategies. Platforms like Global Genes’ RareX are helping capture structured, patient-reported data that complements clinical records.
For life sciences teams, integrating RWD offers several advantages, including richer natural history data for trial design and expanded access to ultra-rare patient populations.
Equally important is patient motivation to contribute data – not only for personal benefit, but for future therapies:
This willingness creates a powerful opportunity for sponsors to build more comprehensive, patient-centered datasets.
What must change to accelerate innovation
To accelerate progress in rare disease drug development, we must ensure:
1. Standardized disease representation
Consistent coding and terminology are essential for identifying patients, measuring prevalence, and scaling research.
2. Integrated data ecosystems
Linking clinical, genomic, and patient-reported data enables a more complete understanding of disease.
3. AI grounded in trusted data frameworks
High-quality inputs are necessary to generate reliable, regulatory-grade insights.
4. Policy evolution
Modernized data-sharing frameworks are critical to enabling cross-sector collaboration.
5. Earlier identification of rare disease signals
Embedding decision support into clinical workflows can reduce time to diagnosis and expand trial-ready populations.
Turning fragmented rare disease data into discovery
For life sciences organizations, the rare disease challenge is just as much infrastructural as it is scientific. The ability to connect, standardize, and operationalize data will determine how quickly new therapies reach patients.
By closing the gap between fragmented data and usable evidence, the industry can move beyond the diagnostic odyssey – and toward a future where discovery happens faster, trials are more efficient, and treatments reach patients sooner.
