Webinar recap: From insights to impact—The evolving rare disease landscape in 2025 and beyond

Rare diseases may affect only a small percentage of the population individually, but their collective impact is staggering. In the U.S. alone, the economic burden of rare diseases was estimated at $997 billion in 2019—a figure that has only grown as healthcare costs rise. But beyond financial strain, the real cost is measured in diagnostic delays, misdiagnoses, and limited treatment options—challenges that patients and families face every day.

To explore how data, technology, and collaboration are shaping the rare disease landscape, I joined experts from Sanofi and Novartis, for a roundtable discussion: “From Insights to Impact—The Evolving Rare Disease Landscape in 2025 and Beyond.”

If you missed the session, here are the top takeaways on how AI, real-world data, and cross-functional collaboration are revolutionizing rare disease diagnosis and treatment.

The right data supports better outcomes

Rare disease patients often fall through the cracks of traditional healthcare systems. Even geneticists and specialists—those trained to recognize rare conditions—may only see a handful of cases in their careers. This lack of exposure fuels the seven-year average diagnostic delay many rare disease patients endure. More staggering is the fact that less than 5% of the estimated 7,000-plus rare diseases have an effective treatment.

Raising awareness about new discoveries in the rare disease space largely falls on biopharma companies, which shoulder the responsibility of educating healthcare professionals (HCPs) on the latest advancements. However, even under the best conditions, this is an uphill battle.

HCPs are already overwhelmed with high patient volumes, administrative tasks, and ever-evolving medical advancements, making it difficult to prioritize learning about rare diseases—which often make up only a small fraction of their daily practice. As a result, many physicians may be reluctant to allocate their limited time to meetings with medical science liaisons (MSLs) or industry experts, even when these discussions could lead to earlier diagnoses and improved patient outcomes.

In fact, HCPs are probably seeing more rare disease patients than they might imagine. An estimated 50% of rare disease patients are likely undiagnosed today, in part due to key indicators of their conditions—genomic markers, lab results, and symptom patterns—are buried in unstructured data or siloed in disparate systems. More widely available structured data assets do not holistically represent specific therapy areas or capture meaningful patient behavior.

Advanced data analytics tools like artificial intelligence (AI) and natural language processing (NLP) built on large language models can help developers and providers understand the rare disease landscape and identify patients more effectively using an assortment of data types.

Artificial intelligence (AI) and natural language processing (NLP) are changing this reality in small ways but could be scaled to go through physician notes and medical records to uncover disease signals long before a formal diagnosis is made.

AI-driven predictive analytics can flag high-risk patients’ years earlier, transforming rare disease care from a reactive to a proactive model. In addition, multi-source data fusion, which combines claims data, genomic insights, and real-world evidence (RWE), can enable a more comprehensive view of patient journeys. The future of rare disease detection isn’t just about collecting more data—it’s about extracting deeper meaning from the data we already have.

With this additional context, providers and biopharma companies can close the awareness gap, accelerate diagnosis, and design more targeted treatments that lead to better outcomes.

Cross-functional collaboration is critical

Data alone isn’t enough—it needs to be activated through collaboration. Rare disease care spans a complex ecosystem that includes healthcare providers, biopharma developers, academic institutions, registries, and advocacy groups. While these stakeholders share a common goal, their insights often remain siloed, leading to misdiagnoses, redundant research, and slow clinical trial recruitment. This fragmentation prevents the healthcare system from recognizing rare disease patients in real time.

Achieving a fuller understanding of the rare disease landscape requires providers and developers to work closely together. Providers have unmatched visibility into patients’ needs and the fundamentals of care, while developers tend to have a better view of high-level trends and emerging technologies.

But even in collaboration, these two groups are likely missing valuable insights. To build the most comprehensive datasets, those operating in the rare disease space should partner with organizations that offer unique perspectives: Specialty registries, patient advocacy groups, academic institutions, and other sources of truth can shed light on otherwise hidden aspects of the patient journey.

Biopharma companies can use data and analytics not only to guide development and understand their competitive landscape, but also to more effectively target, engage, and collaborate with HCPs.

Using predictive analytics on consumer data within a provider’s market, for instance, biopharma developers can identify unmet needs among potential patient populations and shape their messaging accordingly. They could also leverage data to help providers spot clusters of symptoms in patients they might otherwise not be looking for, or fill openings in clinical trials.

Bridging this gap requires AI-driven engagement strategies that allow biopharma to identify which healthcare providers (HCPs) are most likely seeing undiagnosed rare disease patients. Predictive analytics can analyze prescribing patterns, referral networks, and geographic disease prevalence to help field teams prioritize physician outreach. This targeted approach ensures that HCPs receive the most relevant insights on rare diseases, improving patient identification and early intervention.

Data-driven collaboration also extends to clinical trial optimization. AI can analyze real-world patient data to match the right patients with clinical trials faster, significantly reducing the time and costs associated with recruitment. Meanwhile, federated learning and blockchain-based data sharing can help to securely connect patient data across hospitals, trial sites, and advocacy groups, ensuring comprehensive datasets without compromising patient privacy.

Decentralized clinical trials, sales data, key opinion leader insights, and consumer behaviors can all be mined and combined to create powerful intelligence for providers and developers alike. Wherever barriers are present—between marketing and medical affairs, or sales and clinical teams, for instance—they should be taken down.

Technology is brightening the future of rare disease in biopharma

Biopharma’s future in the rare disease space will be defined by the industry’s ability to understand and adapt to a rapidly changing healthcare landscape. Technology—and especially data and analytical tools—allows biopharma organizations to navigate this landscape more effectively.

Our roundtable panelists identified a few of the technologies they think will have the biggest impact on the rare disease space:

• One of the most promising ideas is blockchain technology, which could significantly improve the security and validity of patient datasets by securely connecting disparate data points across multiple institutions while maintaining patient ownership of their health records. This decentralized approach ensures that critical rare disease data is accessible across the healthcare ecosystem, while reducing duplication and enhancing data integrity.
• Another game-changer is generative AI and large language models (LLMs), which are revolutionizing small dataset enrichment. By integrating structured and unstructured data sources—such as clinical notes, imaging data, and lab results—AI can fill in missing attributes, improving the accuracy of predictive models. This means that even in cases where there are only a few hundred documented patients worldwide, AI can extract meaningful insights to guide diagnosis and treatment.
• Beyond AI, multi-omic data fusion can unlock entirely new pathways for rare disease research. By integrating genomic, proteomic, and metabolomic data with traditional claims and EHR records, researchers can uncover previously undocumented disease mechanisms. This approach not only helps in identifying new biomarkers but also accelerates the development of personalized medicine tailored to rare disease patients.

These examples aren’t futuristic fantasies or concepts; they’re all extant technologies that are on the brink of broader adoption. But you don’t even need to wait that long to see how powerful the right data and analytical tools can be for your biopharma or provider organization.

The rare disease space is on the cusp of an AI-driven transformation. The organizations that embrace advanced data analytics, AI-powered collaboration, and multi-source data fusion today will lead the industry tomorrow. This is an opportunity to not only improve patient outcomes but also to redefine how we approach rare disease diagnosis, treatment, and research.