Why Rare Disease Breakthroughs Require High Fidelity Data

Not all rare diseases were created equal. All told, the broad category of diseases classified as rare – or those affecting fewer than 200,000 people – includes about 7,000 different conditions, which impact a total of roughly 30 million Americans. While the life sciences industry has made enormous strides toward reducing that total in recent years, only 5% of rare diseases currently have an approved treatment option.

This is starting to change. Hundreds of new rare disease treatments have entered the market over the past decade, thanks largely to a combination of government incentives through the Orphan Drug Act, strong urging from patient advocacy groups, advances in gene therapies, and other technological innovations.

But, as the industry continues to plunge deeper into rare disease-focused drug development, clinical development teams find themselves chasing ever narrower segments of the population, which makes patient recruitment a significant challenge for the industry.

Finding the Needle in the Haystack

Take one rare blood disorder, paroxysmal nocturnal hemoglobinuria (PNH), as an example. This acquired disease caused by a mutation in bone marrow cells affects between 1 and 1.5 persons per million people. Adding to the challenge, the wide spectrum of symptoms associated with PNH – which includes fairly common things like abdominal pain, headache, back pain, and fatigue – makes it incredibly tough to diagnose. A typical diagnosis often takes years.

For extremely difficult to isolate conditions like PNH, simply finding a patient, let alone the dozens required for a clinical trial, using traditional methods of patient recruitment and key influencer research can be a needle-in-the-haystack hunt.

Thankfully, the last few years of life sciences industry innovation has also extended to patient data analytics. Today, research and development teams trying to identify candidates for PNH research are able to deploy powerful real-world patient data analytics to find signals and patterns of treatment consistent with a PNH diagnosis – sometimes even before the patient is diagnosed.

The process works by using artificial intelligence and advanced big data analytics to track individual patient encounters with the healthcare system at scale, identifying noteworthy patterns of treatment and red-flag symptoms along the way. In the case of Komodo’s Healthcare Map™, which tracks the complete patient journeys of more than 325 million patients by drawing data from hospital and physician networks, healthcare claims processing companies, pharmacies, laboratories, and health insurers, clinical teams are able to quickly identify the patients and providers who could benefit most from their treatments.

Timing Is Everything

These data-analytic technologies are also being used in situations where finding patients at the right stage in their disease progression is critical. As an example, patients with relapsed and refractory multiple myeloma experience a reduced response to anti-myeloma therapy, triggering the need for an adjustment in drug therapy protocols.

For life sciences companies studying this condition, finding the exact right moment when the relapse or development of refractory disease was happening used to require a proactive update from the treating physician. Now, using advanced patient data analytics that capture timely patient engagement, clinical development teams can receive an automated alert the moment a patient enters this phase in the disease progression.

This ability to dig deep into the patient population to find the needles in the haystack and also track incremental progression of the disease in near real-time creates a more refined, high-fidelity snapshot of the patient population. As a result, life sciences teams are able to build a level of precision into their drug development pipelines for rare disease that was never before possible.

To learn why the top rare disease innovators are turning to Pulse alerts for clinical developments, click here.