Five experts in the rare disease and data science fields shared their experiences and thoughts on how advancements in data science can advance the treatment of rare diseases.

David Scheer, president of Scheer & Company, moderated the panel discussion, which was part of the National Organization for Rare Disorders (NORD) Breakthrough Summit 2021.

Presenters included Wade Schulz, MD, PhD, assistant professor of laboratory medicine and computational health care researcher, Yale Medical School, New Haven, Connecticut; Martin Mackay, PhD, chief executive officer, Rallybio; Monica Coenraads, chief executive officer, Rett Syndrome Research Trust (RSRT); and Rachel Soloman, chief data officer, The Hospital for Sick Children, Ontario, Canada.

Continue Reading

Read more coverage of the NORD Breakthrough Summit 2021

The panelists all described the need for the information in electronic health records (EHR) to be aggregated, integrated, and standardized to be more available for research into rare diseases.

“How do we stop looking at research as one thing and clinical care as another and start looking at it as a continuum?” Soloman asked.

“Data is currency and data is a driver and helps us understand who a disease affects, to what extent, the efficacy of treatments, but also it is a rate limiter,” she said. “And so who is able to access it, what we are collecting, or not collecting, and how we are analyzing and communicating it really helped us understand and not understand certain features early in the [COVID-19] pandemic.”

“When it comes to real-world data, we tend to think of the EHR as a static repository of data that we can query and get information and really drive insights and knowledge out of,” Dr. Schulz elaborated. “But it’s really a bit more challenging than that, both in terms of how we extract data from those systems as well as how we design studies that can really leverage these data assets just because of how the data are acquired and documented through the clinical workflow.”

The panelists all commented on the need to incorporate information from patients, their families, and patient groups along with clinical data. Dr. Mackay commented that it was “such a humbling experience to recognize that many of the patient groups knew more about the science and the disease than we did.”

“We have to unpack the assumption that data is for physicians and researchers only. But it is also for patients and their families who are meaningful contributors,” Soloman said.

“Relationship with patients and patient groups is absolutely critical to coming up with the right medicines,” Coenraads stated.

Coenraads also described how the RSRT is going to be piloting a digital natural history study that will recruit families who will provide information themselves in addition to EHR information from hospitals where they have been seen. The study will then use a machine learning platform to provide structure and access to the data in 1 place for patients, while academics and scientists will get free access to aggregate data. The business model would then involve biopharma paying for licenses.

“I’m hoping that by providing this clinical component to the parent-reported data that we will have in the registry, we will have a really rich, comprehensive data set to learn more about the disease to aid in clinical trial design and understand outcome measures better to run studies,” Coenraads said.

By aggregating, organizing, and providing access to data, research will accelerate and new hypotheses will be developed, the panelists said.

“We need to start unpacking those data norms and constructs and just start looking for patterns in data. We won’t always have our hypotheses right up front,” Soloman concluded.


How data sciences are being used in rare disease research and more broadly in healthcare. Presented at NORD Breakthrough Summit: October 18, 2021; Virtual.