A new platform for rare disease patients, powered by AWS

Nonprofit leaders hope that researchers will be able to use the platform to share and access currently siloed patient information.
By Kat Jercich
11:03 AM
A person with a stethoscope around their neck clicks a computer mouse

Epidermolysis bullosa, or EB, is a family of rare skin diseases that affects about 500,000 people worldwide. People with mild forms of EB may experience their disease as largely an inconvenience, while those with more severe forms can have dramatically shortened life expectancy as a result.

There is currently no cure for EB. But Michael Hund, CEO of EB Research Partnership, told Healthcare IT News that "because it's caused by one genetic mutation, this gives us a solvable problem. We believe that this is a solvable problem."

That belief, Hund said, drove EBRP to team up with Amazon Web Services to create a platform that EB parents and patients could use to navigate the disease. 

Hund, who spoke Wednesday at the AWS IMAGINE: Nonprofit Online conference with EBRP cofounder and vice chair Jill Vedder, said that EBRP's venture philanthropy model has already led to 30 clinical trials focused on EBRP. 

But a pervasive issue, said Hund, is data management: Existing information from patient surveys and university studies isn't housed in a central place. Instead, it's "siloed," preventing researchers and individuals from accessing the most up-to-date information. 

When a child is diagnosed, Hund said, parents typically turn to the Internet for guidance. But because EB cases can vary so severely, and because – as with any other medical issue – few people are fact-checking social media or message boards, that guidance can be misleading or even harmful. 

"We started to think to ourselves: 'Look there's some data out there on patients, patients are willing to contribute to research … How can we bring in industry partners?" Hund asked during the keynote. "How can we leverage technology, leverage cloud computing and the power Amazon brings –  not only cloud computing but machine learning and rapid analytics? How can we combine these things?"

"We found a partner in AWS that shared that curiosity, that shared that innovation, that shared that urgency," Hund continued.

EBRP's goal with the platform is to combine de-identified patient data sets with genotype and phenotype information about the disease. The resulting tool, said Hund, will be HIPAA-compliant and secure, allowing researchers to both access and contribute data. 

"Whether you're an investigator, or parent or guardian: Can we connect you with resources or publications?" Hund said. "There's patient libraries that exist, but I don't think anybody has said, 'Let's start from the beginning by combining them all on one platform.'"

The pilot platform is now available at Stanford, where researchers have access to data from 500 patients. By 2021, EBRP hopes to move forward with a direct-to-patient initiative, where the organization can send sequencing kits to homes and have patients enter in data directly. 

Ideally, Hund said, the platform would be as easy to use "as GPS is while driving," directing patients to nearby doctors or resources, and researchers to the best available data.

Eventually, EBRP hopes to expand the platform to the 7,000 rare diseases worldwide using the same model.

"There's so much shared experiences between rare diseases as a cohort, we want to make this available to all rare disease patients," said Hund. "If you're a rare disease patient, you want to be able to be connected to the best information."

 

Kat Jercich is senior editor of Healthcare IT News.
Twitter: @kjercich
Email: kjercich@himss.org
Healthcare IT News is a HIMSS Media publication.

More regional news

A person looks into a vaccine refrigerator

(Photo by Ivan Romano/Getty Images)

By
SUNY Upstate Medical University Cancer Center

SUNY Upstate Medical University Cancer Center

By
Want to get more stories like this one? Get daily news updates from Healthcare IT News.
Your subscription has been saved.
Something went wrong. Please try again.