Atropos Health, maker of a real-world data platform, sits at the intersection of two rapidly growing fields, real-world evidence research and artificial intelligence.
The startup, founded in 2019 as a spin-out of the “Green Button” technology developed at Stanford University, developed a consultation service for doctors powered by publication-grade real-world evidence to guide clinical decisions. The technology can quickly answer clinical questions, such as which drug is most effective for certain cancer patients.
Last fall, the company launched a new operating system, Geneva OS, and a chatbot interface to help generate observational studies rapidly and at scale. That technology, ChatRWD, reduces the time to produce high-quality publication-grade real-world evidence from months to minutes through a chat-based AI co-pilot, according to the company.
Atropos Health is getting closer to its mission of addressing the evidence gap in healthcare and has secured $33 million in series B funding to help fuel its growth.
There’s a severe evidence gap in healthcare treatment guidelines, and physicians often rely on their own judgment and educated guesswork. According to a BMJ study in 2017, only 18% of clinical recommendations made by primary care physicians are based on current evidence. Clinical trials shape most care guidelines and payer decisions, yet they exclude about 70% of the U.S. population, according to Brigham Hyde, Ph.d., CEO and co-founder of Atropos Health.
"The good news is that healthcare data is actually quite plentiful. There's a lot of it. We're 20 years on from Meaningful Use and the adoption of EMRs and adding new data all the time. And, that data now is largely in the cloud. That means the compute layers are available to it. What we built with our technology is designed to convert that data layer into high-quality evidence and do that at great scale," Hyde said during a recent interview.
Observational research on real-world data is not new, Hyde noted, but it typically takes weeks and months to go from the data layer to producing a publication-grade piece of evidence.
At Stanford, Hyde, Nigam Shah, Ph.D., and Saurabh Gombar M.D., Ph.D. developed the Green Button technology by tapping the huge volumes of data lying dormant in the EMRs of millions of patients to build clinical decision-making. That formed the basis for Atropos Health's Geneva OS.
Queries pull on the company’s network of hundreds of millions of de-identified patient records. The system then runs the analysis using statistical methods and then formats the final result into an observational study report. Such studies can traditionally take weeks or months, the company noted.
"We've scaled that and are applying it now the point of care, to research, to policy, and we're also working with life sciences," Hyde said.
Atropos Health plans to use the fresh funding to fuel the company's work in generative AI capabilities. Atropos launched its large language model-driven chatbot, ChatRWD, in beta in Q4 2023 and aims for a full launch this year.
The company says it's the first generative AI application that delivers full observational studies on real-world data in minutes. The user-friendly chat also means a clinician doesn't need a data scientist in the loop.
"This is accelerating what I call the evidence flywheel that already exists. We already do observational research on data, we publish it and those are cited by guidelines, which then influences care. We're just turning that wheel way, way faster," Hyde said.
During a demo of the ChatRWD technology, Hyde asked the chatbot a question related to the use of osteoporosis drugs among female cancer patients. Specifically, he queried about the difference in skeletal-related events in female oncology patients without a history of osteoporosis who received zoledronic acid versus denosumab. The generative AI-powered chatbot then makes suggestions to improve the study design.
Based on that query and pulling from data in Atropos' evidence network, the AI generated a publication-grade study, which the company calls a prognostogram, in under three minutes.
"This is a big step forward," Hyde said. "In the hands of life sciences today, researchers in academia, that's six to eight weeks now happening in minutes. On top of that, you're getting access to great data without data being moved around and you're getting transparency on the quality of it. This is a huge accelerant in life sciences."
While the speed of generative AI tools is important, accuracy and transparency of results are critical factors in healthcare and a core tenant of Atropos Health's foundation, Hyde noted.
Built on the core technology Geneva OS, ChatRWD is an LLM independent framework, which enables institutions to maintain LLM security and integrity, and it's built to eliminate hallucination risk, according to the company.
"Hallucinations are unacceptable in our field of clinical evidence," Hyde said.
The company plans to publish a study next month that further outlines its approach and the benefits to accuracy in the generation of evidence, he noted.
Venture capital firm Valtruis led the series B round backed by existing investors Breyer Capital, Emerson Collective and Presidio Ventures. The company also added new strategic investors including Cencora Ventures, McKesson Ventures and Merck GHI Fund.
The company pocketed $14 million in a series A funding round in August 2022. Atropos has raised $53.8 million total, to date.
The funding round, backed by several strategic investors, is a strong validation of Atropos Health's platform, noted Hyde.
"[Atropos Health's platform] is rapidly becoming the standard not only for automation and user experience, but also as the trusted quality layer between healthcare data, and the critical evidence needed to better inform patient care, research and policy," he said.
The startup also continues to build out its evidence network, a federal data network based on healthcare institution partners that offers access to more than 200 million unique patients de-identified records.
"I think we have an opportunity to be the largest international federated healthcare data network," Hyde said.
Another key priority for Atropos Health is building out strategic partnerships with pharmaceutical companies and vertically integrated healthcare companies to drive evidence adoption in oncology and other specialty areas.
Last fall, the startup inked a partnership with Janssen Research & Development, a Johnson & Johnson company, to help accelerate clinical development with real-world data analysis. Data researchers from both companies will leverage the software platform of Atropos to assess the real-world data sets of Janssen.
Atropos is going to "double down" on its life sciences partnerships, Hyde noted.
"We've taken off on that side of the house," he said. "The same thing that lets us bring that answer back to a doctor also accelerates everything pharma is doing, all the way through R&D with real-world data into medical and commercial. Instead of them having to hire huge teams and take weeks and months to do a study, it's one day."
The company is partnering with global pharmaceutical solutions organization Cencora to bring its real-world evidence platform to oncology and specialty care. Oncologists and researchers can leverage Atropos' evidence network to inform their decision-making process.
“Atropos Health addresses a very real problem that both health systems and pharmaceutical manufacturers face in turning medical data into actionable evidence in a timely, predictable, and cost-effective manner.” said Michelle Snyder, partner at McKesson Ventures, in a statement. “As leading industry players increasingly recognize the value of having the right data, rather than just having the most data, Atropos Health stands out in its ability to organize, understand, and score datasets in order to accelerate research, generate insights, and improve patient outcomes.”
Atropos Health also sees partnership opportunities to support value-based care. Healthcare data analytics company Arcadia joined the company's evidence network, which adds Atropos' evidence generation capability to Arcadia’s data platform. Atropos' life sciences users can, in turn, conduct real-world evidence studies leveraging Arcadia’s longitudinal research dataset.
To help support this work, value-based care veteran Mike Spadafore, managing director of Valtruis, joined Atropos' board of directors.
The company picked up an undisclosed strategic funding round in September from the Sumitomo Corporation VC arm Presidio Ventures, Samsung Next, Gaingels and Audere Capital. Atropos Health said it planned to use that funding to accelerate commercial and development opportunities outside the U.S., with an initial focus on customers in Japan and Brazil.