Brigham Hyde, co-founder and CEO of Atropos Health.
Atropos Health
Doctors usually advise diabetic patients not to suddenly change or stop taking their regular medications before surgery. That's why David Klonoff, an endocrinologist who treats diabetic patients, said last summer that the American Society of Anesthesiologists recommended that patients take GLP-1, a newer diabetes and weight-loss drug like Wegovy and Ozempic, before their scheduled surgery. I was surprised that they issued a guideline stating that agonists should be discontinued. “We were concerned that the downside of that decision was worse than the upside,” Klonoff, medical director of the Diabetes Institute at Mills Peninsula Medical Center in San Mateo, Calif., told Forbes.
Because the drug slows stomach emptying, anesthesiologists say patients taking GLP-1 agonists may vomit during surgery and swallow some of it, potentially causing conditions such as pneumonia. I was worried about something. A national group of anesthesiologists has determined that the risks of potential complications outweigh the benefits of continuing to take the drug. But to Klonoff's frustration, there was little evidence to support this position, other than a few case reports. “There was no data,” he said.
In medical school, doctors learn that randomized controlled trials are the gold standard for determining how to treat patients. But every day, doctors must make life-or-death decisions in situations that cannot be detected in clinical trials. Instead, they must rely on their training, intuition, and experience to triangulate what is best for each individual patient. “We don't have the time, energy, or money to do randomized controlled trials on everything,” Klonoff says.
That's why he joined Atropos Health's Medical Advisory Board. Klonoff thinks the startup has come up with the next best thing. Atropos leverages 200 million patient records to help doctors make decisions based on what's happening in the real world outside of carefully planned clinical trials.
For example, consider a 45-year-old woman with bladder cancer. Which option has fewer complications and better results: radiation therapy or cystectomy? Atropos' software sifts through millions of records to find similar patients and their journeys, then performs statistical analysis within days. Nowadays, advances in generative AI allow statistical analysis to be performed in minutes.
On Thursday, Atropos announced a $33 million Series B round led by Valtroui, the venture growth arm of private equity firm Welsh, Carson, Anderson & Stowe. Forbes values the company at about $250 million. To date, it has raised $54 million. In addition to existing investors Breyer Capital, Emerson Collective and Presidio Ventures, new strategic investors Cencora Ventures (formerly AmerisourceBergen), McKesson Ventures and Merck Global Health Innovation Fund also participated in the round.
Atropos takes a fundamentally different approach than most other health data companies. Rather than moving your medical data to create one giant dataset, query each data pile or node individually. This is known as the federated data model.
The uniqueness of this approach attracted the interest of billionaire venture capital investor Jim Breyer, who was an early investor in Facebook. As he navigated his parents' cancer diagnosis, he recognized how difficult it was for doctors to share data between different hospitals. “There is a huge challenge in giving healthcare professionals, whether they are good doctors or nurses, access to a set of data.Currently, this industry is as siloed as any industry I know. ,” he told Forbes. Mr. Breyer joined Atropos' seed round and board of directors in 2020 and has participated in every financing since then.
The startup is named after one of the three Destinies in Greek mythology, whose scissors have the power to end or extend human life, and hopes that by generating evidence to help clinicians make better decisions, the company can also help improve treatment outcomes and extend people's lives.
“Whenever you try to aggregate all the world's data into one place, the biggest dataset always wins.”
Brigham Hyde, Co-Founder and CEO of Atropos Health
Before launching Atropos, CEO Brigham Hyde spent 10 years founding and running a company that aggregated medical data to sell to pharmaceutical companies for research. Companies collect this data with the goal of creating a complete patient record, which is essential to aid in the development of new medicines. For example, if a patient receives their medicine at one clinic, has surgery at another, and is diagnosed with cancer at another, the data is only valuable if all the pieces of the puzzle are in place.
But as Hyde discovered, simply buying up more data is expensive and time-consuming for companies. Patient privacy is also at risk. The more health data in circulation, the greater the risk for everyone involved under the federal privacy law known as HIPAA. But importantly, Hyde realized that he had one major flaw. “Every time he tries to aggregate all the world's data into one place, the biggest data set always wins,” he told Forbes. Essentially, the dataset is biased toward the entities that provided the most data.
Hyde co-founded Atropos with two other medical informatics experts, Nigam Shah, chief data scientist at Stanford Health Care, and pathologist Saurabh Gombar. The trio built software that runs analyses on individual data sources (in this case, each medical institution) and combines the results in what's called a meta-analysis. Using this technology, the company can provide doctors and researchers with retrospective studies based on how real people responded to different treatments. And because the data never moves, there are fewer privacy concerns. Now, researchers and companies that want to do retrospective analyses don't have to buy the data, “you just buy the answer,” Hyde said.
To date, Atropos has conducted more than 10,000 such retrospective studies, including one in collaboration with endocrinologist Klonoff to look at whether people taking GLP-1 agonists who underwent nine common surgeries were at increased risk of complications. Published this month in the journal Diabetes, Obesity, and Metabolism, an analysis of more than 13,000 diabetes patients, about 2,200 of whom took GLP-1, found no evidence to support stopping the medication before surgery.
Klonoff said he was upset by the results, but that he wasn't ultimately responsible for whether patients lived or died in the operating room. And Girish Joshi, an anesthesiologist and vice chair of the committee that wrote the recommendation, said this particular study was flawed. The number of patients, especially those who died, was too small to suggest a meaningful correlation. About one in every 3,000 patients under anesthesia aspirates or inhales their own vomit. And Joshi said the study used medical diagnosis codes that were too broad to properly dig into this particular phenomenon. Joshi said he's not opposed to the general idea of using retrospective analyses to inform guidance, but the study needs to be better designed and include more patients.
Ultimately, Hyde hopes it can provide that. Atropos will continue to add new patients and new drugs, even if there isn't enough data to convince Joshi just yet. “What's needed is new evidence,” Hyde says. “And that's what we're going to produce.”
More from Forbes
ForbesThis medical AI startup aims to provide doctors with the latest scientific informationBy Katie JenningsForbesWhy $4.6 billion medical records giant Epic is betting big on Generative AIBy Katie JenningsForbesMicrosoft and federal agencies aim to provide medical AI Nonprofit supergroup launches to conquer uncharted territory By Katie JenningsForbes This $2.2 billion startup's GPT-4-powered AI bot solves health insurance mysteries By Katie Jennings
Source link
