So why do some patients react to immuno-oncology drugs when others don’t?
It’s one of several million-dollar questions in medicine that confound companies, researchers, and clinicians alike. And also the really frustrating part? We all know where most of the solutions lie. They’re held in electronic health records (EHRs) and siloed by disparate health systems.
Chicago, Illinois-based Tempus is trying to extract that information at scale. The 2-year-old company lately started offering an operating-system, dubbed Tempus O, made to structure, cleanse, and annotate clinical data.
Tempus O is a a part of an finish-to-finish service which includes full DNA and RNA sequencing at its CAP/CLIA-certified labs. However, the organization found a lot of its clients wanted to pay attention to the information organization component, Chief executive officer Eric Lefkofsky stated, to harness the phenotypic, therapeutic, and outcome and response data.
“People wish to structure this data clinically simply because they think that getting that data in their fingertips will assist them have better choose to their sufferers,” Lefkofsky stated via phone. “And people wish to structure that data for research because, clearly, it’s vital to allow them to understand: Exist particular characteristics leading many people to possess outsized positive or outsized negative responses to the therapeutic?”
It’s harder of computer sounds.
Some medical records happen to be digitized recently, they’re not quite “user-friendly” sources. These were created for medical billing, in the end. Many of the important information is tangled up in free text — individuals hastily written progress notes.
Tempus O taps into some sophisticated workflow tools, including optical character recognition and natural language processing, which extract meaning out of this text. Individuals notes may then be compared and arranged inside a bigger dataset, together with insights from research databases, images, and scans. Lefkofsky stated the organization has additionally developed a group of abstractors that may by hand input data when needed and evaluate the finished work.
Obviously, all this is performed at scale, to power real insights.
“To provide you with some perspective with that, we predict to structure around 400,000-patients price of data within the next 12 several weeks,” Lefkofsky stated.
That’s almost one-quarter from the 1.seven million Americans likely to be identified as having cancer in 2017.
Unsurprisingly, there’s big interest in this kind of software.
In September, Tempus closed a $70 million Series C round co-brought by New Enterprise Associates (NEA) and Revolution Growth. That cash injection introduced the startup’s total funding to $130 million. (Lefkofsky, a serial entrepreneur, has additionally invested a lot of personal money).
It’s only some of the player hanging around. Palo Alto, California-based Syapse closed a $$ 30 million Series D round in November, for as many as $71 million elevated.
While Syapse also activly works to bring fragmented clinical, molecular, treatment, and health outcomes data together, it is centered on the program component. Instead of carry out the sequencing, it’s collaborations with assorted labs.
Inside a This summer interview with MedCity News, Lefkofsky stated a part of his company’s edge continues to be the opportunity to do all of it, to simplify the connection using the client.
Nevertheless the molecular information is generated, both Syapse and Tempus concurs the two data sources have to be examined together.
“As a business, we’re most thinking about the mixture of both clinical data and also the molecular data,” Lefkofsky stated. “When you will find the molecular data you may also answer the ultimate goal question, that is ‘why.’ How come these patients responding well? How come these patients not responding well? For cancers, that’s a molecular question frequently.”
Photo: coffeekai, Getty Images