After being diagnosed with breast cancer, the wife of Groupon co-founder Eric Lefkofsky, he discovered how much the US healthcare system is forcing doctors to access data when making clinical decisions. This led him to create a disease-independent startup, Tempus AI, which is now collecting patient data for a potential Covid-19 drug.
Tempus, a Chicago-based technology company that offers genetic testing and collects clinical information initially targeted at cancer patients. But with the spread of the coronavirus pandemic worldwide, she joins the fight against Covid-19.
Tempus was founded in 2015 by its CEO Eric Lefkofsky, who previously co-founded Groupon, an e-commerce company, in 2008. But although the Tempus mission may seem far from the mission of the online discount market, Lefkofsky had personal reasons for forking. ,
“My wife was diagnosed with breast cancer about five years ago, and I was amazed at how little data was actually used as part of her therapy, mainly because our system makes it difficult for doctors to access data when making clinical decisions in real time, “he said.” It became clear to me that I needed to try to solve this problem, and soon I founded Tempus. “
Tempus, which ranked 6th on the 2020 CNBC Disruptor 50 list, has three main lines of business. It offers genomic tests that doctors can order, the most popular of which is to search for more than 600 genes related to cancer. (He does not offer direct genetic testing, such as 23andMe.)
It also collaborates with academic medical centers and public health systems to organize and aggregate clinical data, primarily data stored in electronic medical records. Clinical data is de-identified to protect patient privacy, and then commonly used in research projects to help understand patterns. Tempus claims to have one of the largest libraries of molecular and clinical data.
There is a crossover between the two enterprises, since the genomic tests carried out in his laboratories in Chicago and Atlanta can be de-identified and added to the molecular and clinical data library.
Finally, he identifies and sells anonymized data to pharmaceutical drug companies.
Tempus is an organization covered by the Health Insurance Mobility and Accountability Act (HIPAA), so all data is cleared on behalf of the patient, date of birth, address and any other data that can identify the person.
At first, Lefkofsky believed that the Tempus platform should be used for a wide range of patient data. For this reason, the Tempus platform has been designed so that the company calls it “disease-independent”, which means that it can be used for patients suffering from many types of diseases, including cancer, depression, diabetes, cardiovascular disease and Covid-19 .
Using Real Covid-19 Patient Data
In April, Tempus began offering the Polymerase Chain Reaction (PCR) test for Covid-19 from its laboratories in Chicago and Atlanta. It is currently expected that its capacity will be increased to approximately 10,000 tests per day in the coming months. The test, proposed by doctors, was launched in parallel with the start of a research project to collect data on patients with Covid-19 and to search for patterns that can help doctors in treating patients, as well as in drug discovery efforts. Procedures can vary from patient to patient, but it is here that Lefkofsky believes that the company’s AI platform can be most beneficial.
“We are focused on collecting evidence about patients with a positive outcome at Covid-19, to help doctors divide patients into risk groups and help to choose the optimal treatment regimen for patients based on their unique phenotypic characteristics,” he said.
Dr. Scott Gottlieb, a fellow at the American Enterprise Institute and a special partner at New Enterprise Associates, is a member of the Tempus Board of Directors. He joined after serving as Commissioner of the Food and Drug Administration, where he led efforts to promote the use of artificial intelligence to create personalized care for patients.
“I was impressed by the company’s ability to rapidly scale sequencing operations to support large academic institutions and oncologists across the country and transfer data to patients and providers,” he said. “They focused not only on server technologies that provide these capabilities, but also interface interfaces are equally important to make this information useful at the time of departure.”
He also launched an initiative that seeks to collect data for 50,000 positive patients with Covid-19. We hope that these data will help to identify the genetic characteristics that cause the specific results of Covid-19, as well as methods of treating this disease.
Gottlieb added that Tempus is working with employers and health care providers to support Covid-19 testing in states where restrictions are lifted and employees are returning to work, so they will know what to look for.
“We need to be more aggressive in detecting and identifying respiratory viruses,” he said. “To this end, Tempus is also working on a next-generation sequencing panel for respiratory pathogens, including Covid-19, to diagnose patients and create a library for research on infectious diseases that can reduce the impact of future epidemics.”
Current Limitations of the Use of AI in Medicine
While Tempus’s use of health data as a way to accelerate research has shown promising results, there are limitations.
Dr. Jose Mori, director of medical innovation at Liberty BioSecurity and advisor to MIT Solve and NASA iTech, said that one of the problems associated with artificial intelligence in the case of the new virus is that at the moment there is not enough data.
“You must have large data sets in order to be able to train, test and validate,” he said. “When you have something new and new, like this, they simply do not exist … It will happen in the end, but not yet.”
More from Disruptor 50:
Meet CNBC 2020 Disruptor 50 Companies
A technology that will dominate everyday life on the other side of the coronavirus
A missed opportunity: why this year there are no black leaders on the list
He added that the technology is as good as the data, and there are currently systemic flaws in the collection and recording of medical data. For example, he said that data in medical centers can be “dirty.”
“This means that the data is not structured so that it can be included in the mathematical system of AI,” he said. He added that health data is also often “parsed” or divided into unrelated compartments. In order for AI to do its job, it requires large amounts of interconnected data that, according to Mori, do not exist in the US healthcare system.
“Health organizations, such as device manufacturers, electronic medical records providers and hospitals, are not interested in sharing data with each other,” he said. “That’s why so many artificial intelligence companies step-by-step develop algorithms and applications.”
In April, Tempus began offering the Polymerase Chain Reaction (PCR) test at its laboratories in Chicago and Atlanta. In the coming months, he is expected to increase his testing capabilities to 10,000 tests per day and hopes to combine data for 50,000 positive patients with Covid-19.
Although this is a difficult moment for the US healthcare system, the medical community and investors have recognized Tempus as a company whose platform shows great promise.
The company raised $ 5 billion from investors such as Bailey Gifford, Franklin Resources and T. Rowe Price. He also partnered with CVS Health to enable oncologists on the CVS Health / Aetna network to use the results of their genome sequencing tests.
The Covid-19 study, like cancer research, has not yet cured. However, Gottlieb said he considers AI tools indispensable for cancer patients, patients with infectious diseases, and others in the future.
“As we collect more data and use AI tools for large data sets, we can guide patients at greatest risk,” he said. “Artificial intelligence can help us solve a number of serious public health problems and improve treatment outcomes for patients with infectious diseases.”
Disclosure: Scott Gottlieb is a member of CNBC.