When a loved one was diagnosed with cancer, entrepreneur Eric Lefkofsky was shocked by how little technology had permeated the clinic and patient care. As a founder of companies such as Groupon and Echo Logistics, among others, Lefkofsky knew from first-hand experience the potential that state-of-the-art software and modern data processing technology had for enhancing cancer treatment. Driven by his experiences, Lefkofsky founded a company that aims to help doctors make data-driven decisions about cancer treatment with a goal of improving patient outcomes.
In pursuit of this goal, Tempus developed a platform that analyzes the detailed profiles of individual cancer patients and compares them with a database of patients that came before to help physicians determine the treatment option that has the highest probability of success. This platform integrates molecular profiles generated by analyzing samples of tumors and healthy tissues in the company’s CLIA lab with pathology reports, physician notes, and treatment outcomes to deliver information tailored to each patient’s needs. Though the exact figures are undisclosed, Tempus says that the size of this set of data eclipses NCI’s Cancer Genome Atlas by a factor of ten to twenty, constituting several petabytes of data. This dataset aggregates publicly sourced data, data obtained through collaboration with academic institutions and community hospitals, and data generated by the company itself.
One of the challenges Tempus faces in developing this tool is contextualizing massive amounts of raw, unstructured data. As processing such a large collection of data by hand is infeasible, Tempus leverages the cutting edge of machine learning software and continually updates its platforms with modified algorithms and new patient profiles. Using this system, Tempus is able to deliver treatment options to physicians, ranked by order of their computed probability of success, within two to three weeks. This report is supplemented by information about anonymous patients with similar genetic and clinical characteristics, and can be updated in real-time as the patient responds to treatment.
While other molecular genetics companies exist, they oftentimes depend upon separate, disparate silos of information, or use their resources in service of research or drug discovery efforts. Tempus expands upon the capabilities of these companies by combining multiple sources of data and delivering physicians with personalized results and focusing on delivering a direct real-world impact. Lefkofsky describes Tempus’ efforts using the metaphor of a computer operating system: Just as the operating system allows the different parts of the computer, including the screen, keyboard, and processor to work in concert with one another, Tempus aims to bring together all of the elements of cancer treatment into a cohesive whole.
Tempus’ focus on helping real patients derives from Eric Lefkofsky’s personal philosophy of results-oriented business practices. Lefkofsky attributes his past successes to this philosophy, citing his belief in solving real problems, taking risks, and adapting his strategies to changing conditions. Having had direct experience with the strongly negative effects cancer can have on individual lives, and mindful of the tremendous number of people affected by the disease, Lefkofsky sees cancer treatment as a problem that his guiding philosophy and knowledge of the technology industry can effectively address.
Tempus is not the only project devoted to helping the lives of others Lefkofsky has devoted himself to. He and his wife, Liz, chair the Lefkofsky Family Foundation, a philanthropic organization dedicated to enhancing the quality of human life by focusing on expanding access to education, improving medical research, and promoting universal human rights. He also serves as a Trustee of Lurie Children’s Hospital of Chicago, and has joined The Giving Pledge, a campaign that encourages wealthy people to contribute a majority of their wealth to charitable causes.
Leave a Reply
Be the First to Comment!
You must be logged in to post a comment.
You must be logged in to post a comment.