Eric Lefkowsky seeks to fight cancer through Big Data

Of all the incredible advancements in medicine that have taken place over the last few decades, one stands above the rest. The ability to sequence the human genome cheaply and quickly has provided researchers and doctors with more data than they can currently handle. The magnitude of unmined data that has been provided by advancements in genomics is difficult for the mind to comprehend. The vast majority of current genomic information has not been tested or used for any medical purpose.

Eric Lefkofsky, the famous co-founder of volume discount giant Groupon, is undertaking the construction of a system that will allow this vast trove of genomic data to be used in real time by oncologists and physicians. Tempus, founded in 2016, seeks to take vast amounts of genomic data, clinical trial data, patient histories and other relevant medical information and combine it into a platform that will inform oncologists and other doctors in real time what the best course of treatment, for any given patient, is likely to be.

But Tempus is no ordinary data management company. Eric Lefkofsky believes that, through the use of advanced statistical methods, including machine learning, Tempus will be able to create a platform that amounts to the ability to perform real-time, on-site meta studies, allowing oncologists to have a granular understanding of treatment phenomena that has never before been possible. Lefkofsky believes that this system will enable the complete customization of treatment regimens. Dosages, duration of administration and even the molecular construction of drugs will be able to be manipulated on the spot, yielding the precise combination of treatment elements that are likely to maximize that individual patient’s chances of survival and recovery.

The heart of the Tempus idea is to create a far more granular understanding of oncological phenomena. Today, most treatment regimens are essentially little more than a one-size-fits-all affair, with virtually every patient suffering from a certain type of cancer receiving effectively the same treatment. This is a blunt, strongly suboptimal approach, but it is the best tool that oncologists have had to work with. However, with the use of genomic data and sophisticated, real-time analysis, it may be possible to break that group of patients into 100 or more subgroups, each with a different treatment regimen designed to maximize patient survival and minimize side effects.

for details: