Medical experts agree that personalized care is crucial in fighting cancer. But that’s easier said than done when more than 14 million people are diagnosed with cancer each year. Our personalized medicine podcast explores how one of the world’s leading cancer researchers plans to use big data and IBM Watson to help oncologists customize treatment options and finally turn the tables on this global scourge.
Behind the scenes : A tour of Dr. Norman Sharpless’ lab at UNC Lineberger Comprehensive Cancer Center
Dr. Norman “Ned” Sharpless is the director of the UNC Lineberger Comprehensive Cancer Center in Chapel Hill, North Carolina. He’s a world-renowned cancer researcher, an oncologist, a mathematician, and the author of more than 100 scientific papers with 10 patents to his name. Lately, he’s become a huge proponent of mastering big data as a way to help personalize cancer treatment. This photo was taken during the Wild Ducks interview on campus at the University of North Carolina at Chapel Hill.
During a tour, Sharpless explains a research method developed in his lab to visualize aging and tumor growth in mice. The image posted on the door shows four “transgenic” mice whose interiors have been illuminated through a process of optical imaging that reveals a type of gene interaction consistent with both aging and cancer suppression. Labs around the world are using the method to research muscle aging, DNA repair, ulcers, arthritis, pulmonary hypertension and several types of cancer.
The Sharpless lab consists of 10 specialists, including assistant professors, post-docs, grad students and technicians. Sharpless introduced us to Will Jeck, the MD/PhD student in the baseball cap who has been working directly on training IBM Watson. Sharpless said that Jeck represents a new guard of game-changing cancer researchers. He’s a data-savvy technologist comfortable exploring huge amounts of data with an open mind.
Sharpless showed us a robot, which helps prepare genomic libraries for sequencing machines, and said it perfectly demonstrates how cancer research has entered the realm of big data. With the old manual “prep” process, a technician would prepare one pipette at a time. It would take up to 5 hours to prep a genomic library. This machine does it in 30 minutes, which vastly increases the amount of sequences the lab can handle as well as the amount of data it produces.
Sharpless is a big proponent of genomic sequencing as a means to personalize cancer care. UNC Lineberger Comprehensive Cancer Center specialists have access to 17 different genome-sequencing machines. The most sophisticated models can sequence the entire human genome for 10 patients in six days. This is equal to around 1 terabyte of data. In the 2013-2014 fiscal year, UNC Lineberger produced over 13,000 genomic libraries and sequenced over 100 trillion base pairs. The Lineberger Center has a cancer genetic sequencing trial called UNCSeq, which has enrolled 1,500 patients to date and 600 in the past year alone.
Sharpless is hardly the only world-renowned researcher roaming the halls at UNC Lineberger. At the end of our tour, he introduced us to Charles Perou. Perou was the lead author of a breakthrough 2012 study that showed breast cancer isn’t just one disease. He and his team identified four genetically varied types of breast cancer and a string of genetic triggers that hasten its spread.
Sharpless shared with us his excitement for a new initiative harnessing the cognitive computing powers of IBM Watson to deliver treatment options for doctors to consider for cancer patients with extremely precarious medical histories. In short, Sharpless and his team are training Watson to provide drug options for doctors to consider for their patients who have not responded well to standard treatment. “That’s the goal,” he says. “I think it’s very doable.
Infographic:See how leading cancer researchers are training IBM Watson to prescribe the most effective cancer treatment options