Project Aims to Develop Database for Guiding Precision Medicine Treatments
January 25, 2016
By Alex Keown, BioSpace.com Breaking News Staff
PITTSBURGH – During his State of the Union Address, President Barack Obama pledged a “new national effort” to find a cure for cancer—a disease that touches most every family. That’s a message that many researchers are taking to heart, including a team of 40 researchers from Carnegie Mellon University and the University of Pittsburgh.
The project, called Big Data for Better Health, is being funded by a $5 million grant from the CURE grant program of the Pennsylvania Department of Health in order to use “machine learning” to develop better ways to analyze data from cancer patients. The project will initially focus on breast and lung cancer using data provided by the University of Pittsburgh.
Following the development of the data points for those cancer types, the methods will be applied to other cancer forms, Ziv-Bar Joseph, a computational biologist at Carnegie Mellon’s School of Computer Science, told BioSpace in an exclusive interview. The data gathering project is in line with the cancer initiative the president promoted, Joseph said.
Precision medicine is an effort that the Obama administration has thrown its weight behind in previous State of the Union speeches. In 2015, Obama pointed out a guest of the first lady who was taking Vertex ’s Kalydeco medication, the first drug designed to counter the genetic cause cystic fibrosis. Such medications are at the heart of the proposed Precision Medicine Initiative, which will allow researchers to examine the cause of cancers and other illnesses on a genetic level in order to provide targeted therapies for individual care.
Joseph said that gathering the large quantity of data will allow the research team to predict cancer outcomes in patients using “molecular data,” which is data gathered from the patients’ cells, including their DNA and the set of proteins they express. The patient data acquired for the project includes age and gender of the individuals, as well as treatments they tried.
“We will be able to see what worked and didn’t work as well as survivability,” Joseph said.
Joseph noted that the data analysis will “build predictive models that will inform treatment directions” and give doctors a stronger idea of which drug to prescribe to treat the patient.
“What we’re doing is building an infrastructure that can be used widely,” Joseph said. That infrastructure could lead to advances in precision medicine, something the Obama administration called for in 2015.
“Precision medicine gives clinicians tools to better understand the complex mechanisms underlying a patient’s health, disease, or condition, and to better predict which treatments will be most effective,” the Obama administration said in a 2015 statement. “Patients with breast, lung, and colorectal cancers, as well as melanomas and leukemias, for instance, routinely undergo molecular testing as part of patient care, enabling physicians to select treatments that improve chances of survival and reduce exposure to adverse effects.”
Joseph said the Big Data for Better Health project will run for three years, but he hopes to see some early hard data points within the year—which will then allow the researchers to examine data on other cancers.
“If we can build a good foundation and know how to find things with a large data set, we can use the same methods on a new cohort of data,” Joseph said.
A new analysis from the Personalized Medicine Coalition released last week shows an upward trend in the number of personalized medicine approvals by the U.S. Food and Drug Administration. According to the analysis, personalized medicines accounted for more than 25 percent of novel new drugs (NNDs) approved in 2015. The analysis said the 13 personalized medicines approved as NNDs in 2015 represent 28 percent of the 45 NNDs the agency approved overall.