Machine learning finds tumor gene variants and sensitivity to drugs in The Cancer Genome Atlas

April 9, 2018, Perelman School of Medicine at the University of Pennsylvania

Matching unique genetic information from cancer patients' tumors with treatment options - an emerging area of precision medicine efforts - often fails to identify all patients who may respond to certain therapies. Other molecular information from patients may reveal these so-called "hidden responders," according to a Penn Medicine study in Cell Reports this week. The findings are published alongside several papers in other Cell journals this week examining molecular pathways using The Cancer Genome Atlas (TCGA).

"Targeted sequencing can find individuals with certain mutations that are thought to confer susceptibility to anti- drugs," said senior author Casey Greene, PhD, an assistant professor of Pharmacology in the Perelman School of Medicine at the University of Pennsylvania. "But many people may lack these mutations, and as machine learning approaches improve they may help guide these patients to appropriate therapies."

Greene and first author and doctoral student Gregory P. Way used machine learning to classify abnormal protein activity in tumors. This branch of artificial intelligence develops computer programs that can use new data to learn and make predictions. The algorithm they devised to search TCGA integrates genetic data from 33 different cancer types. Greene and Way used information from the transcriptome, the grand total of all messenger RNAs expressed within an individual.

They specifically applied their model to the Ras pathway, a family of genes that make proteins that govern cell replication and death. Changes in the normal function of Ras proteins - mutations which are responsible for 30 percent of all cancers - can power cancer cells to grow and spread. These mutations are often referred to as the "undruggable Ras," having beaten back a variety of investigational inhibitor drugs and vaccine-based therapies.

"This model was trained on genetic data from human tumors in The Cancer Genome Atlas and was able to predict response to certain inhibitors that affect cancers with overactive Ras signaling in an encyclopedia of cancer cell lines," Greene said. The upshot is that the transcriptome is underused in bringing precision to oncology, but when combined with machine learning it can aid in identifying potential hidden responders.

The Penn team collaborated with coauthor Yolanda Sanchez, PhD, a cancer biologist from the Geisel School of Medicine at Dartmouth College. They are working together to mesh her identification of compounds that target tumors with runaway Ras activity and data (analyzed by machine learning) to find patients who could benefit from these potential cancer drugs.

"For precision medicine to benefit individuals in real time, we must develop robust models to efficiently test efficacy of potential therapies," Sanchez said. "We can use this very powerful combined approach of machine learning-guided drug discovery using Avatars, which are mice carrying identical copies of a patient's tumors. The Avatars allow our interdisciplinary team to identify the tumors with runaway Ras activity and evaluate and compare multiple therapies in real time."

Explore further: New rapid-fire method using pathology images, tumor data may help guide cancer therapies

Related Stories

New rapid-fire method using pathology images, tumor data may help guide cancer therapies

April 6, 2018
By combining data on pathology images of 13 types of cancer and correlating that with clinical and genomic data, a Stony Brook University-led team of researchers are able to identify tumor-infiltrating lymphocytes (TILs), ...

Even DNA that doesn't encode genes can drive cancer

April 2, 2018
Most of the human genome—98 percent—is made up of DNA but doesn't actually encode genes, the recipes cells use to build proteins. The vast majority of genetic mutations associated with cancer occur in these non-coding ...

Major milestone reached in effort to ID cancers' genetic roots

April 5, 2018
Researchers nationwide have reached a major milestone in describing the genetic landscape of cancer. Scientists at Washington University School of Medicine in St. Louis and other institutions have completed the genetic sequencing ...

Organoids created from patients' bladder cancers could guide treatment

April 5, 2018
Columbia University Irving Medical Center (CUIMC) and NewYork-Presbyterian researchers have created patient-specific bladder cancer organoids that mimic many of the characteristics of actual tumors. The use of organoids, ...

Researchers find novel mechanism of resistance to anti-cancer drugs

October 17, 2017
The targeted anti-cancer therapies cetuximab and panitumumab are mainstays of treatment for advanced colorectal cancer, the second leading cause of cancer-related deaths in the United States. However, many patients have tumors ...

Three-pronged approach is key to precision medicine

March 23, 2017
Combining genetic information from a patient's tumor cells with three-dimensional cell cultures grown from these tumors and rapidly screening approved drugs can identify the best treatment approaches in patients for whom ...

Recommended for you

Single-cell study in a childhood brain tumor affirms the importance of context

April 20, 2018
In defining the cellular context of diffuse midline gliomas, researchers find the cells fueling their growth and suggest a potential approach to treating them: forcing their cells to be more mature.

Aggressive breast cancer already has resistant tumour cells prior to chemotherapy

April 20, 2018
Difficult to treat and aggressive "triple-negative" breast cancer is chemoresistant even before chemotherapy begins, a new study by researchers from Karolinska Institutet and the University of Texas MD Anderson Cancer Center ...

Mechanism that drives development of liver cancer brought on by non-alcoholic fatty liver disease discovered

April 19, 2018
A team of researchers from several institutions in China has found a mechanism that appears to drive the development of a type of liver cancer not caused by alcohol consumption. In their paper published in the journal Science ...

Variants in non-coding DNA contribute to inherited autism risk

April 19, 2018
In recent years, researchers have firmly established that gene mutations appearing for the first time, called de novo mutations, contribute to approximately one-third of cases of autism spectrum disorder (ASD). In a new study, ...

Discovery adds to evidence that some children are predisposed to develop leukemia

April 19, 2018
St. Jude Children's Research Hospital researchers have made a discovery that expands the list of genes to include when screening individuals for possible increased susceptibility to childhood leukemia. The finding is reported ...

Scientists identify 170 potential lung cancer drug targets using unique cellular library

April 19, 2018
After testing more than 200,000 chemical compounds, UT Southwestern's Simmons Cancer Center researchers have identified 170 chemicals that are potential candidates for development into drug therapies for lung cancer.

0 comments

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.