MCG doctors using artificial intelligence to personalize cancer treatment


Cancer doctors at the Medical College of Georgia are using artificial intelligence to analyze genes. The data they gather allows them to create treatment plans for cancer patients that are personalized to their specific needs.

NewsChannel 6’s Ashley Osborne sat down with the lead doctor for the new treatment and shares how they are transforming patient care.

Now, the standard of care is to treat cancer based on tumor type. For example, if patient A and Patient B have the same kind of brain cancer, they will be treated with the same checklist, even though they are not the same person. Doctors at MCG are part of a national trend to change that strategy.

MCG is home to 1 of only 7 labs chosen by the National Cancer Institute for their MATCH trial. MATCH stands for Molecular Analysis for Therapy Choice. This trial aims to test new ways to personalize cancer treatment.  

“Every cancer is very unique. Every patient is very unique,” says Dr. Ravindra Kolhe.

Dr. Kolhe is a molecular pathologist and director of the Georgia Esoteric and Molecular Laboratory in the MCG Department of Pathology. He is also the lead developer of the new test called Augusta OncoTarget. 

Dr. Kolhe is using the Augusta OncoTarget test to help those with the worst forms of cancer.

“It doesn’t matter if the patient has breast cancer, doesn’t matter if the patient has brain cancer or ovarian cancer or lung cancer. We are going for what is driving that tumor. In this setting we already know that what we were doing traditionally has failed,” says Dr. Kolhe.

He envisions the option becoming the standard for an even wider range of cancer diagnoses in the future.

“I think we will go towards more the time of diagnosis and start treating these patients right at the beginning rather than wait until it becomes aggressive, metastatic or resistant,” he says.

Time is critical for cancer patients, so Augusta OncoTarget uses IBM’s Watson (artificial intelligence software) to evaluate the genes much faster than humans. 

“It’s relatively impossible to compare because if you have to do it manually, 5-7 people in the lab would take probably more than a week to do the analysis and Watson for Genomics, it helps us to do maybe in 20 minutes,” Dr. Kolhe explains.

Dr. Kolhe showed NewsChannel 6 the profile of a woman who received the test. Traditional treatment did not work for her breast cancer so they tested her genes, determined why the standard failed, and changed her care to something that did work.

“We’re going to fundamentally change the way we look at these patients,” says Dr. Kolhe.

Copyright 2020 Nexstar Broadcasting, Inc. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.