November 26, 2019 - An artificial intelligence algorithm was able to find previously unseen changes in patterns in CT scans and determine how well patients with lung cancer would benefit from immunotherapy, according to a study published in Cancer Immunology Research.
Researchers from Case Western Reserve University noted that currently, only about 20 percent of cancer patients would actually benefit from immunotherapy, a treatment different from chemotherapy in that it uses drugs to help your immune system fight cancer. Chemotherapy, on the other hand, uses drugs to directly kill cancer cells.
Having an artificial intelligence tool that could match up which patients would best respond to immunotherapy could significantly reduce healthcare costs, researchers stated.
Dig Deeper
Artificial Intelligence Success Requires Human Validation, Good Data
KLAS: Artificial Intelligence Success Requires Partnership, Training
Artificial Intelligence in Radiology Will Require Ethics, Standards
“Even though immunotherapy has changed the entire ecosystem of cancer, it also remains extremely expensive—about $200,000 per patient, per year,” said Anant Madabhushi, director of the Center for Computational Imaging and Personalized Diagnostics (CCIPD).
“That's part of the financial toxicity that comes along with cancer and results in about 42% of all new diagnosed cancer patients losing their life savings within a year of diagnosis.”
In the initial study, the team used CT scans from 50 patients to train the AI algorithm to identify changes in lesions and distinguish patterns most associated with immunotherapy success. The algorithm achieved an area under the curve (AUC) of 0.88.
Additionally, the patterns on the CT scans that were most associated with a positive response to treatment and with overall patient survival were also later found to be closely associated with the arrangement of immune cells on the original diagnostic biopsies of those patients.
This indicates that those CT scans actually appear to be capturing the immune response elicited by the tumors against the invasion of the cancer, and the ones with the strongest immune response would best respond to the immunotherapy.
The algorithm was also able to note changes in the texture, shape, and volume of a given lesion, and not just its size.
“This is important because when a doctor decides based on CT images alone whether a patient has responded to therapy, it is often based on the size of the lesion,” said Mohammadhadi Khorrami, a graduate student working at the CCIPD. “We have found that textural change is a better predictor of whether the therapy is working.
“Sometimes, for example, the nodule may appear larger after therapy because of another reason, say a broken vessel inside the tumor—but the therapy is actually working. Now, we have a way of knowing that.”
Moreover, the study showed that the results were consistent across scans of patients treated at two different sites and with three different types of immunotherapy agents.
“This is a demonstration of the fundamental value of the program, that our machine-learning model could predict response in patients treated with different immune checkpoint inhibitors,” said Prateek Prasanna, a postdoctoral research associate and co-author of the study. “We are dealing with a fundamental biological principal.”
The results demonstrate the potential for AI algorithms to enhance cancer care and reduce costs associated with cancer treatment.
“This is no flash in the pan—this research really seems to be reflecting something about the very biology of the disease, about which is the more aggressive phenotype, and that's information oncologists do not currently have,” said Madabhushi.
Watch this:https://www.youtube.com/watch?v=Z5vxRC8dMvs