AI May Help Predict Colon Cancer Recurrence

Predicting colorectal cancer recurrence may be improved by using artificial intelligence, or AI, to assess tumor imaging, according to recent study results.

Researchers at the Mayo Clinic in Phoenix developed QuantCRC, a deep-learning segmentation algorithm, and said it can be a “powerful” addition to pathological reports of colorectal cancer. In particular, deep-learning algorithms — a more technical term for AI — are used to emulate the way humans assess information, but they are developed to take analyses a step further by finding intricate structures within data.

With AI, a patient’s cancer team can use a program to scan an endless amount of data in a more efficient and potentially accurate manner. This can provide physicians with greater predictive abilities.

In the study published in the journal Gastroenterology, 15 parameters were recorded from 6,468 images of colorectal cancer. According to the Mayo Clinic, colorectal cancer is a term that combines colon cancer and rectal cancer, which begins in the rectum. These parameters were then compared to findings in the pathology report and health records. This information was used to develop a prognostic model using QuantCRC to predict recurrence-free survival (time from treatment assignment to first recurrence or all-cause death).

"QuantCRC can identify different regions within the tumor and extract quantitative data from these regions," Dr. Rish Pai, a pathologist at the Mayo Clinic and senior author on this study, said in a press release. "The algorithm converts an image into a set of numbers that is unique to that tumor. The large number of tumors that we analyzed allowed us to learn which features were most predictive of tumor behavior. We can now apply what we have learned to new colon cancers to predict how the tumor will behave."

Artificial intelligence could also assist in guiding cancer treatment. If predicted correctly, the algorithm would be able to identify a subset of patients who would not have to receive chemotherapy because their risk of recurrence was low. On the other hand, AI may identify patients who would benefit from more intense therapies or follow up because their risk of recurrence is higher.

Using AI to predict colorectal cancer recurrence may benefit several groups including patients with colon cancer, oncologists who treat colon cancer and pathologists who assess colon cancer specimens.

“For patients with colon cancer, the algorithm gives oncologists another tool to help guide therapy and follow up," Pai explained.

In the next steps, Pai noted that he plans to evaluate if QuantCRC can be used to better understand the mechanisms of tumor recurrence and see if it can predict patients’ response to some treatments such as immunotherapy.

For more news on cancer updates, research and education, don’t forget to subscribe to CURE®’s newsletters here.