Using synthetic intelligence (AI), researchers have developed an algorithm that may assist enhance the prediction of colorectal cancer (CRC) recurrence.
The QuantCRC algorithm can establish sufferers with CRC who may have the ability to skip chemotherapy, given a low chance of recurrence, and establish these sufferers at excessive threat for recurrence who could profit from extra intensive remedy or follow-up, the researchers say.
“For sufferers with colon cancer, the algorithm provides oncologists one other software to assist information remedy and follow-up,” Rish Pai, MD, PhD, a pathologist at Mayo Clinic in Arizona, who developed the software, stated in a information launch.
The research is published online within the journal Gastroenterology.
Predicting Tumor Behavior
The software is a deep-learning segmentation algorithm developed utilizing 6468 digitized CRC photos. It quantifies 15 options from a CRC picture and makes use of them to enhance prediction of recurrence.
“QuantCRC can identify different regions within the tumor and extract quantitative data from these regions,” Pai defined.
“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,” Pai stated.
The researchers developed a prognostic mannequin incorporating stage, mismatch restore, and QuantCRC, that resulted in a concordance (c)-index of 0.714 within the inner take a look at and 0.744 within the exterior cohort. Removing QuantCRC from the mannequin diminished the c-index to 0.679 within the exterior cohort.
Using QuantCRC, they recognized prognostic threat teams for recurrence, which offered a hazard ratio of two.24 for low- vs high-risk stage III CRC and a pair of.36 for low- vs high-risk stage II CRC, within the exterior cohort after adjusting for established threat elements.
The predicted median 36-month recurrence fee for high-risk stage III CRC was 32.7% vs 13.4% for low-risk stage III CRC, and 15.8% for high-risk stage II CRC vs 5.4% for low-risk stage II CRC, the researchers report.
QuantCRC offers a “powerful adjunct” to routine pathologic reporting of CRC, and a prognostic mannequin utilizing QuantCRC can enhance prediction of recurrence-free survival, they write.
Looking forward, Pai plans to make use of QuantCRC to raised perceive mechanisms of tumor recurrence and see if it will possibly predict the response to sure therapies, like immunotherapy, he stated.
Funding for this research was offered partly by the Colon Cancer Family Registry, which is supported partly by funding from the National Cancer Institute and National Institutes of Health. Pai reviews consulting revenue from Alimentiv Inc., Eli Lilly, AbbVie, Allergan, Genentech, and PathAI exterior of the submitted work.
Gastroenterology. Published on-line August 16, 2022. Abstract