Article Type
Changed
Tue, 10/11/2022 - 15:27

Using artificial intelligence, researchers have developed an algorithm that can help improve the prediction of colorectal cancer (CRC) recurrence.

The QuantCRC algorithm can identify patients with CRC who might be able to skip chemotherapy, given a low probability of recurrence, and identify those patients at high risk for recurrence who may benefit from more intensive treatment or follow-up, the researchers say.

“For patients with colon cancer, the algorithm gives oncologists another tool to help guide therapy and follow-up,” Rish Pai, MD, PhD, a pathologist at Mayo Clinic, Phoenix, who developed the tool, said in a news release.

The study was published online in the journal Gastroenterology.

The tool is a deep-learning segmentation algorithm developed using 6,468 digitized CRC images. It quantifies 15 features from a CRC image and uses them to improve prediction of recurrence.

“QuantCRC can identify different regions within the tumor and extract quantitative data from these regions,” Dr. Pai explained.

“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,” Dr. Pai said.

The researchers developed a prognostic model incorporating stage, mismatch repair, and QuantCRC that resulted in a concordance (c)-index of 0.714 in the internal test and 0.744 in the external cohort. Removing QuantCRC from the model reduced the c-index to 0.679 in the external cohort.

Using QuantCRC, they identified prognostic risk groups for recurrence, which provided a hazard ratio of 2.24 for low- versus high-risk stage III CRC and 2.36 for low- versus high-risk stage II CRC, in the external cohort after adjusting for established risk factors.

The predicted median 36-month recurrence rate for high-risk stage III CRC was 32.7% versus 13.4% for low-risk stage III CRC and 15.8% for high-risk stage II CRC versus 5.4% for low-risk stage II CRC, the researchers report.

QuantCRC provides a “powerful adjunct” to routine pathologic reporting of CRC, and a prognostic model using QuantCRC can improve prediction of recurrence-free survival, they write.

Looking ahead, Dr. Pai plans to use QuantCRC to better understand mechanisms of tumor recurrence and see if it can predict the response to certain treatments, like immunotherapy, he said.

Funding for this study was provided in part by the Colon Cancer Family Registry, which is supported in part by funding from the National Cancer Institute and National Institutes of Health. Dr. Pai reports consulting income from Alimentiv, Eli Lilly, AbbVie, Allergan, Genentech, and PathAI outside of the submitted work.

A version of this article first appeared on Medscape.com.

Publications
Topics
Sections

Using artificial intelligence, researchers have developed an algorithm that can help improve the prediction of colorectal cancer (CRC) recurrence.

The QuantCRC algorithm can identify patients with CRC who might be able to skip chemotherapy, given a low probability of recurrence, and identify those patients at high risk for recurrence who may benefit from more intensive treatment or follow-up, the researchers say.

“For patients with colon cancer, the algorithm gives oncologists another tool to help guide therapy and follow-up,” Rish Pai, MD, PhD, a pathologist at Mayo Clinic, Phoenix, who developed the tool, said in a news release.

The study was published online in the journal Gastroenterology.

The tool is a deep-learning segmentation algorithm developed using 6,468 digitized CRC images. It quantifies 15 features from a CRC image and uses them to improve prediction of recurrence.

“QuantCRC can identify different regions within the tumor and extract quantitative data from these regions,” Dr. Pai explained.

“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,” Dr. Pai said.

The researchers developed a prognostic model incorporating stage, mismatch repair, and QuantCRC that resulted in a concordance (c)-index of 0.714 in the internal test and 0.744 in the external cohort. Removing QuantCRC from the model reduced the c-index to 0.679 in the external cohort.

Using QuantCRC, they identified prognostic risk groups for recurrence, which provided a hazard ratio of 2.24 for low- versus high-risk stage III CRC and 2.36 for low- versus high-risk stage II CRC, in the external cohort after adjusting for established risk factors.

The predicted median 36-month recurrence rate for high-risk stage III CRC was 32.7% versus 13.4% for low-risk stage III CRC and 15.8% for high-risk stage II CRC versus 5.4% for low-risk stage II CRC, the researchers report.

QuantCRC provides a “powerful adjunct” to routine pathologic reporting of CRC, and a prognostic model using QuantCRC can improve prediction of recurrence-free survival, they write.

Looking ahead, Dr. Pai plans to use QuantCRC to better understand mechanisms of tumor recurrence and see if it can predict the response to certain treatments, like immunotherapy, he said.

Funding for this study was provided in part by the Colon Cancer Family Registry, which is supported in part by funding from the National Cancer Institute and National Institutes of Health. Dr. Pai reports consulting income from Alimentiv, Eli Lilly, AbbVie, Allergan, Genentech, and PathAI outside of the submitted work.

A version of this article first appeared on Medscape.com.

Using artificial intelligence, researchers have developed an algorithm that can help improve the prediction of colorectal cancer (CRC) recurrence.

The QuantCRC algorithm can identify patients with CRC who might be able to skip chemotherapy, given a low probability of recurrence, and identify those patients at high risk for recurrence who may benefit from more intensive treatment or follow-up, the researchers say.

“For patients with colon cancer, the algorithm gives oncologists another tool to help guide therapy and follow-up,” Rish Pai, MD, PhD, a pathologist at Mayo Clinic, Phoenix, who developed the tool, said in a news release.

The study was published online in the journal Gastroenterology.

The tool is a deep-learning segmentation algorithm developed using 6,468 digitized CRC images. It quantifies 15 features from a CRC image and uses them to improve prediction of recurrence.

“QuantCRC can identify different regions within the tumor and extract quantitative data from these regions,” Dr. Pai explained.

“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,” Dr. Pai said.

The researchers developed a prognostic model incorporating stage, mismatch repair, and QuantCRC that resulted in a concordance (c)-index of 0.714 in the internal test and 0.744 in the external cohort. Removing QuantCRC from the model reduced the c-index to 0.679 in the external cohort.

Using QuantCRC, they identified prognostic risk groups for recurrence, which provided a hazard ratio of 2.24 for low- versus high-risk stage III CRC and 2.36 for low- versus high-risk stage II CRC, in the external cohort after adjusting for established risk factors.

The predicted median 36-month recurrence rate for high-risk stage III CRC was 32.7% versus 13.4% for low-risk stage III CRC and 15.8% for high-risk stage II CRC versus 5.4% for low-risk stage II CRC, the researchers report.

QuantCRC provides a “powerful adjunct” to routine pathologic reporting of CRC, and a prognostic model using QuantCRC can improve prediction of recurrence-free survival, they write.

Looking ahead, Dr. Pai plans to use QuantCRC to better understand mechanisms of tumor recurrence and see if it can predict the response to certain treatments, like immunotherapy, he said.

Funding for this study was provided in part by the Colon Cancer Family Registry, which is supported in part by funding from the National Cancer Institute and National Institutes of Health. Dr. Pai reports consulting income from Alimentiv, Eli Lilly, AbbVie, Allergan, Genentech, and PathAI outside of the submitted work.

A version of this article first appeared on Medscape.com.

Publications
Publications
Topics
Article Type
Sections
Article Source

FROM GASTROENTEROLOGY

Disallow All Ads
Content Gating
No Gating (article Unlocked/Free)
Alternative CME
Disqus Comments
Default
Use ProPublica
Hide sidebar & use full width
render the right sidebar.
Conference Recap Checkbox
Not Conference Recap
Clinical Edge
Display the Slideshow in this Article
Medscape Article
Display survey writer
Reuters content
Disable Inline Native ads
WebMD Article