Researchers at the ECOG-ACRIN Cancer Research Group (ECOG-ACRIN) say artificial intelligence can enhance current methods for predicting the risk of head and neck cancer spreading beyond the borders of cervical lymph nodes . A customized deep learning algorithm using standard computed tomography (CT) scan images and associated data provided by patients enrolled in the E3311 Phase 2 trial specifically targeted human papillomavirus (HPV)-associated head and neck cancer. The E3311 validated dataset has the potential to contribute to more accurate staging of disease and prediction of risk.
Benjamin Kann, MD (Dana-Farber Cancer Institute, Brigham and Women’s Hospital, Harvard Medical School) led the ECOG-ACRIN study. He plans to present his findings at the American Society for Radiation Oncology (ASTRO) annual meeting in San Antonio, Texas.
“This type of research is important because it helps identify patients with high-risk, aggressive disease and also helps select patients for treatment escalation,” said Dr. Kann. .
Head and neck cancer and its standard treatments (surgery, radiation, or chemotherapy) are associated with significant morbidity. They affect how a person looks, speaks, eats, and breathes. Therefore, there is great interest in developing low-intensity treatment strategies for patients. For example, in the completed E3311 phase 3 trial, oral surgery followed by low-dose radiation of 50 Grays (Gy) without chemotherapy was associated with very high survival rates and excellent quality of life in patients at intermediate risk of recurrence. (Ferris RL. J. Klin OnkDecember 2021).
Dr. Kann and colleagues have developed a neural network-based deep learning algorithm Diagnostic computed tomography (CT) scans, pathology, and clinical dataThe source was a cohort of participants in the E3311 trial who were assessed at high risk of recurrence by standard pathological and clinical measures.
“Head and neck staging cancer Dr. Kann said:
Factors that determine the stage of cancer include the size of the original tumor, the number of lymph nodes involved, and extranodal extension (when malignant cells spread beyond the boundaries of the neck lymph nodes into surrounding tissues). I have. For E3311, patients were assessed as high risk if the extranodal extension (ENE) was 1 mm or greater. These patients were assigned to chemotherapy and high-dose radiation (66 Gy) after transoral surgery.
Dr. Kann and colleagues obtained, as available, pretreatment computed tomography (CT) scans and corresponding surgical pathology reports from the E3311 high-risk cohort. From the 177 scans collected, 311 nodes were annotated. 71 (23%) had an ENE and 39 (13%) had his ENE ≥1 mm.
The tool performed well in predicting ENE, significantly outperforming reviews by expert head and neck radiologists.
“The deep learning algorithm correctly classified 85% of the nodes as having ENE compared to 70% for radiologists,” said Dr. Kann. “In terms of specificity and sensitivity, deep learning his algorithm he was 78% accurate, while the radiologist he was 62% accurate.”
The team plans to evaluate the dataset as part of an upcoming head and neck therapeutic trial. neck cancerThe algorithm will be evaluated for its potential to improve current staging and risk assessment methods.
“Our ability to develop biomarkers from standard CT scan images is an exciting new area of clinical research that will help individual It gives us hope that we can better tailor treatment to patients,” said senior author Barbara A. Bartness, M.D., Ph.D.
141 Deep Learning Screening for Extranodal Extension: An Evaluation in ECOG-ACRIN E3311, A Randomized, Scaled-down Study of HPV-Associated Oropharyngeal Cancer, plan.core-apps.com/myastroapp2…c7071c5947c71a441519
Provided by ECOG-ACRIN Cancer Research Group
Quote: Head and Neck Cancer Researchers Demonstrate Capabilities of Deep Learning Algorithm in Post-Surgical Setting (October 22, 2022) from https://medicalxpress.com/news/2022-10-neck-cancer-capability Retrieved October 22, 2022-deep-algorithm.html
This document is subject to copyright. No part may be reproduced without written permission, except in fair trade for personal research or research purposes. Content is provided for informational purposes only.