Head and neck cancer researchers demonstrate

Benjamin Kann, MD

image: “This type of research is critical because it can help identify patients with high-risk aggressive head and neck cancer and also aid in the selection of appropriate patients for de-escalation of therapy,” says Dr. Benjamin Kann, who led the study.
outlook more

Credit: Artificial Intelligence in Medicine Program, Brigham and Women’s Hospital

According to researchers at the ECOG-ACRIN Cancer Research Group (ECOG-ACRIN), artificial intelligence can augment current methods to predict the risk of head and neck cancer spreading beyond the borders of the cervical lymph nodes. A customized deep learning algorithm using standard computed tomography (CT) scan images and associated data contributed by patients enrolled in the Phase 2 E3311 study shows promise, particularly for patients with a new diagnosis of Head and neck cancer linked to human papillomavirus (HPV). . The validated E3311 dataset has the potential to contribute to more accurate disease staging and risk prediction.

Benjamin Kann, MD (Dana-Farber Cancer Institute, Brigham and Women’s Hospital, Harvard Medical School) led the study for ECOG-ACRIN. He will present the findings during the American Society of Radiation Oncology (ASTRO) annual meeting in San Antonio, Texas.

“This type of research is critical because it can help identify patients with high-risk, aggressive disease and select appropriate patients for de-escalation of therapy,” said Dr. Can.

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, or breathes. Therefore, there is great interest in developing less intensive treatment strategies for patients. For example, the completed Phase 3 E3311 study showed that low-dose 50 Gray (Gy) radiation without chemotherapy after transoral surgery resulted in very high survival and excellent quality of life in patients at intermediate risk of recurrence (Ferris RL. J Clin Onc. December 2021).

dr Kann and colleagues developed and validated a neural network-based deep learning algorithm based on diagnostic computed tomography (CT) scans, pathology, and clinical data. The source was the cohort of participants in the E3311 study who were identified as being at high risk of recurrence using standard pathological and clinical measures.

“Staging of head and neck cancer is a challenging clinical problem,” said Dr. Can. “In particular, our current efforts to identify extranodal lengthening through human interpretation of pre-treatment imaging have shown generally poor results.”

Factors that determine cancer stage include the size of the original tumor, the number of lymph nodes involved, and extranodal extent — when malignant cells spread beyond the borders of the cervical lymph nodes into the surrounding tissue. In E3311, patients with an extranodal extension (ENE) ≥ 1 mm were classified as high-risk patients. These patients were referred to chemotherapy and high-dose radiation (66 Gy) after transoral surgery.

dr Kann and colleagues obtained pre-treatment computed tomography (CT) scans and corresponding surgical pathology reports from high-risk cohort E3311 when available. Of 177 scans collected, 311 nodules were annotated: 71 (23%) with ENE and 39 (13%) with ≥ 1 mm ENE.

The tool showed high performance in predicting ENE, significantly outperforming reviews by experienced head and neck radiologists.

“The deep learning algorithm accurately classified 85% of the nodes as ENE compared to 70% of the radiologists,” said Dr. Can. “In terms of specificity and sensitivity, the deep learning algorithm was 78% accurate versus 62% for the radiologists.”

The team plans to use the dataset as part of future head and neck cancer treatment studies. The algorithm will be evaluated for its potential to improve current disease staging and risk assessment methods.

“Our ability to develop biomarkers from standard CT scan images is an exciting new area of ​​clinical research and gives hope that we will be able to better tailor treatment for individual patients, including deciding when to start one.” Surgery is best used and in whom the extent can be reduced of treatment,” said the senior author Barbara A Burtness, MD.

dr Burtness is Professor of Medicine and co-director of the Developmental Therapeutics Research Program at Yale Cancer Center, Chair of the ECOG-ACRIN Head and Neck Committee, and Chair of the ECOG-ACRIN Task Force on Advancement for Women.

Summary 141 (Screening for extranodal extension with deep learning: Assessment in ECOG-ACRIN E3311, a randomized de-escalation study in HPV-associated oropharyngeal cancer is a scientific highlight in the session “Variations on a Theme: De-Intensification Strategies for HPV+ Oropharynx Cancer” on Monday, October 24 at 11:15 a.m. Central Time.

ASTRO has Dr. May present his Basic/Translational Science Award for his novel research.

dr Can also be a panelist in one educational session tailored for Head and Neck Cancer Practitioners on Wednesday, October 26 from 8:00-9:00 p.m. Central Time. The session aims to break down barriers in understanding artificial intelligence and encourage future adoption.

For more informations, see the profile of dr. Can be found on the ASTRO website.

Co-authors include Benjamin H Kann, Jirapat Likitlersuang, Zezhong Ye, Sanjay Aneja, Henry S Park, Richard Bakst, Hillary R Kelly, Amy F Juliano, Sam Payabvash, Jeffrey P Guenette, Hugo JWL Aerts, Rathan M Subramaniam, Robert L Ferris, and Barbara A Burtness

This study was supported by the ECOG-ACRIN Cancer Research Group (Peter J. O’Dwyer, MD and Mitchell D. Schnall, MD, PhD, group co-chairs) and the National Cancer Institute of the National Institutes of Health under the following conditions supports award numbers: U10CA180794, U10CA180820, UG1CA233180, UG1CA233184, UG1CA233337, UG1CA233253 and UG1CA232760.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of the press releases published on EurekAlert! by contributing institutions or for the use of information about the EurekAlert system.

READ :  AI creates a sharper image of M87's black hole