A Nationwide Study of Interobserver Agreement and Accuracy in HER2 Immunohistochemistry and the Impact of Artificial Intelligence on HER2 Interpretation

Purpose: Accurate assessment of low levels of human epidermal growth factor receptor 2 (HER2) expression has become increasingly important with the expansion of HER2-targeted therapies to patients with HER2-low and HER2-ultralow breast cancer (BC). Following the introduction of revised College of American Pathologists (CAP) guidelines incorporating the HER2-ultralow (0+) category, we conducted a nationwide multireader study to evaluate interobserver variation in HER2 interpretation and to investigate the impact of artificial intelligence (AI)-assisted image analysis. Materials and Methods: A total of 168 BC cases stained using the VENTANA HER2 (4B5) assay were digitized as whole-slide images (WSIs). Thirty-four board-certified breast pathologists independently reviewed all cases without AI assistance and repeated the assessment after a four-week washout period using AI support. Interobserver agreement, classification performance, and reading time were compared between the two phases. Results: AI assistance significantly improved interobserver agreement (Krippendorff's alpha, 0.7415 to 0.8849; Fleiss' kappa, 0.4492 to 0.7524; p<0.001). The largest gains in diagnostic performance were observed in HER2-ultralow and HER2 IHC 1+ BCs, with sensitivities improving from 48.5% to 73.9% and from 64.9% to 87.7%, respectively (p<0.001). Although overall reading time was not reduced in the unadjusted analysis, AI assistance shortened interpretation time in HER2-ultralow BCs by 3.12 seconds (5.1% reduction). Conclusion: AI-assisted assessment improved the reproducibility and diagnostic performance of HER2 interpretation, with the greatest benefit observed in clinically relevant HER2 low-expression categories. The reduced interpretation time in HER2-ultralow BCs further suggests that AI can improve efficiency in diagnostically challenging cases encountered in routine practice.

Authors

Institutions

Publication Details

Journal
Cancer Research and Treatment
Published
2026-11-02
DOI
https://doi.org/10.4143/crt.2026.0805
Primary Topic
AI in cancer detection
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

A Nationwide Study of Interobserver Agreement and Accuracy in HER2 Immunohistochemistry and the Impact of Artificial Intelligence on HER2 Interpretation

Eun Yoon Cho, Gyungyub Gong, Jinye Yoo, Jiwon Koh et al.
Cancer Research and Treatment
AI in cancer detection
article

A Nationwide Study of Interobserver Agreement and Accuracy in HER2 Immunohistochemistry and the Impact of Artificial Intelligence on HER2 Interpretation

Eun Yoon Cho, Gyungyub Gong, Jinye Yoo, Jiwon Koh, Ho-chang Lee, Yeonjin Jeon, Yoon Jin Cha, Sun‐Young Jun, Gui Young Kwon, Byung-Kwan Jeong, Min Chong Kim, Young Kyung Bae, Ji Min Na, Youjeong Seo, Ji‐Seon Jeong, Young Hee Maeng, Sunyoung Lee, Miseon Lee, Song-Hee Han, Yoon Ah Cho, Hyun Joo Choi, Ae Ri Ahn, Ji Yun Shin, Junjeong Choi, Jeong Yun Shim, Hera Jung, Wan Seop Kim, So Yeon Park, Jee Yeon Kim, Hye Jeong Choi, Somang Lee, Yo Han Jeon, Chungyeul Kim, Hyunwoo Lee, Su Jung Kum, Nora Jee-Young Park, Gyu Young Kim, Hyang Im Lee, Hee Jin Lee, Ji-Young Kim, Hee Sung Kim
article en

Abstract

Purpose: Accurate assessment of low levels of human epidermal growth factor receptor 2 (HER2) expression has become increasingly important with the expansion of HER2-targeted therapies to patients with HER2-low and HER2-ultralow breast cancer (BC). Following the introduction of revised College of American Pathologists (CAP) guidelines incorporating the HER2-ultralow (0+) category, we conducted a nationwide multireader study to evaluate interobserver variation in HER2 interpretation and to investigate the impact of artificial intelligence (AI)-assisted image analysis. Materials and Methods: A total of 168 BC cases stained using the VENTANA HER2 (4B5) assay were digitized as whole-slide images (WSIs). Thirty-four board-certified breast pathologists independently reviewed all cases without AI assistance and repeated the assessment after a four-week washout period using AI support. Interobserver agreement, classification performance, and reading time were compared between the two phases. Results: AI assistance significantly improved interobserver agreement (Krippendorff's alpha, 0.7415 to 0.8849; Fleiss' kappa, 0.4492 to 0.7524; p<0.001). The largest gains in diagnostic performance were observed in HER2-ultralow and HER2 IHC 1+ BCs, with sensitivities improving from 48.5% to 73.9% and from 64.9% to 87.7%, respectively (p<0.001). Although overall reading time was not reduced in the unadjusted analysis, AI assistance shortened interpretation time in HER2-ultralow BCs by 3.12 seconds (5.1% reduction). Conclusion: AI-assisted assessment improved the reproducibility and diagnostic performance of HER2 interpretation, with the greatest benefit observed in clinically relevant HER2 low-expression categories. The reduced interpretation time in HER2-ultralow BCs further suggests that AI can improve efficiency in diagnostically challenging cases encountered in routine practice.

Cancer Research and Treatment
Seoul National University (KR), Hallym University (KR), Inha University (KR), Yonsei University (KR), Konkuk University (KR), Dong-A University Hospital (KR), Seoul Metropolitan Government (KR), Kosin University Gospel Hospital (KR), Asan Medical Center (KR), Samsung Medical Center (KR), Seoul National University Hospital (KR), Seoul National University Bundang Hospital (KR), Kyungpook National University (KR), University of Ulsan (KR), Konkuk University Medical Center (KR), The Catholic University of Korea St. Vincent's Hospital (KR), SNUH SMG-SNU Boramae Medical Center (KR), The Catholic University of Korea Incheon St. Mary's Hospital (KR), Keimyung University Dongsan Medical Center (KR), Jeonbuk National University Hospital (KR), Ulsan University Hospital (KR), Gangnam Severance Hospital (KR), Chonnam National University Hwasun Hospital (KR), CHA University (KR), Pusan National University Yangsan Hospital (KR), Hallym University Dongtan Sacred Heart Hospital (KR), Incheon Catholic University (KR), Inha University Hospital (KR), Chung-Ang University Hospital (KR), The Catholic University of Korea Seoul St. Mary's Hospital (KR), CHA University Gangnam Medical Center (KR), Jeju National University Hospital (KR), Yeungnam University College (KR), Kyungpook National University Chilgok Hospital (KR), Korea University Guro Hospital (KR), CHA University Ilsan Medical Center (KR), Keimyung University Dongsan Hospital (KR), Chung-Ang University Gwangmyeong Hospital (KR), Chung-Ang University Healthcare System (KR), Dong-A University (KR), Keimyung University (KR), Yeungnam University (KR), Chung-Ang University (KR), Jeonbuk National University (KR), Jeju National University (KR), Sungkyunkwan University (KR), Catholic University of Korea (KR), Kosin University (KR)
Openalex Percentile: Top 16%
AI in cancer detection
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon