Overcoming the Interobserver Variability in Lung Adenocarcinoma Subtyping: A Clustering Approach to Establish a Ground Truth for Downstream Applications.

Autor:	Lami K; From the Departments of Pathology (Lami, K. Tanaka, Fukuoka), Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan., Bychkov A; Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan; the Department of Pathology, Kameda Medical Center, Kamogawa, Japan (Bychkov)., Matsumoto K; Surgical Oncology (Matsumoto, Miyazaki, Tsuchiya, Nagayasu), Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan., Attanoos R; The Department of Cellular Pathology, Cardiff University, Cardiff, United Kingdom (Attanoos)., Berezowska S; The Institute of Pathology, Centre Hospitalier Universitaire Vaudois, University of Lausanne, Lausanne, Switzerland (Berezowska)., Brcic L; The Diagnostic and Research Institute of Pathology, Medical University of Graz, Graz, Austria (Brcic)., Cavazza A; The Unit of Pathologic Anatomy, Azienda USL/IRCCS di Reggio Emilia, Reggio Emilia, Italy (Cavazza)., English JC; The Department of Pathology, Vancouver General Hospital, Vancouver, British Columbia, Canada (English)., Fabro AT; The Department of Pathology and Legal Medicine, Ribeirão Preto Medical School, University of São Paulo, Ribeirão Preto, Brazil (Fabro)., Ishida K; The Department of Pathology, Kansai Medical University, Osaka, Japan (Ishida)., Kashima Y; The Department of Pathology, Hyogo Prefectural Awaji Medical Center, Sumoto, Japan (Kashima)., Larsen BT; The Department of Laboratory Medicine and Pathology, Mayo Clinic, Scottsdale, Arizona (Larsen, Smith)., Marchevsky AM; The Department of Pathology, Cedars-Sinai Medical Center, Los Angeles, California (Marchevsky)., Miyazaki T; Surgical Oncology (Matsumoto, Miyazaki, Tsuchiya, Nagayasu), Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan., Morimoto S; The Innovation Platform & Office for Precision Medicine (Morimoto), Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan., Roden AC; The Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota (Roden)., Schneider F; The Department of Pathology and Laboratory Medicine, Emory University, Atlanta, Georgia (Schneider)., Soshi M; BonBon Co, Ltd, Kyoto, Japan (Soshi)., Smith ML; The Department of Laboratory Medicine and Pathology, Mayo Clinic, Scottsdale, Arizona (Larsen, Smith)., Tabata K; The Department of Pathology, Kagoshima University, Kagoshima, Japan (Tabata)., Takano AM; The Department of Anatomical Pathology, Singapore General Hospital, Singapore, Singapore (Takano)., Tanaka K; From the Departments of Pathology (Lami, K. Tanaka, Fukuoka), Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan., Tanaka T; The Department of Diagnostic Pathology, Kobe University Hospital, Kobe, Japan (T. Tanaka)., Tsuchiya T; Surgical Oncology (Matsumoto, Miyazaki, Tsuchiya, Nagayasu), Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan., Nagayasu T; Surgical Oncology (Matsumoto, Miyazaki, Tsuchiya, Nagayasu), Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan., Fukuoka J; From the Departments of Pathology (Lami, K. Tanaka, Fukuoka), Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan.
Jazyk:	angličtina
Zdroj:	Archives of pathology & laboratory medicine [Arch Pathol Lab Med] 2023 Aug 01; Vol. 147 (8), pp. 885-895.
DOI:	10.5858/arpa.2022-0051-OA
Abstrakt:	Context.—: The accurate identification of different lung adenocarcinoma histologic subtypes is important for determining prognosis but can be challenging because of overlaps in the diagnostic features, leading to considerable interobserver variability. Objective.—: To provide an overview of the diagnostic agreement for lung adenocarcinoma subtypes among pathologists and to create a ground truth using the clustering approach for downstream computational applications. Design.—: Three sets of lung adenocarcinoma histologic images with different evaluation levels (small patches, areas with relatively uniform histology, and whole slide images) were reviewed by 17 international expert lung pathologists and 1 pathologist in training. Each image was classified into one or several lung adenocarcinoma subtypes. Results.—: Among the 4702 patches of the first set, 1742 (37%) had an overall consensus among all pathologists. The overall Fleiss κ score for the agreement of all subtypes was 0.58. Using cluster analysis, pathologists were hierarchically grouped into 2 clusters, with κ scores of 0.588 and 0.563 in clusters 1 and 2, respectively. Similar results were obtained for the second and third sets, with fair-to-moderate agreements. Patches from the first 2 sets that obtained the consensus of the 18 pathologists were retrieved to form consensus patches and were regarded as the ground truth of lung adenocarcinoma subtypes. Conclusions.—: Our observations highlight discrepancies among experts when assessing lung adenocarcinoma subtypes. However, a subsequent number of consensus patches could be retrieved from each cluster, which can be used as ground truth for the downstream computational pathology applications, with minimal influence from interobserver variability. Competing Interests: This paper is based on results obtained from a project, 19100971-0, commissioned by the New Energy and Industrial Technology Development Organization (NEDO). (© 2023 College of American Pathologists.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu Plný text ve formátu PDF