AI Algorithm for Lung Adenocarcinoma Pattern Quantification (PATQUANT): International Validation and Advanced Risk Stratification Superior to Conventional Grading

MedComm, vol. 6, pp. e70380

Abstract

ABSTRACT The morphological patterns of lung adenocarcinoma (LUAD) are recognized for their prognostic significance, with ongoing debate regarding the optimal grading strategy. This study aimed to develop a clinical‐grade, fully quantitative, and automated tool for pattern classification/quantification (PATQUANT), to evaluate existing grading strategies, and determine the optimal grading system. PATQUANT was trained on a high‐quality dataset, manually annotated by expert pathologists. Several independent test datasets and 13 expert pathologists were involved in validation. Five large, multinational cohorts of resectable LUAD (patient n = 1120) were analyzed concerning prognostic value. PATQUANT demonstrated excellent pattern segmentation/classification accuracy and outperformed 8 out of 13 pathologists. The prognostic study revealed a distinct prognostic profile for the complex glandular pattern. While all contemporary grading systems had prognostic value, the predominant pattern‐based and simplified IASLC systems were superior. We propose and validate two new, fully explainable grading principles, providing fine‐grained, statistically independent patient risk stratification. We developed a fully automated, robust AI tool for pattern analysis/quantification that surpasses the performance of experienced pathologists. Additionally, we demonstrate the excellent prognostic capabilities of two new grading approaches that outperform traditional grading methods. We make our extensive agreement dataset publicly available to advance the developments in the field.

Authors 16

1. Yuan Wang

   University Hospital Cologne

   Affiliation as printed

   Institute of Pathology University Hospital Cologne Cologne Germany

   Institute of Pathology, University Hospital Cologne, Cologne, Germany

2. Kris Lami

   Nagasaki University

   Affiliation as printed

   Department of Pathology Informatics Nagasaki University Nagasaki Japan

   Department of Pathology Informatics, Nagasaki University, Nagasaki, Japan

3. Simon Schallenberg

   Charité - Universitätsmedizin Berlin

   Affiliation as printed

   Institute of Pathology Charité Berlin Germany

4. Andrey Bychkov

   Kameda Medical Center · Nagasaki University

   Affiliation as printed

   Department of Pathology Informatics Nagasaki University Nagasaki Japan

   Department of Pathology Kameda Medical Center Kamogawa Japan

   Department of Pathology Informatics, Nagasaki University, Nagasaki, Japan

   Department of Pathology, Kameda Medical Center, Kamogawa, Japan

5. Yuanzi Ye

   First Affiliated Hospital of Anhui Medical University

   Affiliation as printed

   Department of Pathology The First Affiliated Hospital of Anhui Medical University Hefei China

   Department of Pathology, The First Affiliated Hospital of Anhui Medical University, Hefei, China

6. Danny David Jonigk Aachen Institute of Pathology

   Universitätsklinikum Aachen · German Center for Lung Research

   Affiliation as printed

   German Center for Lung Research DZL BREATH Hanover Germany

   Institute of Pathology University Hospital Aachen Aachen Germany

   German Center for Lung Research, DZL, BREATH, Hanover, Germany

   Institute of Pathology, University Hospital Aachen, Aachen, Germany

7. Xiaoya Zhu

   Zhejiang University · Sir Run Run Shaw Hospital

   Affiliation as printed

   Department of Pathology of Run Run Shaw Hospital Affiliated to Zhejiang University Hangzhou China

   Department of Pathology of Run Run Shaw Hospital Affiliated to Zhejiang University, Hangzhou, China

8. Sofia Campelos

   IMP Diagnostics (Portugal)

   Affiliation as printed

   IMP Diagnostics Porto Portugal

   IMP Diagnostics, Porto, Portugal

9. Anne Maria Schultheis

   University of Freiburg · University Medical Center Freiburg

   Affiliation as printed

   Institute of Pathology University Hospital Freiburg Freiburg Germany

   Institute of Pathology, University Hospital Freiburg, Freiburg, Germany

10. Matthias B. Heldwein

    University Hospital Cologne

    Affiliation as printed

    Department of Cardiothoracic Surgery University Hospital Cologne Cologne Germany

    Department of Cardiothoracic Surgery, University Hospital Cologne, Cologne, Germany

11. Alexander Quaas

    University Hospital Cologne

    Affiliation as printed

    Institute of Pathology University Hospital Cologne Cologne Germany

    Institute of Pathology, University Hospital Cologne, Cologne, Germany

12. Aleš Ryška

    Charles University · University Hospital Hradec Králové

    Affiliation as printed

    The Fingerland Department of Pathology Charles University Medical Faculty Hospital Hradec Králové Czech Republic

13. Andre Luis Moreira

    New York University

    Affiliation as printed

    Department of Pathology NYU Grossman School of Medicine New York New York USA

    Department of Pathology, NYU Grossman School of Medicine, New York, New York, USA

14. Junya Fukuoka

    Kameda Medical Center · Nagasaki University

    Affiliation as printed

    Department of Pathology Informatics Nagasaki University Nagasaki Japan

    Department of Pathology Kameda Medical Center Kamogawa Japan

    Department of Pathology Informatics, Nagasaki University, Nagasaki, Japan

    Department of Pathology, Kameda Medical Center, Kamogawa, Japan

15. Reinhard Büttner corresponding

    University Hospital Cologne

    Affiliation as printed

    Institute of Pathology University Hospital Cologne Cologne Germany

    Institute of Pathology, University Hospital Cologne, Cologne, Germany

16. Yuri Tolkach corresponding

    University Hospital Cologne

    Affiliation as printed

    Institute of Pathology University Hospital Cologne Cologne Germany

    Institute of Pathology, University Hospital Cologne, Cologne, Germany