EUVIMEDEUVIMED entdecken
Crossref

Development of a deep‐learning model tailored for HER2 detection in breast cancer to aid pathologists in interpreting HER2 ‐low cases

Pierre-Antoine Bannier iD, Glenn Broeckx, Loïc Herpin, Rémy Dubois, Lydwine Van Praet, Charles Maussion, Frederik Deman, Ellen Amonoo, Anca Mera, Jasmine Timbres, Cheryl Gillett, Elinor Sawyer und 5 weitere

Histopathology · 2024-09 · 85 (3) · 478-488

DOI: 10.1111/his.15274 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Histopathology, online erschienen am 14.07.2024. Verfasst von Pierre-Antoine Bannier und 16 weiteren Autor:innen. Beteiligte Einrichtung: Owkin France, Paris, France. Hauptthemen laut MEDLINE: Breast Neoplasms, Erb-b2 Receptor Tyrosine Kinases, Deep Learning, Immunohistochemistry. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 9 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aims

Over 50% of breast cancer cases are “Human epidermal growth factor receptor 2 (HER2) low breast cancer (BC)”, characterized by HER2 immunohistochemistry (IHC) scores of 1+ or 2+ alongside no amplification on fluorescence in situ hybridization (FISH) testing. The development of new anti‐HER2 antibody‐drug conjugates (ADCs) for treating HER2‐low breast cancers illustrates the importance of accurately assessing HER2 status, particularly HER2‐low breast cancer. In this study we evaluated the performance of a deep‐learning (DL) model for the assessment of HER2, including an assessment of the causes of discordances of HER2‐Null between a pathologist and the DL model. We specifically focussed on aligning the DL model rules with the ASCO/CAP guidelines, including stained cells' staining intensity and completeness of membrane staining. Methods and Results We trained a DL model on a multicentric cohort of breast cancer cases with HER2‐IHC scores ( n = 299). The model was validated on two independent multicentric validation cohorts ( n = 369 and n = 92), with all cases reviewed by three senior breast pathologists. All cases underwent a thorough review by three senior breast pathologists, with the ground truth determined by a majority consensus on the final HER2 score among the pathologists. In total, 760 breast cancer cases were utilized throughout the training and validation phases of the study. The model's concordance with the ground truth (ICC = 0.77 [0.68–0.83]; Fisher P = 1.32e‐10) is higher than the average agreement among the three senior pathologists (ICC = 0.45 [0.17–0.65]; Fisher P = 2e‐3). In the two validation cohorts, the DL model identifies 95% [93% ‐ 98%] and 97% [91% ‐ 100%] of HER2‐low and HER2‐positive tumours, respectively. Discordant results were characterized by morphological features such as extended fibrosis, a high number of tumour‐infiltrating lymphocytes, and necrosis, whilst some artefacts such as nonspecific background cytoplasmic stain in the cytoplasm of tumour cells also cause discrepancy.

Conclusion

Deep learning can support pathologists' interpretation of difficult HER2‐low cases. Morphological variables and some specific artefacts can cause discrepant HER2‐scores between the pathologist and the DL model.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Breast NeoplasmsErb-b2 Receptor Tyrosine KinasesDeep LearningImmunohistochemistryBiomarkers, TumorHumansFemalePathologistsIn Situ Hybridization, FluorescenceMiddle Aged

Stichwörter der Autor:innen

artificial intelligencedeep learningher2human epidermal growth factor receptor 2immunohistochemistrymachine learningquantification

Substanzen

Erb-b2 Receptor Tyrosine KinasesBiomarkers, TumorERBB2 protein, human

Autor:innen und Institutionen

  • Owkin France, Paris, France.
  • Department of Pathology, ZAS Hospitals, Antwerp, Belgium.
  • Cancer & Pharmaceutical Sciences, King's College London, London, UK.
  • Guy's & ST Thomas' NHS Trust, London, UK.
  • Biobank Research Group, Lukasiewicz Research Network-PORT Polish Center for Technology Development, Wroclaw, Poland.
  • Department of Clinical and Experimental Pathology, Wroclaw Medical University, Wroclaw, Poland.

ORCID-Profile: Pierre-Antoine Bannier

Zitieren

APA

Bannier, P. A., Broeckx, G., Herpin, L., Dubois, R., Van Praet, L., Maussion, C., Deman, F., Amonoo, E., Mera, A., Timbres, J., Gillett, C., Sawyer, E., Gazińska, P., Ziolkowski, P., Lacroix-Triki, M., Salgado, R., & Irshad, S. (2024). Development of a deep‐learning model tailored for HER2 detection in breast cancer to aid pathologists in interpreting HER2 ‐low cases. Histopathology, 85(3), 478-488. https://doi.org/10.1111/his.15274

Vancouver

Bannier PA, Broeckx G, Herpin L, Dubois R, Van Praet L, Maussion C, et al. Development of a deep‐learning model tailored for HER2 detection in breast cancer to aid pathologists in interpreting HER2 ‐low cases. Histopathology. 2024;85(3):478-488. doi:10.1111/his.15274

BibTeX
@article{Bannier2024,
  author = {Bannier, Pierre-Antoine and Broeckx, Glenn and Herpin, Loïc and Dubois, Rémy and Van Praet, Lydwine and Maussion, Charles and Deman, Frederik and Amonoo, Ellen and Mera, Anca and Timbres, Jasmine and Gillett, Cheryl and Sawyer, Elinor and Gazińska, Patrycja and Ziolkowski, Piotr and Lacroix-Triki, Magali and Salgado, Roberto and Irshad, Sheeba},
  title = {Development of a deep‐learning model tailored for HER2 detection in breast cancer to aid pathologists in interpreting HER2 ‐low cases},
  journal = {Histopathology},
  year = {2024},
  volume = {85},
  number = {3},
  pages = {478--488},
  doi = {10.1111/his.15274}
}
Kostenlos starten →