Multi-Field-of-View Deep Learning Model Predicts Nonsmall Cell Lung Cancer Programmed Death-Ligand 1 Status from Whole-Slide Hematoxylin and Eosin Images.

Autor: Sha L; Tempus Labs, Inc, Chicago, IL USA., Osinski BL; Tempus Labs, Inc, Chicago, IL USA., Ho IY; Tempus Labs, Inc, Chicago, IL USA., Tan TL; Tempus Labs, Inc, Chicago, IL USA.; Department of Pathology, Northwestern University Feinberg School of Medicine, Chicago, IL, USA., Willis C; Tempus Labs, Inc, Chicago, IL USA., Weiss H; Tempus Labs, Inc, Chicago, IL USA.; Department of Neurological Surgery, Northwestern University Feinberg School of Medicine, Chicago, IL, USA., Beaubier N; Tempus Labs, Inc, Chicago, IL USA., Mahon BM; Tempus Labs, Inc, Chicago, IL USA., Taxter TJ; Tempus Labs, Inc, Chicago, IL USA., Yip SSF; Tempus Labs, Inc, Chicago, IL USA.
Jazyk: angličtina
Zdroj: Journal of pathology informatics [J Pathol Inform] 2019 Jul 23; Vol. 10, pp. 24. Date of Electronic Publication: 2019 Jul 23 (Print Publication: 2019).
DOI: 10.4103/jpi.jpi_24_19
Abstrakt: Background: Tumor programmed death-ligand 1 (PD-L1) status is useful in determining which patients may benefit from programmed death-1 (PD-1)/PD-L1 inhibitors. However, little is known about the association between PD-L1 status and tumor histopathological patterns. Using deep learning, we predicted PD-L1 status from hematoxylin and eosin (H and E) whole-slide images (WSIs) of nonsmall cell lung cancer (NSCLC) tumor samples.
Materials and Methods: One hundred and thirty NSCLC patients were randomly assigned to training ( n = 48) or test ( n = 82) cohorts. A pair of H and E and PD-L1-immunostained WSIs was obtained for each patient. A pathologist annotated PD-L1 positive and negative tumor regions on the training samples using immunostained WSIs for reference. From the H and E WSIs, over 145,000 training tiles were generated and used to train a multi-field-of-view deep learning model with a residual neural network backbone.
Results: The trained model accurately predicted tumor PD-L1 status on the held-out test cohort of H and E WSIs, which was balanced for PD-L1 status (area under the receiver operating characteristic curve [AUC] =0.80, P << 0.01). The model remained effective over a range of PD-L1 cutoff thresholds (AUC = 0.67-0.81, P ≤ 0.01) and when different proportions of the labels were randomly shuffled to simulate interpathologist disagreement (AUC = 0.63-0.77, P ≤ 0.03).
Conclusions: A robust deep learning model was developed to predict tumor PD-L1 status from H and E WSIs in NSCLC. These results suggest that PD-L1 expression is correlated with the morphological features of the tumor microenvironment.
Competing Interests: L.S., B.L.O., I.Y.H., C.W., N.B., B.M.M., T.J.T., and S.S.F.Y. are employees and/or shareholders of Tempus Labs. H.W. is an intern at Tempus Labs. T.L.T. was compensated by Tempus Labs for his participation as a pathologist.
Databáze: MEDLINE
Externí odkaz: