» Articles » PMID: 34786761

Development and Validation of a Supervised Deep Learning Algorithm for Automated Whole-slide Programmed Death-ligand 1 Tumour Proportion Score Assessment in Non-small Cell Lung Cancer

Overview
Journal Histopathology
Date 2021 Nov 17
PMID 34786761
Citations 15
Authors
Affiliations
Soon will be listed here.
Abstract

Aims: Immunohistochemical programmed death-ligand 1 (PD-L1) staining to predict responsiveness to immunotherapy in patients with advanced non-small cell lung cancer (NSCLC) has several drawbacks: a robust gold standard is lacking, and there is substantial interobserver and intraobserver variance, with up to 20% discordance around cutoff points. The aim of this study was to develop a new deep learning-based PD-L1 tumour proportion score (TPS) algorithm, trained and validated on a routine diagnostic dataset of digitised PD-L1 (22C3, laboratory-developed test)-stained samples.

Methods And Results: We designed a fully supervised deep learning algorithm for whole-slide PD-L1 assessment, consisting of four sequential convolutional neural networks (CNNs), using aiforia create software. We included 199 whole slide images (WSIs) of 'routine diagnostic' histology samples from stage IV NSCLC patients, and trained the algorithm by using a training set of 60 representative cases. We validated the algorithm by comparing the algorithm TPS with the reference score in a held-out validation set. The algorithm had similar concordance with the reference score (79%) as the pathologists had with one another (75%). The intraclass coefficient was 0.96 and Cohen's κ coefficient was 0.69 for the algorithm. Around the 1% and 50% cutoff points, concordance was also similar between pathologists and the algorithm.

Conclusions: We designed a new, deep learning-based PD-L1 TPS algorithm that is similarly able to assess PD-L1 expression in daily routine diagnostic cases as pathologists. Successful validation on routine diagnostic WSIs and detailed visual feedback show that this algorithm meets the requirements for functioning as a 'scoring assistant'.

Citing Articles

A Thorough Review of the Clinical Applications of Artificial Intelligence in Lung Cancer.

Kotoulas S, Spyratos D, Porpodis K, Domvri K, Boutou A, Kaimakamis E Cancers (Basel). 2025; 17(5).

PMID: 40075729 PMC: 11898928. DOI: 10.3390/cancers17050882.


InterobServer AgreeMent in Pd-l1 evaLuatIoN on cytoloGical samples-SAMPLING project: A multi-institutional, international study.

Acanfora G, Iaccarino A, Cerbelli B, Di Cristofano C, Bellevicine C, Barberis M Cancer Cytopathol. 2025; 133(3):e70003.

PMID: 39992702 PMC: 11849766. DOI: 10.1002/cncy.70003.


The clinical application of artificial intelligence in cancer precision treatment.

Wang J, Zeng Z, Li Z, Liu G, Zhang S, Luo C J Transl Med. 2025; 23(1):120.

PMID: 39871340 PMC: 11773911. DOI: 10.1186/s12967-025-06139-5.


Power line fault diagnosis based on convolutional neural networks.

Ning L, Pei D Heliyon. 2024; 10(8):e29021.

PMID: 38628723 PMC: 11019159. DOI: 10.1016/j.heliyon.2024.e29021.


Correlation Between ImageJ and Conventional Manual Scoring Methods for Programmed Death-Ligand 1 Immuno-Histochemically Stained Sections.

Al Taher R, Abbas M, Halahleh K, Sughayer M Technol Cancer Res Treat. 2024; 23:15330338241242635.

PMID: 38562094 PMC: 10989033. DOI: 10.1177/15330338241242635.


References
1.
Butter R, t Hart N, Hooijer G, Monkhorst K, Speel E, Theunissen P . Multicentre study on the consistency of PD-L1 immunohistochemistry as predictive test for immunotherapy in non-small cell lung cancer. J Clin Pathol. 2019; 73(7):423-430. DOI: 10.1136/jclinpath-2019-205993. View

2.
Chen R, Hou X, Yang L, Zhao D . Comparative efficacy and safety of first-line treatments for advanced non-small cell lung cancer with immune checkpoint inhibitors: A systematic review and meta-analysis. Thorac Cancer. 2019; 10(4):607-623. PMC: 6449246. DOI: 10.1111/1759-7714.12971. View

3.
Simons K, de Jong A, Jukema J, de Vries M, Arens R, Quax P . T cell co-stimulation and co-inhibition in cardiovascular disease: a double-edged sword. Nat Rev Cardiol. 2019; 16(6):325-343. DOI: 10.1038/s41569-019-0164-7. View

4.
Cohen D, Hondelink L, Solleveld-Westerink N, Uljee S, Ruano D, Cleton-Jansen A . Optimizing Mutation and Fusion Detection in NSCLC by Sequential DNA and RNA Sequencing. J Thorac Oncol. 2020; 15(6):1000-1014. DOI: 10.1016/j.jtho.2020.01.019. View

5.
Williams G, Nicholson A, Snead D, Thunnissen E, Lantuejoul S, Cane P . Interobserver Reliability of Programmed Cell Death Ligand-1 Scoring Using the VENTANA PD-L1 (SP263) Assay in NSCLC. J Thorac Oncol. 2019; 15(4):550-555. DOI: 10.1016/j.jtho.2019.11.010. View