Development and applications of computer image analysis algorithms for scoring of PD-L1 immunohistochemistry

Immune checkpoint inhibitors targeting programmed cell death 1 (PD-1) and programmed cell death ligand 1 (PD-L1) have rapidly become integral to standard-of-care therapy for non-small cell lung cancer and other cancers. Immunohistochemical (IHC) staining of PD-L1 is currently the accepted and approved diagnostic assay for selecting patients for PD-L1/PD-1 axis therapies in certain indications. However, the inherent biological complexity of PD-L1 and the availability of several PD-L1 assays – each with different detection systems, platforms, scoring algorithms and cut-offs – have created challenges to ensure reliable and reproducible results based on subjective visual assessment by pathologists. The increasing adoption of computer technologies into the daily workflow of pathology provides an opportunity to leverage these tools towards improving the clinical value of PD-L1 IHC assays. This review describes several image analysis software programs of computer-aided PD-L1 scoring in the hope of driving further discussion and technological advancement in digital pathology and artificial intelligence approaches, particularly as precision medicine evolves to encompass accurate simultaneous assessment of multiple features of cancer cells and their interactions with the tumor microenvironment.

[1]  Arun Ahuja,et al.  Genomic Features of Response to Combination Immunotherapy in Patients with Advanced Non-Small-Cell Lung Cancer , 2018, Cancer cell.

[2]  M. Levitt,et al.  Adverse Effects of Immune Checkpoint Inhibitors (Programmed Death-1 Inhibitors and Cytotoxic T-Lymphocyte-Associated Protein-4 Inhibitors): Results of a Retrospective Study , 2019, Journal of clinical medicine research.

[3]  Razelle Kurzrock,et al.  PD-L1 Expression as a Predictive Biomarker in Cancer Immunotherapy , 2015, Molecular Cancer Therapeutics.

[4]  Puay Hoon Tan,et al.  An automated staining protocol for seven-colour immunofluorescence of human tissue sections for diagnostic and prognostic use. , 2018, Pathology.

[5]  D. Djureinovic,et al.  PD-L1 immunohistochemistry in clinical diagnostics of lung cancer: inter-pathologist variability is higher than assay variability , 2017, Modern Pathology.

[6]  Anagha P. Jadhav,et al.  A Multi-Institutional Study to Evaluate Automated Whole Slide Scoring of Immunohistochemistry for Assessment of Programmed Death-Ligand 1 (PD-L1) Expression in Non–Small Cell Lung Cancer , 2019, Applied immunohistochemistry & molecular morphology : AIMM.

[7]  R. Bourgon,et al.  Atezolizumab in patients with locally advanced and metastatic urothelial carcinoma who have progressed following treatment with platinum-based chemotherapy: a single-arm, multicentre, phase 2 trial , 2016, The Lancet.

[8]  L. Pusztai,et al.  Prospective multi-institutional evaluation of pathologist assessment of PD-L1 assays for patient selection in triple negative breast cancer , 2020, Modern Pathology.

[9]  Hui Yu,et al.  PD‐L1 Immunohistochemistry Comparability Study in Real‐Life Clinical Samples: Results of Blueprint Phase 2 Project , 2018, Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer.

[10]  Andrew H. Beck,et al.  Diagnostic Assessment of Deep Learning Algorithms for Detection of Lymph Node Metastases in Women With Breast Cancer , 2017, JAMA.

[11]  Kyoung-Mee Kim,et al.  PD-L1 expression in gastric cancer determined by digital image analyses: pitfalls and correlation with pathologist interpretation , 2019, Virchows Archiv.

[12]  Johannes Griss,et al.  Digital image analysis improves precision of PD‐L1 scoring in cutaneous melanoma , 2018, Histopathology.

[13]  I. Wistuba,et al.  Immunohistochemical and Image Analysis‐Based Study Shows That Several Immune Checkpoints are Co‐expressed in Non–Small Cell Lung Carcinoma Tumors , 2018, Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer.

[14]  Aaron Y. Lee,et al.  Artificial intelligence and deep learning in ophthalmology , 2018, British Journal of Ophthalmology.

[15]  Arthur L. Samuel,et al.  Some studies in machine learning using the game of checkers , 2000, IBM J. Res. Dev..

[16]  Shidan Wang,et al.  Pathology Image Analysis Using Segmentation Deep Learning Algorithms. , 2019, The American journal of pathology.

[17]  K. Kerr,et al.  Non-Small Cell Lung Cancer, PD-L1, and the Pathologist. , 2016, Archives of pathology & laboratory medicine.

[18]  Souptik Barua,et al.  Spatial computation of intratumoral T cells correlates with survival of patients with pancreatic cancer , 2017, Nature Communications.

[19]  David J. Hawkes,et al.  Automated Classification of Breast Cancer Stroma Maturity From Histological Images , 2017, IEEE Transactions on Biomedical Engineering.

[20]  Marius Ilie,et al.  Assessment of the PD-L1 status by immunohistochemistry: challenges and perspectives for therapeutic strategies in lung cancer patients , 2016, Virchows Archiv.

[21]  K. Harrington,et al.  Nivolumab for Recurrent Squamous-Cell Carcinoma of the Head and Neck. , 2016, The New England journal of medicine.

[22]  Xue Han,et al.  Anti-PD-1/PD-L1 therapy of human cancer: past, present, and future. , 2015, The Journal of clinical investigation.

[23]  Douglas Bowman,et al.  Introduction to Digital Image Analysis in Whole-slide Imaging: A White Paper from the Digital Pathology Association , 2019, Journal of pathology informatics.

[24]  M. Salto‐Tellez,et al.  Critical Appraisal of Programmed Death Ligand 1 Reflex Diagnostic Testing: Current Standards and Future Opportunities , 2019, Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer.

[25]  K. Steele,et al.  Automated image analysis of NSCLC biopsies to predict response to anti-PD-L1 therapy , 2019, Journal of Immunotherapy for Cancer.

[26]  Ce Zhang,et al.  Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features , 2016, Nature Communications.

[27]  Y. Shentu,et al.  Pembrolizumab versus Chemotherapy for PD-L1-Positive Non-Small-Cell Lung Cancer. , 2016, The New England journal of medicine.

[28]  David C. Smith,et al.  Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. , 2012, The New England journal of medicine.

[29]  H. Kohrt,et al.  Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients , 2014, Nature.

[30]  David L Rimm,et al.  Comparison of Biomarker Modalities for Predicting Response to PD-1/PD-L1 Checkpoint Blockade: A Systematic Review and Meta-analysis. , 2019, JAMA oncology.

[31]  Nicolas Brieu,et al.  Deep Semi Supervised Generative Learning for Automated Tumor Proportion Scoring on NSCLC Tissue Needle Biopsies , 2018, Scientific Reports.

[32]  J. Taube,et al.  A Prospective, Multi-institutional, Pathologist-Based Assessment of 4 Immunohistochemistry Assays for PD-L1 Expression in Non–Small Cell Lung Cancer , 2017, JAMA oncology.

[33]  A. Rao,et al.  Spatial and phenotypic immune profiling of metastatic colon cancer. , 2018, JCI insight.

[34]  A. Madabhushi,et al.  Artificial intelligence in digital pathology — new tools for diagnosis and precision oncology , 2019, Nature Reviews Clinical Oncology.

[35]  Jaime Rodriguez-Canales,et al.  Automated Multiplex Immunofluorescence Panel for Immuno-oncology Studies on Formalin-fixed Carcinoma Tissue Specimens. , 2019, Journal of visualized experiments : JoVE.

[36]  Michael Barnes,et al.  Whole tumor section quantitative image analysis maximizes between-pathologists' reproducibility for clinical immunohistochemistry-based biomarkers. , 2017, Laboratory investigation; a journal of technical methods and pathology.

[37]  Inge M. N. Wortel,et al.  Eight-Color Multiplex Immunohistochemistry for Simultaneous Detection of Multiple Immune Checkpoint Molecules within the Tumor Microenvironment , 2018, The Journal of Immunology.

[38]  M. Salto‐Tellez,et al.  Automated Tumour Recognition and Digital Pathology Scoring Unravels New Role for PD-L1 in Predicting Good Outcome in ER-/HER2+ Breast Cancer , 2018, Journal of oncology.

[39]  Jakob Nikolas Kather,et al.  Deep learning can predict microsatellite instability directly from histology in gastrointestinal cancer , 2019, Nature Medicine.

[40]  N. Katabi,et al.  Inter‐ and intraobserver agreement of programmed death ligand 1 scoring in head and neck squamous cell carcinoma, urothelial carcinoma and breast carcinoma , 2019, Histopathology.

[41]  J. Taube,et al.  Reanalysis of the NCCN PD-L1 companion diagnostic assay study for lung cancer in the context of PD-L1 expression findings in triple-negative breast cancer , 2019, Breast Cancer Research.

[42]  C. Marquette,et al.  Effect of mutant variants of the KRAS gene on PD-L1 expression and on the immune microenvironment and association with clinical outcome in lung adenocarcinoma patients. , 2018, Lung cancer.

[43]  T. Chan,et al.  The evolving landscape of biomarkers for checkpoint inhibitor immunotherapy , 2019, Nature Reviews Cancer.

[44]  R. Sullivan,et al.  Fatal Toxic Effects Associated With Immune Checkpoint Inhibitors: A Systematic Review and Meta-analysis , 2018, JAMA oncology.

[45]  Holger Lange.  Digital Pathology: A Regulatory Overview , 2011 .

[46]  W. Weichert,et al.  Multicentric analytical comparability study of programmed death-ligand 1 expression on tumor-infiltrating immune cells and tumor cells in urothelial bladder cancer using four clinically developed immunohistochemistry assays , 2019, Virchows Archiv.

[47]  Boris Sepesi,et al.  Effect of neoadjuvant chemotherapy on the immune microenvironment in non–small cell lung carcinomas as determined by multiplex immunofluorescence and image analysis approaches , 2018, Journal of Immunotherapy for Cancer.

[48]  Karl Garsha,et al.  Fully automated 5-plex fluorescent immunohistochemistry with tyramide signal amplification and same species antibodies. , 2017, Laboratory investigation; a journal of technical methods and pathology.

[49]  Nico Karssemeijer,et al.  Using deep convolutional neural networks to identify and classify tumor-associated stroma in diagnostic breast biopsies , 2018, Modern Pathology.

[50]  Arkadiusz Gertych,et al.  Spatial Mapping of Myeloid Cells and Macrophages by Multiplexed Tissue Staining , 2018, Front. Immunol..

[51]  Navid Farahani,et al.  A Practical Guide to Whole Slide Imaging: A White Paper From the Digital Pathology Association. , 2018, Archives of pathology & laboratory medicine.

[52]  R. Herbst,et al.  Quantitative Assessment of the Heterogeneity of PD-L1 Expression in Non-Small-Cell Lung Cancer. , 2016, JAMA oncology.

[53]  M. Gurcan,et al.  Digital pathology and artificial intelligence. , 2019, The Lancet. Oncology.

[54]  K. Dreyer,et al.  Artificial intelligence in radiology: the ecosystem essential to improving patient care. , 2020, Clinical imaging.

[55]  Helen Pitman,et al.  Artificial intelligence in digital pathology: a roadmap to routine use in clinical practice , 2019, The Journal of pathology.

[56]  H. Grote,et al.  PD-L1 immunostaining scoring for non-small cell lung cancer based on immunosurveillance parameters , 2018, PloS one.

[57]  K. Higgins,et al.  Reproducibility of PD-L1 immunohistochemistry interpretation across various types of genitourinary and head/neck carcinomas, antibody clones, and tissue types. , 2018, Human pathology.

[58]  S. Fox,et al.  Intra- and Interobserver Reproducibility Assessment of PD-L1 Biomarker in Non–Small Cell Lung Cancer , 2017, Clinical Cancer Research.