Deep Semi Supervised Generative Learning for Automated Tumor Proportion Scoring on NSCLC Tissue Needle Biopsies
Ansh Kapil, Armin Meier, Aleksandra Zuraw, Keith E. Steele, Marlon C. Rebelatto, Günter Schmidt & Nicolas Brieu Scientific Reportsvolume 8, Article number: 17343 (2018)
The level of PD-L1 expression in immunohistochemistry (IHC) assays is a key biomarker for the identification of Non-Small-Cell-Lung-Cancer (NSCLC) patients that may respond to anti PD-1/PD-L1 treatments. The quantification of PD-L1 expression currently includes the visual estimation of a Tumor Cell (TC) score by a pathologist and consists of evaluating the ratio of PD-L1 positive and PD-L1 negative tumor cells. Known challenges like differences in positivity estimation around clinically relevant cut-offs and sub-optimal quality of samples makes visual scoring tedious and subjective, yielding a scoring variability between pathologists.
In this work, we propose a novel deep learning solution that enables the first automated and objective scoring of PD-L1 expression in late stage NSCLC needle biopsies. To account for the low amount of tissue available in biopsy images and to restrict the amount of manual annotations necessary for training, we explore the use of semi-supervised approaches against standard fully supervised methods. Concordance measures computed on a set of slides unseen during training provide evidence that our automatic scoring method matches visual scoring on the considered dataset while ensuring repeatability and objectivity.
This study presents the first deep learning based system for an accurate and automated scoring of PD-L1 expression on needle biopsy samples, opening the door for a computer-aided assessment of patient eligibility for an anti-PDL1 immune checkpoint inhibitor treatment. The proposed work also introduces the first application of deep semi-supervised learning in the field of digital pathology image analysis. By learning from both labeled and unlabeled images, this innovative approach substantially improves over fully supervised networks which are more commonly used in this field.