Mitosis domain generalization in histopathology images - The MIDOG challenge

The density of mitotic figures (MF) within tumor tissue is known to be highly correlated with tumor proliferation and thus is an important marker in tumor grading. Recognition of MF by pathologists is subject to a strong inter-rater bias, limiting its prognostic value. State-of-the-art deep learning methods can support experts but have been observed to strongly deteriorate when applied in a different clinical environment. The variability caused by using different whole slide scanners has been identified as one decisive component in the underlying domain shift. The goal of the MICCAI MIDOG 2021 challenge was the creation of scanner-agnostic MF detection algorithms. The challenge used a training set of 200 cases, split across four scanning systems. As test set, an additional 100 cases split across four scanning systems, including two previously unseen scanners, were provided. In this paper, we evaluate and compare the approaches that were submitted to the challenge and identify methodological factors contributing to better performance. The winning algorithm yielded an F₁ score of 0.748 (CI95: 0.704-0.781), exceeding the performance of six experts on the same task.

Can AI predict epithelial lesion categories via automated analysis of cervical biopsies: The TissueNet challenge?

The French Society of Pathology (SFP) organized its first data challenge in 2020 with the help of the Health Data Hub (HDH). The organization of this event first consisted of recruiting nearly 5000 cervical biopsy slides obtained from 20 pathology centers. After ensuring that patients did not refuse to include their slides in the project, the slides were anonymized, digitized, and annotated by expert pathologists, and finally uploaded to a data challenge platform for competitors from around the world. Competing teams had to develop algorithms that could distinguish 4 diagnostic classes in cervical epithelial lesions. Among the many submissions from competitors, the best algorithms achieved an overall score close to 95%. The final part of the competition lasted only 6 weeks, and the goal of SFP and HDH is now to allow for the collection to be published in open access for the scientific community. In this report, we have performed a "post-competition analysis" of the results. We first described the algorithmic pipelines of 3 top competitors. We then analyzed several difficult cases that even the top competitors could not predict correctly. A medical committee of several expert pathologists looked for possible explanations for these erroneous results by reviewing the images, and we present their findings here targeted for a large audience of pathologists and data scientists in the field of digital pathology.

A Partial Label-Based Machine Learning Approach For Cervical Whole-Slide Image Classification: The Winning TissueNet Solution. Annu Int Conf IEEE Eng Med Biol Soc 2021;2021:2127-2131. [PMID: 34891709 DOI: 10.1109/embc46164.2021.9631009]

Cervical cancer is the fourth most common cancer in women worldwide. To determine early treatment for patients, it is critical to accurately classify the cervical intraepithelial lesion status based on a microscopic biopsy. Lesion classification is a 4-class problem, with biopsies being designated as benign or increasingly malignant as class 1-3, with 3 being invasive cancer. Unfortunately, traditional biopsy analysis by a pathologist is time-consuming and subject to intra- and inter-observer variability. For this reason, it is of interest to develop automatic analysis pipelines to classify lesion status directly from a digitalized whole slide image (WSI). The recent TissueNet Challenge was organized to find the best automatic detection pipeline for this task, using a dataset of 1015 annotated WSI slides. In this work, we present our winning end-to-end solution for cervical slide classification composed of a two-step classification model: First, we classify individual slide patches using an ensemble CNN, followed by an SVM-based slide classification using statistical features of the aggregated patch-level predictions. Importantly, we present the key innovation of our approach, which is a novel partial label-based loss function that allows us to supplement the supervised WSI patch annotations with weakly supervised patches based on the WSI class. This led to us not requiring additional expert tissue annotation, while still reaching the winning score of 94.7%. Our approach is a step towards the clinical inclusion of automatic pipelines for cervical cancer treatment planning.Clinical relevance- The explanation of the winning Tis-sueNet AI algorithm for automated cervical cancer classification, which may provide insights for the next generation of computer assisted tools in digital pathology.

Chromogenic Multiplex Immunohistochemistry Reveals Modulation of the Immune Microenvironment Associated with Survival in Elderly Patients with Lung Adenocarcinoma

With underrepresentation of elderly patients with lung adenocarcinoma (LADC) in anti-PD-1/PD-L1 clinical trials, better understanding of the interplay of PD-L1 and tumor-associated immune cells (TAICs) could assist clinicians in stratifying these patients for immunotherapy. One hundred and one patients with LADCs, stratified by age, were included for analysis of PD-L1 expression and density of TAICs expressing CD4, CD8, and CD33, by using multiplex chromogenic immunohistochemistry (IHC) assays and automated digital quantification. The CD4⁺/CD8⁺ ratio was significantly higher in elderly patients. In patients <75 years, the density of CD4⁺, CD8⁺, and PD-L1 in TAICs showed a positive significant correlation with PD-L1 expression in tumor cells (TCs), while a lower correlation was observed in the elderly population. In the latter, a high CD4⁺/CD8⁺ ratio, and combined PD-L1 expression ≥1% TCs with a low CD8⁺ density, low CD33⁺ density, and a high CD4⁺ density correlated to worse overall survival. We identified differences according to age in the CD4⁺/CD8⁺ ratio and in correlation between PD-L1 expression and the density of TAICs in LADC patients. Distinct groups of tumor microenvironments had an impact on the OS of elderly patients with LADC.

Automated brightfield multiplex immunohistochemistry to quantify biomarkers related to immune senescence: Relationships with survival in non-small cell lung cancer patients

e20500

Background: Elderly patients have an eroded immune characterized by a progressive decline in immune surveillance that favors infection and cancer development. Tumor cells can escape immune surveillance by upregulating inhibitory immune checkpoint such as PD-L1. High expression of PD-L1 was reported in association with CD8+T-cell exhaustion and increased levels of CD33+ myeloid-derived suppressor cells. Although low CD4/CD8 ratio is associated with increased mortality, the status of the CD4+T-cells as a clinical marker of immunosenescence is less well characterized in the field of aging. The aim of this study was to determine the presence of immunosenescence biomarkers according to age in non-small cell lung cancer (NSCLC) patients and to evaluate them as predictive biomarkers of patients’ outcome. Methods: One hundred NSCLC patients, matched by age (50 patients < 70 years, 50 patients ≥70 years) were included. An automated 4-Plex optical IHC assay was developed on the Discovery ULTRA automated stainer using monoclonal antibodies PD-L1 (SP263), CD4, CD8, and CD33. The stained slides were scanned with Nanozoomer HT 2.0 Scanner, and analyzed with Calopix software. Results: The CD4/CD8 ratio and PD-L1 expression in tumor and immune cells were significantly lower in elderly NSCLC patients ≥70 years than in age-paired patients, while absolute count of CD33+ was increased. Patients with CD4/CD8 ratio higher than two, high PD-L1 density and low CD33+ frequency achieved increase in median disease-free survival. Conclusions: Distribution of PD-L1, CD4, CD8, and CD33 cells was influenced by age in NSCLC patients. The proportion of CD8 + CD28- T cells, CD4+ T cells and CD4/CD8 ratio may be used as predictive biomarkers of anti-PD-L1 therapy efficacy in NSCLC patients. The automated 4-Plex IHC assay together with its respective digital analysis could serve as a tool for further characterizing tumors and their microenvironment and provide a better understanding of which patients may benefit from immunotherapy.