A Computational Tumor-Infiltrating Lymphocyte Assessment Method Comparable with Visual Reporting Guidelines for Triple-Negative Breast Cancer

Abstract

Background: Tumor-infiltrating lymphocytes (TILs) are clinically significant in triple-negative breast cancer (TNBC). Although a standardized methodology for visual TILs assessment (VTA) exists, it has several inherent limitations. We established a deep learning-based computational TIL assessment (CTA) method broadly following VTA guideline and compared it with VTA for TNBC to determine the prognostic value of the CTA and a reasonable CTA workflow for clinical practice. Methods: We trained three deep neural networks for nuclei segmentation, nuclei classification and necrosis classification to establish a CTA workflow. The automatic TIL (aTIL) score generated was compared with manual TIL (mTIL) scores provided by three pathologists in an Asian (n = 184) and a Caucasian (n = 117) TNBC cohort to evaluate scoring concordance and prognostic value. Findings: The intraclass correlations (ICCs) between aTILs and mTILs varied from 0.40 to 0.70 in two cohorts. Multivariate Cox proportional hazards analysis revealed that the aTIL score was associated with disease free survival (DFS) in both cohorts, as either a continuous [hazard ratio (HR)=0.96, 95% CI 0.94À0.99] or dichotomous variable (HR=0.29, 95% CI 0.12À0.72). A higher C-index was observed in a composite mTIL/aTIL threetier stratification model than in the dichotomous model, using either mTILs or aTILs alone. Interpretation: The current study provides a useful tool for stromal TIL assessment and prognosis evaluation for patients with TNBC. A workflow integrating both VTA and CTA may aid pathologists in performing risk management and decision-making tasks.

In Situ Hybridization of PRRSV-1 Combined with Digital Image Analysis in Lung Tissues of Pigs Challenged with PRRSV-1

“…In order to investigate the applicability of the assay for quantitative measurements, we first used automatic, software-based positive cell counting on slides that were digitalized in a slide scanner instrument. Computational cell counting is widely used in human pathology, especially for Ki67 index and tumor-infiltrating lymphocyte assessments, and it is proved to be comparable to or, in some cases, even more accurate and reproducible than manual counting [ 26 , 38 , 39 ].…”

Prognostic Capability of TNBC 3-Gene Score among Triple-Negative Breast Cancer Subtypes

“…The TNBC 3-gene score was evaluated as continuous variable while TIL count was categorized into high and low, using a cutoff value of 20% since it has been proved as a prognostic biomarker of survival in TNBC [ 18 , 27 , 28 ]. The boxplot graphic and the Student’s t-test were used to analyze the differences between groups.…”

Development of Training Materials for Pathologists to Provide Machine Learning Validation Data of Tumor-Infiltrating Lymphocytes in Breast Cancer

“…Anticipating the influx of artificial intelligence and machine learning algorithms (AI/ML) to assess TILs [17][18][19][20][21][22], we began the High Throughput Truthing (HTT) project in collaboration with an international team of pathologists, clinical scientists, and leadership from the Working Group [23]. Our goal is to create a dataset of digital slide data with pathologist annotations for the validation of computational pathology models (e.g., AI/ML) for stromal TILs (sTILs) assessment that will be fit for a regulatory purpose as a medical device development tool [24].…”