Abstract:An emerging predictive parameter of immunotherapy response is the patient's tumor immune status, generally classified as "inflamed", "immune excluded" and "desert". Historically, classification was performed semi-quantitatively with somewhat subjective parameters as evidenced by the recent Delphi Workshop consensus. There are a variety of approaches to classifying immune status, most based upon analysis of H&E images (counting tumor infiltrating lymphocytes, TILs), or counting subpopulations of immune cells using immunologic staining. However, use of multiplexed or multiple stains has recently been explored to better quantify the counts and types of immune cells in individual tissue compartments. While the immune microenvironment represents an important parameter in predicting patient response, other regions such as the extracellular matrix (ECM) may yield complementary information, as collagen-rich ECMs may present a barrier to drug diffusion and the orientation of fibers may direct the migration of malignant cells. Here we present a proof of concept virtual staining enhanced image analysis pipeline, which converts autofluorescence signals from a single tissue section into virtual H&E and Masson's Trichrome along with virtual detection of pan cytokeratin (AE1/AE3/PCK26) and CD45 (LCA) positive cells. We apply image analysis to the panel of virtual stains to study the spatial and case-by-case heterogeneity of tumor collagen frameworks and immune phenotype. Using a standard slide scanner (Axio Scan Z1, Zeiss), multiple autofluorescence images were captured from unstained sections (4-6um thick) of lung tissue. The virtual staining was performed by four deep neural networks trained in a supervised learning fashion. Select chemical stains were performed on previously scanned tissues and reviewed by pathologists side-by-side with virtual stains to ensure consistency and quality control. Four perfectly registered WSI virtual stain images were generated from each tissue section. The multi-stain results were rendered from a variety of lung cancers, including tissue microarray slides. Image analysis was performed using HALO (Indica Labs) and custom python scripts. We identified unique areas of tumor and adjacent stroma with heterogeneous immune phenotype and collagen characteristics, suggesting that an interplay might exist that could be utilized to improve patient selection or prognosis when deploying advanced AI tools. Future work includes applying this virtual stain technique retrospectively to cases with known immunotherapy treatment responses to determine the prognostic significance of the combined immune/ECM phenotype. Citation Format: Serge Alexanian, Yair Rivenson, Ning Xuan, Brian Cone, Zihang Fang, Sean Meyering, Luis A. Carvalho, David Palacios, Raymond Kozikowski. Combination analysis of tumor-associated collagen frameworks and tumor immune phenotype of lung carcinomas using virtual staining [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular s); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl) nr 914.

Digital Analysis of Immunostaining of ZW10 Interacting Protein in Human Lung Tissues

Automatic Immunoscoring of Immunohistochemistry Images of Human Lung Cancer Tissue Samples: A Screening Tool

Optimization of whole slide imaging scan settings for computer vision using human lung cancer tissue

Immunocytochemical detection by a fully automated immunostainer for assisting subclassification of pulmonary tumors on ThinPrep slides in real-life clinical samples: A single-institution experience

Comparing Digital Histology Slides with Multiple Staining Based on Decoloring and Dyeing Technique

Automated whole-slide images assessment of immune infiltration in resected non-small-cell lung cancer: towards better risk-stratification

Automated quantitative analysis of Ki-67 staining and HE images recognition and registration based on whole tissue sections in breast carcinoma

Automated Whole Slide Imaging for Label-Free Histology using Photon Absorption Remote Sensing Microscopy

Rapid Whole-Organ Characterization Via Quantitative Light-Sheet Microscopy

Abstract 914: Combination analysis of tumor-associated collagen frameworks and tumor immune phenotype of lung carcinomas using virtual staining

Autofluorescence Virtual Staining System for H&E Histology and Multiplex Immunofluorescence Applied to Immuno-Oncology Biomarkers in Lung Cancer

Accurately Matching Serial Whole Slide Images for Tumor Heterogeneity Analysis

Seven-colour multiplex immunochemistry/immunofluorescence and whole slide imaging of frozen sections

Deep Learning-Based H-Score Quantification of Immunohistochemistry-Stained Images

A Deep Learning Approach for Histology-Based Nucleus Segmentation and Tumor Microenvironment Characterization

Staining Condition Visualization in Digital Histopathological Whole-Slide Images

Digital Pathology in the Detection of Infectious Microorganisms: An Evaluation of Its Strengths and Weaknesses Across a Panel of Immunohistochemical and Histochemical Stains Routinely Used in Diagnostic Surgical Pathology

Introduction to Digital Image Analysis in Whole-slide Imaging: A White Paper from the Digital Pathology Association

An Automatic Darknet-based Immunohistochemical Scoring System for IL-24 in Lung Cancer

Spatial analysis of tissue immunity and vascularity by light sheet fluorescence microscopy

Assessing the Practical Application of Imaging Mass Cytometry for Visualizing Tumor Immune Microenvironment Heterogeneity