Artificial intelligence-based virtual staining platform for identifying tumor-associated macrophages from hematoxylin and eosin-stained images
Arpit Aggarwal, Mayukhmala Jana, Amritpal Singh, Tanmoy Dam, Himanshu Maurya, Tilak Pathak, Sandra Oršulić, Kailin Yang, Deborah J. Chute, Justin A. Bishop, Farhoud Faraji, Wade L. Thorstad, Shlomo A. Koyfman, Scott Steward-Tharp, Qiuying Shi, Vlad C. Sandulache, Nabil F. Saba, James S. Lewis, Germán Corredor, Anant Madabhushi
Abstract
BACKGROUND: Virtual staining is an artificial intelligence-based approach that transforms pathology images between stain types, such as hematoxylin and eosin (H&E) to immunohistochemistry (IHC), providing a tissue-preserving and efficient alternative to traditional IHC staining. However, existing methods for translating H&E to virtual IHC often fail to generate images of sufficient quality for accurately delineating cell nuclei and IHC+ regions. To address these limitations, we introduce VISTA, an artificial intelligence-based virtual staining platform designed to translate H&E into virtual IHC. METHODS: We applied VISTA to identify M2-subtype tumor-associated macrophages (M2-TAMs) in H&E images from 968 patients with HPV+ oropharyngeal squamous cell carcinoma across six institutional cohorts. M2-TAMs are a critical component of the tumor microenvironment, and their increased presence has been linked to poor survival. Co-registered H&E and CD163 + IHC tissue microarrays were used to train (D1, N = 102) and test (D2, N = 50) the VISTA platform. M2-TAM density, defined as the ratio of M2-TAMs to total nuclei, was derived from VISTA-generated CD163 + IHC images and evaluated for prognostic significance in additional training (D3, N = 360) and testing (D4, N = 456) cohorts using biopsy or resection H&E whole slide images. RESULTS: High M2-TAM density was associated with worse overall survival in D4 (p = 0.0152, Hazard Ratio=1.63 [1.1-2.42]). VISTA outperformed existing methods, generating higher-quality virtual CD163 + IHC images in D2, with a Structural Similarity Index of 0.72, a Peak Signal-to-Noise Ratio of 21.5, and a Fréchet Inception Distance of 41.4. Additionally, VISTA demonstrated superior performance in segmenting M2-TAMs in D2 (Dice=0.74). CONCLUSION: These findings establish VISTA as a computational platform for generating virtual IHC and facilitating the discovery of novel biomarkers from H&E images.