Interpreting and Validating a Deep Learning Model Predictive of Spatial Morphologic-Molecular Patterns in Lung Adenocarcinoma, Using Ground Truth Immunohistochemistry Images
The American Journal of Pathology · 2026-11
DOI: 10.1016/j.ajpath.2026.10.009 ↗Kurz gefasst
Fachartikel in der Zeitschrift The American Journal of Pathology, 2026-11. Verfasst von Vibha R Rao und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH. Die Arbeit nennt 84 Literaturangaben.
Abstract (Originalsprache)
Lung adenocarcinoma (LUAD), the most common subtype of non-small cell lung cancer, exhibits profound histological and molecular heterogeneity. While genomic profiling has identified key oncogenic drivers and immune signatures, its use is limited by cost, technical demands and tissue availability. In addition, spatial transcriptomics provides spatially resolved molecular insights but remains challenging and time-consuming. To address this gap, we developed XpressO-Lung, an explanatory deep learning model that predicts gene expression heterogeneity spatially in tumor and its microenvironment on hematoxylin and eosin based diagnostic (Dx) whole-slide images (WSIs) by learning associations between tissue morphology and the corresponding bulk-transcriptomic data. Utilizing 200 LUAD cases from The Cancer Genome Atlas, XpressO-Lung predicted spatial expression patterns of NAPSA, TP53I3, CD8A, TTF1, KRT7, CDKN2A, FOXO1, KEAP1, RB1 and TP53 on Dx-WSIs with AUCs ranging from 0.64 to 0.92. The predicted spatial gene expression patterns aligned with the known morphologic interactions of the tumor and its microenvironment, capturing biological events directly on Dx-WSIs. These spatio-morpho-molecular associations were further validated using immunohistochemistry on an external set of clinical samples at Dartmouth Health, demonstrating concordance between model-predicted spatial patterns and observed histomorphologic features. By coupling predictive performance with spatial interpretability of gene expression on Dx-WSIs, the XpressO-Lung model bridges histopathology and bulk-transcriptomics, enabling explainable spatio-morpho-genomic analyses to advance biomarker discovery, therapeutic stratification and precision oncology in LUAD.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH.
- Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH; Dartmouth Cancer Center, Dartmouth Health, Lebanon, NH.
- Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH; Dartmouth Cancer Center, Dartmouth Health, Lebanon, NH. Electronic address: Xiaoying.Liu@dartmouth.edu.
- Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH. Electronic address: Shrey.Sukhadia@dartmouth.edu.
Zitieren
APA
Rao, V. R., Workman, A. A., Palisoul, S. M., Limoge, C. J., Rawal, P. K., Vaickus, L. J., Zanazzi, G. J., Lu, L., Liu, X., & Sukhadia, S. S. (2026). Interpreting and Validating a Deep Learning Model Predictive of Spatial Morphologic-Molecular Patterns in Lung Adenocarcinoma, Using Ground Truth Immunohistochemistry Images. The American Journal of Pathology. https://doi.org/10.1016/j.ajpath.2026.10.009
Vancouver
Rao VR, Workman AA, Palisoul SM, Limoge CJ, Rawal PK, Vaickus LJ, et al. Interpreting and Validating a Deep Learning Model Predictive of Spatial Morphologic-Molecular Patterns in Lung Adenocarcinoma, Using Ground Truth Immunohistochemistry Images. The American Journal of Pathology. 2026. doi:10.1016/j.ajpath.2026.10.009
BibTeX
@article{Rao2026,
author = {Rao, Vibha R and Workman, Adrienne A and Palisoul, Scott M and Limoge, Cassandra J and Rawal, Pratha K and Vaickus, Louis J and Zanazzi, George J and Lu, Liang and Liu, Xiaoying and Sukhadia, Shrey S},
title = {Interpreting and Validating a Deep Learning Model Predictive of Spatial Morphologic-Molecular Patterns in Lung Adenocarcinoma, Using Ground Truth Immunohistochemistry Images},
journal = {The American Journal of Pathology},
year = {2026},
doi = {10.1016/j.ajpath.2026.10.009}
}
EUVIMED