Back

Eco-evolutionary Guided Pathomic Analysis to Predict DCIS Upstaging

Xiao, Y.; Elmasry, M.; Bai, J.; Chen, A.; Chen, Y.; Jackson, B.; Johnson, J. O.; Gillies, R. J.; Prasanna, P.; Chen, C.; Damaghi, M.

2024-10-26 cancer biology
10.1101/2024.06.23.600274 bioRxiv
Show abstract

Cancers evolve in a dynamic ecosystem. Thus, characterizing cancers ecological dynamics is crucial to understanding cancer evolution and can lead to discovering novel biomarkers to predict disease progression. Ductal carcinoma in situ (DCIS) is an early-stage breast cancer characterized by abnormal epithelial cell growth confined within the milk ducts. In this study, we show that ecological analysis of hypoxia and acidosis biomarkers can significantly improve prediction of DCIS upstaging. First, we developed a novel eco-evolutionary designed approach to define habitats in the tumor intraductal microenvironment based on oxygen diffusion distance. Then, we identified cancer cells with metabolic phenotypes attributed to their habitats, including CA9 for hypoxia responding phenotype, and LAMP2b for acid adapted phenotype. Traditionally these markers have shown limited predictive capabilities for DCIS progression, if any. However, when analyzed from an ecological perspective, their power to differentiate between non-upstaged and upstaged DCIS increased significantly. Second, we discovered distinct niches with spatial patterns of these biomarkers and used the distribution of such niches to predict patient upstaging. The niches were characterized by pattern analysis of both cellular and spatial features. With a 5-fold validation on the biopsy cohort, we trained a random forest classifier to achieve the area under curve (AUC) of 0.74. Our results affirm the importance of tumor ecological features in eco-evolutionary-designed approaches for novel biomarkers discovery. SignificanceOur results affirm the importance of spatial and ecological features in eco-evolutionary-designed biomarkers discovery studies in the era of digital pathology.

Matching journals

The top 10 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.