Supplementary Figure S1 from Deep Learning Predicts Subtype Heterogeneity and Outcomes in Luminal A Breast Cancer Using Routinely Stained Whole-Slide Images

Kurian, Nikhil Cherian; Gann, Peter H.; Kumar, Neeraj; McGregor, Stephanie M.; Verma, Ruchika; Sethi, Amit

doi:10.1158/2767-9764.28286739

Supplementary Figure S1 from Deep Learning Predicts Subtype Heterogeneity and Outcomes in Luminal A Breast Cancer Using Routinely Stained Whole-Slide Images

journal contribution

posted on 2025-01-27, 11:20 authored by Nikhil Cherian Kurian, Peter H. Gann, Neeraj Kumar, Stephanie M. McGregor, Ruchika Verma, Amit Sethi

Supplementary Figure S1. Scatterplot of LumA proportion by transcriptomic analysis versus percentage of tumor image patches classified as LumA by the DNN model - including held-out cases with nonLumA PAM50 assignment (n = 256). Best-fitting regression line shown, with 95% confidence band. Horizontal lines depict quartile thresholds for number of cases.

History

ARTICLE ABSTRACT

A deep learning model, trained using transcriptomic data, inexpensively quantifies and fine-maps ITH due to subtype admixture in routine images of LumA breast cancer, the most favorable subtype. This new approach could facilitate exploration of the mechanisms behind such heterogeneity and its impact on selection of therapy for individual patients.