Sexually dimorphic computational histopathological signatures prognostic of overall survival in high-grade gliomas via deep learning.

in Science advances by Ruchika Verma, Tyler J Alban, Prerana Parthasarathy, Mojgan Mokhtari, Paula Toro Castano, Mark L Cohen, Justin D Lathia, Manmeet Ahluwalia, Pallavi Tiwari

TLDR

  • The study used deep learning on H&E-stained slides to identify sex-specific histopathological attributes of the TME and create sex-specific risk profiles for HGG.
  • The results showed sex-specific C-index values for the female and male cohorts, suggesting the potential of machine learning-based methods for patient-centric prognostic risk assessment.
  • The study's findings have implications for the development of personalized treatment plans for HGG patients, considering sex as a key factor.

Abstract

High-grade glioma (HGG) is an aggressive brain tumor. Sex is an important factor that differentially affects survival outcomes in HGG. We used an end-to-end deep learning approach on hematoxylin and eosin (H&E) scans to (i) identify sex-specific histopathological attributes of the tumor microenvironment (TME), and (ii) create sex-specific risk profiles to prognosticate overall survival. Surgically resected H&E-stained tissue slides were analyzed in a two-stage approach using ResNet18 deep learning models, first, to segment the viable tumor regions and second, to build sex-specific prognostic models for prediction of overall survival. Our mResNet-Cox model yielded C-index (0.696, 0.736, 0.731, and 0.729) for the female cohort and C-index (0.729, 0.738, 0.724, and 0.696) for the male cohort across training and three independent validation cohorts, respectively. End-to-end deep learning approaches using routine H&E-stained slides, trained separately on male and female patients with HGG, may allow for identifying sex-specific histopathological attributes of the TME associated with survival and, ultimately, build patient-centric prognostic risk assessment models.

Overview

  • The study aimed to identify sex-specific histopathological attributes of the tumor microenvironment (TME) in high-grade glioma (HGG) and create sex-specific risk profiles to prognosticate overall survival.
  • The researchers used an end-to-end deep learning approach on hematoxylin and eosin (H&E) scans to segment viable tumor regions and build sex-specific prognostic models for overall survival prediction.
  • The study demonstrated the potential of machine learning-based methods using routine H&E-stained slides to develop patient-centric prognostic risk assessment models.

Comparative Analysis & Findings

  • The study found that the mResNet-Cox model yielded sex-specific C-index values for the female (0.696, 0.736, 0.731, and 0.729) and male (0.729, 0.738, 0.724, and 0.696) cohorts across training and three independent validation cohorts.
  • The results suggest that deep learning approaches may allow for identifying sex-specific histopathological attributes of the TME associated with survival.
  • The study's findings demonstrate the potential of end-to-end deep learning approaches to develop patient-centric prognostic risk assessment models for HGG.

Implications and Future Directions

  • The study's results highlight the importance of considering sex as a key factor in developing personalized treatment plans for HGG patients.
  • Future studies should explore the potential of integrating sex-specific histopathological attributes into clinical decision-making for HGG patients.
  • The development of sex-specific prognostic models using machine learning-based methods may be a valuable tool in predicting patient outcomes and guiding treatment strategies.
Read Full Article