The discovery of molecular subtypes has corroborated the heterogeneity of breast cancers. But in clinical routine, treatment selection relies on measuring the aggressiveness of the tumor via histopathology, which is routinely based on grading, a prognostic yet laborious, poorly reproducible, and subjective procedure defined more than 30 years ago, before molecular subtypes existed. In this paper, we present an interpretable deep learning-based computer-aided diagnosis method to automate mitotic count, one of the components of grading; the method complies with current guidelines and automates hot-spot finding on full digital whole-slide images of routinely prepared pathologic slides stained with hematoxylin and eosin. We use the computer model to query the current “one-size-fits-all” mitotic count procedure by investigating subtype-specific prognostic value of mitotic count using an external independent large-scale multicentric dataset of 1,709 breast cancers. We show that for human epidermal growth factor-2 (HER2) positive tumors, mitotic count was not prognostic. In hormone-receptor positive/HER2 negative cancers, an average mitotic count density different from clinically used cut offs predicted worse recurrence free survival in multivariable analysis, adjusted for known clinical prognosticators. We provided some first experimental evidence that the prognostic role of mitotic count should be reviewed.

Ovarian cancer is a complex disease with poor outcome affecting women worldwide. The lack of successful therapeutic options for ovarian cancer patients results in the strong need to identify new biomarkers for patient selection. The development of outcome predictors based on gene expression is important not only for patient stratification but also to recognize categories of patients that are more likely to respond to particular therapies. In this paper, we proposed a new deep learning survival model trained on the high-dimensional transcriptomic data for the task of ovarian cancer prognostication. We validated our deep learning survival model on an independent clinical and molecular datatset. Finally, we illustrated the way our model can be interpreted, by calculating the contributions of the input features to the network outputs. We demonstrated how these contributions can be related to the molecular pathways to uncover biological processes associated with ovarian cancer patients survival.

Background: Between 17 to 25% of recurrent or metastatic endometrial carcinoma (EC) are MSI-H which results in improved response to immune checkpoint inhibitors (ICI). However, little is known regarding chemotherapy sensitivity in MSI-H EC, especially response to first-line platinum-based treatment.