Emilian Jungwirth

Metabolic pressure shifts signaling pathways of nuclear receptors, including the bile acid receptor FXR, which are sensitive to nutritional inputs. We performed an FXR ChIP-seq–centered multiomic analysis of liver biopsy samples from individuals with or without obesity, who were treated with either placebo or the FXR agonist obeticholic acid, to define metabolic adaptions of FXR signaling pathways. FXR occupied substantially more DNA binding sites in individuals with obesity, and FXR activation by OCA robustly changed the transcriptional output. Integration of ChIP-seq and RNA-seq data showed that mitochondrial function and substrate oxidation were the top metabolic pathways selectively modulated by FXR activation in individuals with obesity. FXR activation restored compromised substrate oxidation by enhancing β-oxidation and oxidative phosphorylation along with antagonizing ROS production. In line with this, the amount of reduced glutathione in patients with obesity normalized after OCA treatment. In summary, FXR signaling profoundly differs in patients with obesity, consisting of changes in DNA binding profiles and transcriptional programs, which enhance energy substrate utilization in this patient cohort.

Studies have shown that colorectal cancer prognosis can be predicted by deep learning-based analysis of histological tissue sections of the primary tumor. So far, this has been achieved using a binary prediction. Survival curves might contain more detailed information and thus enable a more fine-grained risk prediction. Therefore, we established survival curve-based CRC survival predictors and benchmarked them against standard binary survival predictors, comparing their performance extensively on the clinical high and low risk subsets of one internal and three external cohorts. Survival curve-based risk prediction achieved a very similar risk stratification to binary risk prediction for this task. Exchanging other components of the pipeline, namely input tissue and feature extractor, had largely identical effects on model performance independently of the type of risk prediction. An ensemble of all survival curve-based models exhibited a more robust performance, as did a similar ensemble based on binary risk prediction. Patients could be further stratified within clinical risk groups. However, performance still varied across cohorts, indicating limited generalization of all investigated image analysis pipelines, whereas models using clinical data performed robustly on all cohorts.