AI-based Predictive Signaling Pathway Profiling in Cardiac Fibrosis Suggests a Novel Combinatorial Treatment...

Area of Science:

• Cardiovascular research
• Biomedical engineering
• Computational biology

Background:

• Cardiovascular disease (CVD) is a leading cause of death globally.
• Myocardial fibrosis, marked by excessive extracellular matrix (ECM) deposition, drives cardiac dysfunction.
• Current anti-fibrosis therapies are limited by an incomplete understanding of the fibrotic process.

Purpose of the Study:

• To leverage artificial intelligence (AI) to identify key signaling pathways in cardiac fibrosis.
• To develop an optimized, temporally-regulated therapeutic strategy for cardiac fibrosis.
• To differentiate therapeutic targets for acute versus chronic fibrotic conditions.

Main Methods:

• Trained an AI model on 6,528 cardiac fibrosis publications to identify 10 nodal signaling pathways.
• Utilized single-cell RNA sequencing to analyze pathway activity in mouse models and clinical samples.
• Developed and tested a temporal inhibition strategy for JAK-STAT and TGF-β signaling.

Main Results:

• AI identified JAK-STAT and TGF-β as critical pathways in cardiac fibrosis.
• JAK-STAT signaling peaked early in acute myocardial infarction (MI) and chronically in transverse aortic constriction (TAC).
• Fibroblast-specific Jak1/2 deletion reduced TAC fibrosis but not MI fibrosis, indicating differential pathway roles.
• A temporal inhibition strategy for JAK-STAT and TGF-β improved cardiac function and reduced fibrosis post-MI.

Conclusions:

• An AI-driven, temporally-regulated inhibition of JAK-STAT and TGF-β signaling is a promising therapeutic strategy for cardiac fibrosis.
• This approach effectively reduces pathological fibrosis while preserving necessary acute compensatory scarring.
• The strategy involves ruxolitinib (RUX) for JAK-STAT inhibition and pirfenidone (PFD) for TGF-β inhibition post-MI.