Artificial intelligence has identified potential new therapies for idiopathic pulmonary fibrosis (IPF), a complex and debilitating lung disease. Researchers have developed a deep learning generative model called UNAGI, which can track disease progression and identify potential therapies by analyzing cell types and disease-related genes.
The UNAGI model predicted that nifedipine, a calcium channel blocker commonly used to treat hypertension, might block scar tissue formation in IPF. Experiments on human lung tissue samples confirmed this prediction, suggesting that nifedipine could be a potential therapeutic option for IPF.
Additionally, UNAGI identified two histone deacetylase (HDAC) inhibitors, apicidin and belinostat, as potential therapeutic options for IPF. While apicidin's development has been hindered by safety concerns, belinostat is already approved as a cancer therapy.
Another promising development is the discovery of rentosertib, an AI-discovered TNIK inhibitor that has shown safety and tolerability in a phase 2a trial for IPF. Rentosertib works by disrupting profibrotic pathways and reducing extracellular matrix accumulation.
Furthermore, a novel PDE4B inhibitor called nerandomilast has slowed lung function decline in patients with progressive pulmonary fibrosis (PPF) in a phase 3 trial. Nerandomilast has shown consistent efficacy regardless of background antifibrotic therapy.
The use of AI in IPF research has the potential to revolutionize the field by accelerating drug discovery, enabling personalized medicine, and improving patient outcomes. By rapidly analyzing vast amounts of data, AI can identify potential therapeutic targets and design new therapies, offering new hope for patients with IPF.
