In 2020, a Yale-led team created a high-resolution atlas of all the cells in the human lung, an ambitious project that yielded insights into how cells are affected by the disease Idiopathic Pulmonary Fibrosis (IPF), which induces progressive scarring of lung tissue.
The researchers, led by Dr. Naftali Kaminski, are now taking that project one step further, expanding the atlas to identify signatures specific to different stages of the disease and then, with the help of artificial intelligence, exploring compounds that might reverse those signatures.
They hope the new advancements might inspire new therapies for treating pulmonary fibrosis in a way that is much faster and relevant to humans than relying on animal studies.
They are calling the new project the Pulmonary Fibrosis (PF) Connectome.
“Drug companies have libraries of thousands of compounds that could be used for therapeutics,” said Kaminski, the Boehringer Ingelheim Pharmaceuticals Inc. Professor of Medicine (pulmonary) and section chief of pulmonary, critical care, and sleep medicine at Yale School of Medicine. “By identifying compounds that reverse the cell and disease stage fibrosis signatures, we could prioritize those that are likely to stop or even reverse pulmonary fibrosis, not only slow it down.”
Once they’ve identified around 100 high-priority compounds, he said, researchers will test these on lung slices, cultured in the lab from human lungs.
In patients with pulmonary fibrosis, increased scarring of the lung makes breathing progressively more difficult and prevents oxygen from reaching the bloodstream. The disease is often fatal. Across the country, an estimated 200,000 people suffer from IPF, and 40,000 die annually from the disease.
Current FDA-approved therapies work by slowing disease progression. Kaminski, who has been at the forefront of pulmonary fibrosis research for years, said that the scientific understanding has increased dramatically in recent years. But these insights have had limited impact on new therapies because “much of drug development uses models that have little semblance to the human lung.”
“Using human tissue from stages of progressing disease to determine the potential antifibrotic efficacy of compounds could overcome this limitation,” he said.
Three Lakes Foundation, a nonprofit organization dedicated to better understanding IPF, is providing significant funding for the two-year PF Connectome project, as part of the multimillion budget for the Three Lakes Consortium for Pulmonary Fibrosis (TLC4PF).
Kaminski said that the research team hopes to emerge with three to six promising drug candidates for a clinical study. He was recently part of a team that discovered that a cancer and Alzheimer’s drug, saracatinib, could be a potential therapy for IPF using methods similar to the ones that will be employed in the PF Connectome.
“We would never have considered this drug,” he said, “but because of computational analysis, we could build a rationale. If it were not for COVID-19, this trial would have started a year ago, but I’m glad to say that now we are recruiting.” (Clinical trial information is available here.)
Using human tissue in the search for effective drugs offers a marked contrast to relying on mouse models. That approach can take decades to reach human patients due to the slow testing and approval process. Between the acceleration of single-cell RNA sequencing — which allows scientists to profile in detail thousands of individual cells — the discovery of genetic biomarkers for fibrosis, and new insights into treatment pathways revealed by the IPF cell atlas, researchers are at a tipping-point moment, Kaminski said.
This research could also advance understanding of the long-term consequences of COVID-19, Kaminski added, noting that a small percentage of COVID-19 patients develop fibrotic complications in their lungs. While the connection between the two diseases is not yet fully understood, he said there are important parallels that his research might help uncover.
Other members of the research team include Ziv Bar-Joseph, a professor of computational biology and machine learning at Carnegie Mellon University; Geremy Clair, a scientist at the Pacific Northwest National Laboratory; Jun Ding, an assistant professor at McGill University; Dr. Oliver Eickelberg, a professor at the University of Pittsburgh’s Department of Medicine; and Xiting Yan, assistant professor of pulmonary and biostatistics in Yale School of Medicine’s Center for Precision Pulmonary Medicine (P2MED).