A novel artificial intelligence (AI) tool uncovered an effective treatment for a patient with idiopathic multicentric Castleman’s disease, according to a recent study published by Mumau et al in The New England Journal of Medicine.
Background
Idiopathic multicentric Castleman’s disease is a rare cytokine storm disorder with a poor survival rate and few treatment options. Cytokine storms are characterized by an exaggerated and harmful response by the immune system in which too many inflammatory cytokines are released and can damage the body’s tissues and organs. Those with idiopathic multicentric Castleman disease can experience swelling of the lymph nodes, inflammation throughout their body, and life-threatening multiorgan failure.
The process of using an existing drug for a purpose other than its initial intent is called drug repurposing. Many diseases may appear very different in symptoms, prognosis, or etiology, but could share some underlying links in the body like common genetic mutations or molecular triggers. Therefore, some diseases can be treated with the same drug.
Study Methods and Results
In the study, researchers used an AI technique known as machine learning to evaluate 4,000 existing medications.
The AI tool determined that the tumor necrosis factor (TNF) inhibitor adalimumab—a monoclonal antibody that is approved by the U.S. Food and Drug Administration to treat conditions ranging from arthritis to Crohn’s disease—was the top-predicted new treatment that was likely to be effective for idiopathic multicentric Castleman disease.
Parallel experiments also found that the TNF protein was potentially playing a key role in idiopathic multicentric Castleman disease. The researchers detected elevated TNF signaling levels in patients with the most severe forms of the disease. Further analysis showed that immune cells from patients with idiopathic multicentric Castleman disease produce more TNF when activated compared with those of healthy individuals.
As a result of the findings, the researchers assigned a patient with idiopathic multicentric Castleman disease who didn’t respond to multiple prior lines of therapy to receive adalimumab.
“The patient in this study was entering hospice care, but now he is almost 2 years into remission,” detailed senior study author David Fajgenbaum, MD, Associate Professor of Translational Medicine and Human Genetics and a physician at the University of Pennsylvania as well as co-founder of the nonprofit Every Cure. “This is remarkable not just for this patient and [idiopathic multicentric Castleman disease], but for the implications it has for the use of machine learning to find treatments for even more conditions,” he continued.
Conclusions
The researchers noted that although idiopathic multicentric Castleman disease is relatively rare—with about 5,000 diagnoses in the United States each year—the findings could save the lives of many more patients. They suggested that the patient involved in the study could be one of the first to receive life-saving treatment with the help of an AI prediction system, which could potentially be applied to other rare conditions.
The researchers highlighted the importance and power of combining a range of scientific approaches vs using AI, laboratory work, or clinical research methods alone. They plan to launch a clinical trial in 2025 to explore the effectiveness of another repurposed drug—a Janus kinase 1/2 inhibitor—in patients with idiopathic multicentric Castleman disease.
“There are probably a few hundred patients in the United States and few thousand patients around the world who, each year, are in the midst of a deadly flare-up like this patient had been experiencing,” underscored Dr. Fajgenbaum. “More research is needed, but I’m hopeful that many of them could benefit from this new treatment,” he concluded.
Disclosures: The research in this study was supported by the National Heart, Lung, and Blood Institute; the FDA; the National Center for Advancing Translational Sciences (NCATS); the Advanced Research Projects Agency for Health; the Castleman Disease Collaborative Network; Every Cure; BioAegis; and the NCATS Biomedical Data Translator Initiative; the Chan Zuckerberg Initiative; Lyda Hill Philanthropies; Arnold Ventures; the Elevate Prize Foundation; and the Carolyn Smith Foundation. For full disclosures of the study authors, visit nejm.org.
