Researchers from the Keck School of Medicine of USC are receiving up to $6.8 million for a two-year research project to develop new computational models and support tools that could accelerate access to cell and gene therapies for children with rare diseases. The team will develop a new framework that combines detailed data about the biological features of each therapy and how patients respond to them. By using artificial intelligence (AI) to study these connections, the project aims to better understand how specific features of a therapy relate to patient outcomes. The research is funded by the Advanced Research Projects Agency for Health (ARPA-H) UNIfying Cell Therapy Outcome prediction and Regulatory Navigation (UNICORN) project, led by ARPA-H Program Manager Daria Fedyukina, Ph.D.
Scientific advances have led to the development of life-saving gene and cell therapy treatments for childhood cancers and genetic diseases. These personalized therapies engineer or genetically modify a patient’s own cells to target diseases at the cellular level. But unlike conventional drugs that can be produced and studied in large batches, cell and gene therapies involve living cells and are typically created one patient at a time in tightly controlled lab settings, making the process costly and complex. As a result, clinical development typically involves smaller patient populations and more limited datasets, making traditional trial models harder to apply.
Designing smarter therapies
UNICORN combines advanced cell analysis technology developed by the team with machine learning tools to identify biological patterns and therapy product features linked to treatment response. This approach aims to enable the development of a regulatory decision-support tool that guides interpretation of product-related evidence when limited data makes conventional measures difficult to establish, enabling patients and families to access new treatments sooner.
“This project reimagines how we develop therapies for rare diseases,” said Mohamed Abou-el-Enein , MD, PhD, principal investigator of the new project and executive director of the USC/Children’s Hospital Los Angeles (CHLA) Cell Therapy Program . “By working closely with colleagues across the field, we are building a scientific foundation that helps translate complex biological data into clearer evidence and guides the creation of more effective therapies.”
At the heart of UNICORN is an advanced cell-analysis platform that Abou-el-Enein’s laboratory built to study and improve chimeric antigen receptor (CAR) T cell therapies, which train the body’s immune cells to attack blood cancers such as leukemia and lymphoma. The platform uses a large panel of protein markers to measure the physical and functional properties of individual CAR T cells at the same time, enabling researchers to track how they evolve during manufacturing and identify characteristics that may be linked to stronger potency. That initial work helped secure the ARPA-H funding, which will now expand the platform’s use to characterize other cell and gene therapy products and examine how their properties relate to patient outcomes.
“The technology driving this work was developed here in our lab at USC. Now it’s part of a unique effort that combines advanced cell analytics with artificial intelligence and machine learning workflows to create a blueprint for developing small-batch therapies that others can learn from and build on,” Abou-el-Enein said.
Building the framework
To build UNICORN, the team will first focus on creating a strong foundation of high-quality data. They will collect detailed information on cell and gene therapy manufacturing and measure key characteristics of the therapeutic products that may affect their quality and consistency.
Next, they will collect patient data and samples at multiple time points across a range of pediatric diseases, in collaboration with researchers at several U.S. academic centers. These patient samples will be analyzed with the team’s cell-analysis platform, and the resulting data will help train the AI model to identify biological patterns associated with patient outcomes.
The model will be applied to several types of gene and cell therapies, including CAR T cell therapy, hematopoietic stem cell (HSC)-derived therapies, which modify the immature stem cells that give rise to all types of blood cells, and other pediatric gene-edited products, which repair genetic defects in a child’s own cells.
A learning system for rare diseases
The project is ambitious, particularly because it seeks to build meaningful AI models from relatively small patient populations that would translate to a regulatory decision-support tool. To address this, the team collects and analyzes data from patients at multiple time points, increasing the depth and value of each case studied.
“Every child treated adds new data that strengthens the model for the next one. It’s a living, learning system designed to get smarter with each patient,” Abou-el-Enein said.
UNICORN is also supported by an electronic quality and data management platform called Bluecord, implemented at USC and supported in part by prior funding from the California Institute for Regenerative Medicine (CIRM). This system enables standardized sample tracking, secure data integration across participating centers, and structured linkage of product and clinical information.
“Our vision is simple: when a child’s life depends on one therapy, we should be able to move forward with confidence because we’ve built the systems and the evidence to guide us,” Abou-el-Enein said.
About this research
This work will be carried out in the Abou-el-Enein Laboratory at the University of Southern California (USC) in collaboration with several academic partners across the United States. The project has been highlighted in a Nature Medicine Correspondence , which outlines the scientific vision behind the UNICORN framework.
This research is funded, in part, by the Advanced Research Projects Agency for Health (ARPA-H). The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the U.S. Government.