St. Jude Develops AI-Powered Tool to Accelerate Dual-Target CAR-T Cell Therapy Design
核心洞察
St. Jude researchers developed a computational tool that screens thousands of tandem CAR designs in days, dramatically accelerating development of dual-target immunotherapies.
The AI-informed algorithm successfully optimized CAR-T cells targeting pediatric brain tumor proteins B7-H3 (搜索) and IL-13Rα2 (搜索), completely clearing tumors in four out of five mice.
The computational approach overcame key challenges in tandem CAR design, including poor surface expression and suboptimal cancer-killing ability that have limited effectiveness against solid tumors.
St. Jude Children's Research Hospital scientists have developed a computational approach that promises to make designing dual-target CAR-T cell immunotherapies significantly faster and more effective. The breakthrough addresses a critical limitation in treating solid and brain tumors, where single-target CAR-T cells have shown limited success compared to their effectiveness against blood cancers.
The research team created an AI-informed algorithm that can screen approximately 1,000 theoretical tandem CAR designs in a matter of days, a process that would traditionally take many years in laboratory settings. The findings were published in Molecular Therapy.
Overcoming Single-Target Limitations
While CAR-T cells have successfully treated some blood cancers, they have not been as effective in treating solid and brain tumors. Cancer cells do not uniformly express the same proteins, so CAR-T cells targeting a single antigen can miss malignant cells that do not express that protein, leaving them to regrow the tumor and cause difficult-to-treat relapses.
"We have developed and validated a computational tool that can significantly accelerate the design of tandem CAR constructs with improved surface expression and anti-tumor function," said co-corresponding author Giedre Krenciute, PhD, from St. Jude's Department of Bone Marrow Transplantation & Cellular Therapy.
Scientists have attempted to create CARs that target two proteins simultaneously, but have encountered significant problems, including poor CAR expression on T cell surfaces and suboptimal cancer-killing ability.
Computational Pipeline Success
The computational pipeline predicted an improved design for a tandem CAR targeting pediatric brain tumor-related proteins B7-H3 (搜索) and IL-13Rα2 (搜索). The original unoptimized version of the bi-specific tandem CAR failed to reach the T cell surface, preventing it from contacting target proteins on tumor cells to perform its cancer-killing function.
After confirming that the computationally optimized CAR expressed on the T cell surface, researchers tested it against several single-targeted CARs in mice with heterogeneous tumors containing cells with both targets, one target or the other, or neither target, mimicking clinical tumor heterogeneity.
"Our most compelling result is that we completely cleared tumors in four out of five mice with the CAR-T cells that had the computationally optimized tandem construct," said co-first author Michaela Meehl from St. Jude's Department of Bone Marrow Transplantation & Cellular Therapy. "By contrast, all heterogeneous tumors treated with single-targeted CAR-T cells grew back."
AI-Informed Algorithm Design
The scientists trained their AI-informed algorithm on structural and biophysical features of known effective CARs, including predicted properties such as protein folding stability, tendency to aggregate, and other structural and functional features. The program summed these features into a single "fitness" score predicting CAR expression and functionality, with the highest-scoring designs further optimized to improve protein binding ability.
"Systematic experimental dissection allowed us to first pinpoint the region within the tandem CAR that was problematic for expression and function," said co-corresponding author M. Madan Babu, PhD, FRS, St. Jude Senior Vice President of Data Science. "This was important and helped guide our efforts as we developed a computational approach for CARs that cleared tumors in our in vivo models more effectively than any single-targeted CAR we tested."
Broad Applicability Demonstrated
The research team demonstrated that their computational method could improve the design of several other tandem CARs in laboratory testing. In all cases, the computationally optimized versions killed cancer cells better than non-optimized tandem CARs, providing evidence that the design of other bi-specific tandem CARs can benefit from this computational method.
"We designed this computational tool to be broadly applicable to many different CARs," said co-first author Kalyan Immadisetty from St. Jude's Department of Bone Marrow Transplantation & Cellular Therapy.
The research represents a significant step toward successfully treating challenging tumors, particularly pediatric brain cancers, by enabling researchers to screen and create better tandem CARs more efficiently than traditional experimental approaches.