AI-guided CAR designs and targeted pathway modulation to enhance multi-antigen CAR T cell durability and overcome...

Area of Science:

• Immunotherapy
• Artificial Intelligence
• Molecular Biology

Background:

• CAR T-cell therapy faces challenges with persistence and antigen escape in hematologic malignancies.
• Developing CAR T-cell constructs that resist dysfunction and improve durability is crucial for therapeutic success.

Purpose of the Study:

• To develop an AI-guided CAR T-cell platform integrating structure-based design and targeted protein degradation.
• To enhance CAR T-cell persistence, broaden antigen coverage, and achieve durable anti-tumor efficacy.

Main Methods:

• AI-guided design and in vitro screening to develop a predictive model (CARMSeD) for CAR construct optimization.
• Incorporation of a PROTAC-based module to selectively degrade AKT3, enhancing mitochondrial fitness and T-cell memory.
• Development of a trispecific CAR T platform with a secretable bispecific engager for broader antigen targeting.

Main Results:

• Optimized bispecific CD20/CD19 CAR T cells showed improved persistence and anti-tumor activity.
• The PROTAC module promoted central memory differentiation and reduced mTOR signaling.
• The trispecific platform achieved potent tumor eradication, including in antigen-negative models and patient samples.

Conclusions:

• This next-generation AI-guided CAR T strategy enhances T-cell persistence and broadens antigen targeting.
• The integration of structure-based optimization and intracellular modulation offers a promising approach for durable cancer therapy.
• The platform demonstrates significant potential across various hematologic malignancies and solid tumor models.