CAR-T Engineering and Functional Optimization
Using the CellOS virtual cell model to predict how candidate gene perturbations may affect T cell states and function-related molecular pathways, we help screen engineering targets linked to exhaustion, expansion capacity, and tumor-killing activity, supporting the development of in vivo CAR-T cell therapies.
Project Background
A leading Chinese cell therapy company focused on in vivo CAR-T sought to use the AIVC platform to improve T cell resistance to exhaustion and in vivo persistence, screen delivery approaches, and accelerate validation of candidate engineering strategies.
The Challenge
| 1 | In vivo T cell–targeted delivery is complex, making it hard to balance delivery efficiency and specificity |
| 2 | Variability in CAR expression and T cell functional states makes intervention outcomes hard to predict |
| 3 | T cell exhaustion and in vivo persistence are difficult to optimize together |
| 4 | Long in vitro and in vivo functional validation cycles drive up candidate screening costs |
Solution
| 1 | Build a T cell state model with CellOS to decode the molecular features of different subsets |
| 2 | Combine with delivery experiment data to assess the link between T cell subsets and CAR expression |
| 3 | Use AuraPert to screen candidate targets related to anti-exhaustion and functional persistence |
| 4 | Validate candidate targets with in vitro and in vivo functional experiments and iteratively refine the model |
Core Value
Decode T cell subset differences to support in vivo delivery strategy research
Screen anti-exhaustion candidate targets to support CAR-T functional optimization
Reduce inefficient functional screening and lower R&D validation costs
Accelerate validation of candidate approaches to support next-generation CAR-T development
Case Gallery
