ㅤ
Article by Youngsoo Han, Ilaria Baldassarri & Katherine Cunningham
Lab-Grown Breast Tumors to Accelerate Personalized CAR-T Therapy
Source Publication:
Breast cancer-on-chip for patient-specific efficacy and safety testing of CAR-T cells, Cell Stem Cell, 2024, Maulana et al., Peter Loskill lab
The long-term goal of chimeric antigen receptor (CAR)-T cell therapy is to treat tumors, such as in breast cancer, where immune cells often struggle to enter the tumor, survive within its immunosuppressive microenvironment, and recognize appropriate targets. In this study, the researchers built a three-dimensional "breast cancer-on-a-chip" model that recreates this hostile microenvironment to answer two key questions: can CAR-T cells overcome the engineered barriers and break into a lab-grown tumor, and can this system help predict therapeutic efficacy and safety?
What did these researchers do?
The researchers identified key features of human breast tumors that influence CAR-T cell trafficking and efficacy to mimic them within a microphysiological system (MPS). Cancer cells were grown inside tiny chambers with underlying channels containing endothelial cells as blood vessels. The researchers could then flow CAR-T cells through these artificial vessels to determine whether the cells could successfully reach and destroy the tumor. CAR-T cells were engineered to target a protein found on some breast cancer cells, ROR1, and evaluated by their ability to cross the endothelium, infiltrate the tumor, and kill cancer cells, measured by levels of the target antigen they displayed. Researchers also tested whether a drug, Dasatinib, could prevent CAR-T cell activity from becoming too aggressive. While initial experiments were conducted using cancer cell lines, which are valuable for benchmarking but not for capturing the complexity of patient tumors, the researchers concluded their work by incorporating mini-tumors derived from patient specimens – also known as organoids – into the chip model to demonstrate potential for personalized medicine. In doing so, they evaluated whether the system could predict patient-specific responses by measuring how effectively CAR-T cells eliminated tumors with different levels of antigen expression.
Why is this important?
Currently, CAR-T therapy works well for blood cancers but often fails to treat solid tumors where cells cannot penetrate the tumor tissue. Doctors currently lack reliable predictive human models of a patient’s tumor response, including severe side effects such as cytokine release syndrome (CRS). The MPS developed here addresses this need, as standard lab dishes are too simple to capture the clinical challenges immune cells face once infused in patients, and mouse models require significant time and resources with limited capacity for human mimicry. This platform provides a scalable human model that recreates key aspects of the tumor microenvironment (TME) contributing to clinical barriers. By combining efficacy testing with safety monitoring, it offers a way to evaluate both whether a therapy is likely to work and whether it may trigger dangerous immune overreactions before moving it to the clinic.
How did the researchers do this?
The researchers designed a microfluidic platform featuring two interfacing channels separated by a porous membrane to simulate the blood vessel lining through which the tumor cells infiltrate. In the tumor compartment, they embedded breast cancer spheroids within a hydrogel to mimic the stiff extracellular matrix (ECM) encasing tumor tissue. To replicate vascular circulation, they perfused medium along the vascular channel. CAR-T cells could then be added to the perfused medium to allow migration towards the tumor. Several analytics were used to monitor system changes over time, including microscopic tracking of individual CAR-T cells as they breached the vessel wall, and routine sampling of the liquid outflow to detect spikes in inflammatory cytokines representative of CRS observed in clinical trials. To quantify therapeutic efficacy, they measured the tumor cell death and determined that killing capacity was strictly dependent on the density of the antigen on the surface of cancer cells. To address therapeutic safety concerns, they tested whether the drug Dasatinib could function as a tunable off-switch to pause immune attack. Finally, they integrated tumor organoids into the chip to demonstrate that the platform could be used to investigate specific patient responses.
Schematic representation of the breast-cancer-on-a-chip platform and its applications
What comes next?
While this tumor-on-a-chip model captures a physical barrier to CAR-T cell trafficking – the endothelium – and successfully highlights the importance of antigen density for therapeutic efficacy, it does not yet include many of the immune-suppressive cell types that make solid tumors particularly difficult to treat, such as tumor-associated macrophages and regulatory T cells. Without these, the experimental model could overestimate CAR-T cell efficacy in eliminating the tumor. Safety evaluations related to dangerous cytokine storms could be similarly limited by the lack of key immune players within the TME. Expanding the system to incorporate these suppressive components – and extending experiments to assess CAR-T persistence beyond the current eight-day window – will be essential for addressing the major challenges of solid tumor immunotherapy. Finally, increasing the number of patient-derived samples will be necessary to validate whether this platform can reliably predict treatment responses across a broader patient population.