Local CAR-T as a novel therapeutic strategy for CNS progression in NSCLC

Career Development Award
Elliott Brea, MD, PhD
Dana-Farber Cancer Institute
Boston
MA

Lung cancer is the leading cause of cancer deaths, with patients with EGFR-mutant NSCLC frequently developing brain metastases that are difficult to treat with existing therapies. Dr. Brea and his team are studying CAR-T cells that target TROP2, a protein commonly found on EGFR-mutant lung cancer cells. Targeting TROP2 can be diffficult because TROP2 is found on healthy cells.  To overcome this, Dr. Brea proposes two approaches:  (1) delivering CAR-T cells directly into the central nervous system (CNS). The CNS does not express TROP2 but is a site where tumors often reside, and (2) genetically editing the CAR-T cells to prevent them from entering healthy organs like lungs and gut. If successful in preclinical testing, this could provide a new targeted treatment option for patients with EGFR-mutant lung cancer with brain metastases.

Research Summary

Lung cancer is the leading cause of cancer-related deaths, and patients with a specific type called EGFR-mutant non-small cell lung cancer (NSCLC) often develop spread to the central nervous system (CNS). These CNS metastases are difficult to treat and are not well controlled by existing therapies like radiation or even high doses of targeted pills such as Osimertinib. Many clinical trials exclude patients whose cancer has spread to the CNS, leaving few treatment options for this vulnerable group. 

We are developing a promising new approach using the patient’s own immune system, called CAR-T cell therapy. We have demonstrated that CAR-T cell that targets a protein called TROP2, which is commonly found on EGFR-mutant lung cancer cells, is effective against EGFR-mutant NSCLC. One barrier to translation to trials is that CAR-T cells can also harm healthy tissues that have TROP2. To overcome this, we propose two key innovations: first, we will deliver the CAR-T cells directly into the CNS in patients with metastatic CNS, where tumors reside but TROP2 is not expressed in normal CNS, avoiding toxicity. Second, we will edit the CAR-T cells to remove a molecule that allows them to enter normal tissues like the lungs and gut, further reducing side effects. Our research will test this strategy in patient derived preclinical models to show that it is both safe and effective. If successful, this could lead to a new, highly targeted treatment for patients with EGFR-mutant lung cancer with CNS metastases.

Technical Abstract

EGFR-mutant non-small cell lung cancer (NSCLC) frequently metastasizes to the central nervous system (CNS). While EGFR inhibitors such as Osimertinib or radiation can be effective initially, patients often develop further CNS progression with limited durable options. Moreover, patients with active CNS disease are often excluded from clinical trials, leaving a critical unmet need for more effective and CNS-active therapies. We have developed a novel TROP2-directed chimeric antigen receptor (CAR)-T cell therapy for EGFR-mutant NSCLC. TROP2 is a clinically validated target in this setting, with high expression in EGFR-mutant tumors and preserved expression in CNS metastases. 

Our preclinical studies show that TROP2 CAR-T cells display potent cytotoxicity, superior to TROP2-directed antibody-drug conjugates. However, systemic delivery of CAR-T cells targeting epithelial antigens carries substantial risk of on-target/off-tumor toxicity. To address this, we propose a combinatorial strategy: (1) localized intraventricular delivery of TROP2 CAR-T cells to the CNS, and (2) CRISPR/Cas9-mediated knockout of the integrin aEß7, a homing molecule that mediates T cell trafficking to epithelial tissues. This approach is designed to retain efficacy while reducing systemic toxicity. We will evaluate TROP2 and aEß7 expression in patient CNS metastases, optimize TROP2 CAR-T designs using patient derived ex vivo and vascularized tissue platforms, and test safety and efficacy in EGFR-mutant CNS xenograft models and humanized TROP2 transgenic mice. This proposal advances a rationally engineered CAR-T strategy for a patient population with few options and lays the groundwork for a CNS-directed clinical trial.