CRISPR-Edited Dual Stem Cell Therapy Offers Fresh Hopes for Brain Metastases

When cancer metastasises to the brain, it often spells a death sentence for patients; brain metastases cause half of all deaths from melanoma.

Khalid Shah, Professor at Harvard Medical School and Director of the Center for Stem Cell and Translational Immunotherapeutics, has taken aim at metastatic brain melanoma (MBM) in his latest research. The work specifically targets tumours in the leptomeninges, one of the thin layers of tissue that coats the brain and spinal cord.

»One of the bigger problems in brain metastases is when the tumours go to the leptomeninges. Leptomeningeal tumours are the worst tumours in the brain, and at that point, [almost] no-one survives,« Shah says of why he chose this particular disease.

Shah and his team at the Brigham and Women’s Hospital and Harvard Stem Cell Institute have developed an innovative therapy to treat deadly leptomeningeal metastases (LM). The new therapy is made up of two types of stem cells: one that delivers a cancer-killing oncolytic virus, and another that has been edited using CRISPR-Cas9 to be resistant to that virus, allowing it to deliver immune modulators to the tumour.

Published recently in Science Translational Medicine, the approach overcomes key obstacles that have prevented the clinical translation of this type of therapeutic modality. As Shah remarks, it also has implications for the treatment of other metastases.

»The innovation in this paper is how we managed to deliver the immune modulators. We created CRISPR-edited stem cells, which are resistant to the oncolytic virus delivered by the other population of stem cells. This approach is also very much applicable to all tumours that end up in the brain.«

Creating a precision model for a deadly disease

Accurate disease models are crucial when attempting to generate novel therapies for a complex disease like MBM. With no precision models available at the start of the project, the team needed to generate their own.

First, Shah and his team investigated expression of phosphatase and tensin homolog (PTEN), a regulator of cell survival, in advanced melanomas, finding that higher levels of PTEN expression were associated with higher overall survival. Loss of PTEN expression was seen in late-stage metastatic melanomas, particularly in brain metastases, meaning that to adequately study the disease, they would need to establish a PTEN-deficient model.

Further analysis demonstrated that populations of immune cells also differed between the PTEN-expressing and PTEN-deficient melanomas, and between primary and metastatic melanomas, indicating immunosuppressive mechanisms are at play in the poor clinical outcomes of MBM patients.

To create their model, the team used PTEN-deficient mouse melanoma cell lines to establish both primary and leptomeningeal metastatic melanoma tumours in live mice. After ensuring the profile of immune cells in the mouse tumours reflected the same composition and immunosuppressive nature of MBM in humans, the team were ready to develop and test their new therapy.

The twin stem cell approach: fixing delivery issues using CRISPR

Oncolytic viruses have been used to target cancers before. However, delivery challenges and the highly immunosuppressive environment of brain tumours means that oncolytic viruses haven’t progressed well as a treatment modality for brain metastases.

The new therapy overcomes those key issues, using two different populations of stem cells. One group of cells carries the oncolytic herpes simplex virus (oHSV), which is engineered to attack cancer cells, and delivers it to the tumour site within the brain. The other group of stem cells is edited using CRISPR-Cas9 reagents, delivered via a lentiviral vector, to knock out the cell surface adhesion molecule nectin-1 (CD111).

Because CD111 encodes the entry receptor for the oncolytic virus, its deletion renders the second population of cells resistant to the virus. This allows the resulting SC^N1KO population to survive when they travel to the tumour site and perform a critical job - delivering immune modulators. Other therapies have been developed in which the immune modulator and virus are delivered in the same cell. However, when the virus infects the cancer cell, it kills the cell and the immune modulator is destroyed in the process.

Shah’s approach allows the immune modulators to be delivered in resistant cells so they can do their job effectively. After screening several immune modulators in vivo to identify the best performer, the team chose granulocyte macrophage-colony stimulating factor (GM-CSF), an inflammatory cytokine. GM-CSF works in concert with the SC-oHSV population of cells to recruit immune cells to the tumour, improving tumour clearance.

»We sorted out a viral delivery issue, so we can now use stem cells to deliver the virus. We also sorted out how to deliver immune modulators in patient settings using a different type of cell that is resistant to the virus,« Shah explained.

The team also introduced immune checkpoint blockade, transducing the SC^N1KO cells with a single chain variable fragment (scFv) against PD-1 to combat the immunosuppressive programmed cell death-1 ligand 1 (PD-L1), which is expressed on PTEN-deficient melanoma cells.

Superior in vivo efficacy to existing oncolytic viral approaches

After early experiments using mouse stem cells to treat the original mouse model, the real test of the therapy was in humanised mice using allogeneic adult mesenchymal stem cells. Patient-derived PTEN-deficient MBM tumour cells were first injected intracranially into bone marrow–liver–thymic (BLT) humanised mice.

After establishment of LM in the mice, the dual stem cell therapy was delivered to the tumour site via intrathecal injection. The treatment was able to significantly suppress tumour growth and increase survival of the mice, without causing a loss of body weight. Subsequent immune profiling analysis revealed that the therapy activated a strong immune cell response, with increases in CD45+ cells, both CD4+CD3+ and CD8+CD3+ T cells, and CD11c+ dendritic cells.

These results show that, in animal models, the new treatment has greater therapeutic efficacy than existing oncolytic viral treatments. Because the cells used are allogeneic and can be easily banked under good manufacturing practice (GMP) requirements, this approach is a strong candidate for an ‘off-the-shelf’ cellular immunotherapy. There are also fewer safety concerns than other types of allogeneic cell therapies.

»These are bone marrow-derived cells, which are usually less immunogenic, particularly in the brain, and they don’t get rejected. Another advantage is that they are readily available ‘off-the-shelf’ for patients,« Shah commented.