Immunotherapy Combo Reduces Solid Tumors in Mice

Oncolytic viruses are bioengineered, benign viruses created specifically to infiltrate and replicate within cancer cells. If scientists modify an oncolytic virus to carry a particular gene, it can function as a Trojan horse, smuggling the gene inside and forcing infected cells to express the gene for a desired surface protein that flags them as cancerous to T cells.

The authors chose CD19 as the cancer flag for the virus because of its demonstrated ability to attract CAR T cells in approved treatments for leukemia and lymphoma, which naturally present the marker. 

Solid cancers are more difficult to treat with CAR T cells than liquid cancers for a number of reasons. Tumors are difficult to penetrate, and once inside, immune cells face a toxic, immunosuppressive environment. In addition, solid tumors are heterogeneous, meaning that each cell has its own mosaic of surface proteins, many of which are shared with healthy cells. In contrast, CD19 doesn’t typically appear outside of blood malignancies.

So far, scientists have not identified a unique antigen common to all cells within solid tumors.

Saul Priceman, a tumor immunologist at City of Hope and the senior author on the study, says he realized that if the team could find a way to make solid tumors produce CD19, these cancers could then be targeted by CAR T cells. So his group engineered an oncolytic vaccinia virus carrying a payload of CD19. “We thought, ‘If it works for blood cancers, why can’t we make it work for solid tumors?’” Priceman tells The Scientist. “And that’s what we set out to do.”

See “The Next Frontier of CAR T-Cell Therapy: Solid Tumors”

The team began with in vitro studies using a variety of human cancer cells in culture. First, they showed that their modified vaccinia virus could in fact deliver the CD19 gene into tumor cells, leading to the presence of the protein on their surfaces. As the viruses replicated, some tumor cells burst, spilling the virus onto nearby cells that escaped the first wave of infection. After 24 hours, nearly all the cells had been successfully modified.

Ten days later, the team added CAR T cells designed to destroy cells bearing CD19 to the culture. Together, the deaths caused by the virus and the subsequent attack by T cells resulted in a loss of between 60 percent and 70 percent of tumor cells over the next few days. 

“The whole idea that you can turn a solid tumor into a quasi–B cell [liquid cancer] by introducing a CD19 antigen was clever and really novel,” says Katy Rezvani, a transplant immunologist at the MD Anderson Cancer Center who was not involved in the study. “The experiments were very well done, very convincing.”

Next, Park and his colleagues applied their combined immunotherapy to several mouse models, focusing on both the efficacy of the treatment and how best to deliver it.

They began with an immunosuppressed model to test the therapy on human breast cancer tumor cells and human T cells in a pared down immune environment. Ten days after injecting the virus directly into a subcutaneous tumor, roughly 70 percent of the cells harbored CD19. Mice that received both the virus and T cells showed marked tumor regression compared to mice that received mock T cells. 

See “Tactical Maneuvers”

The second model, known as a syngeneic, immunocompetent mouse model, used mouse cancer cells and immune cells rather than human cells, but allowed the team to study the full suite of immune cells involved in fighting cancer. 

Among mice that received only the virus, 22 percent recovered from the cancer due to virus-induced cell death alone, and the authors noted very few off-target infections of healthy cells. When combined with CAR T cells, the number of mice experiencing complete tumor regression jumped to roughly 60 percent.

Lastly, Park and his colleagues used an immunocompetent model to study an emerging question in the field: how best to deliver these therapies. Treatments for liquid cancers are administered by IV, but debate exists over whether a systemic or local injection would be best for solid tumors. Both the viruses and the T cells might benefit from being injected directly, as the results in the study were strongest at higher viral concentrations (greater than 10⁵ plaque-forming units per mouse). However, many solid cancers metastasize, making them difficult to access.

Trying out various injection protocols, the researchers found that mice with metastatic tumors that received a regional injection of viruses and T cells into the peritoneal lining surrounding the main body cavity had a lower tumor burden and generally survived longer than controls did. These findings suggest that oncolytic viruses may be amenable to IV drips; currently, the only oncolytic virus therapy approved by the US Food and Drug Administration relies on direct injection into a tumor. “Our future [research focus] is whether we can give this virus systemically and safely so those viruses can seek out all the tumors wherever they may be in the body,” Priceman tells The Scientist.

Andrea Schmidts, a CAR specialist at Harvard Medical School who was not involved in the study, says she found the authors’ rigorous use of mouse models and their testing of different delivery methods “a very important quality marker” of the study. “The combination of having both these models is very robust,” Schmidts says.

The team has already begun designing a Phase I clinical trial, which could begin in the next year or two. Priceman and Park say the future of cancer treatment lies in these combinatorial therapies, and they don’t necessarily intend to stop at only two different methods. For example, they’re currently considering combining their new therapy with checkpoint inhibitor drugs that unmask efforts by tumors to cloak themselves from surrounding immune cells.

Based on their preliminary work, Priceman says, “there are hints that blocking checkpoints might amplify the therapeutic response.” 

See “Deploying the Body’s Army”