CAR-T Spotlight - Episode 6

Moving a Step Closer to Using CAR T-Cell Therapy in Solid Tumors

A recent paper in Nature Cancer outlines a novel strategy for overcoming barriers to CAR T-cell therapy in glioblastoma by pretreating the vascular microenvironment.

Every day, chimeric antigen receptor (CAR) T-cell therapy seems to break more barriers in blood cancers—for example, an FDA approval in multiple myeloma seems not a question of if, but when. But significant hurdles remain to deploying this technology against solid tumors, and that’s especially true when it comes to glioblastoma—the deadly form of brain cancer that killed Arizona Senator John McCain and Beau Biden, son of the now President-elect.

Glioblastoma is famously hard to treat because the tumors tend not to be self-contained; they have tendrils that extend into the brain and may not be easily removed. Most patients have a prognosis of about 15 months; very few live 5 years.

A recent paper in Nature Cancer offered a potential solution for letting CAR T-cell therapy do its job in glioblastoma. This approach calls for pretreating the cells that line the blood vessel walls, which normally do offer the right kind of structure to allow the CAR T-cells to reach the tumor; by paving the way, this pretreatment allows the immunotherapy to attack the cancer.

The work involved a series of deliberate, painstaking steps, which the study’s co-corresponding author Yi Fan, PhD, discussed in a recent interview with The American Journal of Managed Care® (AJMC®), to find the right enzyme to target to create the tumor microenvironment. Then, the team implanted glioblastoma tumor cells in the mice; they found that by limiting PAK4, they reduced vascular abnormalities and appeared to allow T cells to penetrate the tumor. “Interestingly, PAK4 knockout robustly stimulated T cell infiltration into the tumors, probably contributing to the inhibition of tumor growth,” the authors wrote in the paper.

They reported that approximately 80% of the PAK-knockout mice survived for at least 60 days after the experiment ended; by contrast, all the wild-type mice died within 40 days of having the tumor cells implanted, despite having a T-cell infusion.

“We suggest that the therapeutic efficacy associated with enhanced T cell delivery and homing by PAK4 inhibition can be further improved when combined with other immunotherapy approaches capable of increasing T cell persistence by inhibiting T cell exhaustion and/or by stimulating T cell activation in the tumor microenvironment.,” the authors wrote.

This could have implications beyond glioblastoma and even immunotherapy. Returning blood vessels to their normal state by reducing oxygen deprivation, known as hypoxia, could improve results across a range of cancer treatments, from targeted therapy, to radiation, to chemotherapy. The work builds on Fan’s earlier work on vascular detransformation, which proposes a new approach in fighting cancer.

Fan, who is an associate professor of Radiation Oncology in the Perelman School of Medicine at the University of Pennsylvania, discussed how they study addressed the central issue that confronts scientists who work with CAR T-cell technology: blood vessels are needed to carry T-cells to the tumor, but the vascular structure surrounding the cancerous tumor is typically inhospitable for this purpose. So, it must be altered.

In 2019, Fan’s paper, “Vascular Detransformation for Cancer Therapy,” proposed that endothethial transformation “fuels cancer progression” and can cause resistance to treatment. New therapies that return the vascular system around the tumor to its precancerous condition “may serve as an effective next-generation antivascular approaches in cancer treatment.”

Experiments outlined in the Nature Cancer paper represent the next step of putting the idea behind vascular detransformation into action. Unlike leukemia cells, which Fan said “are really good at migration and proliferation,” and thus create a tumor microenvironment where T cells can work, genetic reprogramming is needed for the endothelial transformation required to allow CAR T cells to reach their target in glioblastoma.

But just what programming would work?

“So, we screened the whole human catalogue—more than 500 kinases,” Fan said. “We screened one by one—it’s actually a lot of work.” The team found that by limiting PAK4, it could regulate blood vessel activation in endothelial cells, and the team then developed a PAK4-knockout mouse model for one type of experiment.

And then experiments confirmed the screening tests. When the PAK4 inhibitor was added to the CAR T-cell therapy in one experiment, Fan said, “you can see the tumor becomes wrinkled.”

AJMC® asked, given the difficulty of CAR T-cell therapy with solid tumors, why keep trying this approach despite all the challenges? Aren’t there other strategies?

“CAR T is one of the most exciting immunotherapy approaches available,” Fan said. After all the successes in leukemia—many of which were first seen at Penn Medicine—the challenge of treating other cancers, an figuring out how to use CAR T-cell therapy for very different tumor cells and entirely different microenvironments—remains the “big, big question” that scientists can’t resist chasing. Plus, the need is huge. What’s good about the strategy explored in the Nature Cancer paper, Fan said, is that is can likely be replicated for other solid tumors.

“I think this is a pretty common mechanism for most solid tumors, so probably we can also use our strategy for treating other types of cancer including breast cancer, pancreatic cancers—all kinds of different solid tumors,” he said.

AJMC® discussed with Fan the labor-intensive mechanics of trying kinases one by one and asked—if more than 1 kinase if found to be effective in enabling the effectiveness of CAR T-cell therapy, would it make sense to target multiple kinases in this vascular detransformation approach?

“That's a brilliant idea,” Fan said. Even though the targeting PAK4 greatly increases the efficacy, “I don’t think you will be at 100% efficacy. So, in the future probably, I would target more than 1 of the kinases at the same time.”

So, AJMC® asked, where do we go from here? What are the next steps in advancing this concept?

From this point, Fan expects to see more preclinical work and setting the stage for commercialization of this concept. “I think maybe 1-2 years from now probably we can say some good results from clinical trials. So that's what we are expecting,” he said.


Ma W, Wang Y, Zhang R, et al. Targeting PAK4 to reprogram the vascular microenvironment and improve CAR-T immunotherapy for glioblastoma. Nat Cancer. Published online November 30, 2020.