A new analysis of patients who took part in clinical trials for Kite Pharma’s CAR-T cell therapy Yescarta has revealed genetic signatures of toxicity risk, potentially providing a path forward for future CAR-Ts to be engineered to be safer.
The researchers also singled out a gene that seems related to the ability of CAR-T cells to replicate inside the body, an area they are now exploring further.
The results were published today in Science Immunology.
The study was led by Mark Leick, M.D., an oncologist at Massachusetts General Hospital, who started the project in 2018 when he joined the lab of leading CAR-T researcher Marcela Maus, M.D., Ph.D.
“This is something that the field has been thinking about for a long time,” Leick told Fierce. While it was known that some CAR-T cells are more toxic than others, and that the characteristics of a patient’s cancer can influence how they respond to treatment, “there's a lot of other features about toxicity that we don't entirely understand.”
Because all approved CAR-Ts are made using cells taken from the patient themselves, the patient’s own DNA may play a role in how the therapy performs.
CAR-T cells “carry all the ancestral DNA modifications that you're born with,” Leick said. “The hypothesis of our study was that those DNA changes that you're born with affect how the CAR-T cells behave.”
To get a handle on what might cause toxic side effects to arise in some patients but not others, Leick and his colleagues tapped into data from Kite’s Zuma-1 and Zuma-7 trials of axi-cel, now known as Yescarta. These trials tested the CAR-T in pretreated lymphoma patients, though Zuma-1’s population generally had more severe disease.
Leick started by looking at genes known to be related to a rare hyperinflammatory disease found in children called hemophagocytic lymphohistiocytosis (HLH), he said, because researchers have long recognized similarities between HLH and cytokine release syndrome, a common but serious side effect of CAR-T therapy.
From this analysis, the gene for syntaxin binding protein 2 (STXBP2) stood out. Six patients in Zuma-1 had mutations in STXBP2 that knocked out its function, and all of them experienced toxicity after treatment. But no such relationship was found in the Zuma-7 trial, which Leick thinks stems from the more severe condition the Zuma-1 patients were in.
“Zuma-1 patients are much more inflamed, and we think that's why the signal was most robust for this finding for STXBP2 in Zuma-1,” Leick explained.
The team then zoomed out and scanned for genetic red flags across the entire genome. This revealed potentially protective variants of a gene called ADAMTSL3 that were associated with lower toxicity. The screen also found that versions of the PTPN22 gene—a regulator of T-cell activity—appear linked to the ability for CAR-T cells to proliferate once they’re infused.
Overall, “this study convincingly demonstrates that the underlying germline genetics of the patient act as a major, intrinsic determinant of both CAR-T cell expansion and clinical toxicity,” Carl June, M.D., who pioneered the first CAR-T therapy at the University of Pennsylvania but was not involved with the new study, told Fierce.
The connection between PTPN22 and CAR-T expansion, confirmed across Zuma-1 and Zuma-7, is “particularly striking,” June added.
The finding suggests that “intentional knockouts or knockdowns of PTPN22 could be systematically incorporated into manufacturing protocols to enhance cell expansion, reduce the required cell dose and potentially overcome T-cell exhaustion in solid tumors,” June said.
A key limitation of the study is that it focuses only on patients of European descent, which was the vast majority of those enrolled in the Kite trials.
“We just did not have the numbers of non-Europeans to conduct any serious type of analysis,” Leick told Fierce. “It's obviously something that we would love to do.”
As more and more patients are treated with the growing number of approved CAR-T products out there, more data for such an analysis could be available soon.