The Cell Therapy That Rests Between Kills

CAR T cell therapy revolutionized blood cancer treatment. It has spent a decade failing to crack solid tumors.

At this year's AACR annual meeting in San Diego, researchers from the University of Pennsylvania presented the first human data for a different kind of CAR T—one designed to solve the problem that kept all the others stuck. The approach, called KIR-CAR, separates the act of finding a tumor cell from the act of attacking it. The engineered T cell rests between kills, rather than burning constantly until it exhausts itself. (Penn Medicine)

The results are early: nine patients, four stabilizations, one partial response. This is not an approved therapy. It is not a vindication. But it is the first concrete signal that T cell exhaustion—the core reason CAR-T has been locked out of solid tumors—might be solvable through design rather than luck. (Penn Medicine)

The exhaustion problem

Conventional CAR T cells are always on. Once they recognize their target, they attack and keep attacking. In the complex, immunosuppressive environment of a solid tumor, that relentless activation eventually depletes the cells. They lose their potency, stop killing, and the tumor resumes growing.

The Penn team's design addresses this by borrowing from natural killer cells. Rather than a single receptor chain that does both jobs—find and attack—the KIR-CAR uses two separate chains. One locates the tumor antigen. The other, once the first has done its job, triggers the kill. When there's no target nearby, both chains sit idle and the T cell rests. (Penn Medicine)

"The CAR turns on when it finds its target, kills it, and then rests, rather than constantly burning energy," said Janos L. Tanyi, the study's principal investigator and an associate professor of obstetrics and gynecology at Penn's Perelman School of Medicine. (Penn Medicine)

The design also appears to reduce off-target toxicity. Because the T cell is not continuously engaged, it has fewer opportunities to mistake healthy tissue for tumor. (Penn Medicine)

The patients

The trial enrolled nine people with advanced ovarian cancer, mesothelioma, or cholangiocarcinoma—cancers selected because they express mesothelin on their cell surfaces, and because treatment options for all three are limited and often grim. Ovarian cancer recurs in roughly 70 percent of patients after standard therapy. These were people who had already received an average of four prior lines of treatment. (Penn Medicine)

The primary goal was safety, and the safety profile was clean. No dose-limiting toxicities occurred. Three of nine patients developed low-grade cytokine release syndrome, a known CAR T side effect, but none higher than Grade 2. Critically, there were no cases of ICANS—the neurotoxicity that has complicated some other CAR T programs. (Penn Medicine) (Verismo/PR Newswire)

Efficacy signals emerged at the highest dose level. One patient out of three at that dose achieved a partial response per RECIST criteria, still ongoing more than three months after treatment as of the September 2025 data cutoff. Four patients overall showed disease stabilization. (Verismo/PR Newswire)

Carl June, the Penn CAR T pioneer who led the original development of the technology that became Kymriah, noted in the AACR opening plenary that the field is now in a race to extend cell therapies into solid tumors—the frontier where the revolution never arrived. (Penn Medicine)

A Penn spinout with Korean ownership

SynKIR-110 was developed by Verismo Therapeutics, a company founded by current and former Penn faculty and the Center for Cellular Immunotherapies. In December 2024, Verismo became a wholly-owned subsidiary of HLB Innovation, a South Korean pharma company listed on KOSDAQ under ticker 024850. (PR Newswire)

Verismo is also running a second KIR-CAR program, SynKIR-310, in a Phase I trial for blood cancers using a CD19 binder. The company has $28 million in prior investment from HLB. (PR Newswire) (Verismo/PR Newswire)

The trial remains ongoing, with enrollment continuing at Penn's Abramson Cancer Center and three additional sites to establish a recommended Phase II dose. (Verismo/PR Newswire)

What this is not

Nine patients is a small number. One partial response, from the highest dose cohort, does not establish a dose-response relationship. The data cutoff predates the AACR presentation by seven months, leaving open whether that response has since progressed. Phase I trials are designed to find a safe dose, not prove efficacy.

The mechanism is also not yet validated in the sense that matters most for the exhaustion thesis: whether the T cells actually rest and recover in human patients over the longer term, and whether that rest translates into durable tumor control rather than just delayed progression.

What the data does establish is that the multi-chain architecture behaves differently from conventional CAR T in terms of safety, and that biological activity is present at doses where conventional designs were already failing.

The blueprint exists. The question now is whether it works.