Equipping NK Cells with CARs

Chimeric antigen receptor (CAR) T cells have occupied the spotlight of late, especially with the late-August approval of the first such therapy, tisagenlecleucel (Kymriah; Novartis), for children and young adults with relapsed/refractory acute lymphoblastic leukemia (ALL). Another strategy—engineering CARs on natural killer (NK) cells instead—has received considerably less attention, but to some researchers in the field, it's an appealing alternative.

A chief advantage to using NK cells is that “they aren't thought to cause graft-versus-host disease [GvHD],” explains Jeffrey Miller, MD, deputy director of the University of Minnesota's Masonic Cancer Center in Minneapolis. On the other hand, the risk of GvHD is ever present with allogeneic T cells; even when autologous T cells are used, complications such as cytokine release syndrome may occur. As such, Miller says, CAR NK cells offer the possibility of “treating multiple patients with an off-the-shelf, donor-generic product that's potentially safer, which would make widespread use of this immunotherapy more feasible.”

“It's logistically cumbersome to manufacture CAR T cells for each patient,” says Katy Rezvani, MD, PhD, of The University of Texas MD Anderson Cancer Center in Houston. “Even at our well-equipped center, we have a waiting list of more than 50 patients for just one CAR T-cell therapy trial; some may never get to enroll, because of their rapidly advancing disease.”

Rezvani and her colleagues recently published a preclinical study of NK cells outfitted with a CD19-targeting CAR, as well as the gene for IL15—to improve proliferation and survival, because NK cells otherwise lack persistence in the body—and a kill switch to mitigate toxic side effects, should these arise. The cells were derived from umbilical cord blood, a “clearly advantageous source” for product scalability, Rezvani points out, due to the existence of cord blood banks around the world.

“With the cell isolation and expansion strategy we've developed, it takes just 2 weeks to make approximately 10⁸ to 10⁹ CAR NK cells, which is quite a massive dose,” she adds.

The researchers reported promising results: Their CAR NK cells efficiently wiped out primary leukemia cells in vitro and showed potent antitumor activity, including complete responses in a mouse model of lymphoma. Adding the ability to produce IL15 was crucial, Rezvani notes; CAR NK cells without this growth factor were able only to transiently control, not eradicate, tumors.

“I like that they [Rezvani's team] went after CD19, a well-validated target,” Miller says. “This should enable them to make some direct comparisons with what's going on in the CAR T-cell therapy space.”

Besides the CAR strategy, “a bunch of other approaches” to better manipulate NK cells in immunotherapy are being pursued, mainly by small biotechs, Miller observes. Several years ago, his group identified a functionally distinct subset of adaptive memory NK cells that demonstrate long-term persistence in vivo, and began figuring out how to clinically exploit them, in collaboration with San Diego, CA–based Fate Therapeutics. In March, the company launched a phase I trial of its first-in-class adaptive memory NK cell therapy, FATE-NK100, for patients with acute myeloid leukemia (AML). More recently, Miller's team reported that pharmacologic inhibition of the kinase GSK3 enhances this NK cell subset's antitumor activity.

With Oxis Biotech in Tampa, FL, Miller is also evaluating OXS-3550, which utilizes a “trispecific killer engager” technology he developed. The drug binds CD16 on NK cells and CD33 on tumor cells, bringing them into close proximity and thereby unleashing tumor-directed cytotoxicity. It also features an IL15 linker to sustain NK cell proliferation.

Meanwhile, a phase I/II trial of Rezvani's CAR NK-cell therapy is under way. To date, two patients with non–Hodgkin lymphoma have received infusions, and a third with chronic lymphocytic leukemia is enrolled; the study is also open to patients with ALL. Preclinically, she and her team are assessing CAR NK cells in AML, multiple myeloma, pancreatic cancer, and ovarian cancer—targeting mesothelin in the latter two tumor types.

“Our goal, to make this a truly off-the-shelf therapy, is to have a bank of frozen cord blood–derived CAR NK cells, ready to be thawed and infused at any time,” Rezvani says. “We're still figuring out the best method of cryopreservation, however.”

Ultimately, “we've been waiting for someone to do it first,” Miller says of CAR NK-cell therapy, “and that's what I appreciate about [Rezvani's] work. I think people will be closely watching what unfolds.” –Alissa Poh