The cancer fighters known as CAR T cells have proved their prowess in recent years. Three therapies using the altered T cells against lymphoma or leukemia have won U.S. Food and Drug Administration approval, and hundreds of trials are now unleashing them on other malignancies, including solid tumors. But the cells may soon have company. Researchers have equipped other immune guardians—natural killer cells and macrophages—with the same type of cancer-homing receptor, and the natural killer cells have made their debut in clinical trials.

CAR T cells—their name comes from the chimeric antigen receptor, or CAR, added to help the immune cells target cancer cells—inspired the new work. CAR natural killer (CAR NK) cells could be safer, faster to produce, and cheaper, and they may work in situations where T cells falter. CAR-carrying macrophages also have potential advantages, and one firm plans to launch the first clinical trials of these cells next year.

Although they aren’t likely to replace CAR T cells, these alternative cancer fighters “could be an addition to the armamentarium of cell therapies,” says hematologist and oncologist Katy Rezvani of the University of Texas MD Anderson Cancer Center in Houston. She is leading the first trial of CAR NK cells in the United States, which began in 2017, and organizing another that is due to start this year.

Making CAR T cells involves removing patients’ own T cells and genetically altering them to attack cancer cells that carry a specific immune-stimulating molecule, or antigen. (All of the CAR T treatments approved so far target the CD19 protein on cancerous B cells, a type of immune cell.) The cells have produced impressive results in clinical trials—in one study, they triggered remissions in 83% of children with previously untreatable acute lymphoblastic leukemia. But some patients who have already undergone chemotherapy or radiation treatment may not have enough T cells left to donate. And these powerful immune warriors can trigger a potentially fatal flood of the immune system molecules known as cytokines or turn against normal body cells.

Perhaps the biggest shortcoming of CAR T cells, though, is they don’t work well against solid tumors, says hematologist and oncologist Saar Gill of the University of Pennsylvania. Tumors rebuff T cells that try to enter, inhibit immune cells that do make it inside, and can curb production of antigens targeted by CAR T cells. Researchers are trying several approaches to improve CAR T cells’ performance against solid tumors. But NK cells are a tempting alternative, scientists say.

“Natural killer cells are our first line of defense against cancer cells,” Rezvani says. They scan other cells in the body and destroy any that are infected or otherwise abnormal, including tumor cells. Researchers have been trying to harness the cancer-fighting activity of NK cells that don’t carry CARs for more than 20 years, notes translational immunologist Jeffrey Miller of the University of Minnesota in Minneapolis. But upgrading them by adding CARs seems to boost their potency.

Earlier this year, for instance, stem cell biologist Dan Kaufman of the University of California (UC), San Diego, and colleagues reported that in mice, CAR NK cells perform about as well against ovarian tumors as CAR T cells do—and substantially better than unaltered NK cells. Mouse trials also suggest CAR NK cells may not cause some of the side effects of CAR T cells, such as excess cytokine release and neurological damage. CAR NK cells might also be less vulnerable to some of tumors’ tricks for avoiding attacks. Because NK cells rely on other receptors to recognize tumor cells, not just the CAR, they may be able to detect a tumor even if it alters its antigens. In addition, Kaufman points out, it may be feasible to give patients multiple doses of CAR NK cells and hammer away at tumors, whereas the cost of CAR T cells limits patients to a single dose.

The first trials of CAR NK cells started in China in 2016 in patients with several kinds of cancers—early results from one suggest the cells are safe. Rezvani and colleagues’ initial trial is pitting the cells against several varieties of lymphoma and leukemia. A European trial, which is testing CAR NK cells in patients with the brain cancer glioblastoma, launched this year. In the upcoming MD Anderson trial, patients with B cell lymphoma, a type of blood cancer, will receive stem cell transplants and chemotherapy before CAR NK cells, which the researchers hope will mop up any remaining cancer cells.

“I think the future is bright for CAR NK cells, but we are at the very beginning,” says hematologist Mitchell Cairo of the New York Medical College in Hawthorne. One unknown is the best source for the cells. T cells from someone other than the patient can trigger a potentially fatal immune complication known as graft-versus-host disease, in which the transplanted cells attack the recipient’s own tissues. But NK cells from a donor do not appear to cause that response, which opens a range of options. Although sieving NK cells from donors’ blood is a possibility, the procedure is expensive and can harm the donors. Both MD Anderson trials instead rely on NK cells isolated from umbilical cord blood and then implanted with CARs. Donated umbilical cord blood is abundant and plenty of NK cells can be grown from it.

In contrast, the Chinese and European trials generated enough NK cells by turning to a cell line derived from a person with a type of lymphoma. These cells are staples of clinical trials, and despite their cancerous origin, they appear to be safe, says immunologist Torsten Tonn of the Technical University of Dresden in Germany, one of the researchers participating in the glioblastoma trial. Kaufman and colleagues are also exploring another possible source of NK cells: induced pluripotent stem cells, which are produced by nudging adult body cells to return to an unspecialized state.

All these approaches could lead to off-the-shelf CAR NK cells that avoid the need to extract and modify a cancer patient’s own cells. The patient’s immune system will eventually reject any foreign NK cells, Miller notes. But before that happens, Rezvani and other researchers think the donor NK cells will have a window of time during which they can combat cancer cells. The question, she says, is whether they will persist long enough to benefit patients.

Like NK cells, macrophages can destroy cancer cells, but the catch is that most macrophages inside a tumor are traitors, which help the tumor by quashing immune attacks against it, for example. “Tumors acquire macrophages to support their own growth and turn them into their minions,” Gill says. But he and graduate student Michael Klichinsky have discovered that the procedure for equipping macrophages with a CAR prevents them from switching sides. The duo helped found a company, Carisma Therapeutics in Philadelphia, Pennsylvania, that expects to begin clinical trials of CAR macrophages next year.

At least in the lab, adding a CAR to macrophages boosts their tumor-fighting abilities, just as it does for other immune cells, postdoc Meghan Morrissey of UC San Francisco and colleagues have also reported. But tumor immunologist Kim O’Neill of Brigham Young University in Provo, Utah, who leads another group trying to improve the cells’ tumor-killing abilities, suggests macrophages could do the most good by recruiting other immune cells. T cells, for example, respond to the cellular debris leftover when a macrophage digests a tumor cell, so he envisions that patients would receive CAR macrophages along with CAR T cells. Like great detectives, even the most powerful cancer-fighting cells might benefit from a talented sidekick.