When Lieping Chen, M.D., Ph.D., was training to be an oncologist in the 1980s, the lack of effective cancer treatments made it a depressing job. “That’s why I quit clinical practice,” he says. But Chen soon shifted to research, focusing on the role of the immune system in cancer.

For decades Chen and other scientists have grappled with the mystery of why our immune systems—such brave warriors when it comes to other diseases—mostly sit on the sidelines while cancer invades the body. Though our bodies produce T cells, the immune system’s soldiers, in response to cancer, these cells don’t seem to have much of an effect. Scientists have yearned to find some way to “switch on” the immune system to attack cancer cells, an approach that could target the disease precisely while sparing normal cells, resulting in greater effectiveness and fewer of the side-effects of conventional treatments. And because the immune system “remembers” invaders it has encountered, the response to immunotherapy should be long-lasting.

On June 2 in the New England Journal of Medicine, a multi-center team that included Chen and two Yale colleagues reported that in an early clinical trial of BMS-936558 (a drug developed by Bristol-Myers Squibb) enrolling 296 patients, tumors had regressed significantly in about one in three to one in five patients with non-small-cell lung cancer, kidney cancer, or melanoma. The results, which received wide media coverage, were presented the same day at the annual meeting of the American Society of Clinical Oncology in Chicago, Ill.

For reasons that are not fully understood, the small number of patients with colorectal or prostate cancer in the trial did not respond to BMS‑936558. But the response in lung cancer patients is remarkable, says Professor of Medicine Mario Sznol, M.D., because previously immunotherapies showed effectiveness only in melanoma and kidney cancers. Chen points out that the patients in this trial had all previously been heavily treated with therapies that likely depleted their immune systems; if administered to patients in earlier stages of cancer, the drug should be even more effective, he says.

The other exciting finding of the study, says Sznol, who was instrumental in bringing the trial to Yale, is the consistent, durable regression of cancers: 20 of 31 responses lasted for at least one year, a result only rarely observed with traditional chemotherapy or other forms of therapy that target specific characteristics of tumors.

Since similar results were seen at all nine centers participating in the trial, Sznol considers it highly unlikely that the drug’s effectiveness is a fluke. Scott N. Gettinger, M.D., associate professor of medicine, who recruited many lung cancer patients for the trial, agrees, saying the response in those patients “opens the door to looking at other solid tumors.” The majority of his patients experienced few or no side effects, he says, without the typical toxicities of chemotherapy such as hair loss, precarious drops in blood-cell counts, altered taste, and profound fatigue. “This is the best-tolerated drug I have ever given to patients with lung cancer,” Gettinger says.

The long road to these results began in 1991, when Chen, then at the Mayo Clinic, discovered a protein called B7-H1 that had the power to suppress the activity of T cells. He did not know the T cell molecule responsible for this effect, but in 2000, researchers from Harvard Medical School’s Dana-Farber Cancer Institute and the University of Kyoto found that B7-H1 binds to another protein called programmed death 1 (PD-1), a “checkpoint” receptor on the surface of T cells that suppresses their activity.

The B7-H1 protein is now usually called PD-L1, to reflect its role as a binding partner, or “ligand,” for PD-1. Chen wondered if PD-L1 binding might be one of the mechanisms that allows cancer to evade an immune response, and in experiments with mice lacking the protein, he found that the PD-1 pathway is crucial to the normal functioning of the immune system: a prolonged immune response can be very damaging to the body, so tissues express PD-L1 to turn the response off. Chen dubbed the pathway the “peacekeeper” for its protective role, but he found that while PD-L1 is expressed in healthy tissue at low levels during an immune response, it is over-expressed in many human tumors.

“Cancer steals this mechanism,” Chen, professor of immunobiology and medicine, says. In the presence of cancer, an immune response is triggered properly, and T cells flock to the tumor. But the T cells contribute to their own demise by releasing signaling molecules called cytokines, which induce the tumor to produce PD-L1. When PD-L1 binds to PD-1 on T cells, those cells are either inactivated or die off. BMS-936558 prevents this immune suppression by blocking the PD-1 binding site and keeping PD-L1 at bay.

Tests on tumor samples taken from a subset of patients in the newly published trial revealed that tumors that did not express PD-L1 did not respond to treatment, while 36 percent of PD-L1-positive tumors showed a response, indicating that PD-L1 may be a useful biomarker for identifying patients who will most benefit from anti-PD-1 drugs.

The PD-1 pathway is only one inhibitory strategy tumors may employ to silence the immune system. Last year the FDA approved the drug ipilimumab (marketed as Yervoy), which blocks another checkpoint molecule expressed on the surface of T cells known as ctla-4, for the treatment of advanced melanoma. Ipilimumab has a response rate of about 11 percent, but anti-PD-1 therapies have been shown to be safer and more efficacious. Chen believes that this is due to limited PD-L1 at sites of inflammation, including tumors. The ctla-4 ligand, on the other hand, controls an earlier stage of immune system activation, so blocking its receptor has led to severe toxicities. There are side-effects of PD-1 blockage—in the current trial, three patients died of drug-related lung inflammation—but Chen thinks they can generally be successfully managed.

Jedd D. Wolchok, M.D., Ph.D., of Memorial Sloan-Kettering Cancer Center, who was not involved in the current trial, says that the new results are important because, along with the success of ipilimumab, they are the second clear demonstration that unleashing the immune system can produce durable regressions in cancer.

Although BMS-936558 was quite effective when used alone, the Yale researchers believe that it will be most useful combined with other therapies. Sznol is currently collaborating with Wolchok on a clinical trial that targets both PD-1 and CTLA-4, and Chen believes that adding drugs that block T cell checkpoints could even reverse the low success rate of cancer vaccines seen thus far.

“We’re just at the beginning of a sort of paradigm shift in the treatment of cancer,” Sznol says, who believes that immunotherapy will surpass standard treatments like chemotherapy over the next five years. As for BMS-936558, Chen is hoping to see FDA approval in two to three years, but “it should be approved tomorrow,” Sznol says: “It’s better than the approved agents. I have patients who have benefited enormously from this drug. For some patients it’s a home run. For the field it may be a triple.” And while current treatments prolong life for months or a couple of years, Sznol says, “these new therapies may eventually offer the possibility of cure.”