Second only to smoking, obesity is a top risk factor for cancer. But cancer doctors have noticed something surprising: overweight patients sometimes respond better than other patients to powerful drugs that harness the immune system to fight tumors. Now, researchers tracing the complex effects of obesity on cancer are glimpsing a possible explanation: Obesity weakens the immune system and favors tumor growth by boosting the very same molecules those drugs target.

“For the most part, everybody assumes obesity is always bad. But [with these drugs], there was a net positive,” says cancer immunologist William Murphy of the University of California (UC), Davis, who, with UC Davis oncologist Arta Monjazeb, led the work reported today in Nature Medicine . Murphy thinks the finding could point to ways to make the drugs more effective in all cancer patients.

Called checkpoint inhibitors, the drugs work by blocking the activation of PD-1, a protein on the surface of immune sentinels called T cells. The body naturally triggers PD-1 to dampen immune responses, but tumors can also stimulate PD-1 to protect themselves. Lifting this molecular “brake” allows the T cells to attack the cancer cells. PD-1 inhibitors have caused untreatable tumors to vanish for years in people with melanoma, lung cancer, and some other cancer types.

But only a minority of patients respond to the drugs, and a study early this year in The Lancet Oncology showed that the responders disproportionately include people who are overweight. In 330 advanced melanoma patients given a PD-1 inhibitor, researchers at MD Anderson Cancer Center in Houston, Texas, found that those who were male and overweight lived much longer on average: nearly 27 months compared with 14 months for patients with a normal body mass index (BMI).

Now, Murphy’s team has firmed up this clinical observation in the lab and identified a possible basis. After confirming that tumors grow faster in obese mice, his team studied the T cells of obese mice, monkeys, and people. They found that the cells were what immunologists call “exhausted.” They were slow to proliferate and had stopped making secreted proteins that stimulate other immune system helpers. They also displayed more PD-1 than average, meaning cancer cells could more easily suppress them and grow unhindered.

Leptin, a hormone made by fat cells, is one factor in the PD-1 excess, Murphy’s group found. Overweight animals and people produce high levels of the hormone, which normally signals the brain that the animal has had enough to eat. But leptin also affects the immune system, and the UC Davis team suspects it triggers a signaling pathway that increases PD-1 on T cells.

The PD-1 excess also has a paradoxical benefit: In obese mice, it makes T cells unusually responsive to PD-1 inhibitors, Murphy’s team reports today in Nature Medicine . Once the drugs released this brake, the T cells sprang back into action. Nourished by glucose and other nutrients abundant in an overweight animal’s tissues, they worked better at curbing tumors than in normal weight animals.

The finding suggests an “unexpected” benefit of obesity for cancer patients, says Harvard University immunologist Lydia Lynch. Her group reports in Nature Immunology today on a different way obesity impairs the immune system’s ability to attack tumors: by hampering a type of immune cell called natural killer cells that seek out and destroy abnormal cells.

Murphy plans to explore whether giving normal weight mice with cancer a high-fat diet or leptin in order to mimic some effects of obesity could boost their response to PD-1 inhibitors. But such treatments for cancer patients could also have harmful effects, cautions tumor immunologist Suzanne Ostrand-Rosenberg of The University of Utah in Salt Lake City, who also studies how obesity spurs tumor growth. “It’s a balance here, a very careful balance,” Ostrand-Rosenberg says.

Whereas the UC Davis team’s findings suggest obese patients may respond better to PD-1 drugs, normal weight patients can also benefit and it’s too early to make treatment decisions based on BMI, says MD Anderson melanoma researcher Jennifer McQuade, lead author on The Lancet Oncology study. “Ultimately, we need an integrative analysis to understand the contributions of BMI, sex, age, and how these interact with each other,” McQuade says.