Cancer cells have a well-documented trick for surviving: they flag themselves with proteins that tell the immune system to leave them alone. Cedars-Sinai researchers just found aging cells appear to run the same con, using one of the very same molecular flags that oncologists have spent the last decade learning to block in tumors.

The study, published September 10, 2026, in Cell Metabolism, identifies a protein called PD-L2 as a mechanism that helps senescent cells, the damaged, no-longer-dividing cells that accumulate as the body ages, dodge the immune system's cleanup crews. Blocking that protein in mice reduced how many of these harmful cells built up in tissue, along with the inflammatory chemicals they release.

What senescent cells actually do to the body

Senescent cells occupy a strange middle ground in biology. They're damaged enough that the body stops them from dividing further, a safeguard against runaway cell growth like cancer, but they don't die off the way damaged cells normally would. Instead they linger in tissue indefinitely, releasing a cocktail of inflammatory molecules known collectively as the senescence-associated secretory phenotype, according to background provided by Cedars-Sinai's Center for Translational Geroscience. That secretion pattern disrupts the healthy cells around it and has been tied to a long list of age-related problems, including metabolic dysfunction and declining physical fitness.

Under normal circumstances, the immune system is supposed to identify and clear these cells out. The Cedars-Sinai team's central question was why that clearance process seems to break down as people age, letting senescent cells pile up instead of getting swept away.

Borrowing a concept straight from cancer immunology

The answer they landed on borrows directly from oncology. Immune checkpoint proteins, PD-L2 and its better-known relative PD-L1, are the same molecular flags cancer cells use to switch off immune attacks against themselves, a mechanism that led to an entire generation of checkpoint-inhibitor cancer drugs over the past decade. Selim Chaib, research assistant professor of medicine at Cedars-Sinai and the study's first and co-corresponding author, said the team wanted to test whether a similar evasion process might explain senescent cell persistence during ordinary aging, not just in tumors.

It did. "Our findings suggest that PD-L2 may help aging cells stay in the body when they would normally be removed by the immune system," Chaib said, according to the university's announcement of the findings. The team found PD-L2 was elevated in isolated senescent human cells and rose further with age across several human tissue types, giving the checkpoint-evasion theory a plausible mechanism rather than just a correlation.

What happened when the researchers removed it

To test whether PD-L2 was actually driving the buildup, rather than just showing up alongside it, the team turned to genetically engineered mice lacking the PD-L2 gene entirely. Old PD-L2 knockout mice accumulated fewer senescent cells than old mice with a normal, functioning copy of the gene, according to the study's abstract published in Cell Metabolism. Separately, blocking PD-L2 in laboratory experiments reduced both the number of aging cells present and the levels of inflammatory substances those cells were producing.

That combination, fewer senescent cells and less of the inflammatory signaling they generate, is exactly the outcome researchers in this field have spent years trying to achieve through other means, most notably a class of drugs called senolytics that selectively kill off senescent cells. What's different here is the target: rather than killing the cells directly, blocking PD-L2 appears to let the immune system do the clearing work it was supposed to be doing all along.

A blood marker that already tracks with treatment

The study's human data adds a practical dimension beyond the mouse work. Soluble PD-L2 levels in blood increased with age and dropped after patients received senolytic treatment, according to the paper's findings as summarized in its Cell Metabolism abstract. That pattern suggests PD-L2 could double as a measurable biomarker, giving researchers a blood-based way to track how much senescent-cell burden a person is carrying and whether a given treatment is actually reducing it.

That's a meaningful gap in the field right now. Senescent cell buildup is currently difficult to measure directly in living patients without invasive tissue biopsies, which has slowed efforts to test senolytic and senomorphic drugs in human trials. A reliable blood marker tied directly to the underlying biology, rather than a downstream proxy, would give researchers a faster, less invasive way to evaluate whether an anti-aging therapy is working.

Where this fits in a fast-moving field

Cedars-Sinai has positioned itself as a hub for this kind of research, running parallel studies on how exercise, senolytic drugs, and metabolic interventions affect senescent cell accumulation in humans. Existing senomorphic drugs already in use for other conditions, including metformin and rapamycin, work by suppressing the inflammatory signals senescent cells put out rather than clearing the cells themselves, part of a broader toolkit researchers are assembling to manage the effects of cellular aging even before a single definitive therapy exists.

PD-L2 blockade would represent a genuinely different mechanism from either of those existing approaches: not suppressing inflammation and not directly killing cells, but restoring the immune system's own ability to recognize and remove the cells causing the problem in the first place. The Cedars-Sinai team's next steps will likely involve testing whether PD-L2-targeted approaches, potentially adapted from existing checkpoint-inhibitor drug technology already approved for cancer, can achieve the same senescent-cell reduction in humans that they achieved in mice. Given how mature checkpoint-inhibitor manufacturing already is in oncology, repurposing that toolkit toward aging biology, rather than building an entirely new drug class from scratch, could meaningfully shorten the road from this laboratory finding to an actual human trial.