Researchers at Stanford Medicine and collaborating institutions have developed a way to convert natural killer cells, a fast-acting part of the immune system, into a specialized form that can push into solid tumors and keep attacking cancer once inside. In laboratory tests and in mice, the modified cells infiltrated tumors far better than conventional natural killer cells and slowed the growth of several cancers, including tumors derived from human melanoma and head and neck cancers. The results, described by the team as a proof of concept, come from experiments in animals and cell cultures rather than from treatment in patients.
Cell therapies that harness the immune system have already changed the outlook for some cancers of the blood and lymphatic system. Solid tumors have proven much harder to treat, because immune cells struggle to enter dense tumor tissue and the tumor environment can release signals that blunt nearby immune defenses. The Stanford strategy aims squarely at both of those obstacles by reprogramming natural killer cells before they are deployed.
Why solid tumors resist immune cell therapies
Natural killer cells were first identified in the 1970s and are named for their ability to recognize and destroy abnormal cells, including cancer cells and virus-infected cells, without needing to be primed against a specific target first. That speed makes them attractive for cancer therapy, but circulating natural killer cells often fail to penetrate solid tumors in useful numbers. The senior author, John Sunwoo of the Stanford School of Medicine, said the team focused on a tissue-resident form of the cell that settles into tissues and adapts to the local environment, rather than the version that patrols the bloodstream. The study was published in Science Translational Medicine.
Turning circulating cells into tissue-resident killers
To create the more aggressive cells, the researchers isolated circulating natural killer cells from human blood donors and exposed them to combinations of biological signals. A key ingredient was transforming growth factor beta, or TGF-beta, a signaling protein produced by many cell types, including tumor cells. The amount and timing of that signal proved decisive. Sunwoo described it as a balance in which just enough TGF-beta produced tissue-resident cells with strong toxic activity against cancer, while too much left the cells resident but dysfunctional and unable to kill. Brief exposure worked far better than prolonged exposure.
Finding the right cellular recipe
The most effective approach involved briefly exposing the natural killer cells to short-lived human epithelial tumor cells that delivered a temporary burst of active TGF-beta, and direct physical contact between the cells was essential rather than mere proximity. The team then compared the two resulting types of tissue-resident cells and found that only the strongest cancer-killing version carried a surface protein called CD39, alongside the shared markers CD49a and CD103. Those potent cells also contained more of the molecular machinery used to destroy targets, including perforin, which punches holes in target cells, and granzyme A, a toxic molecule delivered through them.
What happened in the mouse experiments
When the modified cells were injected into mice, they slowed the growth of several types of solid tumors over days and weeks. The effect grew stronger when the cells were paired with cetuximab, a monoclonal antibody that marks certain cancer cells for immune attack and is already approved to treat metastatic colorectal cancer and advanced head and neck squamous cell carcinoma. A single dose of the combination suppressed tumor growth in mice far more effectively over a month than either treatment alone, and the researchers reported no apparent adverse effects, with combination-treated mice appearing healthy even as others became sick. Sunwoo cautioned against reading too much into mouse results, calling the work proof of concept.
The appeal of an off-the-shelf treatment
Natural killer cells offer a practical advantage that could make this approach easier to scale than many existing cell therapies. They generally do not trigger an immune reaction when transferred between people, whereas most current immune cell treatments must be manufactured individually from each patient’s own cells. That means a therapy based on these modified cells could in principle be produced in large batches, frozen and made available to many patients. According to the researchers, cells collected from a single donor could yield roughly 20 treatment doses in about two weeks, which Sunwoo described as making cell therapy far more accessible.
What comes next in the clinic
Sunwoo and his colleagues are preparing a Phase I clinical trial to test the combination therapy in people with advanced squamous cell carcinoma, a trial that could begin by the end of the year pending approval from the Food and Drug Administration. Phase I trials are designed primarily to evaluate safety and dosing in a small number of patients, so the study would be an early step rather than a confirmation of effectiveness. Sunwoo has also filed to patent a method for producing and expanding large numbers of the modified cells, technically known as cytotoxic tissue-resident natural killer cells. The Stanford Medicine account of the research notes that contributors came from Ohio State University and Washington University School of Medicine, with funding from the National Institutes of Health.
How to weigh an early-stage advance
The findings represent encouraging laboratory and animal results, not an available treatment, and the gap between promising mouse experiments and approved therapies is wide and often difficult to cross. Even so, the work addresses a long-standing bottleneck in cancer immunotherapy by tackling the twin problems of getting immune cells into solid tumors and keeping them active once there. A detailed summary of the study underscores that the next real test will be whether the approach proves both safe and effective in human patients.
This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.
More from Morning Overview
