Skip to main content

Stanford Team Uses Engineered Immune Cells as Cancer Diagnostics


NEW YORK (360Dx) – A team led by scientists at Stanford University has engineered immune cells to serve as cancer diagnostics.

In a study published this week in Nature Biotechnology, the researchers demonstrated the ability of engineered macrophages to detect tumors as small as 25 mm3 and with sensitivity superior to that offered by endogenous protein or nucleic acid biomarkers.

The researchers have licensed the technology to the diagnostic startup Earli, said Stanford professor Sanjiv Gambhir, senior author on the paper and a cofounder and board member of Earli. The South San Francisco, California-based company was launched in 2018 to commercialize another cancer diagnostic technology developed in Gambhir's lab, this one using mini-circles of DNA that could be injected into subjects and activated by tumor cells to produce a plasma protein indicating the presence of cancer.

Both technologies have their roots in the notion that endogenous biomarkers like plasma proteins or cell-free DNA secreted or shed by tumors may in many cases not be present in sufficient quantities to detect cancers in their earliest stages. This has led Gambhir and others to explore whether synthetic biomarkers might help address this issue.

"For many years we've been studying this [endogenous cancer marker] problem, and we have come at it in different ways," he said. "We have developed ways to stress the tumor environment so that it forces the cells to release more biomarkers. We have built mathematical models to look at what amount of signal in the blood one can expect to detect for a given small-sized tumor somewhere in the body."

Another recently launched firm in the synthetic biomarker space is Glympse Bio, a spinout from the lab of Massachusetts Institute of Technology researcher Sangeeta Bhatia, which is using nanoparticles linked to peptides specific for different endogenous proteases. By injecting these agents into a patient and then measuring the levels of different peptide cleavage products, the company can assess the activity levels of different proteases, which can be linked to various disease states.

The macrophage-based approach presented in the Nature Biotechnology study takes as inspiration the burgeoning cancer immunotherapy field, Gambhir said, though rather than using a person's immune system to fight cancer, it uses it for early detection of cancer.

To achieve this, the researchers leveraged the fact that macrophages exhibit two distinct metabolic phenotypes upon confronting disease. One, called M1, is a pro-inflammatory state that is the most common macrophage response. Less common is an anti-inflammatory state called M2. This is the state macrophages adopt when confronting many solid tumors.

Using this knowledge, Gambhir and his team engineered macrophages so that they would express the Gaussia luciferase (Gluc) protein upon conversion to M2 status. They injected these engineered macrophages into mice with various types and stages of cancers, and the macrophages traveled to these cancers, whereupon the macrophages produced secreted Gluc, which the researchers could then detect using a blood test or via imaging.

Gambhir noted that using macrophages, as opposed to devices like the DNA mini-circles, solves one of the major issues facing synthetic biomarker approaches — delivering the synthetic markers to the tumor.

"One of the challenges is you have to get the [biomarker] molecules delivered to the cancer cells if they exist somewhere in your body," he said. "But immune cells are naturally patrolling your body and making their way to tumors. We don't have to engineer tumors, they already find them. And we know that because when we remove tumors from people and look at what is there, in addition to tumor cells, there are many different immune cell populations, including macrophages."

In the Nature Biotechnology study, the researchers were able to detect tumors as small as 25 mm3, which they wrote is well below the limit of detection for approaches like PET imaging (200 mm3) and cell-free DNA (10,000 mm3 or more according to some estimates). They also found that the approach performed with high specificity under inflammatory conditions, which they noted "is a significant confounding disease state for cancer diagnostics."

That said, a number of challenges remain regarding clinical implementation of such an approach. Calling the study a "proof-of-principle," Gambhir highlighted a number of areas where he and his colleagues are looking to improve the technology.

While it is relatively specific for cancers, "it is not as specific as we would like," he said. "There are things that are not tumors that can cause M2 polarization [in macrophages], for example, certain kinds of wound healing."

Gambhir noted that the fact that the secreted Gluc protein can be imaged would allow clinicians to follow up a positive result with imaging tests that would add another layer of specificity helps with this issue somewhat. He added that in the future the researchers plan to add more complex circuitry into the engineered macrophages that would allow for a more tumor-specific response.

They are also exploring whether other kinds of immune cells can be engineered for this purpose. For instance, engineering longer-surviving immune cells could allow for a model similar to immunization wherein patients are "given a set of cells that live in their body forever, and when they encounter a problem they ramp up, multiply, divide, and then produce [a] signal for us to detect and image," Gambhir said.

While immune cell engineering is currently too complicated and expensive a technology for large-scale screening, Gambhir said he imagined the approach could be used in patients at high risk of developing cancer either due to family history or previous cancer diagnoses.

"It's not something that is simple to do, but it is doable," he said, adding that he expected developments in immunotherapy would continue to advance cell engineering technology.

Ultimately, he said he envisioned immunodiagnostics and immunotherapeutics merging to a point at which immune cells are engineered to both detect and kill tumors.

In the near term, Gambhir and his colleagues plan to test their macrophage-based diagnostic in additional mouse tumor models and to engineer additional circuitry into the cells to improve their specificity for cancer.

They are also looking to expand their work to diseases beyond cancer, like multiple sclerosis.

"For any disease in which immune cells like macrophages make their way to the diseased cells, we believe we can reengineer the immune cells to sense that environment and produce a signal," he said.