T cells, colored blue here, move in to attack.
Share this story
In addition to fighting off invaders that arrive from outside the body, the immune system is also able to identify cells that have gone bad inside the body. Even though cancer cells look a lot like normal ones, immune cells can often tell the difference—enough that people who receive long-term treatments of immunosuppressive drugs have a higher incidence of cancer.
But the immune system clearly has its limits, or cancer wouldn't be a problem. Cancer cells evolve ways to avoid detection or use the immune system's own signals to tamp down its activity. A number of researchers have been looking for ways to reestablish the immune system's superiority, boosting it in a way that it once again clears out cancer cells. One option for doing so has been to simply boost the cells that already recognize a tumor by isolating them and growing them in large numbers in culture.
This doesn't consistently work, however, as it can be hard to identify and isolate tumor-specific immune cells. A team of researchers has figured out a way of taking stem cells, converting them into immune cells, and directing them to attack one type of cancer.
When most people think of immune function, they tend to think of the cells that make antibodies, called B cells. The antibodies they make stick to invading cells and viruses, keeping the invaders from infecting cells and helping to attract immune cells to destroy them. But there's a second branch of the immune system that targets human cells after they become infected. The T cells recognize strange or unfamiliar proteins on the surface of human cells. When they spot them, the T cells attack and kill the infected cell while boosting the rest of the immune response.
Cancer cells can also have odd-looking proteins on their surface. These proteins can be recognized by T cells, which will then attempt to kill the cancer cell. In many cases, these anti-cancer T cells are present in the bodie they're just not active in large enough numbers to keep the cancer in check. Some researchers have reasoned that getting more of these cells into the body could give the immune system an edge, so they set about isolating them, growing large numbers outside the body and then putting them back into the bloodstream.
For some cancers, initial trials of this technique suggested it could have promise. But it's not always easy to identify and isolate the immune cells that are targeting a cancer, nor is it necessarily easy to grow large numbers of them in culture.
The alternative is to use stem cells, which do grow extremely well in culture and can be available in large numbers. The problem is that stem cells don't start out recognizing anything in particular. In the body, both T and B cells mature through an elaborate process that gets rid of any cells that recognize normal, healthy human proteins and selects for those that actually have a reasonable chance of recognizing an invader. It's simply not possible to put stem cells through a similar process in a culture dish.
The new paper involves a clever solution to most of these problems. The authors started with existing T cells, which would have already gone through the selection process that prevents them from attacking normal cells. They then used the techniques that have been developed to induce them to adopt a stem cell fate to grow them up into large numbers.
Most of the results, however, would recognize something other than the cancer involved. So the authors engineered a hybrid receptor, with one part that recognizes a common leukemia protein and another part that plugs in to the normal T cell receptor system. They inserted these into their stem cells, converting all of them into cells that could recognize leukemia. They then converted the stem cells into T cells.
There were a couple of oddities about the resulting T cells. To begin with, they could recognize both the leukemia protein and whatever was recognized by their normal receptor. Because they expressed the receptors during their normal maturation process, they didn't mature normally. That said, they did recognize leukemia, and they generated a robust response to cancerous cells, clearing them from mice they were tested in. Essentially, the technique mass-produces cells that are trained to kill any cell that has a specific protein. If only cancer cells have the protein, only cancer cells get killed.
That makes the technique very flexible, because you can essentially direct it to any protein provided you find another protein that sticks to it—an antibody, its normal partner, etc. The challenge is that it's very difficult to find a protein that's exclusively expressed on cancer cells, meaning there's a real risk that these cells could accidentally end up targeting healthy tissue. Some extensive safety testing would seem to be in order.
Nevertheless, the work is a nice demonstration of the potential of induced stem cells as well as the degree to which we can manipulate the immune system. Not every type of cancer seems likely to be vulnerable to immune attack, but techniques like these may ultimately help control the ones that the immune system does attack.
Disclosure: About 15 years ago, I think I went to a party at the last author's apartment. I didn't realize this when choosing the cover the paper, and it did not influence my coverage.
Nature Biotechnology, 2013. DOI:
John became Ars Technica's science editor in 2007 after spending 15 years doing biology research at places like Berkeley and Cornell.