Therapies that make use of a patient’s own T cells, a component of the immune system, are showing promise in the clinic. Credit: David Scharf/Corbis

When immunologist Michel Sadelain launched his first trial of genetically engineered, cancer-fighting T cells in 2007, he struggled to find patients willing to participate. Studies in mice suggested that the approach — isolating and engineering some of a patient’s T cells to recognize cancer and then injecting them back — could work. But Sadelain did not blame colleagues for refusing to refer patients. “It does sound like science fiction,” he says. “I’ve been thinking about this for 25 years, and I still say to myself, ‘What a crazy idea’.”

Europe proposes joint Moon trips with Russia Microsoft billionaire takes on cell biology Watson's Nobel medal sells for US$4.1 million

Since then, early results from Sadelain’s and other groups have shown that his ‘crazy idea’ can wipe out all signs of leukaemia in some patients for whom conventional treatment has failed. And today, his group at the Memorial Sloan Kettering Cancer Center in New York City struggles to accommodate the many people who ask to be included in trials of the therapy, known as adoptive T-cell transfer.

At the American Society of Hematology (ASH) meeting held in San Francisco, California, on 6–9 December, attendees heard dozens of talks and poster presentations on the promise of engineered T cells — commonly called CAR (chimaeric antigen receptor) T cells — for treating leukaemias and lymphomas. The field has been marred by concerns over safety, the difficulties of manufacturing personalized T-cell therapies on a large scale, and how regulators will view the unusual and complicated treatment. But those fears have been quelled for some former sceptics by data showing years of survival in patients who once had just months to live.

“The numbers are pretty stunning,” says Joseph Hedden, an analyst for the London-based market-research firm Datamonitor Healthcare. “Companies have clearly decided that it’s worth the pitfalls of how much this therapy is going to cost to develop.”

At least five major pharmaceutical companies have invested in developing CAR-T-cell therapy over the past three years. Such interest from industry is a dramatic turn for a field that once consisted of a handful of academic medical centres. Small biotechnology firms have also sprung up to develop CAR T cells, including Kite Pharmaceuticals of Santa Monica, California, which raised US$127.5 million when it went public in June. And investors pumped $310 million into another CAR-T-cell company, Juno Therapeutics of Seattle, Washington, this year. “There is no doubt there has been a shift,” says Juno chief executive Hans Bishop.

Most of these efforts focus on killing the cancerous, antibody-producing B cells behind some leukaemias and lymphomas. Researchers do this by engineering T cells to recognize a protein on the surface of most B cells — CD19 — and attacking cells that display it (see ‘Call to arms’). Finding proteins that are expressed only on cancer cells can be difficult, and CD19 represents a compromise: the treatment sometimes wipes out all B cells, cancerous and healthy alike, but patients can survive without them.

At the ASH meeting, Sadelain and his colleagues reported that this approach left no signs of cancer in all six patients with lymphoma who were enrolled in one trial. In another presentation, immunologist Carl June of the University of Pennsylvania in Philadelphia showed that targeting CD19 reduced cancer burden in 9 of 23 patients with chronic lymphocytic leukaemia. In a more aggressive disease called acute lymphoblastic leukaemia, 27 of 30 patients had no signs of cancer after therapy and the CAR T cells remained in their blood two years later.

But studies also highlight the risks of revving up immune responses. In April, at least five CAR-T-cell trials were halted after a series of patient deaths linked to unusually high levels of a protein called interleukin-6, which promotes inflammation, as well as other inflammatory molecules. Interleukin-6 is part of the body’s normal response to infection. But the intense immune onslaught launched by CAR T cells can send interleukin-6 levels soaring. The trials resumed after investigators adjusted their protocols to better monitor and treat the problem.

These safety risks, as well as the difficulty of manufacturing CAR T cells, are still putting many drug companies off, says Andrew Baum, the London-based head of global health-care research for Citi, an investment bank headquartered in New York City. “The bulk of the multinationals are standing back and watching, rather than getting engaged here,” he says.

When CAR T cells do reach the market, they will not be cheap. Baum says that some sponsors are tentatively planning to price their therapies higher than bone-marrow transplants, which can exceed $500,000. The cost may be so high, he says, that companies are forced to set up a reimbursement scheme in which they are paid only when a patient benefits from the treatment. Baum estimates that peak sales of CAR-T-cell therapies will reach $10 billion annually, although that amount will depend on what competing therapies emerge and whether the treatment can be extended to other cancers.

For now, Sadelain, a scientific founder of Juno Therapeutics, hopes that the attention from industry will spur the field. He remembers his postdoc days, when he struggled to insert genes into T cells and colleagues asked him why he was bothering. “We’ve never had this kind of investment in the field before,” he says. “It’s hard to believe — sometimes I still pinch myself.”