The year-old Novartis-Penn Center for Advanced Cellular Therapeutics in Philadelphia supplies cancer fighting T cells to multiple hospitals, visible out the window.


For experimental cancer therapy, a struggle to ensure supply keeps up with demand

A transformative cancer therapy based on modified immune cells has lured doctors, companies, and patients alike, but many are hitting a frustrating roadblock: generating enough of these chimeric antigen receptor (CAR)-T cells to meet surging demand. The situation is fluid, with shortages cropping up in some places and easing in others. Doctors, meanwhile, are grappling with how best to distribute the experimental therapy among very sick patients in clinical trials.

“How do I allocate the resource in a way that’s fairest to everybody and that treats the most patients and potentially saves the most lives?” asks Stephan Grupp, a pediatric oncologist at the Children’s Hospital of Philadelphia (CHOP) in Pennsylvania. Grupp has offered CAR-T therapy to more than 150 children with late-stage acute lymphoblastic leukemia (ALL)—and worries that because of supply limitations, he can’t help more.

CAR-T cell therapy took the cancer world by storm in the summer of 2010. It involves removing a patient’s immune cells, genetically modifying them to fight their particular cancer, then transfusing them back. The approach is risky—some have even died from it—but for blood cancers in particular, it’s been remarkable, saving patients at the 11th hour and keeping some in remission for years.

For patients, getting the most anticipated new treatments is never easy. Clinical trials are tightly controlled and not everyone is eligible. But for this personalized approach, the difficulties are multiplied. From the beginning, CAR-T cells were tough to produce. Unlike a drug, each batch is designed for a specific patient. Production involves genetic engineering and working with live cells, and it is still mostly done by hand, by highly trained technicians. Preparing cells for a single patient can take weeks and cost tens of thousands of dollars.

The demand took off when doctors began reporting impressive results in leukemia and, later, lymphoma, with response rates ranging from 40% to 50% in lymphoma to more than 90% in some leukemias. More researchers joined the fray, keen to test CAR-T therapy in patients with other cancers. Companies joined in, too, anticipating a burgeoning market. Dozens of trials are underway, and two CAR-T cell products, for childhood and young adult ALL and aggressive B-cell lymphoma, may be approved later this year by the U.S. Food and Drug Administration (FDA).

“The promise of this stuff outpaced the typical approach to development,” Grupp says. With about 50 scientists and technicians working furiously to produce cell therapies for trials throughout the University of Pennsylvania (UPenn), CHOP can treat about five children each month. That number “rules my life,” says Grupp, who receives two or three calls a week about yet another child whose parents hope to secure the therapy. He engages in a delicate dance, trying to keep some children stable while treating the sickest, before they’re too sick to benefit. So far, he believes, he has reached everyone referred from within the United States in time.

At the National Cancer Institute in Bethesda, Maryland, hematologist James Kochenderfer’s waiting list is driven by the two or three CAR-T cell products a month he’s able to secure from the agency’s facility. That’s not enough to accommodate all the adults eligible for the seven CAR-T trials he’s running in blood cancers. He usually enrolls on a first-come, first-served basis.

Demand for the therapy is also spurring competition for the researchers and technicians who create the cells. “Everyone is losing [people] to everyone else,” says immunologist Bruce Levine of UPenn, who directs the cell production facility. He guards against the poaching of his staff by companies that can pay a higher salary, in part by stressing a connection with patients at the hospitals he can see out his window. “We have patients come over on a regular basis” who received cells made in UPenn’s facility. “No company can offer that.”

One center that says it’s keeping up with demand is Baylor College of Medicine in Houston, Texas, in part because of a stroke of luck. “We kind of overbuilt,” opening a vast facility in 2010, says Adrian Gee, who runs it. Seattle Children’s Hospital in Washington recently broke ground on a building that in a few years will triple or quadruple its cell therapy capacity, now about 10 batches of CAR-T cells a month.

Another way to generate cells for more patients is to shorten the time it takes to make them. Right now the time from “vein to vein” ranges from about 2 to 4 weeks, depending partly on the technique. Scientists are experimenting with more efficient approaches. Rebecca Gardner, a pediatric oncologist at Seattle Children’s, says the hospital is shifting to one that shaves a week or two off its 3- to 4-week time frame.

Ultimately, the supply problem “can be solved with money,” says Ronald Levy, a lymphoma specialist at Stanford University in Palo Alto, California. And no one has more money to funnel into CAR-T therapy than the companies. Novartis, for example, spent $43 million on a manufacturing facility in Morris Plains, New Jersey, and last week it released results from a lymphoma trial in which cells were frozen and flown to and from patients in 10 countries.

But some researchers wonder whether the companies will be ready to accommodate the surge in demand expected if FDA approves the first CAR-T therapies. The strain on supplies would increase if doctors want to offer CAR-T therapy to patients “off-label,” to those who fall outside the approved indication but might still benefit. Unlike a traditional drug, every order needs to be placed through the companies—and it’s not clear whether they and FDA will support off-label use. If Novartis’s product is approved for leukemia patients up to 28 years old, say, and “you have a 28.1-year-old, does that mean you can’t treat them?” asks David Maloney, an oncologist and immunotherapist at the Fred Hutchinson Cancer Research Center in Seattle. “I don’t know what’s going to happen.”

For now, scientists are pondering how to best allocate the therapy. At Seattle Children’s, pediatrician and bioethicist Douglas Diekema was drafted by colleagues to offer ethical guidance on what to do if the hospital can’t make enough CAR-T cells for everyone in planned trials in brain and other solid tumors, as well as more leukemia trials. “A year from now we’ll probably have six to seven trials,” up from three today, Gardner says.

Last month, Diekema and his colleagues published a paper online in the Journal of Medical Ethics describing a “triage plan” for selecting volunteers for CAR-T trials. They argued that, when possible, doctors should focus on the likelihood and magnitude of benefit, treating the sickest patients first. In a second paper still under review, the team will discuss how to allocate CAR-T therapy across clinical trials. There, the ethical calculus is different, including whether the disease affects many people versus just a few.

“I did get a 10 p.m. phone call” recently, saying “we may need to implement” the patient allocation strategy, Diekema says. But the hospital, in the end, managed to get the cells to everyone. “We dodged it” this time, Diekema says. Now, like many others, he’s waiting for the next crunch.