On CAR-T’s Edge, Seattle Researchers Plot to Bypass Novartis Therapy

(Page 2 of 3)

a “me too” product, but to “going long” with a strategy that, if successful, would leapfrog Kymriah and solve the relapse problem.

Jensen runs SCRI’s Ben Towne Center for Childhood Cancer Research, about a 15-minute drive from Gardner and the hospital when Seattle’s downtown roads aren’t choked with traffic. At the center, researchers have developed a menu of T cell variations that Gardner and her colleagues have begun testing in small human trials this year. One that attracted extra scrutiny from a government oversight committee last year turns a patient’s T cells not just into CAR-T cancer killers but also into cancer mimics that alert the immune system—essentially, the first CAR-T vaccine.

Within 36 months, says Jensen, the Seattle researchers should have enough data to decide if a bigger trial, designed to gain FDA approval, is warranted.


The CAR-T technology that turned Greta Oberhofer’s T cells into a cure has actually made progress toward commercial use—but not in pediatric ALL. Juno Therapeutics (NASDAQ: JUNO), also of Seattle, has rights to the technology and has turned it into a treatment for adults with non-Hodgkin lymphoma. It is called JCAR017, and after encouraging clinical results earlier this year, Juno has a big study planned. If all goes well, Juno could ask the FDA for approval in 2018.

Pushing JCAR017 has become Juno’s top priority. Last year, five patients died in a Phase 2 study of Juno’s then-leading product, a different CAR-T product for adults with ALL. Juno has ended all work on the product, which was originally developed at Memorial Sloan Kettering Cancer Center in New York.

But Juno has no stated plans to develop JCAR017 for pediatric ALL. Jensen, who is a scientific co-founder of Juno and holds stock, says Novartis has raised the bar. Another approval in the same indication—specifically, kids and young adults with ALL who have relapsed twice— would require doing well in a randomized trial head-to-head with Kymriah, Jensen says, and that could mean paying the full $475,000 for each trial participant using the Novartis product.

Juno’s regulatory filings acknowledge that Seattle Children’s is pressing on with PLAT-02 independently from Juno. Juno did not address repeated questions about its own specific plans for pediatric ALL. Spokesman Chris Williams wrote that the PLAT-02 Phase 1 data “are valid, but differences in the [manufacturing process] preclude use for regulatory approval of a new indication. These data do remain informative for future trials in pediatrics.”

CAR-T is no cookie-cutter pharmaceutical treatment. The cell manufacturing is complex, unique for each patient, and takes three to four weeks to turn around—a lag that many developers are trying to shrink.

Novartis says once its product is available through 30 or more cancer centers by the end of 2017, it should be able to handle the capacity of every pediatric ALL patient who meets the criteria. Novartis spokeswoman Julie Masow declined to specify Novartis’s capacity. Every year, there are more than 600 kids and young adults in the U.S. whose ALL comes back after multiple treatments. Some subset would be eligible for Kymriah.

While Seattle Children’s continues to treat the same patient population in the PLAT-02 trial (100 patients and counting), Jensen agrees that “there only needs to be one commercially available CAR-T for relapsed-refractory remissions.” Novartis has pledged not to charge for patients who don’t respond to treatment within a month, a small step toward what some call value- or outcomes-based pricing. But Kymriah, like the Seattle Children’s CAR-T and others tested in adults, often starts to fade after the one-month burst of activity, which means that Novartis is going to get paid for a lot of treatments that stop working after several months.

And that’s key to the Seattle Children’s leapfrog idea for pediatric ALL. Jensen and colleagues have seen biological warning signs in patients that seem to predict relapse. Here’s one example: When CAR-T is working well it kills all cells, cancerous or not, that display the target protein on their surface. For kids like Greta, the protein is called CD19. If healthy B cells that produce CD19 start to rebound within a couple months, it’s a sign that the CAR-T cells could be fizzling out. (Greta’s CAR-T cells have stayed active, and she has needed supplements of antibodies that her missing B cells would normally produce.)

What if Seattle Children’s could give those patients different CAR-T treatments or the vaccine in various combinations to avoid relapse?


To find out, they have launched a complicated multi-part study, which drew both praise and concern last year from a National Institutes of Health committee that reviews plans for genetic engineering tests in humans.

Jensen, Gardner, and their colleague Colleen Annesley went before the committee to describe one part of the plan. It’s called PLAT-03, with a goal of enrolling 30 patients. (The committee voted in favor, and enrollment began this summer.)

PLAT-03 is the first trial of its kind. Jensen and lab mates are making two batches from a patient’s T cells. One batch becomes CAR-T cells (CAR stands for chimeric antigen receptor)—the kind of killers that Greta Oberhofer received, hunting down cells that display CD19. The second batch, however, become tumor mimics. They produce CD19 to stimulate the CD19-hunting T cells that might be losing steam, like waving extra red capes in front of a tired bull.

The technical term is “antigen presenting cell.” This is the first time a T cell has been used as an antigen-presenting cell in a human medicine. Seattle Children’s calls them T-APCs. (The pioneering prostate cancer vaccine sipuleucel-T (Provenge), also developed in Seattle, used different immune cells to rev up T cells.)

[A previous version of this story described Provenge as “failed” because its owner, Dendreon, went bankrupt in 2014. But Provenge’s latest owner—a new, private version of Dendreon—says it continues to provide the treatment.]

The idea behind T-APCs comes from a glaring limitation of CAR-T therapy in blood borne cancers. After spending their first weeks maniacally hunting down cancer—often with such vigor that the patient suffers dangerous side effects—the live CAR-T cells often lose steam, or “persistence.” Loss of persistence, or related warning signs, could also be a precursor to a relapse. In PLAT-03, patients will receive a dose of the T-APCs once a month up to six times, like booster shots. (Some of the patients will effectively have to start over, getting a new dose of their CAR-T cells and the T-APCs.)

But it gets more complicated. Seattle Children’s will start patients … Next Page »

Single Page Currently on Page: 1 2 3 previous page