Deaths in Cancer Trials Remain Mystery for Promising Therapyby
Optimism from CAR-T results tempered by serious side effects
Investors looking for answers at blood-disease conference
An experimental cancer therapy has shown dramatic results, including long-term cures, for patients with severe disease yet has also left puzzling questions about deaths that have occurred during some tests.
Drugmakers are trying to understand how the healing power of CAR-T, a therapy that tweaks a patient’s own protective cells to make them destroy cancer, could be causing severe side effects.
“It’s a double-edged sword,” said Lewis Silverman, clinical director of the Pediatric Hematologic Malignancies Center at the Dana-Farber/Boston Children’s Cancer and Blood Disorders Center. “You have to get through some bad side effects because that’s actually a sign that the CAR-T is doing something.”
Some of the newest data on the therapy’s use in blood cancers will be presented at the American Society of Hematology meeting in San Diego starting this weekend. Among the companies there will be the industry’s top players -- Juno Therapeutics Inc., Kite Pharma Inc. and Novartis AG. A key question concerns the risk of brain swelling, which happened to five leukemia patients in a Juno trial, who later died, as well as one in a Novartis study.
“It’s a slippery slope between efficacy and side effects,” said Jason McCarthy, an analyst at Maxim Group. “Side effects are absolutely a measure of efficacy.”
Kite said in September that 76 percent of patients with aggressive lymphomas responded to its therapy in a trial, and tumors disappeared in 47 percent. But two patients died after a severe immune reaction, called cytokine release syndrome, that’s been seen in tests of similar treatments.
The deaths are a reminder that an overactive immune system is linked to many serious conditions, such as lupus and rheumatoid arthritis. Doctors say side effects are part and parcel of the way CAR-T fights cancer -- by ratcheting up the body’s defenses.
Each CAR-T therapy has unique characteristics that can lead to differences in how cells behave in patients, Kite Chief Medical Officer David Chang said. The company has continued development of a therapy, KTE-C19, that began almost a decade ago at the U.S. National Cancer Institute, he said.
“We are pleased with the outcomes to date and are committed to making KTE-C19 available to those patients in need,” he said in an e-mail.
Immune cells make toxins to kill cancer and are likely to release more of them when tumors are large. Indeed, the more cancer a patient starts with, the greater the extent of side effects is likely to be, according to Stephan Grupp, head of the cancer immunotherapy program at Children’s Hospital of Philadelphia and lead investigator in one of the Novartis trials.
‘Pounds of Leukemia’
“I don’t think you have to get really sick for this stuff to work,” Grupp said. “You have to get really sick to clear up what could be pounds of leukemia.”
Ahead of the conference, Bluebird Bio Inc. presented early data from a small study that looked promisingly safe. Bluebird’s CAR-T for the blood cancer multiple myeloma didn’t cause any severe cases of cytokine release or serious brain side effects among 11 patients, the company said Wednesday.
Why patients responded so severely to the Juno CAR-T, called JCAR015, remains a mystery. The trial was halted in July after three therapy-related deaths, which Juno said were linked to a conventional chemotherapy drug called fludarabine. The U.S. Food and Drug Administration quickly gave a green light to restart the study without that drug.
Then, last month, two more patients died in the trial after showing similar symptoms. Juno halted the study again and now faces questions about how to prevent further deaths.
“There are many issues that need to be investigated” from the Juno trial, said Kenneth Anderson, a Harvard Medical School oncologist. Juno’s production of the therapy, the genes altered in the cells, and the severity of patients’ illness before treatment may all be linked to the brain injury, he said. Their disease, adult acute lymphoblastic leukemia, is also an especially dangerous cancer. “Those are among the issues -- there might be others.”
The risks need to be seen in the context of patients who otherwise would almost surely die sooner, Anderson said.
“They really have no other options,” he said.
Among more than 100 adult ALL patients that Juno has treated, less than 10 percent have died from treatment-related causes, Chief Medical Officer Mark Gilbert said Nov. 23 in a conference call. That compares to 11 percent to 23 percent of patients who die from their last-ditch chemotherapy drugs in other trials, he said.
“People have become a little bit jaded at how good the clinical results have become,” said Robert Nelsen, managing director at Arch Venture Partners, the third-largest shareholder in Juno, according to data compiled by Bloomberg. In a study of JCAR015 presented last year at the ASH conference, 82 percent of patients with ALL saw their tumors wiped out.
“If you shut down the trials you may have what’s approaching a 100 percent death rate,” he said. “So that’s what people need to keep in mind.”