Juno scrambles to avoid being crushed by a hold, but experts question its best case scenario
“Fludarabine in some ways acts like a dose amplifier,” says Juno CEO Hans Bishop, explaining the company’s theory of what went wrong.
About the last thing a biotech wants to see in the leadup to a widely heralded regulatory filing is a clinical hold from the FDA after patients in a pivotal trial are killed by your treatment. The instinctive response is to get out in front of the disaster, see if you can identify a factor other than your drug that could have caused the deaths, and reassure investors that you have a plan – or soon will have one – that will get you back on track ASAP.
Then, without burning through too many hours between the hold and the news, you host a press conference and put out a release and hope your stock doesn’t get eviscerated in the reaction.
By that measure, the past 24 hours at Juno Therapeutics must have been a blur of nonstop action. They not only immediately identified the cause of death in their lead CAR-T study, cerebral edema, they also ID’d a prime suspect, newly added fludarabine, which investigators were using to help prep patients for a better response to the treatment.
Solution: Drop the fludarabine, get the documentation back to the FDA and wait for the green light. All while assuring investors that this could all be over soon.
But it may not be anything nearly as easy as that at Juno. And there are major implications here for the whole field.
Sally Church, a Novartis vet with a doctorate from King’s College, writes the highly regarded Biotech Strategy Blog, where she specializes in covering cancer treatments. She’s been following the CAR-T field closely for 4 years, and she was immediately puzzled by Juno’s position.
The chemo fludarabine was used in the study to complete a round of lymphodepletion, eliminating diseased B cells to assist the newly inserted CAR-T cells to repopulate and do their work. It’s the same type of conditioning patients receive for hematopoeitic stem cell transplants. It’s also a procedure used in other CAR-T trials, with no history of cerebral edemas. And there’s no record of fludarabine being linked to cerebral edemas, she writes in her blog. Church gave me access to the article, which you can find here. (Note: There is a pay wall. If you’ve got the budget, we highly recommend you sign up.)
Then she asked a group of hematologists what they knew of a connection. Response: Nothing. And she queried Dr. Stephan Grupp (CHOP), who “probably treated more pediatric ALL patients with CAR T cell therapy than anyone”:
“We have not seen significant cerebral edema in any of our pediatric ALL patients on the CHOP/Penn study treated with CTL019, many of whom received fludarabine. CNS-related side effects have been seen, including confusion, aphasia and seizures in a small number of patients, but these effects have all resolved. Obviously, it is always hard to attribute toxicity to the chemo vs. the CAR T cells when both are given around the same time, but we believe the self-limited CNS side effects we have reported are more likely to be due to the T cells.”
Both Juno as well as rival Kite Pharma use CD-28 as a co-stimulation domain, she adds. Penn/CHOP/Novartis use 4–1BB. There are differences in patient responses, but no cases of cerebral edema have been reported by Kite so far, though Kite has had the same limited CNS issues that Juno has reported.
It’s not unusual to see a difference of opinion among scientists and companies engaged in rival studies. The point here is that there’s thoughtful consideration of a much bigger mystery about what went wrong for Juno. And a mystery is exactly what Juno doesn’t want to explore – especially when your competition at Kite announced that it had fully enrolled its rival study on the same day Juno stunned investors with news of the clinical hold.
Further confounding all of this, Bishop reported a few days ago that he had sold $4.5 million in stock; a small part of his holdings, but not the kind of thing investors like to see when bad news follows soon after.
Juno has already acknowledged that its prediction of a 2017 market launch will have to be delayed a year. And for now, despite its confident assertion that this can all be resolved in short order, there’s reason to believe that this could go on for much longer than Juno wants.
— John Carroll@JohnCendpts