Biology Atlas

Map / Drugs

drug · intermediate · partly written

JCAR015

JCAR015 was a CD19 CAR T cell product for adult acute lymphoblastic leukemia whose development stopped in 2017 after deaths from brain swelling.

Also called: Juno JCAR015

This describes published medicine for readers who want to understand how it works and what happened. It is not medical advice and says nothing about what any person should take.

What happened in one breath

A CD19 CAR T product for adult leukemia worked, then five patients died of brain swelling across two trial holds. Juno stopped the programme on 2 March 2017, citing patient factors, the chemotherapy and product factors, timing and better pipeline candidates. 1

Reported, not yet replicated

The trial and the first hold more

ROCKET was a phase 2, single-arm trial of JCAR015 in adults with relapsed or refractory B-cell ALL. Juno reported complete responses in an early analysis. Then three patients died of cerebral edema, all in the cohort that received fludarabine plus cyclophosphamide and none in the cyclophosphamide-only cohort, according to the company. FDA's hold was announced in July 2016. The company called fludarabine the most likely and most modifiable factor, while saying causes could be multifactorial. 123

Reported, not yet replicated

The second hold and the end more

The hold was lifted about two weeks later for a protocol using cyclophosphamide alone. Two more patients died of cerebral edema and Juno placed a voluntary hold on 23 Nov 2016. The programme was stopped on 2 March 2017. The regimen the two later patients had received was not stated in any page opened. The primary documents (Juno's own release, FDA hold letters) were not opened; this section rests on news reports and a call transcript. 142

Reported, not yet replicated

What the numbers and the analysis said deep

As reported from a conference presentation, in 38 adults severe neurotoxicity occurred in 56% and severe cytokine release syndrome in 22%; the report also gives complete remission or incomplete count recovery of 59% at day 28 and a median overall survival of 8.1 months. The company's post hoc analysis described early rapid CAR T expansion and rising IL-15, and in the fatal cases endothelial damage and blood-brain barrier breakdown rather than CAR T cells in the brain. High-intensity lymphodepletion had an odds ratio of 7.25 (95% CI 1.14 to 53.98), the only significant factor; age under 30 and two or fewer prior regimens had intervals crossing 1. It is exploratory, not proof. 5

Reported, not yet replicated

Compare: the same disease, a different cohort deep

A single-centre cohort of 53 adults with ALL treated with a different CD19 CAR (19-28z) reported 83% complete remission, severe cytokine release syndrome in 26% and one death from it; lower disease burden predicted fewer severe events. The two products differ in design and population, so this shows that adult ALL is not by itself the problem, not what caused JCAR015's deaths. 6

Reported, not yet replicated

Modality

CAR T cell therapy

Targets

Pathways

Indications

Identifiers

Nct rocket

NCT02535364

Mechanism

An autologous CAR T cell product aimed at CD19, given intravenously after lymphodepleting chemotherapy, like tisagenlecleucel. Efficacy was real: Juno reported complete responses in an early phase 2 analysis. The construct details were not opened for this record. 25

Reported, not yet replicated

Delivery route

Must survive

As for autologous CAR T generally: manufacture, then expansion after lymphodepletion. Here the hazard was the host's response to the rapid expansion: in the fatal cases a company analysis described endothelial damage and blood-brain barrier breakdown rather than CAR T cell infiltration of the brain. 5

Reported, not yet replicated

Side effects

Severe neurotoxicity and cytokine release syndrome

In 38 adults reported at a conference: severe neurotoxicity 56% and severe cytokine release syndrome 22%. 5

Reported, not yet replicated

Fatal cerebral edema

Five deaths from cerebral edema in the ROCKET trial (three in 2016, two more in November 2016). 134

Reported, not yet replicated

Verdict

Outcome

failed

Why

Five patients died of cerebral edema, the trial was put on hold twice and the company stopped development on 2 March 2017. The company named patient-specific factors, the conditioning chemotherapy and product factors, and gave timing (a new phase 1 would have been needed) and better pipeline candidates as further reasons. No single cause was isolated. 1

Reported, not yet replicated

Failure mode

ROCKET (phase 2, NCT02535364, started Aug 2015; the registry lists it as terminated with enrolment 82). After early results with complete responses, three patients died of cerebral edema; FDA placed the trial on hold in July 2016. The company called the fludarabine added to the conditioning regimen the most likely and most modifiable factor while saying causes could be multifactorial. The hold lifted about two weeks later for a cyclophosphamide-only protocol, and two more patients died of cerebral edema; Juno placed a voluntary hold on 23 Nov 2016. The removal of fludarabine therefore did not prevent further deaths. In a company post hoc analysis, high-intensity fludarabine plus cyclophosphamide lymphodepletion was the only factor with a significant association (odds ratio 7.25, 95% CI 1.14 to 53.98); this is exploratory and rests on a news report of a conference talk. Other CD19 products also reported neurologic toxicity and serious or fatal cerebral edema, so brain swelling was not unique to this product. 1527348

Reported, not yet replicated

Teach this

In one sentence

JCAR015 is a CD19 CAR T product that showed responses in adult leukemia and was stopped after five deaths from brain swelling. 1

Reported, not yet replicated

The five-minute version

Juno's ROCKET trial first lost three patients to cerebral edema (hold, July 2016), resumed without fludarabine, lost two more (second hold, November 2016) and stopped on 2 March 2017. The company's own analysis pointed to intense lymphodepletion and fast expansion with blood-brain barrier damage, but the cause was never isolated. The same CRS and neurotoxicity biology sits under the approved products too. 15

Reported, not yet replicated

The common wrong idea

Removing the suspected culprit (fludarabine) should have made the trial safe.

Two more patients died after the protocol change, so the cause was bigger than one drug; the toxicity comes from the expansion biology itself.

The one citation to hand over: 1

Still open here

Not yet written: Assays used, Off targets.

Linked from

Sources

  1. Juno halts development of CAR T cell cancer immunotherapy candidate JCAR015. Genetic Engineering & Biotechnology News (2017). news, not peer reviewed; opened 2026-11-17 all records citing this
  2. Juno Therapeutics conference call transcript on the ROCKET clinical hold (SEC exhibit 99.1). Juno Therapeutics via SEC EDGAR (2016). web, not peer reviewed; opened 2026-11-17 all records citing this
  3. FDA halts Juno CAR-T trial after three patient deaths. Fierce Biotech (2016). news, not peer reviewed; opened 2026-11-17 all records citing this
  4. Juno halts CAR-T trial for second time following additional patient deaths. BioSpace (2016). news, not peer reviewed; opened 2026-11-17 all records citing this
  5. JCAR015 experience informs future CAR T studies (SITC 2017 report). OncLive (2017). news, not peer reviewed; opened 2026-11-17 all records citing this
  6. Park, J. H., et al.. Long-Term Follow-up of CD19 CAR Therapy in Acute Lymphoblastic Leukemia. New England Journal of Medicine (2018). paper; opened 2026-11-17 all records citing this
  7. JCAR015 in Adult B-cell Acute Lymphoblastic Leukemia (ROCKET), NCT02535364. ClinicalTrials.gov. database, not peer reviewed; opened 2026-11-17 all records citing this
  8. FDA Summary Basis for Regulatory Action: Yescarta (axicabtagene ciloleucel). FDA (2017). regulatory, not peer reviewed; opened 2026-11-17 all records citing this
Where it sits on the map
History of this record (1)

This record has been changed 1 time, on the dates below. The note kept against each change is written for the people who maintain the atlas, in their own shorthand, so it is not shown here; it is in the record itself.

Year
2017
Regulatory status
discontinued
Status
partly written
Last updated
2026-11-17
Record
content/drugs/jcar015.json