Export

Spotlight – Putting a number on CAR-T deaths

Date June 26, 2017

The toxicity of CAR-T therapies took on fresh importance with the discontinuation of Juno’s Rocket study of JCAR015 owing to deaths from cerebral oedema. Little wonder that when Novartis presented its update of the Eliana trial of CTL019 one spotlight fell on safety.

The good news, according to Saturday’s update at the European Haematology Association meeting, is that there were no new unexpected deaths, and no cerebral oedema. How this plays out will be important: precise safety data are hard to pin down, but an exhaustive EP Vantage search of various sources has yielded a definitive list of CAR-T deaths not due to disease progression (see tables below).

This involves not just CTL019, but all of the three leading players’ CD19-targeted projects. One little-appreciated finding is there have been seven deaths – two from cerebral oedema – in trials of Juno’s JCAR014. This has resonated little with investors, likely because Juno has long maintained that it does not intend to pursue JCAR014 to registration.

However, the close similarity with Juno’s lead – JCAR017 differs from JCAR014 only in a manufacturing step – should push CAR-T followers to pay close attention to this issue. That said, reported deaths on JCAR017 have been substantially lower, the data show.

18 Juno study deaths not due to disease progression 
Indication  Study  Trial ID  Description of grade 5 event  Due to CAR?  Source 
JCAR017 (Juno/Seattle Children's Hospital) 
Lymphoma  Transcend  NCT02631044  Diffuse alveolar damage (subject refused mechanical ventilation)  Yes  Asco 2017 
Lymphoma  Transcend  NCT02631044  Multiple organ failure  No  Asco 2017 
JCAR015 (Juno/Memorial Sloan Kettering) 
Adult ALL  MSKCC study (Jae Park)  NCT01044069  Severe hypotension, (ventricular arrhythmia?), CRS  Yes  Juno S-1, Dec 2014 
Adult ALL  MSKCC study (Jae Park)  NCT01044069  Status epilepticus, CRS (history of neurotox w prior CAR-T)*  Yes  Juno S-1, Dec 2014 
Paed/adult ALL  MSKCC study (Kevin Curran)  NCT01860937  Survived CRS, died 4mth later from multiviral pneumonia  No  Report to GTSAB 
Adult ALL  MSKCC study (Jae Park)  NCT01044069  Sepsis, multi-organ failure  Yes  Ash 2015 
Adult ALL  Rocket  NCT02535364  Neurotox, later specified as cererbral oedema  Yes  Juno Q3 2016 SEC filing 
Adult ALL  Rocket  NCT02535364  Cerebral oedema  Yes  Juno 7 Jul 2016 call (revised) 
Adult ALL  Rocket  NCT02535364  Cerebral oedema**  Yes  Juno 7 Jul 2016 call (revised) 
Adult ALL  Rocket  NCT02535364  Cerebral oedema  Yes  Juno Q4 2016 SEC filing 
Adult ALL  Rocket  NCT02535364  Cerebral oedema***  Yes  Juno Q4 2016 SEC filing 
JCAR014 (Juno/Fred Hutchinson) 
Adult ALL  Hutch study  NCT01865617  CRS  Yes  Juno S-1, Dec 2014 
Lymphoma  Hutch study  NCT01865617  Encephalopathy and pontine haemorrhage  Yes  Ash 2015 
Adult ALL  Hutch study  NCT01865617  Cerebral oedema  Yes  7 Jul 2016 call (revised) 
CLL  Hutch study  NCT01865617  Pulmonary aspergillosis after severe CRS  No  Ash 2015 
Lymphoma  Hutch study  NCT01865617  CRS or neurotox  Yes  Juno Q4 2016 SEC filing 
Adult ALL  Hutch study  NCT01865617  CRS or neurotox  Yes  Juno Q4 2016 SEC filing 
CLL  Hutch study  NCT01865617  CRS, cerebral oedema  Yes  Juno Q4 2016 SEC filing 
Note: ~500 subjects combined have been dosed JCAR015, JCAR017 or JCAR014, including 150 in adult ALL. *FDA-imposed clinical hold. **Second FDA-imposed clinical hold. ***Voiluntary clinical hold, leading to discontinuation of JCAR015. 

The toxicity of CAR-T therapies has been problematic ever since the first signs of truly robust activity were seen – it is known that cytokine release syndrome and neurotoxicities correlate with efficacy and tumour burden.

However, nurses have become very skilled at dealing with these types of toxicities, and several important centres have developed detailed plans to minimise their effects. Nonetheless, the specific finding of cerebral oedema posed very serious questions, and ultimately did for JCAR015, which until recently was Juno’s lead.

The issue also hit Kite’s KTE-C19, a construct very similar to JCAR015, and this group’s stock took a tumble last month when a patient died from cerebral oedema in a safety extension cohort of the pivotal Zuma-1 lymphoma study (Kite investors see an uncomfortable parallel with Juno, May 8, 2017).

Nine KTE-C19 study deaths not due to disease progression 
Indication  Study  Trial ID  Description of grade 5 event  Due to CAR?  Source 
Lymphoma  NCI phase I-IIa trial, group 1  NCT00924326  Influenza pneumonia 18 days after receiving CAR   No  Kite S-1, Dec 2014 
Lymphoma  NCI phase I-IIa trial, group 2  NCT00924326  Unknown cause 16 days after receiving CAR, presumed heart arrhythmia  No  Kite S-1, Dec 2014 
Lymphoma  Zuma-1, phase I  NCT02348216  Gr 4 encephalopathy & CRS, gr 5 intracranial hemorrhage & severe thrombocytopenia  No  Ash 2015 
Lymphoma  Zuma-1, phase II  NCT02348216  Hemophagocytic lymphohistiocytosis   Yes  Ash 2016 
Lymphoma  Zuma-1, phase II  NCT02348216  Cardiac arrest in setting of CRS, later specified as anoxic brain injury  Yes  Ash 2016, FR doc 2017-07800 
Lymphoma  Zuma-1, phase II  NCT02348216  Pulmonary embolism  No  Ash 2016 
Adult ALL  Zuma-3  NCT02614066  Hypotension, heart failure, hypoxemia, acidosis in dose expansion, organ failure & CRS  Yes  Ash 2016 
Paed ALL  Zuma-4  NCT02625480  Fungal infection  No  Ash 2016 
Lymphoma  Zuma-1, safety expansion cohort  NCT02348216  CRS, multi-organ failure, leading to cerebral oedema   Yes  Q1 2017 call 
Note: ~300 subjects have been dosed KTE-C19, including NCI studies. CRS=cytokine release syndrome. 

The data need to be seen in the context of several important caveats. Firstly, they comprise all disclosed CD19-directed CAR-T trial deaths that were not due to disease progression, whether these were deemed to be due specifically to the CAR-T cells or not – a subjective decision made by the specialist in question.

Early deaths on trials of KTE-C19 and JCAR015 were in the academic setting, and will have involved academic rather than commercial manufacturing. In evaluating the frequency of deaths the total number of patients treated must be borne in mind.

And of course it must be remembered that subjects on CAR-T studies are very ill, so are already at high risk of death. While this does not reduce the need for clinical rigour, several sources have commented to EP Vantage off the record that Juno might – scientifically if not reputationally – have been rash to discontinue JCAR015.

Extreme care must also be taken in extrapolating JCAR014 deaths to JCAR017; the deaths caused Juno to continue its ALL trial only with the lowest JCAR014 dose (2x105 cells/kg), and to scrap two higher levels. Juno has yet to choose a dose of JCAR017 to take into its pivotal lymphoma trial.

All that said, taking into consideration the caveats the data should speak for themselves.

After EP Vantage compiled the lists Kite and Juno cooperated fully in confirming the data as accurate. Novartis, however, did not respond to requests for clarification; as such, it is possible that there have been more deaths on CTL019 trials, and the list comprises only the seven that came to light in scientific presentations.

Seven CTL019 study deaths not due to disease progression 
Indication  Study  Trial ID  Description of grade 5 event  Due to CAR?  Source 
Lymphoma  Penn study (Stephen Schuster)  NCT02030834  Encephalitis  Yes  ASH 2015 
Adult ALL & lymphoma  Penn studies (Noelle Frey)  NCT02030847 & NCT01029366  CRS, influenza B  Yes  ASCO 2016 
Adult ALL & lymphoma  Penn studies (Noelle Frey)  NCT02030847 & NCT01029366  CRS, Pseudomonas infection  Yes  ASCO 2016 
Adult ALL & lymphoma  Penn studies (Noelle Frey)  NCT02030847 & NCT01029366  CRS, Strepotrophomonas infection  Yes  ASCO 2016 
Paed ALL  ELIANA  NCT02435849  Cerebral haemorrhage  Yes  ASH 2016 
Paed ALL  ELIANA  NCT02435849  Encephalitis  Poss  EHA 2017 
Paed ALL  ELIANA  NCT02435849  Fungal infection  Poss  EHA 2017 
Note: 283 subjects have been dosed to CTL019. 

Across CAR-T projects, the cerebral oedema deaths tally with those reported at December’s meeting of the US Recombinant DNA Advisory Committee, which additionally detailed two non-fatal cerebral oedema cases, one in a multiple myeloma study of an anti-BCMA CAR.

Several of the other neurotoxicity-related deaths disclosed – brain haemorrhage, for instance – sound uncomfortably close to cerebral oedema. But it should be stressed that after the JCAR015 incident most groups went back and reanalysed all previous toxicities to ascertain whether cerebral oedema might have been involved, and to rule out such a possibility.

Mechanistic mystery

Meanwhile, it is one of the many mysteries of CAR-T therapy that there is still no agreement on what precise mechanism might be responsible for causing cerebral oedema, though it is largely accepted that, being a cytokine-mediated event, it is something that arises from cytokine release syndrome.

Some doctors have told EP Vantage that this likely involves activated CAR-T cells crossing the blood-brain barrier, and then undergoing some kind of secondary activation event in the cerebrospinal fluid. However, many dismiss the simple answer that this results from expression of CD19 in the CNS.

An even deeper mystery is what aspect of CAR-T therapy might give rise to this kind of stimulation. Some have suggested the role of the co-stimulatory domain that each construct uses, others the manufacturing process, others still the indication studied, but given that all the CAR constructs differ in multiple ways, and that the numbers are still small, there is no way to be sure.

The FDA is separately compiling its own database of CAR-T toxicities. The provisos notwithstanding, investors should at the very least be aware of the numbers as they emerge.

This story was updated to include additional CTL019 deaths, and a baseline subject number from Juno.

To contact the writer of this story email Jacob Plieth in London at jacobp@epvantage.com or follow @JacobPlieth on Twitter

This content is written, edited and published by EP Vantage and is distributed by Evaluate Ltd. All queries regarding the content should be directed to: news@epvantage.com

EP Vantage is a unique, forward-looking, news analysis service tailored to the needs of pharma and finance professionals. EP Vantage focuses on the events that will define the future of companies, products and therapy areas, with detailed financial analysis of events in real-time, including regulatory decisions, product approvals, licensing deals, patent decisions, M&A.

Drawing on Evaluate, an industry-leading database of actual and forecast product sales and financials, EP Vantage gives readers the insight to make value-enhancing decisions.

EP Vantage SM ©2017 EP Vantage Ltd