Follow Us GraphicFacebook IconTwitter IconLinkedIn Icon
Search Graphic

Expert Financial Analysis and Reporting

Novartis Releases Three Month Interim Data for JULIET Trial of CTL-019 in r/r DLBCL

Key Points:

  • Comparison of three month data for CTL-019 in the JULIET trial with  Axi-Cel in ZUMA-1 indicates that safety and efficacy are remarkably comparable for the two products. At this point, there seems to be no differentiation between the two.
  • Some analysts are strong propoents of Kite raised the issue that Novartis had manufacturing issues in JULIET that could give Kite a significant marketing advantage. Novartis quickly issued a press release that debunked this argument.
  • Some analysts are also suggesting that Axi-Cell will come to market nine months before CTL-019. I continue to feel that there is a good chance that they will be approved at the same time. There is some chance that Axi-Cell could be approved a few weeks or months before CTL-019.
  • The spin war from aggressive biotechnology analysts supporting Kite has begun in earnest. Novartis is covered by pharma analysts who are more sedate.
  • My primary investment interest in the CAR-T space is Cryoport which is a major beneficiary of the frenetic development activity.

Comparison of Data for JULIET and ZUMA-1

Investors have seen a significant amount of data from the ZUMA-1 phase 2 trial of Kite’s CAR-T product Axi-Cel in the treatment of r/r DLBCL. They have been eagerly awaiting data from the phase 2 trial of Novartis’ comparable product CTL-019 in the JULIET trial in the same cancer. Novartis announced last week that it will present data on CTL-019 on June 14 at the International Conference on Malignant Lymphoma (ICML).  Yesterday, an abstract from the conference was published with three month results from JULIET. These can be compared to three month results from ZUMA-1 that were released in September 2016. The following table compares a number of key data points from each trial. The results indicate that the products have a very similar efficacy and side effect profile.


Comparison of Three Month Interim Data
CTL-019 in JULIET Axi-Cel in Zuma-1
Number of Patients 51 51
Best Response at Any Time (%)
 ORR 59 76
 CR 43 47
 PR 16 29
Response at 3 months (%)
 ORR 45 39
 CR 37 33
 PR 8 6
CRS Syndrome (%)
 Any grade 57
 Grade 3 17
 Grade 4 9
 Grade 3/4 26 18
CNS Side Effects (%)
 Grade 3/4 13 34
Cerebral Edema 0 0
Deaths Due to Adverse Events 0 2



Questions Were Raised About Novartis Having Manufacturing Issue

Following the release of this data, some analyst supporters of Kite suggested that Novartis had meaningful manufacturing issues. They noted that the global JULIET trial enrolled 141 patients in 27 centers, but only 85 received CTL-019 cells. The interpretation was that on an intent to treat basis JULIET treated 60% of patients and ZUMA-1 treated 92%. The abstract did not explain this nor did it mention manufacturing turnaround times. These analysts went on to say that they believed that this was in part because Novartis had a much slower manufacturing turnaround time which resulted in patient withdrawals or deaths before they could be treated.

Novartis Responded to this Issue in a Press Release

Novartis indicated that, of the 43 JULIET patients who discontinued before infusion, the majority did so due to rapid progression of their disease or deterioration in their clinical status. Only 6% (9 of 141) of enrolled patients discontinued due to inability to manufacture an adequate dose of CART cells. Novartis also noted that during the JULIET trial, an ongoing number of process improvements resulted in a manufacturing success rate of 97% for the last 30 patients. This effectively debunks the negative spin from some analysts.

Novartis also said that it uses cryopreserved leukapheresis. This affords physicians the flexibility to schedule apheresis at a time that is best for their patients. This means that leukapheresis can happen at variable times in the study, even before enrollment. They do not believe that measuring “vein to vein” time is the most appropriate measure of the time required to manufacture and reinfuse the cells.

The actual cell processing time is about 12 days. In JULIET trial, the time from start of manufacture to finished product shipment was 29 days which in addition to cell manufacturing includes quality control. Novartis emphasized that they have a rigorous quality and release testing process to ensure that CTL-019 is GMP compliant.  They anticipate that the time from the start of manufacturing to product release will be 22 days at time of commercial launch.

NVS is currently scaling up from a few 1,000 patients to tens of thousands of patients. Initially, about 30 centers will be qualified to administer CTL019.

When Will Axi-Cel and CTL-019 Be Approved?

In my report of June 7, I laid out three scenarios:

  • The FDA holds an AdCom for Axi-Cel and CTL-019 on the same day and goes on to approve both products at the same time. I think that the FDA will want to compare the data from these two products side by side. This is my most likely scenario.
  • The FDA holds an AdCom for Axi-Cel but holds one later for CTL-019. In this event, Axi-Cel could come to market a few months before CTL-019 in this indication.
  • The FDA doesn’t require an AdCom meeting for CTL-019.In this event, CTL-019 could come to market at the same time as Axi-Cel or a few months later.

Some of the same analysts who spuriously raised the manufacturing issue with Novartis also claim that Axi-Cel will come to market for r/r DLBCL nine months before CTL-019. The reason for this was not cited. The only thing I am aware of is that the three month data on Axi-Cel was presented in September 2016 which is about nine months before the release of three month data for CTL-019. I assume that this is how they came up with the nine month lead. If so, his argument makes little sense to me. The ZUMA-1 trial started 3 months and 13 days before JULIET and had a similar design and enrollment period.

Another very important thing to consider is that the AdCom meeting for CTL-019 in the treatment of r/r pediatric ALL will take place on July 12. The manufacturing and quality assurance issues are the same for this indication and r/r DLBCL and some safety issues are the same. These issues usually take up a good part of the review time for a BLA and may not have to be repeated in considering the BLA for CTL-019 in r/r DLBCL. If so, the FDA will only have to focus on the clinical data which one would suspect might speed up the review process.




Tagged as , , , , , + Categorized as LinkedIn, Smith On Stocks Blog


  1. Thomas Cunningham says:

    You want to report on a winner which will turn cancer research on a

    different axis report on Calithera CALA


You must be logged in, or you must subscribe to post a comment.