Expert Financial Analysis and Reporting

Investment Perspective

I have been watching Kite and Juno for nearly two years and have occasionally written on the companies. I am almost through with a report on Kite that should be published soon. As people who follow me know, I am very much attracted to potentially paradigm changing technologies like CAR-T and other related, engineered autologous T cell technologies. There have been remarkable treatment results for CAR-T drugs in trials involving a small number of patients with refractory/ relapsed B-cell malignancies such as acute lymphoblastic leukemia (ALL) and diffuse large B cell lymphoma (DLBCL). CAR-T therapies have produced what appears to be close to cures in patients who had exhausted all other treatment options and often had life expectancies of a year or less.

However, there is another key aspect to any drug and that is safety. The side effect profile of the CAR-T cells is as troubling as efficacy is promising. We have seen severe, life threatening side effects related to cytokine release syndrome and neurological toxicities that have led to deaths. I have been astounded at how little attention has been paid to this drawback by analysts and investors and have just shaken my head in disbelief at the multi-billion valuations given to Juno and Kite. The side effect risk element is brought home by the announcement yesterday by Juno of three deaths in a trial they are conducting. Even though I have not been active in these stocks, I thought I would address this situation.

Clinical Hold in Juno’s ROCKET Trial in r/r ALL

Juno announced yesterday that there have been three deaths in the ROCKET trial in which its CAR-T product, JCAR-015, is being studied in adult patients with relapsed or refractory B-cell acute lymphoblastic leukemia (r/r ALL). As a result, the FDA has placed a clinical hold on the trial. ROCKET is an open label phase 2 trial. Juno management has guided that it expected that positive interim results in this trial would result in approval in 2017. Obviously, this is not going to happen.

The three treatment related deaths were due to severe neurotoxicity (cerebral edema). Two of these deaths occurred in the last week and one in May. The ROCKET treatment regimen starts with a chemotherapy pre-conditioning program, the purpose of which is to deplete the existing T-cell population in the body before the CAR-T cells are infused. This allows the CAR-T cells to expand much more rapidly.

The ROCKET trial was intended to enroll about 50 patients and to date about 20 have been enrolled. About two thirds of the 20 patients received a pre-conditioning regimen of cyclophosphamide and one third received a combination of cyclophosphamide and fludaribine. Earlier phase 1 studies have suggested that the addition of fludaribine leads to more rapid expansion and greater persistence of CAR-T cells which suggests enhanced efficacy. All of the three deaths occurred in the 6 or 7 patients who received fludaribine plus cyclophosphamide and no deaths occurred in the 13 to 14 patients given just cyclophosphamide. Based on this, management has strongly indicated that they believe the primary cause of the deaths is that the the flu + cy pre-conditioning program is too potent and that they can address the issue by just removing fludaribine from the pre-conditioning regimen.

Juno says that the FDA has requested that the Company take the following steps as part of the process to remove the clinical hold and resume enrollment in the trial:

1. A revised protocol in which fludaribine is removed from the pre-conditioning program,
2. A revised informed patient consent form,
3. Revised investigator brochure,

Juno management stated that they plan to submit the requested materials this week and anticipate an expedited response. The FDA has 30 days to respond and provide feedback. Management seems to indicate that this matter can be quickly resolved and allow the trial to continue, but I think that this is unlikely. It is almost a certainty that the FDA will want to do an in-depth analysis of the three patient deaths. After all, the current hypothesis is that fludaribine was the causative factor, but this is just a hypothesis. There are a large number of other issues to consider such as whether there is something unique about adult r/r ALL, patient characteristics and many other factors. This is likely to take much longer than 30 days.

More on the ROCKET and Other CAR-T Trials Sponsored by Juno

Juno management said that it continues to believe that the ROCKET trial can be completed without enlarging the trial size from the expected 50 patients. It cannot give guidance on when the trial will be restarted, but they seem to suggest that this will be relatively soon. They assume that 13 or 14 patients already treated using a cyclophosphamide only regimen will be included in this 50 patient enrollment.

The clinical hold affects only the JCAR-015 ROCKET trial and has no impact on other on-going or planned CD-19 directed CAR-T trials from Juno. The company is continuing trials of JCAR-014 and JCAR-017 trials in pediatric ALL, CLL and NHL patients. Importantly these trials also use the cy + flu preconditioning regimen.

Management has suggested that these severe side effects are more likely in adult r/r ALL. To support this hypothesis during the conference call, management provided a preliminary update on JCAR-017 in patients with DLBCL. Of 13 patients treated using the cy + flu preconditioning regimen, two patients (15%) had severe neurotoxicities and none had severe cytokine release syndrome.  Management points to this as being supportive of the hypothesis that there is more risk for adult r/r ALL patients than DLBCL patients.

However, Novartis reported results for 15 patient’s treated for DLBCL in December 2015. In 15 patients, four patients (27%) developed cytokine release syndrome (CRS) of grade 3 or higher. During CRS, patients typically experience varying degrees of flu-like symptoms with high fevers, nausea, muscle pain, and in some cases, low blood pressure and breathing difficulties. Neurologic toxicity occurred in two patients (13%), including one grade three episode of delirium and one death probably related to cerebral edema. This is a very small amount of data, but it suggests that severe neurotoxicity side effects occur in DLBCL as well as r/r ALL.

Reaction from Street Analysts

The consensus opinion from Street analysts in my judgment is that the issue is manageable and that Juno is taking reasonable steps to address this issue and allow the ROCKET trial to continue in the near future. They are suggesting that approval will be pushed back only slightly from 2017 to 2018. They also believe that the efficacy and safety experience thus far for JCAR-014 and JCAR-017 using the cy/flu regimen, support unimpeded continuation of those trial efforts. In short, the consensus view seems to be that this is just a bump in the road.

My Take

I am far less sanguine. The patient deaths strongly emphasize that while this is an extremely potent therapy that it is accompanied by very serious life threatening side effects. Drugs are ultimately judged on the balance of efficacy versus side effects. I think that there has been over-emphasis on efficacy while turning a blind eye to these dangerous side effects. The side effect risk may not be adequately incorporated in the stock prices even after the hit they took today.

I have no direct insight into whether the FDA will move expeditiously to lift the clinical hold on the ROCKET trial. However, it is my experience that managements almost always underestimate how long it takes the FDA to assess issues. I think that with 3 of 20 patients enrolled in ROCKET having died that this will take much, much longer than 30 days to resolve. I have no way of knowing, but I would be very surprised if this is resolved in 6 months and not terribly surprised if it took a year. I emphasize that this is an intuitive, not a fact based judgment.

Novartis, Juno and Kite are all rushing to get CAR-T products directed against B-cell malignancies approved. Novartis seems to have the clear clinical lead and plans to submit a BLA for pediatric r/r ALL early in 2017. Kite has said that it expects approval on KTE-C19 in DLBCL in early 2017 and Juno had said that it expected approval of JCAR-015 in adult r/r ALL in 2017. Products from all three of these products have the same mechanism of action and it is probable that they will be somewhat comparable products with comparable therapeutic profiles in all forms of B-cell malignancies, but at this early date there is no way of knowing.

In My judgment, first mover advantage will be extremely critical to success. Juno and Kite are already at a considerable disadvantage to Novartis with its huge commercial and financial resources and Novartis also may have first mover advantage as well. It could be the case that even if all three products were approved at the same time, Novartis would likely be the clear winner. I would caution that this may be an overly bold statement since we haven’t seen the results for any of the key phase 2 trials they are pending their hopes on for registration. For Juno to fall a year behind Novartis is a very significant negative.