Asco – Juno throws down another lymphoma marker for Novartis

As cell therapy-focused investors await the announcement of Novartis’s CAR-T data in the lymphoma study Juliet, at Asco Juno put down another marker in this key indication. The battle is important because lymphoma represents a much bigger space than childhood ALL, for which Novartis recently filed CTL019.

The good news for Juno is that the data in the Transcend study of what is now its leading CAR-T hope, JCAR017, look comparable to that generated by Kite’s KTE-C19 in the pivotal Zuma-1 trial. The data are, of course, important for followers of Kite, whose KTE-C19 filing in lymphoma has just been accepted by the US FDA.

Thus, with all the obvious across-study caveats, the most immediate comparison for Transcend will be made with Kite’s Zuma-1 study. At the last data cut Zuma-1 generated a 41% six-month overall remission rate (ORR), with 36% of patients in complete response (CR; Six-month data allay the worst fears for Kite, February 28, 2017).

The Transcend data are less mature, and involve a different mix of patients. Kite had made much of the fact that Zuma-1 had enrolled a high proportion of extremely aggressive non-Hodgkin’s lymphoma subjects, and that only some 20% had had prior transplant; in Transcend significantly more patients were post-transplant.

At a comparison of the Kite and Juno datasets at Asco today, discussant Dr Andrew Evens, of Tufts Medical Center, said Transcend represented a more “real-world” population.

With those provisos, Juno’s efficacy data look good: 76% of 54 evaluable patients across three dose levels went into remission initially, with the three-month ORR and CR rates running at 51% and 39% respectively.

Across-trial comparison of KTE-C19 and JCAR017 lymphoma studies
Zuma-1 (KTE-C19, Kite) Transcend (JCAR017, Juno)
Total patients ORR CR Total patients ORR CR
Best response 101 82% 54% 54 76% 52%
3 months  62 44% 39% 41 51% 39%
6 months 101 41% 36% NA NA NA
Treatment-related deaths Hemophagocytic lymphohistiocytosis Diffuse alveolar damage
Anoxic brain injury
Cerebral oedema (in safety expansion cohort)

There was one death not due to disease progression – the first in a trial of JCAR017. This was due to diffuse alveolar damage in an elderly patient who had refused mechanical ventilation for progressive respiratory failure.

In Kite’s Zuma-1 trial there were five deaths not due to disease progression, but only three were deemed to be treatment-related, including – most worryingly for Kite investors – a death from cerebral oedema in a subsequent safety expansion cohort. In a broad biotech selloff today, Juno stock fell 10%.

Juno confirmed that Transcend largely used a commercial manufacturing process. Of 88 patients in whom apheresis was carried out, JCAR017 was successfully manufactured in 86, though in eight of these it failed to meet specifications, Juno said.

The group also highlighted the fact that it saw suggestions of improving efficacy with higher dosing. Transcend was not a registrational trial; it was primarily a dose-finding test, on the basis of which a pivotal study is to be initiated later this year.

It is an important test for JCAR017, a construct based on work at Seattle Children’s Hospital that uses a defined cell composition to improve uniformity during expansion. It became Juno’s lead when JCAR015, a construct very similar to Kite’s KTE-C19, was discontinued after five patients died of cerebral oedema in a trial in adult ALL.

Awaiting Juliet

Both Kite and Juno’s datasets are important to Novartis, whose CTL019 was filed before KTE-C19, but for the smaller indication of paediatric ALL. The US FDA action date on KTE-C19 is November 29, while that for CTL019 is thought to be September 29.

Little is known about CTL019’s activity in lymphoma, for which results from the pivotal Juliet trial are to be presented at the International Conference on Malignant Lymphoma in Lugano, Switzerland, on June 14.

For now Novartis followers can content themselves with very promising – but also very early – Asco data from a related Novartis construct, CTL119, in chronic lymphoblastic leukaemia. CLL is related to lymphoma in the types of cells affected, but it occurs largely in the blood rather than the lymph nodes.

Those data are striking because CLL has tended to be a tougher indication for CAR-T; here, patients who had failed to go into remission after taking Abbvie/Johnson & Johnson’s Imbruvica second line were given CTL119 on top of Imbruvica, and eight of nine went negative for minimum residual disease (MRD) within three months.

MRD-negative status is a measure of stringent remission in the bone marrow, though radiologic responses in the spleen and lymph nodes are less clear-cut, and might require longer follow-up; for now it looked like CR rate was running at around 50%, said Dana-Farber’s Dr Jennifer Brown, discussing the data today.

CTL119 aims to prevent rejection by the immune system by using a humanised rather than mouse-derived binding domain. That said, these refractory CLL patients were likely immunocompromised, so rejection was unlikely an issue; it is this not clear why CTL119 and not CTL019 had been used.

It has separately been suggested that Imbruvica can help expansion of the CAR-T cells. Dr Brown said this trial offered hope for treating high-risk CLL patients, such as those with 17p deletion.

Investor focus now turns to Novartis’s June 14 Juliet readout. Juno remains behind both of its rivals, but – crucially – Transcend data show that it is by no means out of the game. 

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

Share This Article