Good morning and I’d like to thank the organisers for giving us an opportunity to present a summary of our data that will be presented on Monday morning by Nirali Shah. What I’ll be presenting are results of a phase I clinical trial using CD22 targeted chimeric antigen receptor T-cells in children and young adults with refractory ALL. Most people in the room are familiar with the success with the CD19 chimeric antigen receptor expressing T-cells in B-ALL with complete remission rates approaching 80-90% in treated patients now in phase II clinical trials. We’re learning now that one of the limitations of this approach is loss of CD19 expression occurring in potentially a substantial number of patients although that has not been systematically analysed. CD22 is a well-defined alternative target in B-ALL and there are existing monoclonal antibody toxin conjugates that have shown efficacy in B-ALL. So, based on this, a number of years ago within the NCI we developed a CD22 targeted chimeric antigen receptor. I don’t have time to talk about that in detail but suffice it to say that although there are a number of binders that seem active in the CD19 CAR space it was very difficult to come up with an active antibody-based binding domain that was functional as a chimeric antigen receptor and so we cycled through a lot of binders that were inactive before we finally arrived at M971 which was active as a chimeric antigen receptor.

So we took this into the clinic and opened a phase I dose escalation trial using these anti-CD22 CAR T-cells. A fairly standard relapsed refractory ALL population, obviously expression of CD22 required, age is shown there. We did allow CNS disease but CNS3 was exclusionary. All patients received the same preparative regimen shown there; the dose escalation is shown below. I’ll point out that dose level 2, 1x106, is the dose that has been used in most of the CD19 CAR trials, we started half a log lower than that as part of the safety in the trial. These are the characteristics of the patients treated thus far in the dose escalation in the initial experience and the expansion cohort. I’ll point out a couple of things: all of these patients have received a prior allogeneic transplant, 11 of these patients have received CD19 CAR and 9 of these patients were CD19 negative or had reduced expression of CD19 on the surface of their ALL. Most of these patients were high burden disease patients, 12 of them greater than or equal to an M2 marrow at enrolment. We had 3 patients with extramedullary disease.

So this is the experience from the standpoint of toxicity. We expanded dose level 1 due to a grade 3 diarrhoea that was self-limiting. We then escalated to 1x106, treated 3 patients with no DLT. As per protocol design we then escalated to 3x106, we had one patient who developed grade 4 hypoxia necessitating early initiation of steroids with rapid reversal. Although we could have continued with that dose level, based on the response in dose level 2 and the one DLT we made a decision to go back down to 1x106 and we’ve treated additional patients at that dose level. We have had one grade 5 event due to sepsis that occurred after resolution of CRS.

We’ve seen an 80% complete remission rate at dose level 2 or greater with a maximum grade of cytokine release syndrome of 2. This is not based on CTC grading, this is based on publication from blood that includes all of the clinical aspects of cytokine release syndrome. This is the waterfall plot for the patients treated thus far. Dose level 2 and 3 are in green and blue and so, again, a very high remission induction rate once we achieved the biologically active dose. The hash bars represent the development of cytokine release syndrome.

We have seen sustained remission, we have three ongoing remissions, one patient at greater than a year, another patient at six months and another patient at three months. This obviously means that we have seen relapses and, of interest, these relapses have been associated, at least in the four patients treated at the biologically active dose, with changes in expression of CD22. It appears to be different than with CD19 - where CD19 is lost completely in these patients we saw largely a decrease in site density that we think drops the CD22 expression below the threshold for CAR activity. We have seen one patient with CD19 negative who completely lost CD22 expression on her blasts.

So, in conclusion, I think this is the first successful salvage CAR therapy for CD19 negative B-ALL. The preliminary experience certainly suggests that the potency is at least comparable to the phase I experience with CD19 at the same dose level. There is a suggestion that response tended to correlate with dose level. We did see CRS with the maximum grade of 2. This evening there will be a poster presentation by Haneen Shalabi on the experience with the neurotoxicity. When we wrote this trial we included comprehensive evaluation for neurotoxicity; we did not see any severe or irreversible neurotoxicity thus far. We’ve seen that relapses associated with the changes in CD22 expression and at least our bias is that this may not be best used as a salvage therapy but we’re beginning to think about how this could be potentially included with CD19 in the upfront CAR treatment.

The acknowledgements shown here, this trial was fully developed within the NCI and as the IND sponsor is the Centre for Cancer Research we did have lentiviral vector production produced by Lentigen as part of a CRADA. After trial initiation Juno licensed this CD22 CAR construct; we have an ongoing CRADA that supports data transfer and qualitative assays but it is still an NCI sponsored trial.