AACR 2014

Study shows benefits in use of AG-221 in advanced blood cancers

Dr Eytan Stein - Memorial Sloan Kettering Cancer Center, New York, USA

What I’m going to be talking about now is the clinical safety and activity in a phase I trial of AG-221 which is a first in class potent inhibitor of the IDH2 mutant protein in patients with IDH2 mutant positive advanced haematological malignancies. These are my disclosures.

I want to give you a little bit of background because this is the first time an agent like this has been in the clinic. So, a number of years ago, when screening patients with acute myeloid leukaemia, it was noticed that patients have mutations in genes for IDH1 and IDH2. The normal function of IDH, which is isocitrate dehydrogenase, is to convert isocitrate to alpha-ketoglutarate in the Krebs cycle to produce energy for cellular metabolism. What was noted is that this mutant IDH2 protein, the mutant IDH2 enzyme, instead of converting isocitrate to alpha-ketoglutarate, what ended up happening is that it actually converts alpha-ketoglutarate to a substance called 2HG or β-hydroxyglutarate. What’s important to note about that is that in normal haematopoiesis what happens is that you’ve got a haematopoietic stem cell that gives rise to a common myeloid progenitor. That common myeloid progenitor matures and becomes a neutrophil which helps your body fight infection. In acute leukaemia what happens is that maturation process stops, typically at the myeloblast stage of development and the thought was that through a lot of preclinical investigation that these elevated levels of 2HG, or β-hydroxyglutarate, caused epigenetic reprogramming of the cell such that the differentiation was blocked at the myeloblast stage of development, hence producing leukaemia. The hypothesis was that if you could block the mutant enzyme, thereby decreasing the levels of 2HG, you could therefore allow the myeloblast then to continue to mature into a mature neutrophil and essentially reverse the leukaemic phenotype.

So just a little bit of background on how common this mutation is. If you look at the top part of the panel up here you can see that IDH2 mutations are found in about 10-15% of acute myeloid leukaemia, about 5% of myelodysplastic syndromes and myeloproliferative neoplasms and about 25% of a rare kind of lymphoma called angioimmunoblastic non-Hodgkin’s lymphomas. IDH1 mutations, which we’re not going to talk about today, tend to predominate in solid tumours, in low grade gliomas they’re about 70% of low grade gliomas, about 50% of chondrosarcomas but they’re also found in AML and MDS and MPN.

So based on a lot of preclinical work that I’m going to be presenting later at the bigger session we initiated a single arm phase I open label study of AG-221 in continuous oral daily dosing in 28 day cycles. The dose escalation started at 30mg twice a day. The patient population were relapsed or refractory AML and MDS or patients older than age 60 who were unable to be treated with conventional therapy because of comorbid medical conditions. They had to be IDH2 mutant positive to get into the trial and standard modified three by three dose escalation design was used. The key objectives were to assess the safety and tolerability, determine the MTD and the recommended phase II dose, determine any dose limiting toxicities, PK and PD profiling of the patients who were treated, and to characterise if there’s any differentiation effect as I suggested earlier, because that’s what is hypothesised to happen, and any clinical activity.

So this is the study status; the study was initiated in September 2013. As of March 20th 2014 which is the data cut-off date there were 22 patients enrolled and 16 patients remain on study. After the first three dosing cohorts, you can see there was 30 twice a day, 50 twice a day and 75mg twice a day, we added in a fourth dosing cohort of 100mg once a day, the reason being that we noted that the pharmacokinetic profile of the drug was such that it might be amenable to once a day dosing. Currently there are in parallel twice a day and once a day dose escalation that is ongoing.

In terms of safety there are 22 patients who are evaluable for safety, there have been no dose limiting toxicities. AG-221, there have been two possibly related serious adverse events – grade 2 hyperleukocytosis and differentiation syndrome, which clinically actually might not be bad, and grade 3 confusion in the setting of respiratory failure in a patient with septic shock. There have been 4 deaths with 30 days of study drug termination, all 4 complications of disease-related sepsis, 3 in cohort 1, 1 in cohort 4. This is unfortunately not unusual for patients with relapsed refractory acute myeloid leukaemia who are often very, very, very ill.

So, consistent with the preclinical hypothesis, after the administration of AG-221 you can see on the left hand side you can see that there was a high exposure of AG-221 in plasma at cycle 1 day 15 and cycle 2 day 1. You can see on the right hand side that the amount of 2HG, which is the substance that is predicted to cause this epigenetic rewiring, was knocked down or inhibited by greater than 90%.

So these are the responses that we’ve seen so far. In the 30mg twice a day cohort, of the patients who were evaluable you can see that there was one complete remission and one complete remission with an incomplete platelet count recovery. In the 50mg twice a day cohort there were two complete remissions, one complete remission with an incomplete platelet count recovery and one partial remission, although a partial remission with normalisation of the patient’s neutrophils, and one patient with progressive disease. So overall what this shows is that out of the seven patients who were evaluable for efficacy, five of the patients achieved a complete remission or a complete remission with incomplete platelet count recovery. What that means for those of you, because I’ve gotten questions about this before, what that means is that they’ve cleared the leukaemic blasts from their bone marrow, there is no leukaemia in the bone marrow, their platelet count just has not yet risen to above 100,000. So the overall response rate was six out of seven patients and the CRs and CRPs, which is the incomplete platelet count recovery, was five out of seven patients.

This is a summary of the responses. At the later session I’m going to walk everyone through this but I think what I want you to take home from this is that if you look at cohort 1 on the top, the time for the patients who responded, the time to response, was about late cycle 3 or early cycle 4 while with an increase in dose you can see that the response has happened much quicker with those four patients who had some sort of response responding at the latest by cycle 3.

So conclusions of this early data is that AG-221 is safe and well tolerated to date; β-hydroxyglutarate is inhibited at greater than 90% of baseline in patients with R140Q mutations and approximately 50% inhibition in the one evaluable patient who has got an R172K mutation. Consistent with the preclinical models of IDH2 inhibition these were a profound differentiation effect; I’m going to show that in much greater detail at the session later today. Five out of seven patients who completed one cycle achieved a complete remission or a complete remission with incomplete platelet count recovery. Dose escalation continues; the expansion cohorts are going to begin in late 2014. These data provide early validation of mutant IDH2 as a therapeutic target in AML and MDS.

So in terms of what we’re going to be doing, we’re going to be further characterising the safety, the PK/PD the response rate in AML. I told you already we’re dose escalating on two schedules, a BID dose schedule and a Q day dose schedule. There are going to be these planned expansion cohorts to get a little bit more experience with the agent. We’re hoping down the road to evaluate this in combinations and in earlier lines of treatment, so not just treating relapsed and refractory heavily pre-treated patients because, as you saw, there were a number of patients who unfortunately passed away before they might have gained clinical benefit from the drug. We’re also going to be exploring the activity in other IDH2 mutant hematologic malignancies and IDH2 mutant solid tumours.

There is a parallel phase I study of the IDH1 mutant inhibitor, AG-120, that has been initiated in IDH1 mutant solid tumours and hematologic malignancies. I just want to also thank all the patients who agreed to take part on the study. It’s not easy for a patient to decide they’re going to go on a phase I first in man study, it is nerve-wracking for them, so I really want to thank them for participating and I’m happy to take any questions.