Syndax Pharmaceuticals, Inc. (SNDX) Earnings Call Transcript & Summary

Good afternoon and welcome to the Morgan Stanley 18th Annual Global Healthcare Conference. I'm one of the biotech analysts here. My name is David Lebowitz. And before I get going, I have to go through the requisite disclosures. Please note that this webcast is for Morgan Stanley clients and appropriate Morgan Stanley employees only. Webcast is not for members of the press. If you are a member of the press, please disconnect and reach out separately. For important disclosures, please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. And with that, I'd like to introduce. From Syndax Pharmaceuticals, I have CEO, Briggs Morrison; President and Chief Operating Officer, Michael Metzger; and CFO, Daphne Karydas. And I guess before we really dig in here, if you could kind of give a top-level description of the company, the evolution. And I would say, there's certainly been quite a bit of evolution in the company over the last year kind of to bring us up to speed with where things are today.

Sure. I'd be happy to, David. Thanks so much for the opportunity. Syndax was founded more than 10 years ago based on a business model of trying to license in molecules that we think could have potential for application in the treatment of patients with cancer. The first drug that was licensed in was a drug called entinostat, which did not succeed in the Phase III trial earlier this year and is really no longer a focus of the company. The company then licensed in 2 additional molecules. One is a CSF-1R antibody that we are studying for the treatment of chronic graft-versus-host disease, and we hope to have Phase I data from that program at Scientific Congress at the end of the year and begin potentially a registration Phase II study in the early part of next year. And then our third molecule that we licensed in is a small molecule inhibitor of the interaction of menin with MLL1 for -- used for mixed lineage leukemias and for NPM1 leukemias. That molecule is also in Phase I. We had some early data from that program at AACR earlier this year, and we intend to present the data on the completed Phase I trial early in 2021.

And -- thank you for that. I guess we'll start there with the third molecule, the Menin-MLL-r, SNDX-5613. I guess could you tell us about this molecule, the specific target and why it's attractive place to go in hematology?

Sure. So the so-called mixed lineage leukemias have been known for many, many years. These are leukemias that can present as either acute myelogenous leukemia, AML, or as acute lymphoblastic leukemia, ALL. The thing that characterizes all of these mixed lineage leukemias is a recurrent cytogenetic abnormality that leads to the production of a fusion protein. This fusion protein is the pathognomonic of MLL leukemias. The fusion protein is made up of 2 pieces, if you will. On the left-hand side or the amino-terminus comes from a protein called MLL1, and the right-hand side, or the carboxy-terminus, comes from a variety of different donor proteins, all of which appear to be in a transcription factor family. That rearranged protein was described probably 30 years ago. The thing that made this rearrangement druggable was when people figured out that in order for this fusion protein to transform cells, the amino-terminus has to bind to a protein called menin. Crystal structure of menin has been solved, and it's very clear exactly how MLL1 binds to menin, and it turns out there is a what we refer to as a druggable pocket in menin that you can make small molecule inhibitors that will bind to menin and displace the MLL fusion protein and thereby lead to a loss of leukemic activity by the fusion protein. And we and others have presented nonclinical data or preclinical data to show that these small molecule inhibitors, when they bind to menin and displace the fusion protein, lead to antileukemic effects in the animal models of human leukemia. There's a second form of leukemia known as NPM1 AML, represents about 1/3 of all AML patients. And although the mechanism is not quite worked out in as excruciating detail as it is for the fusion protein, it's quite clear that the NPM1 leukemias also are dependent upon this menin MLL interaction, and again, we and others have published data on small molecule inhibitors that block the menin-MLL interaction and have antileukemic effects in NPM1 leukemia. So that's why we've been excited about testing our molecule in both MLL-r leukemias and NPM1 leukemias.

Excellent. About how many -- I guess who else is playing in this space? I know you've been in here a couple of years since you got in through business development activities. What other companies are out there? I guess how would you compare your molecule to theirs?

So the one other company that is in the clinic and, I would say, roughly, in the same stage of development is a company called Kura. I can't really say very much about how our molecule compares to their molecule simply because there hasn't been a scientific presentation of the characteristics of their development candidate. They, like us, have published a preclinical data on molecules that are related to their development candidates, but the actual development candidate is -- the scientific data on that has not been presented. So I can't really compare. I can say that -- but those are the 2 that appear to be, if you will, roughly, in the same stage of development, both in Phase I with molecules that block the menin-MLL interaction.

So you recently presented some data at AACR earlier this year. Could you run us through what was presented and why people seem to find it quite exciting?

Sure. So the session that we presented at was a session called New Drugs on the Horizon. It's a special session at AACR, where the organizers of AACR reach out to companies that have novel mechanisms that they think might be of interest to the AACR community. And so they asked us if we wanted to present on our menin-MLL story, the 5613 story. And so our base assumption -- we said, yes, of course, we'd be very excited to. Our base assumption is that we would present the preclinical characterization of our development candidate. So just as I said earlier, we had published papers on a related compound, but it wasn't actually our development candidate. So the idea was we would present the full picture of our development candidate. It turned out that our Phase I trial had progressed to an interesting point. So we presented some Phase I data as well. The preclinical data really showed that the drug works in MLL-r and NPM1 leukemias. And then we did some additional -- additional work was presented that showed how we tried to figure out roughly what kind of PK exposures we would need to see in patients in order to see efficacy. And that was a preclinical model where we gave varying doses of our drug characterizing the PK exposure and trying to figure out what's the exposure that led to efficacy in the animal models. So based upon that, we started to describe our clinical program, and our clinical program is a relatively straightforward Phase I dose escalation trial, in which patients with any refractory leukemia at the time -- the trial was open to any patient who had leukemia refractory to any existing standard therapy. So they didn't have to have MLL-r or NPM1 mutations. It could be anybody. And what we presented were 6 patients, the first 6 patients treated on the trial. The very first patient was the patient who did not have one of these mutations, did not end up with drug exposures that, we think, based upon the preclinical data we needed for efficacy, did not have any dose-limiting toxicities and went off the trial because their leukemia progressed. The second patient put on the trial was a patient who did have an MLL rearrangement. It was a woman with MLL-r leukemia that was refractory to chemotherapy and another targeted therapy. This patient actually had exposures that exceeded those that we thought we needed for efficacy based upon our preclinical models and, in fact, had a complete response with incomplete recovery of the rest of her blood counts at the very first evaluation at day 28. The patient then went on to improve to a complete response. So I think that was the patient that was probably the most exciting to people to see that this mechanism, which I've described to you, and this fusion protein has been known for a very long time, people have just worked out this mechanism and to show in people that the mechanism could actually give therapeutic benefit to a patient, I think, was exciting to us, clearly exciting to the patient and exciting to the scientific community. In total, there were 6 patients that we presented. Three of them did not have one of the 2 mutations and so we did not expect them to respond. The other 3 did have MLL-r rearrangements: one, I just described to you, the patient who had a complete response; one, who at their first evaluation on day 28 had a partial response; and one, who did not respond, but we believe probably had levels that were not sufficient to generate the efficacy. So I think that was the first demonstration that this mechanism can work in people, and I think that's what's gotten people excited about the program.

Now you were talking about reaching certain levels to have clinical benefit. And one of the factors that you anticipated going in that could affect and does seem to impact is the use of CYP3A4 inhibitors concurrently. And I guess what did you see in your data with respect to that?

So -- right. A standard workup for a molecule before you go into the clinic is to try to characterize its metabolism and try to understand how the drug is metabolized. And we knew going into the clinic that our drug, in terms of hepatic metabolism, is primarily metabolized by CYP3A4. So we anticipated that if a patient was on a drug that inhibits the enzymatic activity of CYP3A4, then the drug levels from our drug should be higher because its metabolism has slowed. So I described to you that very first patient, who did not have a mutation, did not have very good exposures and went on to progress; the second patient had a doubling of the dose, received a dose twice what the first patient received, but their exposures were probably somewhere between 7 and 8x as high. And that patient was on a CYP3A4 inhibitor, which led us to the hypothesis that probably that patient had very high exposures, very adequate exposures for efficacy because, number one, the dose was doubled, but then the exposure was increased by somewhere between three and fourfold by virtue of the fact that they were on a drug that inhibits CYP3A4. It's not uncommon, as you and many of our investors know, for drugs to be metabolized by CYP3A4. And it's not uncommon that you have to adjust the dose of your drug if the patient happens to be on a strong inhibitor of CYP3A4. That's what we've seen. And so we've split the trial into 2 arms, one where we characterize the drug given in combination with a strong CYP3A4 inhibitor, and one arm where they're not on an inhibitor, so we can fully characterize this drug interaction.

With that in mind, I guess, when you move into the -- you're moving into the next phase of the trial, and the agency saw this data and also agreed with you to allow the protocol to be changed, you're no longer an all-comers trial, it's now more specific, how has the study progressed at this point? How is it different now, I guess, than it was before? And what does this mean for what we'll see when additional data is presented?

Yes. So as you say, we've been very fortunate to be able to work very closely with our reviewing division at FDA. They've seen the data as it's evolved. They agreed to the splitting of the trial into what we call arm A and arm B: arm A, where they're not on a CYP3A4 inhibitor; arm B, where they're on a strong inhibitor. We've also, earlier this year, presented at a pediatric oncology drug advisory committee. The goal of that committee was to try to think about how can you get some of these agents into pediatric development faster. We do know that the MLL-r, particularly ALL, is seen in infants who have ALL and is a particularly bad prognosis. So the pediatricians have been quite excited to see if we can test this mechanism in children. So we had presented some of that data at the pediatric advisory committee, and after that, an agreement with FDA that we could put pediatric patients onto our existing trial down to the age of 1 month. The second change that occurred was that, as you noted, the initial trial was an all-comers trial, any refractory acute leukemic, whether AML or ALL, having NPM1 mutations, MLL mutations or anything else. We went back to the agency and said, to be honest, as the trial has progressed, we think we're in a therapeutic exposure. Obviously, patient #2 benefited, so that's a dose that's probably therapeutic and -- or in that patient was therapeutic. So can we now focus the trial really on the patients who are most likely to benefit, which are the MLL-rs and NPM1s? They agreed to that. And the third change that we made recently was -- and again, this is not that uncommon in Phase I oncology trials. If you have a dose level where you have cleared that dose level from a safety point of view and you've seen some efficacy, when you go to the next highest dose level as part of the dose escalation scheme, you can backfill, if you will, that cohort that looks like it has efficacy and has been deemed safe by the monitoring committee. So that backfill has also been allowed now. We're working our way through all the administrative process to open up these amendments. What I think that says in terms of what you might see when we present the Phase I, perhaps there'll be a larger sample size because of these backfills. I think because of the backfills, it also enhances our ability to get NPM1 patients on as well as MLL-r patients and potentially some additional pediatric patients. We have been treating pediatric patients through the compassionate use process, the single patient IND, and we've learned quite a bit about the PK of our drug in children. But now they can formally go onto the Phase I trial so we may have some pediatric patients to present at the end of Phase I as well.

Now I guess -- so we should probably expect to see patients in the next update from different -- the different groups, the ALL, MLL-r, ALL -- the AML, NPM1 and ALL, MLL-r and some pediatric patients?

Yes. I think, again, just to set expectations, what's happened as a trial has evolved, because of patient #2, who had a complete response and had an MLL-r rearrangement, there is some enthusiasm by the investigators, obviously, to put patients on the trial who have an MLL-r rearrangement. Those patients have a bad prognosis. If they're refractory patients that don't have other available therapies, knowing that there's some signal that our drug works in that population, there's enthusiasm to put those patients on them. So I do think when the trial reads out, there'll potentially be some enrichment for the MLL-r patients, and yet I do believe there will also be some NPM1s and potentially some pediatric patients as well.

Excellent. To put this in perspective, how are these patients currently treated? And what represents an advancement over current therapy?

So the current treatment for patients with acute leukemia is generally divided into those patients who are candidates for intensive chemotherapy, in which case their normal treatment would be to get intensive chemotherapy to induce a remission, additional chemotherapy to try to consolidate that remission, and if there are candidates, they can go on to bone marrow transplant. There's another group of patients who have been referred to as the unfit patients. These are patients who, either due to age or comorbidities, are not really candidates for induction chemotherapy. And in those patients, more and more we're seeing them treated with -- the standard of care is evolving to be a combination of venetoclax plus azacitidine. Once patients have received either induction chemotherapy and bone marrow transplant or venetoclax and azacitidine and they become refractory, there really is no standard of care for those patients. They generally go either on clinical trials or to some salvage chemotherapy, none of which provide very much benefit, unfortunately, to these patients. So I think the bar in terms of a signal of efficacy is not that high because it really is nothing for treating these patients. In the NPM1 population, same exact treatment paradigm, once they become refractory, again, there's not that much you can do. The NPM1 patients, a subset of them will also have an additional mutation besides NPM1, either in FLT3 or IDH1 or 2, as an example, those patients would likely get treated with a targeted agent against the mutation that they have. But again, if they fail that, they're now refractory, and there's really very little that can be done for them.

About how many patients are potential -- could this be potentially applicable to?