Right, so welcome to the next Fireside Chat here with Tango Therapeutics. With us today, we have Barbara Weber, CEO. Barbara, welcome. Thanks for joining us.
Thanks for inviting us.
I'm Michael Schmidt, Senior Biotech Analyst with Guggenheim. And I'll just ask some questions in the Fireside Chat format today. And so before we dive into that, though, Barbara, just remind us, give us a quick high-level overview of the company. And I know the foundation of the company is really the theme of synthetic lethality that's underlying the pipeline. And sort of, yeah, if you just highlight an overview real quick.
Yeah, so synthetic lethality is actually a kind of a, I would say, a genetic trick to find targets that primarily target cancer cells while sparing normal cells. That's how we've built our pipeline. We are now in the clinic with our lead molecule, TNG462, which is targeting PRMT5. The synthetic lethal component of that is MTAP deletion. Targeting PRMT5 in solid tumors that have MTAP deletions. We released a sort of curtain raiser data on 462 last year showing early signs of efficacy, great durability, and tolerability. We have plans for this year for a full data release.
Great. Well, why don't we start then with PRMT5 and dive a little bit deeper here. So yeah, perhaps just remind folks of the history of PRMT5 inhibitors in general. I know there's been a whole wave of the first-generation assets. What are the differences to the second-generation inhibitors?
Yeah. And I think that that, in a way, created some headwinds for us a little bit because the first generation of PRMT5 inhibitors were not synthetic lethal mechanisms of action. And so there was no therapeutic index between normal cells and cancer cells. There was, however, I think, in the proof of concept in those molecules, in a few patients who were sensitive, there were good responses and there were long durable responses similar to what we see now with the second generation. But it did leave, I think, sort of a shadow in people's minds about the target. I think now the data coming from the second-generation inhibitors that are selective for cancer cells with MTAP deletion, and I think the front runners are us, Amgen and BMS, have shown that we, in particular, have a good therapeutic index, a very wide therapeutic index.
And it's yielding pretty exciting efficacy data in the sense in particular that both partial responses and stable disease are very durable, which I think is exciting. It's not something that has normally been seen in the past with TKIs. And again, especially for 462, the tolerability profile is really excellent. So I think these promise to be excellent drugs. And I think we have the potential to have the best in class of the front runners.
Great. So I'll come back to that in a second. And then maybe just in general, how should investors think about the market opportunity for PRMT5 inhibitors in general?
Yeah, the market opportunity is very big. 10%-15% of all solid cancers have MTAP deletions, and if you just look at the two biggest indications, lung cancer and pancreatic cancer, about 15% of lung cancer is MTAP deleted. That's somewhere in the range of 20,000-25,000 patients a year, and for pancreatic cancer, there's been particularly some work recently from Dana-Farber and other places suggesting that the incidence of MTAP deletion in pancreatic cancer is close to 35%. That's another 15,000 patients. If you add GBM and we're going into the clinic with a new brain penetrant molecule to see if we can make a difference in that tumor, that's another 7,000, so we're up over 50,000 patients a year with MTAP deletions just in those three indications.
OK, so very significant, and so yeah, last year was really interesting. We've seen data from three different programs throughout the year: Amgen, AMG 193, the Bristol drug, and then obviously also your own programs, and so before we dive into your program in more detail, just remind us of the key learnings for the field from these data readouts.
I think the data from all of them are very consistent. They're at different stages. The Amgen and BMS went into the clinic about a year before TNG462. So the data they showed was a bit of a bigger data set and more mature. But they showed some really important things. One is that there doesn't appear to be a histology selectivity for sensitivity. That across the board, a number of tumors where there's high medical need are sensitive. And that's not just lung and pancreatic cancer, but esophageal cancer, gastric cancer, a number of really difficult cholangiocarcinoma, a number of difficult-to-treat cancers. And that those people who benefit really have the potential to stay on the drug for long periods of time.
Yeah. And so again, no preference for histology, long duration responses, it seems, in very difficult patients. And then maybe just shifting to your own data. So you disclosed late last year data from your original program, 908, and then also from 462, your second more selective inhibitor. And yeah, maybe just remind us of the key findings and how those data have informed your decisions for additional development.
Yeah. So TNG908 was the first molecule we went into the clinic with around the same time that Amgen and BMS went into the clinic. That's, as you point out, a less potent and less selective PRMT5 inhibitor than TNG462. But it is brain penetrant. And so the major focus for 908 was glioblastoma. We saw over the course of that trial, we put more than 20 glioblastoma patients in, that there was not a clinical signal in glioblastoma. And I think the reason for that was clear. We did have samples from patients showing that the exposure of 908 in the CSF was simply not sufficient to get to what we believed would be the therapeutic threshold. So we discontinued 908 and moved forward with a new molecule that has the potency and selectivity equivalent to 462, but is brain penetrant.
We'll be going into the clinic in the coming months with that molecule. From the standpoint of 462, that molecule is not brain penetrant, but it is very potent and selective. It has a fantastic tolerability profile. It has a long half-life. It has all the things that make it, I think, a really good drug. That's where we've put the focus on. Since the data update in November, we've been really focusing on enrolling lung and pancreatic cancer patients. That's going very well. Those are the data that we expect to be able to provide this year, both the full analysis of the dose escalation cohort and the updated dose expansion cohort patients that will be primarily pancreatic and lung.
Okay. And so I think you obviously saw the data in November. Again, it was a really early data cut. I think there was a little bit of disappointment on the street by the data. But if I remember, the BTC, the cholangiocarcinoma patients looked quite active in that cohort. But yeah, again, I think data seemed immature. But yeah, sort of key takeaways from the data, specifically on 462, and your continued confidence in the molecule and differentiation, perhaps, from Amgen and Bristol?
Yeah, 100%. And I think one of the things you point out was there was a lot of confusion and, to some extent, disappointment about the cholangiocarcinoma data, which I guess surprised me a little bit because we put those data out there as a way to say, this is the one tumor type where we have enough of a single histology to make a comparison to Amgen and BMS and look at the data. It's considerably better. But then people were like, well, this doesn't matter because you can't register. It's too small of an indication. And you must have bad data everywhere else. That's why you're showing us this. Not the case. So that's why we're focusing really hard on getting enough lung and pancreatic cancer patients to make a convincing, not just externally, for making sort of a convincing case that we potentially have the best-in-class molecule.
We are focusing on getting the data we need to make a commitment to phase 3 studies to start next year.
Right. And lung and PDAC, obviously the most commercially relevant indication, so it makes sense. And so you've been enrolling these tumor-specific dose expansion cohorts at, I believe, 200, 250, and 300 milligrams once a day. And so yeah, maybe talk to us a little bit about where you are in terms of philosophy around dose selection, sort of what drove selection of those particular dosing strengths. And then yeah, maybe in general, how has enrollment progressed relative to expectations so far since the November data update?
Yeah. So the dose selection is really based on tolerability profiles. All three of those are active doses. At 200 milligrams, we have almost no side effects, no dose reductions. People do really well. At 300 milligrams, we get over the threshold where you start to see a significant amount of bone marrow toxicity, which is the known on-target toxicity for PRMT5. So we're really just trying to titrate into that space where we get the optimal amount of drug in with the minimal amount of tolerability. So we know it's going to be 200 or 250. We already have enough patients on that when we have the data, when the data mature over the next couple of months, we'll be able to make the call. But it'll be one of those two doses.
Understood. And yeah, then on the sort of anticipated data disclosures in 2025, just help us sort of set expectations, what investors should expect there in terms of, I don't know, patient numbers or follow-up, things of that nature.
So at the time of the data release last October, we had, I think, 59 patients in dose escalation. We have considerably more than that now. And most of them are lung and pancreatic cancer. The majority, I mean, as you might imagine, the patients that are the fastest and easiest to enroll are pancreatic cancer patients because there are so many of them. And there are so few treatment options. So that is certainly a pace ahead of the lung cancer cohorts. But that is enrolling as well. So we're confident we're going to have what we need this year. We'll probably have pancreatic before we have lung. But that will work.
Right. And is the focus on response rate in pancreatic cancer, I know duration or PFS is another important outcome as response rates have generally been very low for other therapies. So how do you think about the type of data you might be able to share?
I think regardless, the right measurement for these targets, and probably in the end, any target is PFS and OS, right? That's the regulatory endpoint. That's certainly what matters to patients and what makes things commercially viable. In the past, stable disease has not been particularly helpful, right, because those patients generally have not done well and stayed on for very long. It's clear with this target and from everybody's data that that's not the case. So for us, people will be looking for response rates. There's no question about that. And we expect ours will be in the range of where you've seen, particularly BMS. We think there are major competition. But the really important thing to follow is going to be the PFS.
Right. And then just longer term, do you believe there is a single agent opportunity to market for 462? Or is it going to be a combination play longer term?
No, I absolutely do believe that in both second-line lung and second-line pancreatic cancer, single agent therapy is a very viable path forward. Those are the data we're generating now to design and adequately power those phase 2 studies. Moving into first-line patients, I mean, I think those will be combinations. The bar is higher there, and the likelihood that right now any molecule will replace the combinations that are being used in first-line lung and first-line pancreatic cancer is pretty low.
Right. But you are obviously planning to initiate combination studies as well with 462 this year. So walk us through some of those and the strategy behind maybe some of the particular combinations you're going to explore.
Right. So those combinations fall into sort of two buckets. One is a set of combinations we're doing just to assess combination tolerability and get a dose for some of the regimens that are already being used in first-line, so whether it's pancreatic cancer with FOLFIRINOX or gem and Abraxane, we just want to get enough patients so we can understand the combinability. Same for the chemotherapy regimens and for pembro for lung cancer, so those will be relatively small studies, small numbers of patients just to assess tolerability, and honestly, 462 is so well tolerated as a single agent, we don't expect to see anything significant, but obviously, we're doing those studies, and then the really exciting ones for me are the combinations with the two RevMed molecules, the panRAS and the G12D.
And those will both be to get at a dose for assessed tolerability, but also to continue moving forward for efficacy. Because I think it's very possible that the combination of a RAS inhibitor and a PRMT5 inhibitor could unseat the chemotherapy regimens in pancreatic cancer. And that would obviously be huge. I mean, I'm an oncologist. I've lost friends to pancreatic cancer. And those are horrible regimens.
Yeah. And obviously, KRAS is very prevalent in pancreatic. Are you looking at these combinations in lung cancer as well?
We are. The numbers are, of course, smaller, right? So basically, every MTAP-deleted pancreatic cancer patient has a RAS mutation. About 40% of those are G12D. So those are the two sets for the two RevMed molecules. In lung cancer, it's a much smaller number. But it could also be very meaningful. What, 30% or so of lung cancer has a RAS mutation. And I think 5%-7% of those are G12D. That's a number that's equivalent, say, to the ALK inhibitor population. So if the data are really strong, that's also a viable path forward.
Do you think the frequency of RAS mutation in lung cancer, is it similar in the MTAP deletion background? Or is it different?
It's the same. Those are completely independent events. So whatever the frequency in the two independent populations are, they'll just be the same.
OK, great. And then yeah, maybe just a few questions on 456. So you mentioned at the beginning, it is sort of meant to be a follow-on product to the 908 molecule. And so yeah, maybe talk about some of the specific properties of this drug and your confidence, perhaps, in its path forward.
So the key factor, it turns out, to be the potency and selectivity of 456 versus 908 because the brain penetrance is sort of similar. The reason why we believe we'll be able to get much more drug into the CNS with 456 is because of that selectivity. We can dose higher without hitting and having normal cell toxicity. So based on all the preclinical work we've done and the modeling, we expect that to be the case. And we should be able to sort that out in the coming year.
OK. And yeah, maybe talk about your development plans for 456 and sort of next steps.
So the primary focus for 456 is in glioblastoma, both of the monotherapy where we expect to see activity, assuming we get sufficient drug into the CSF, and in combination with the abemaciclib, which is based on preclinical data of strong synergy, which goes back again to the genetic alteration of CDKN2A deletion. So we'll be looking at both of those in the phase 1 study, both the monotherapy and the combination with the Abema.
OK, great. And what drove the combination strategy with the abemaciclib specifically?
There were really two things. One is CDKN2A deletion, at least preclinically, drives sensitivity to a Abema, CDK4/6 inhibitor, and there are data from a clinical study of glioblastoma done at the Dana-Farber, suggesting, well, showing actually that there's a PFS benefit to single agent Ab ema in glioblastoma of roughly doubling the standard of care, so the combination of those two data sets, preclinical and clinical, suggested that this might be an interesting combination. It could be relevant, essentially, to any MTAP-deleted tumor, but where we're testing it first is in combination with 456.
OK, great. And so when do you expect to start phase 1? Just remind us for.
That'll be starting in the coming months.
OK, great. And then maybe just on the topic of combinations, so there's obviously your peers are looking at different combinations as well. There are combination studies ongoing with MAT2A inhibitors as well. So where do you see those fit into the picture? And is that something that you're interested in longer term?
As you know, the question of whether MAT2A inhibitors and PRMT5 inhibitor combinations are going to be valuable or not has been going on even inside Tango for a really long time. We, based on a lot of preclinical work, came to the conclusion over a couple of years that with a really good PRMT5 inhibitor, you can fully inhibit the enzyme, and by adding a MAT2A inhibitor, you're probably going to get only added toxicity, but as I think is widely known, IDEAYA has an ongoing study, and we'll be anxious to see those data, and we also have a collaboration that should we decide to do that, that will be an option for us. I think it will complicate additional combinations because then you're talking about already starting with a triplet, but let's see and be driven by the data.
Right. And then I think we're seeing some additional market entrants into this space. I think there are PRMT5 inhibitors in the works from AstraZeneca. BeiGene formally has one as well. And so how do you view your positioning longer term in this space and sort of this growing field?
Clearly, the front runners are Amgen, BMS, and us. I think we're at least a year at that point ahead of AstraZeneca. They have a good molecule. They're also exploring in hematologic malignancies. It's not brain penetrant. So I think that's the one to keep an eye on. BeiGene and Abbisko also going into the clinic this year. Others coming behind that, IDEAYA, Gilead. So I think it's not surprising, given the broad interest. And now, in my mind, the full clinical validation of the target, that there's going to be a lot coming after. But we still have the advantage of being among the front runners. And we just need to stay there.
OK, great. And so then maybe switching gears, the last couple of minutes, I just wanted to touch on your TNG260 molecule, your CoREST inhibitor. And so maybe just stepping back, remind us of the mechanism for this drug.
Yeah. So that's an interesting and completely separate story and an interesting spin on synthetic lethality. But based on our preclinical data and strong retrospective clinical data, it's been shown that STK11 loss of function mutations make lung cancers resistant to pembrolizumab and other checkpoint inhibitors. And they do that by the way they modify the immune environment of the cell, decreasing PD-L1 expression, removing T-cell infiltrates, all those things that you need to respond to a checkpoint inhibitor. And so our trial is with a Co-REST inhibitor that reverses at a molecular level those things that render lung cancers resistant to checkpoint inhibitors, particularly pembrolizumab. The study that we are doing has been with pembro plus 260 in the beginning. We don't expect single agent activity from 260. The purpose of it is to allow pembro to work. So you need it in combination.
We have gone into the expansion phase of that trial in lung cancer, and we expect to have clinical data on that program this year.
OK. Can you just walk us a little bit more through the design of the phase 1 study and some different considerations as you sort of have designed the study?
The design of the study was slightly modified by the FDA, but I think it still was what we wanted to accomplish, which is that in the beginning, we were required to give three weeks of single agent 260 and then add pembrolizumab. Once each dose was cleared, we were allowed to start the combination right from the beginning in every patient, and so I think now in expansion, all the patients are getting the combination from day one.
OK. So it sounds like you have identified a recommended phase 2 dose.
That's right.
OK.
That's correct.
Is the study enrolling presumably checkpoint inhibitor experienced patients? Talk a little bit about the histologies, perhaps.
The primary focus is on lung cancer. By definition, almost all of those patients have had and failed pembrolizumab in the past. It's not required. But most of them have had it.
Right.
We know historically that those patients who progress on pembro and are re-challenged with it again do very poorly.
Right. And there have always been historically other studies in this space, in this post-pembro lung cancer category and combinations as well. And so yeah, sort of what signal, what efficacy signal do you think would validate the mechanism and the concept of the drug?
I think just like not unlike the PRMT5 situation, you want to see a significant improvement in the PFS of these patients to make this an approvable combination. And so you want to see at least a doubling of what the baseline PFS would be in those patients. So that's our bar.
And is docetaxel the historic comparator here through the four to five months? OK. Great. Thanks. And then yeah, so maybe then just to wrap up, as you think about enhancing the pipeline further longer term, any plans for additional pipeline developments?
Sorry.
New pipeline drugs that you're thinking of, how do you think?
Yeah, so we have a pretty robust pipeline coming behind, and you can expect to be seeing some of that also in the coming year or so.
Great. Thanks, so with that, I think we can wrap up, so again, looking forward to seeing sort of phase 1B data on 462, the PRMT5 inhibitor, as well as next steps, and then initial data on the CoREST program as well.
Absolutely.
Thank you so much.
It's going to be a big year.
Yeah. Looking forward to it. Appreciate it. Thank you.