Let's go ahead and get started. Next up, really happy to have the CEO of Foghorn Therapeutics, Adrian Gottschalk, here with us. Adrian, over to you for an overview of the company, where things stand today, then we'll get into it.
Sure. Thanks, Gavin, for having me here in Miami, where the weather's better than Boston. Pleasure to be with you. Yeah, so maybe just a quick, quick intro and overview of Foghorn. We're a clinical stage biotech company. Precision oncology has sort of been the mainstay for the last few years, focused in the area of chromatin regulation, which we believe is super relevant in a raft of different cancers, but potential other areas of biology as well, so up to 50% of cancer has some sort of dependency or genetic tie back to the biology in which we're interested. We've been hard at work over many years developing a range of different programs with a mix of classical enzymatic inhibitors, but also more recently in the field of protein degradation.
Our 909 study, which is in partnership with Lilly, and I'm sure we'll talk about that program and the SMARCA2 target, is really exciting. We're looking forward to a decision sometime in the first half of next year based on where we are with our clinical study, a decision from Lilly as to whether to go into expansion, and a lot of other interesting programs coming up behind that in the proprietary pipeline, so a lot of exciting stuff in the coming year.
Awesome. I guess just a high-level question, given that you're partnered with Lilly, what are your plans on data disclosure? So not just for the phase I-B escalation portion, but also if you go into expansion, would you plan to update the street on that? I guess what about next steps for development too? What's your current thoughts there?
Yeah, so maybe I'll just start on this program, just take a step back for folks to set some context. So this is our 909 program. It's an enzymatic inhibitor. It's going after the SMARCA2 program target, pardon me, that is synthetic lethal with these SMARCA4 mutations, predominantly in non-small cells where the interest is, but there's potentially other cancers. So from a data disclosure perspective, the threshold here is really what's material for us and what we would need to disclose. A decision to go or not go, but hopefully to go from Lilly is material to us. That is something we would need to communicate to our existing and prospective shareholders.
Certainly, if we're in the mind to do financing around that time, we'd also want to communicate at least top-line efficacy and safety tolerability, along with what the intention is to what comes next in the dose expansion. And we've been on the record previously that for dose expansion, the intent is to move very quickly into the front-line setting. Right now, we're treating patients that are predominantly fourth-line or third-line, non-small cell and other histologies. But given the unmet need, and we can talk through some of that, there's a very clear path in expansion to go straight into a combination with pembrolizumab and chemo in first-line metastatic non-small cell patients. There's potentially other combinations in non-small cell that the collaboration would be interested in. But that's effectively the path.
And so we would be updating the public as to a go/no-go decision and then at least some top-line data, as it were, roughly contemporaneous with that.
Yeah, that makes sense. So there has been another SMARCA2 program in the clinic that's Prelude with their degrader. They recently announced that they're not advancing it. Do you think that reads across negatively to the target?
No. To take a step back from what our colleagues at Prelude did, they had both a VHL degrader and a Cereblon degrader. And we've actually analyzed their compounds. I think the fundamental, the positive from what they did, and now I'm referring to their VHL degraders, they showed some very nice proof of concept on this target. They showed with what we believe is a degrader that's inadequate in coverage. I'll say more on that in a second, that you could get a couple of partial responses in lung, some in esophageal. I think they had a gastric cancer partial response as well. The fundamental challenge, and just to take a step back on this target, you have to hit the SMARCA2 target very hard. Whether that's with an enzymatic inhibitor, you want to hit it at an IC90 or better.
If it's a degrader, you really want to be at a DC90 on a sustained basis. The challenge for their VHL drug was it is a once-a-week intravenous molecule. And by their own admission, in their preclinical data, I think in their clinical data, as well as the press release they put out, they just had inadequate target coverage. And we know from all the preclinical models we do, if you aren't sustainably inhibiting or for that matter, sustainably degrading at a DC90 or better, you're not going to get the regressions. You'll get some TGI, but not the really stellar responses. I think the challenge for their Cereblon molecule, and again, we haven't seen any clinical data, but we've made both their molecules from their patents. These are slides that you can find in our investor deck in the appendix. They get to about 80% degradation.
We actually replicate their data, but we actually want to step further, and we did a bunch of the kinetic analysis that we do on all our degraders. Speed matters, and it's directly related to how deep degradation you can get, i.e., DC90s, and then the challenge, or ironically, actually, their VHL degrader was better than their Cereblon degrader in target coverage if you were able to dose it continuously, so going back to your question of a negative read-through, I look at this as nice proof of concept, but really clarifying that if you're not hitting this target hard enough, you're not going to get the results, and ostensibly, they've stopped it, and what they've said publicly was they recognize you need to go into front-line, you need to do combinations. It can be complex. It can be expensive. I actually agree with all of that.
That's why we have our partners with Eli Lilly.
That's perfect. What does the dose response look like as you kind of recapitulate a lot of this preclinically in your hands? Like as you're going from IC50, 80, 90, 95, higher, what does that look like for this target?
Yeah, so when we've looked at both from an inhibitor and degrader perspective, as you vary the concentration, the ICs, or in the case, the DC, the Dmax, you'll start seeing varying degrees of. I can't give you the exact mathematical correlation part because I'm tired, but as you look across those Dmaxes, so just as more of an illustrative point, if you're at a DC60, you'll get some TGI. You'll get a deeper TGI with the DC80. Once you get in the DC90 plus, you start getting regressions. And the same thing we've seen with the ICs. Like if you're covering it in the IC50, you get some TGIs. You don't really get the regressions till the IC90. And it's consistent with, again, seven, eight years of actually working on this target. You have to hit it hard.
There's certainly a dose-response effect as you increase the IC90 or the DC90 respectively for an inhibitor or degrader.
For the phase I, I know you're not giving granular trial updates. I guess more qualitatively, how quickly has enrollment been going? What's the level of interest and focus on this from investigators?
Yeah, so enrollment has gone exceedingly well. And all credit to our colleagues at Lilly who are responsible for the day-to-day operations. So maybe some just basic facts. So Lilly operates the study on a day-to-day basis. We have 15, one five sites open in the United States. There's five sites open in Japan. There's sites that are being initiated and being prepared to open over the next month, month and a half in Germany, France, Spain, and South Korea. Enrollment has been at a pretty good clip. We've been dosing since October of last year. We've not hit a maximum tolerated dose. We recently dose escalated. I anticipate we may dose escalate again here in December.
I think I've said more recently, we now believe we are getting into therapeutic range based on PK and some other parameters and observations, which means we're now starting to backfill some of these cohorts, not exclusively with lung, but more enriching for non-small cell and the more specific loss of function SMARCA4 mutations. So without getting into specific numbers of patients, there's a tremendous queue to get patients in. So again, and Lilly's done a tremendous job. When we open a new cohort, usually there's multiple patients already screened and ready to go into the next escalation cohort, in this case now backfilling at a couple different doses. So very pleased with that.
That's in part informing the high-level guidance on our part that we believe the collaboration with Lilly will be in a position in the first half of next year to make a decision as to whether to go or not go into expansion. Could that shift a little bit? Sure. If we keep going up and things continue to look clean and we keep doing that, maybe stuff shifts a bit. That's our best guidance right now.
Yeah, that makes sense. Have you said roughly when you got into that efficacious range within dose escalation? You haven't put out the doses, but I guess just like where did you start relative to where that range is?
Yeah, I think that's sort of happened over the last couple of months or so. Most of the times when you look at these phase Is, and it's like forecasts, all forecasts are wrong. Some forecasts are useful. I would say similarly for dose escalations, you generally say, we think it's within six to eight doses. Sometimes you hit again in there and it's four or five. Sometimes it's nine or ten. I think we're falling kind of smack into the middle of that is the way to sort of think about it.
Yeah, that makes sense. And for the backfill cohorts, have you actually started enrolling in those yet or is that?
Yes. So we've started backfilling patients or Lilly has, again, because we believe we're now in a range where it's appropriate to add more patients to start trying to look for a signal over the noise.
Okay. Maybe you can just kind of discuss the characteristics of the patients you think you'll be getting in those backfill cohorts.
Right. So I'll speak more specifically to lung because that's the focus. Look, not surprisingly, we're getting esophageal and some gastric and a variety of other tumors. There's a range, as you know, of histologies whereby patients have these SMARCA4 mutations. What's been interesting to us, but obviously to Lilly, is non-small cell lung. That's not to say there won't be other tumor types, but that's if you're wanting to look for a commercial opportunity that's substantive, that's at least where it is initially. So the type of patient we're getting are third, fourth-line non-small cell lung patients. We are focused on finding those patients that have true loss of function. Not all SMARCA4 mutations are equal. We need those that create meaningful deleterious impacts on the protein such that it's non-functional. So that's the type of patient.
Maybe just to put that in context, because I often get asked like, what does good look like and whatnot? Just to throw a few numbers out at you, and this is unfortunately scary and terrible for the patients. If you're in a front-line setting, first-line, chemo plus checkpoint, wild type, 40% response, 15-month OS, eight or so month PFS. If you're SMARCA4 mutant, that response rate's cut in about half to about 20, 21%. Your OS is about eight months. Your median PFS is two and a half, maybe three months. That's first-line, right? We're talking about fourth-line, third, fourth-line patients where the median PFS and OS tend to collapse onto each other, month and a half, two months. Response rate, 5%, zero. Anecdotally chatting with the physician who treats these patients who said they have trouble getting these patients into studies.
And I said, "Oh, is this an informed consent issue?" No, they're so sick that unfortunately they pass away before sometimes you can even get them in a study. So again, the material piece here and what's going to matter is what Lilly decides. So my comment in some ways is irrelevant, but if we're seeing duration of response in the four-plus month range and we're seeing responses better than 15% or so, not to push that to registrational, but to say this is real, let's now jump into the first-line setting in combination. I think that's where the bar is. That would be my bar, but obviously that's up to my colleagues at Lilly.
Yeah, that makes sense. So you mentioned the Class 1 versus Class 2 mutations for kind of true loss of function versus missing mutations. Do you think there'll be differential efficacy in those patients?
I personally do. Look, if we see efficacy out of Class 2 mutations that are non-loss of function or non-truncating or not impacting the catalytic pocket of the enzyme, that'll be a pleasant surprise to the upside. My belief has always been to really get at the synthetic lethality. You want that loss of function, whether that's that it's truncated of the protein, whether it's rendered the catalytic pocket inert. So I'm more focused on what we see out of Class 1 mutants. If we see efficacy in some ways in Class 2 or in combination with other therapies in Class 2, that's all upside as far as I'm concerned.
That makes sense. What do you think about speed of onset of responses and efficacy in general?
I can't comment at this point given the study, but I think generally my going in assumption is you'll want to start seeing something like we're scanning patients at two months, four months. Sometimes they get interim scans or whatnot, but I think you're going to want to start seeing shrinkage within a couple of months. Obviously, you'd like to see some PRs along the way. I think that's a reasonable expectation to want to see something within the first few months. Again, if you're having that impact, as I just said, in these fourth-line patients where unfortunately they progress within a month and a half to two months, I think if we're getting out to four plus months, whether it's just duration initially or knock on wood, some PRs, that's going to be pretty exciting.
What do you think about safety for this target? I mean, you have some experience with SMARCA2/4 previously.
Right. Yeah. So I can't comment. What I can say so far is we have not hit our maximum tolerated dose within the study. What we know from our experiences, as you mentioned on our prior program, which 286 and SMARCA2/4, is we ran into what I would describe as just general tolerability issues where the patients just didn't want to be compliant with the drug. General malaise, muscle fatigue, weakness, some dyschezia, some skin rashes, other things that just I'd say it's like chemo like. We're like, I don't want to take the drug. I don't feel good when I'm taking it, and so that's when we'll know we've sort of hit our tolerability limit if we hit that. At this point, again, we haven't hit that, so.
Yeah, that makes sense. I mean, I guess after this escalation portion, what are your plans as you think about the front-line combination versus the refractory setting?
Right. So again, just based on some of the numbers I threw out at you a few moments ago, there is an absolute direct path straight into that front-line non-small cell setting. So the initial goal, but obviously to be finalized by Lilly, is a combination study potentially varying the type of chemo with pembrolizumab. There's also clear data showing an overlap with KRAS mutation. It's actually the only main when you look at some of the GWAS stuff, KRAS is the most meaningful overlap. It's about three for every 10 patients SMARCA4, roughly three have a KRAS mutation. And those patients, if you thought some of the data I mentioned previously was bad, it's even worse if you have both a KRAS and a SMARCA4 mutation. So Lilly obviously has their own set of KRAS assets.
I think that's a question of when they will want to combine just given that their mutant versus pan-KRAS assets are at different stages in clinical development, but there's definitely interest in combinations there. We know this is in the public domain. We know there's potentially some CDK4/6 interest. Obviously, Lilly has a CDK4/6 asset. So there's a lot of different things that we can actually go do, but I think the primacy is get into the first line, get into a combination with pembrolizumab.
That makes sense. Maybe we can zoom out and talk about the pipeline. I mean, you're working on a lot beyond just SMARCA2. You have an ARID1B program, which also hits the BAF complex. You have CBP EP300. Maybe you could just kind of frame where those are today.
Yeah, super excited by the proprietary stuff. As I like to say, the best part of Foghorn is we're unique. Sometimes the worst part is you're unique because we're breaking new ground in areas that when people say, well, what are the clinical data for an ARID1B program or CBP? The short answer is no one's ever done it in this type of way. I'll work backwards just very quickly. ARID1B, we're targeting to have in vivo proof of concept next year. We just disclosed about a month or two ago some really breakthrough data where we've shown now we've got not only Cereblon, but also VHL degraders for ARID1B. To our knowledge, no one has been able to actually target this in a selective way. So I think that could be super, super interesting a bit earlier than the rest of the pipeline.
Bookend that with CBP, which is in the process of going through non-GLP tox. We're now doing that with our long-acting formulation technology, which is roughly once a week subcutaneous. That could be relevant in a synthetic lethal context with EP300, but also maybe in ER- positive breast, w e're still generating some of those in vivo data that is tracking to an IND in the middle part of next year, obviously subject to financing and other things. The program, again, if you gave me nine seconds on the clock, the only one I'd talk about would be EP300. And that's in part because of the clinical validation that's already occurred from a competitor who's got a dual CBP/ EP300, which unfortunately has some tolerability and tox liabilities when you hit both of those targets. We've dialed that out with our selective EP300 program. Beautiful molecules.
We think we're really close to a pre-development candidate. Best case we're starting some studies middle part of next year. Worst case, it's a little later in the year. Very excited about that program as well. So a lot of stuff going on at Foghorn.
Awesome. I think we'll actually wrap it up right there since we're at time. Adrian, thanks so much for joining.
Gavin, thank you.