Good afternoon, everybody, and welcome back at the Guggenheim Healthcare Innovation Conference. I am Dan Makayati, a research analyst here at Guggenheim, and it's my pleasure to introduce the next company to you, Surrozen. From the company, we have Craig Parker, the CEO. Welcome, Craig.
Thank you.
Why don't we start?
Chuck Williams as well.
We have also Chuck, yes, in the audience. CFO from the company. Craig, don't we start with an overview of the company? If you can tell us a little bit of your story and the key milestone that we can expect in the next month.
Of course. Yeah, so we were founded around the end of 2015, early 2016, by most of the preeminent Wnt biologists in the world. Wnt is a combination of wingless and integrated. These turned out to be the same gene discovered by two different people. They just jam those words together to form Wnt. But it's a key pathway in embryonic development, in tissue regeneration, in maintenance of tissues. And people have considered it for many years, in fact, decades, as an interesting target for drug discovery because of its role in regeneration and maintenance of tissue structure. But no one had ever really been able to drug it with any specificity. And our breakthrough, and indeed the antibodies that we discovered, are able to mimic naturally occurring proteins that modulate the pathway.
But we can do it in either a cell, and we'll talk about SZN-043 in the liver, either a truly cell-specific or a tissue-specific way. So that's really the fundamental principle of the company, is to engage this very powerful biological pathway, but in a tissue-specific or cell-specific way to drive regeneration in diseases where the tissue has been broken down by the underlying pathology.
Got it. And as you were saying, Wnt has been a very difficult pathway to drug, with a lot of challenges, especially from the safety profile. So can you tell us a little bit about your approach? You are using both enhancer and activators. So if you can describe the difference between them? And also, how are you confident in the possibility of your platform to overcome the classical risks that are associated to this target?
Yeah, so normally in our industry, when you find a naturally occurring pathway, a naturally occurring protein, you just make a recombinant version of the molecule. Erythropoietin, insulin, growth hormone, there are many examples of that. Wnt proteins that are involved in this regeneration process or stem cell renewal process are very promiscuous. So there are 19 of them. They bind to 10 different receptors, and they're quite insoluble. So there's really no way to just copy one of these naturally occurring proteins and get a drug. There are multiple other modulators of the pathway. There's another protein called R-spondin. So what we did is looked at specific tissues and said, what seems to be lacking? Which part of the modulation of this pathway seems to be lacking?
And then we made bispecific antibodies that were either specific for one of the receptors, so they took some of the promiscuity out of it, or we could design in targeting to a cell, as in the case with SZN-043, where we can add a moiety to the antibody so it binds specifically to only hepatocytes. So we can harness the pathway, but only in a tissue or cell-specific way. And those are the two different platforms we have. And we look at the tissue, and we say, what's missing? Is it the ligand that's missing, or is there lots of ligand, and maybe it's the receptor that's not being expressed at high enough levels? Now, we're not modulating the expression, but we're modulating the turnover of the receptor. So in the presence of ligands, if you have more receptor, you're going to have stronger signaling.
So despite the ubiquitous function of Wnt, there is some level of specificity between different tissues that allow you to target specifically.
Yes.
When we think about the SWAP and the SWEETs, what's the difference there? If you can go through.
Yeah, of course. So the SWAP is just mimicking the ligand. So these proteins are called Wnts, W-N-Ts. The receptors are called Frizzled, named after the phenotype of the fruit fly that it was observed in initially. And that's our SWAP platform. It's mimicking the ligand by binding to exactly the same receptors that Wnt bind to. But we can bind to individual Frizzled receptors. Where these Wnt proteins are promiscuous and bind to multiple receptors, we can bind to just one of those. So that's copying the ligand in the tissue. And if we look at the tissue, and it looks like there's not a lot of ligand being expressed, then we try to copy the ligand. In the liver, this has been published by some of our founders, in the injured liver, it looks like there's plenty of Wnt ligand.
So that doesn't seem to be limiting the regenerative response. So we took this R-spondin approach that acts on the receptor, copies exactly one of the binding sites of a naturally occurring R-spondin protein, but adds in this binding to hepatocytes. So we're only activating Wnt signaling, it turns out, in mature hepatocytes. So it's even better than just hepatocytes. It's only mature hepatocytes.
So somehow, it's almost trying to adopt a precision medicine approach.
It is. Oh, it's very much precision medicine.
Understand.
But tissue or cell precision, not disease precision. So what's interesting about this pathway is, and you can look in the literature, there are many, many diseases, not where there's a genetic mutation, because this is an embryonic development pathway. So there won't be a lot of human diseases that are mutations of this pathway, because they would have never been viable.
Makes sense.
But what we know is, if you look in adult tissue, for example, or in knockout mouse models, that if you don't have this pathway, or if you downregulate this pathway, that you don't get adequate regeneration.
OK.
And so that's true in lung, as we mentioned, central nervous system. There actually is really interesting literature. We're not pursuing that. Liver, intestine, epithelium, there are some sort of classical Wnt-dependent tissues.
The choice for a SWAP or a SWEET, is it related to the type of disorder? It sounds like it's not motivated by the disorder, but rather by the organ that is affected the most.
Yeah. So we've had to try to answer that question, because without tool molecules, it's difficult to try to answer those questions. So what we do is we look in the literature, see if there's any evidence that the ligand seems to be missing or expressed at low levels. In the liver, it's at high levels. So then what we do is we get disease tissue, compare it to normal tissue, and do some just standard molecular biology methods, like in situ hybridization or RNA expression, and look for what seems to be missing. And that's how we pick which one of our platforms is in the human disease tissue, is look for what seems to be missing. And then, of course, take that into cell and in vivo models.
We actually will pursue both side by side, the SWAP and the SWEET, and see which one, despite the evidence that seemed to be pointing one way or the other, see which one works the best.
Makes sense. And this is a good segue for your leading candidate, SZN-043. That's a SWEET, right?
Correct.
So it's an enhancer. And you are targeting severe alcoholic hepatitis.
Yes.
That's a prime indication. So if you can dig a little bit, if we can dig into the rationale for this specific disorder, so I would like to hear from you both the scientific rationale as well as any commercial competitive reasoning around the choice.
Yeah, I mean, so it starts with, is there a biological rationale for this pathway in this disease?
Yeah.
One of our founders, Roel Nusse at Stanford, has actually spent 35 years elucidating the importance of Wnt molecules in both the development of the liver, zonation of the liver, and response of the liver to injury. So if you asked any Wnt biologists, like, tell me the three most Wnt-responsive tissues, they would probably say skin, intestine, and liver. So that was an obvious starting point for us.
Got it.
There are well-characterized animal models of liver injury. You can give an animal a hepatotoxin and see if you can regenerate liver. We were able to do that, and then we said, OK, we have this proliferative effect on liver, where we're stimulating proliferation of hepatocytes. Where might that be useful clinically, and where might we be able to see a signal quickly, so as we just talked about, if you choose MASH, or a very slow developing disease, it might take you five-plus years to know whether you had any kind of a clinical signal or not. We were looking for something where the biology aligned with the potential to see an early signal, and where there was very high unmet need, and it turned out severe alcoholic hepatitis is that disease.
It's been well documented that if you look at biopsies from people with severe alcoholic hepatitis, and I'll describe what that disease is in a second, and you look at, from those biopsies, proliferation in the liver and Wnt-mediated proliferation, there's a correlation with better survival if that subject has higher, and this is not pharmacologic stimulation, if they just happen to have more hepatocyte proliferation that was Wnt-mediated. So we all respond when we drink by expressing Wnt proteins, and that's why our liver can regenerate. But patients with this disease who had higher Wnt expression and more hepatocyte proliferation had better clinical outcomes. So for us, that was sort of like the stars aligning was the biology is known to be really important in the liver. We can hit hepatocytes, and only hepatocytes, so we're not stimulating fibroblasts, for example.
OK.
There's evidence that in this disease, higher Wnt signaling leads to better clinical outcomes. You asked, what's the regulatory path look like? 90-day survival is the key endpoint that matters clinically and that matters to regulators for this disease. We could see it quickly if we did have a treatment effect.
OK. And from a clinical perspective, what's the staging of this disorder? Because we were discussing, like, it's very different from NASH, where you have this chronic time course.
Yes.
What is for severe alcoholic hepatitis?
Yeah, so severe, so it's a clinical diagnosis. There's not one test for severe alcoholic hepatitis. It's a clinical diagnosis of clearly lower liver function, jaundice, for example, history of alcohol disorder. So this is defined by the grams of alcohol per day. So for a man, it's 60 grams of alcohol per day. Just if you want to turn that into drinks, that's about four to five drinks a day for more than six months would be considered what used to be called harmful alcohol use. And there are some other clinical criteria. For the severe, they use something called a MELD score, which is used in transplant. It stands for Model for End-Stage Liver Disease. This is a compilation of a couple of different serum markers. And for us, it's a score of 21 or higher is considered severe.
OK.
These are patients who are hospitalized. These are patients with a history of alcohol use disorder, very low liver function, frequently poor kidney function because of the communication between the liver and the kidney. They get hospitalized when they have a severe form of the disease. Typically, it's four to 10 days, although we had a subject in our first study who was hospitalized for 30 days. He's like very sick patients with very poor functioning livers. Alcohol, if you don't know, is a hepatotoxin. It directly kills hepatocytes. Frequently, these patients will have binged. They would have had a history of alcohol use disorder, and they would have binged and ended up in an emergency room based on that.
How important is to target specifically the hepatocytes and spare other cells?
Yeah. So you don't want to stimulate fibroblasts. So these patients may or may not have cirrhosis, depending on how long they've had alcohol use disorder, but frequently, they'll have cirrhosis. So you don't want to give them more cirrhosis. So you don't want to stimulate fibroblasts or myofibroblasts or stellate cells. And you don't want to stimulate immature hepatocytes, because they don't function very well. So if all you did was you sort of increased the reservoir of hepatocyte stem cells in the liver, it probably wouldn't lead to better liver function short term. Now, it might lead to it long term, but it won't lead to it short term. And we saw this in every animal model we looked at. We were only hitting mature hepatocytes, which were functional almost immediately. So we saw functional benefits within days of these animals getting their kidneys.
It's a pretty rapid effect.
It's very rapid. It's within 24 to 36 hours.
Got it and in terms of unmet need, so I guess, I mean, there is no approved treatment as far as I know.
Correct, yeah.
But people are treated with corticosteroids. How effective are?
A minority. Yeah.
How effective? Oh, that's interesting. So if you can tell us a little bit, like, which patients are eligible for steroid treatments, and how effective are they?
Yeah, so there is an art to treating these patients. Steroid use varies quite considerably by investigator, and by geography, and by site. There are some people we have investigators who say, I never use steroids. We have investigators who say, I use steroids in about 40% of my patients.
OK.
In terms of an algorithm for using steroids, you would exclude anyone who had an active ongoing infection, because the steroids are just going to worsen that infection, or you thought was at a high risk of infection. Maybe they were immunosuppressed from some other agent or some other disease. That cuts out about half the patients. The people who like steroids will use them in about 40-ish% of their patients. The data on steroids says you can reduce the inflammation in the liver, which was the objective, and they have a short-term benefit. Seven days to even out to 30 days, there seems to be a benefit in things like bilirubin reductions, MELD score improvements, and even survival. That does not persist through 90 days.
So there's now been definitive analysis of large data sets that says steroids do not impact 90-day survival at all. So we allow steroid use in our study, because what we're interested in is 90-day survival, ultimately. And we don't think steroid use is going to impact that. And if you tried to protocolize steroid use, tell investigators you can only use it in this patient, not that patient, I think it'd be very hard to enroll a study. So we allow steroid use.
OK. As a follow-up to that, is it known if steroids affect Wnt pathway?
There's no reason based on the pathway that steroids should affect the Wnt pathway, or that Wnt should affect steroid metabolism, for example.
OK.
So there's no theoretical drug-drug interaction or pathway interaction.
And is there any other agents known to impact Wnt that has been used in these patients, like that can provide some kind of?
Not from that direction.
Not from that.
The direction that we have to exclude, for example, are drugs that may be metabolized by Wnt target genes, or drugs that are hepatotoxic. So there are multiple antibiotics. Tazobactam and Piperacillin, for example, is a known hepatotoxin. So like any clinical trial, we want to exclude people where that would increase the risk.
Makes sense.
And so yes, but that's not directly that molecule impacting the Wnt pathway. It's the Wnt pathway potentially impacting that molecule.
We were discussing a little bit about the regulatory endpoints. The accepted regulatory endpoints, is it the 90 days, or is the MELD?
I want to be careful with acceptance. We're inferring what FDA has said based on what the company called Direct is doing. We have not had an end of phase two meeting with the agency yet. We're still in phase one.
Got it.
But what they've said publicly is they had that meeting.
Historically.
Just very recently, they were given guidance. They did what they called a failed study. I mean, they showed a trend towards a survival benefit in a fairly large phase two, but they were given guidance that a single phase three with a 90-day mortality endpoint would be sufficient for approval. Now, of course, your data has to end up being robust, and you have to end up having an acceptable benefit risk profile, et cetera. But yeah, we think that's the right regulatory path, is a single registrational study with a demonstrable who knows what that means until you meet with the agency, is that 30%, 40%, 50% reduction in mortality at day 90. This was called the landmark analysis of the actual event rate at day 90.
Got it. And you showed very promising, interesting data in Milan at the EASL conference.
Yes.
Can you summarize them and tell us why they are so exciting?
Yeah, so we've actually updated that data since then. So at Milan, what we showed was that in healthy volunteers and people who had some moderate level of liver impairment, before we went into these very sick patients, we wanted to make sure the drug was safe, and someone who had had a history of some kind of liver disease but wasn't actively symptomatic.
Yeah.
So in those people, we saw very good safety. We saw target engagement, so we can measure target engagement for this molecule. And we saw activation of the Wnt pathway. And we saw an improvement in hepatic function. These are tests that are used in the hepatology community. They're a bit technical, so not worth going into since we just have a few minutes. But these are well-characterized tests that can show, in our case, Wnt signal activation and improvement in liver physiology. Subsequently, and we just disclosed this last week, we've now moved on to severe alcoholic hepatitis patients. We finished the first cohort of six.
Yeah.
We had no drug-related serious adverse events. We had 100% survival at day 30. We haven't had a chance to follow them all for 90 days yet. We saw what I think is a single-arm open-label study. So there's not a control group. But we saw a majority of patients actually have ALT and AST improvements, so reductions in their transaminases and reductions in their bilirubin. So we'll hopefully have an opportunity maybe at EASL to present that data in more detail this year, but very encouraging so far.
The next milestone would be the presentation of the second cohort?
We haven't given guidance on the packets of data we'll release along the way. I think it's probably likely that after each cohort, we will. We established a very robust phase two-like safety review process for this study, because these are really sick patients. We have a formal independent safety review committee that meets after each cohort. They approved us moving to the next highest dose. They would do that again after the next six patients. So again, we haven't given guidance on the timing of that, but I think it's probably likely we'll do that.
In terms of benchmark for efficacy, like, people are looking at a direct drug, so what's your view on that?
I mean, of course, you'd like to see 90-day survival better than what's in the literature. And we're working with an academic investigator who has a very large database, more than 2,000 subjects of severe alcoholic hepatitis patients. Now, not all of them are going to meet our criteria, but we've gone into that database, applied our inclusion/exclusion criteria to look at what you should expect in this population. Now, he wants to publish this. I'm not going to say what all that data says. But it's not dissimilar to the literature, which says 90-day survival is about, sorry, mortality is about 25%-30% in this population.
Got it.
30-day is about 15%-20%.
Thinking ahead to a possible registration study, there, I guess, you will need a placebo arm.
Yes, absolutely.
How should we think about placebo response for this disorder?
I think that's. I think when he publishes that data, that'll be as good a comparator as far as expectations. Direct Biologics, which just did a randomized phase two study with 100 patients per arm, so I think that's probably a pretty good, most recent data to look at in a randomized controlled setting. They had, I want to say it was about 27% mortality at day 90 in their placebo arm. So pretty consistent with the literature of between 25% and 30%.
Okay, like a reference.
MELD score, no one's published a large database of MELD scores beyond about day seven.
There is not really a benchmark data.
So there's not a good comparator. Yeah.
Got it. OK, now moving on to your other interesting candidate, SZN-413. So that's a SWAP. And I think you are directing more towards ophthalmology, so retinopathy disorders. Can you explain, like, provide the rationale for that?
Yeah.
And then it's also a very exciting time for Wnt in the field. If you can tell me a little bit why there is so much excitement right now?
Yeah. Well, I mentioned that because this is a pathway that's involved in embryonic development. There isn't much human genetic evidence for Wnt signaling being the underlying pathology in the disease. There is in the eye. So there's a very, very rare disease called familial exudative vitreoretinopathy. It's also called Norrie disease, one form of it. And these kids are born with what looks like an age-related macular degeneration. In fact, many of them are born blind. And they actually lose their hearing. And there are actually some cerebellar manifestations later in life. We won't get into that. But so there's human genetic evidence, and there's mouse genetics that says that Frizzled 4, in particular, is key to maintaining the integrity of the blood vessels in the retina. And if you have mutations in the signaling of Frizzled 4, you have this, what looks like wet AMD-like disease in children.
So that was the initial basis for going into going after Frizzled 4 with a swap. We're trying to mimic the Wnt ligand that binds to Frizzled 4. And we can do that with a swap molecule. We made one, had really impressive data in animal models of retinopathy. We licensed it to Boehringer Ingelheim. They recently just paid us another $10 million because they've continued to advance the molecule. But the whole field has really kind of opened up with our only direct competitor, which was a company called AntlerA, licensed their drug to EyeBio. Merck just bought them for $3 billion, total $1.25 billion and a quarter upfront.
That's very interesting.
We were about even with them before we spent the time to negotiate a license with BI, and I mean, big pharma doesn't always work at the pace we do, so probably about two years behind Merck, but they just had some very exciting phase one data in diabetic macular edema that they had, I think it was a 12-letter improvements in visual acuity.
That's impressive.
With that molecule. So that's alone without VEGF. So Frizzled 4 alone without VEGF looks to be as good as anything that's come before in terms of early.
Have you completed the toxicology studies for?
For BI?
Yeah.
They don't. There's not a lot of disclosure from BI about exactly their product.
Not from your compound.
Oh, OK. Our license with BI is quite narrow. We can make other Frizzled 4 molecules. We have Frizzled 4 molecules that contain VEGF inhibition as well. So maybe that's the most exciting thing we have right now, is both Frizzled 4 activation to improve this vessel integrity and a VEGF component to the molecule to inhibit VEGF signaling.
So why don't we talk about that in this last minute that we have?
Yeah, so I think what's exciting about that is you could inject both, ultimately. Anyone who follows the field knows that the field's moving towards less and less frequent injections. These patients do not like getting frequent injections. Like, no one does Lucentis weekly, and even some of the molecules are moving to every three or four months now. Even though you could give a Frizzled 4 and an Eylea, for example, in different injections, that wouldn't be desirable from the patient's perspective. If you could hit both those pathways with one molecule, that might not only be good for neovascular AMD, where VEGF is probably best suited, but sorry, for Diabetic macular edema. For neovascular, there are vessels in the back of the eye that aren't hit by Frizzled 4, so you'd really want to combine them.
We think for having the most breadth of application across wet AMD, diabetic macular edema, retinal vein occlusion. There's something called uveitic macular edema. Hitting multiple targets is going to be a way for one molecule to do it all. It looks like Frizzled 4 in DME is the best molecule. We think we're going to have something that's both the best and has the widest application across all the potential retinal diseases. We also have a molecule for the front of the eye for cornea, which is really exciting. It turns out Wnt signaling is really important in a lot of different tissues in the eye.
That's very exciting.