Odyssey Therapeutics, Inc. Common Stock Investor update

Good afternoon, everyone, and thank you for joining us. Earlier today, Bruce Sands presented an update from our phase II-A monotherapy trial of OD-001 in ulcerative colitis here at UEG Week in Barcelona. His slot was only 10 minutes long, so we wanted to give you the rest. The complete induction data set, biomarkers of colonic and systemic inflammation, and mucosal transcriptomics from patient biopsies with enough time to walk through each piece of data and then take your questions. Next slide, please. Just before we begin, a reminder that today's presentation contains forward-looking statements, including statements about the clinical development, efficacy, safety, and therapeutic potential of OD-001 and some of our other product candidates, our regulatory plans, and the timing of future milestones. The actual results may differ materially, so please review the risk factors in our SEC filings. We will also show data from other companies' trials.

Those comparisons are for context only. There have been no head-to-head studies, and the data we present today are from an ongoing study, and the trial remains subject to change until there is a database lock. Joining me today is Dr. Tony Okapari, who was recently named our Chief Medical Officer. Tony has been with Odyssey since day one, and he has led the phase II trial, so he brings real continuity and expertise to the next stage of development. I have known Tony for over 30 years, and I could not be more pleased to be working with him and having him lead our clinical work. Also with us is Collin Todd, our SVP of Strategy and Business Development. We plan to spend about 48 minutes on the scientific materials and leave about 20 minutes at the end of the presentation for your questions.

In May, we reported top-line data from this trial, 27% clinical remission at 20 milligrams BID. The primary endpoint was met. Activity in patients who had already failed advanced therapies and a clean safety profile. We also told you that a broad suite of biomarker and transcriptomic analyses was underway, and that we enrolled additional patients at the 10-milligram low dose. Today, we are delivering on both. You will see the complete 10-milligram cohort, now 15 patients in total. Additional subgroup analyses by endoscopic severity and symptomatic remission over time, objective biomarkers of colonic and systemic inflammation, and transcriptomics on paired mucosal biopsies from baseline and week 12. Let me just say this once up front. This was an open-label study, active only proof-of-concept trial, and nothing we show today replaces a randomized placebo-controlled trial.

That trial, our phase II-B, starts this year, and we will read out at the end of 2027. What today's data gives us is more confidence in three things. First, there are multiple converging lines of evidence that inhibiting RIPK2 scaffolding underlies the clinical benefit that we observed with patient-level correlations across clinical efficacy, histology, biomarkers, and transcriptomics. Second, the 20-milligram BID as the lowest maximally effective dose, the dose that phenocopies a RIPK2 knockout. Third, efficacy across patient subsets with no serious adverse events, no discontinuations due to adverse events, which furthers our conviction that OD-001 can be competitive with approved and investigational UC therapies. Let me briefly start with the disease, the target, and how we drug it, and why going upstream has the potential to deliver differentiated outcomes for patients. First, IBD is two diseases, ulcerative colitis and Crohn's disease.

They differ in where they occur and how they progress, but they share the same pathogenic sequence. The intestinal barrier breaks down, gut bacteria translocate across it, and both innate and adaptive arms of the immune system become dysregulated. There are now about a dozen approved drugs across roughly half a dozen mechanisms. Despite that, placebo-adjusted induction remission rates in UC remains below 30%. Most patients do not achieve remission, and many that do ultimately lose response and cycle to the next drug. This is quite personal to me. My son has lived with Crohn's disease for 17 years, and I have watched him go from one therapy to the next, to the next. Patients need new therapies that are more effective, better tolerated, and that could change the natural history of disease. We believe that takes new mechanisms and combinations.

We are starting in ulcerative colitis because it is confined to the colon and supported provably by endoscopy, which makes trials faster and cleaner than in Crohn's. The important thing is that the biology we target is shared by both. IBD develops in three steps. In the first step, the epithelial barrier is breached. Bacteria enter the lamina propria. In step two, innate immune cells, meaning monocytes, macrophages, and neutrophils, and the like, sense bacterial cell wall components through one of two receptors, NOD1 or NOD2. RIPK2 transduces that signal, and RIPK2 activation also amplifies toll-like receptor signaling. The output of this is cytokines that you know well, IL-23, TNF, and TL1A. Neutrophils activated through RIPK2 continue to damage the epithelium, so the loop feeds itself.

In step three, adaptive immune cells, T cells in particular, are recruited to the sites of inflammation that have been activated in the gut. Let us have a quick look at where our previous late-stage drugs sit. The anti-TL1A, the anti-TNF, and the IL-23s, the JAK inhibitors, S1P1 modulators, vedolizumab, all act downstream on individual cytokine components of the adaptive immune response. OD-001 acts at the point where microbial sensing first becomes inflammation, so that rather than blocking one cytokine, the aim of inhibiting RIPK2 is to turn off several of these cytokine sources. That has two implications we will come back to. For monotherapy, the patient who has failed to develop against one downstream cytokine has never been exposed to this mechanism.

For combinations, OD-001 doesn't overlap with any mechanism on this side, and it acts on the innate rather than the adaptive immunity, which makes it a natural partner for them. The case for RIPK2 rests on over 20 years of both animal and human data. In animals, colitis models require microbes. Knocking out RIPK2 reduces inflammatory cytokines and improves intestinal histology, and blocking RIPK2 signaling pharmacologically does the same. Importantly for safety, RIPK2 knockout mice are developmentally normal. In humans, the RIPK2 pathway activation is increased in intestinal tissue from patients with both ulcerative colitis and Crohn's disease. Activated, ubiquitinated RIPK2 is observed across ulcerative colitis patients, including those carrying non-function loss-of-function variants. This is not limited to genetically defined sufferers. Variants in ATG16L1, an autophagy gene, and CARD9, both linked to this pathway, increase IBD risk, and protective CARD9 variants reduce it.

Now with our phase IIa clinical efficacy data with OD-001, there is now proof of concept that RIPK2 works in ulcerative colitis. On the left now I'm showing you how RIPK2 is activated. Bacterial peptidoglycan is recognized by NOD1 or NOD2. RIPK2 must scaffold with a protein called XIAP, which is ubiquitin. That complex activates TAK1, which drives NF-kappa B and MAP kinase signaling, and ultimately inflammatory cytokines. Hold on to the MAP kinase arm, which is something that we'll come back to. You can see that RIPK2 is activated based on ubiquitin status in both Crohn's disease and ulcerative colitis, and that the RIPK2 gene signature, which we designed, tracks with clinical severity from inactive to severe disease. It's something that we tracked in the clinical trials. Importantly, RIPK2 drugs must block scaffolding to fully inhibit NOD2 signaling. This is the design principle behind OD-001.

RIPK2 is a kinase, and the first-generation programs were kinase inhibitors. In 2018, two independent groups, one from Genentech, which published in "Molecular Cell," and one from an academic group, which published in the "EMBO Journal," showed that RIPK2's kinase activity is dispensable for NOD2 signaling. What matters is its scaffolding interaction with a protein called XIAP. Kinase inhibitors that block signaling did so to the extent that they blocked this interaction. If you want to fully shut down this pathway, one must block scaffolding. That's what we set out to do from the first day of this program. OD-001 is an allosteric scaffolding inhibitor. It binds to RIPK2 away from the interaction site with XIAP, and it holds the protein in a conformation such that it reduces the affinity for XIAP.

In the middle panel, it shows you this directly as OD-001 binds to RIPK2, which is shown in the red, XIAP binding to RIPK2, which is shown in the black, falls to zero. The right panel shows you the functional consequences of this interaction in human macrophages stimulated with peptidoglycan, OD-001 blocks production of IL-23, TNF, and TL1A in a concentration-dependent way to near complete inhibition. Three of the most validated targets in IBD are all inhibited by one oral molecule. The question that we get often is whether RIPK2 scaffold inhibitors are all the same. The answer is they're not. On the left is the pharmacology describing this. This is a three-component system. The inhibitor and XIAP can bind to RIPK2 at the same time, forming a ternary complex. What determines how well a compound blocks signaling isn't simply how tightly it binds to RIPK2.

It's how much it changes the affinity of RIPK2 for XIAP once it's bound. That is the Ki on this slide. Because it's a three-component system, the result depends on how much XIAP is present in the assay. A compound could look like a strong scaffold inhibitor at low XIAP concentrations and can lose much of that effect at physiological ones. On the right, we tested OD-001 against three RIPK2 inhibitors developed by others and disclosed in the patent literature. Here we're measuring TL1A production by human macrophages stimulated with peptidoglycan as the output. OD-001 drives TL1A production to near zero. Comparators were A plateaus with most TL1A production intact, whereas comparators D1 and D2 are less potent and plateau with roughly one fifth of TL1A production remaining.

How completely one blocks scaffolding determines how completely you block signaling, and that's why we dose for maximum inhibition in the clinic. Let me turn to the trial itself. OD-001 is the first innate immune-targeted therapy and the first RIPK2 scaffold inhibitor to enter clinical development for ulcerative colitis. We designed this phase II-A to answer a go/no question, is the drug safe and active, and is the signal strong enough to justify further development? We set the bar before the trial got started. The primary endpoint was changed from baseline in 3-component modified Mayo score of 12, power to detect a 2-point change. For the key secondary endpoint, which was clinical remission, the trial was designed to detect a placebo-adjusted improvement of at least 15 percentage points over a historic placebo rate of 10%. In other words, we were looking for a remission of roughly 25%.