Hi, everyone. My name is Maury Raycroft. I'm one of the Biotech Analysts at Jefferies. Like to welcome Brian Sullivan, the CEO of Celcuity. We're doing a fireside chat format. Brian, thanks for joining us. You're welcome. Maybe for those who are new to the story, if you can give a brief intro to the company. You just had a big update recently. Talk about that as well. Sure. I started the company about 12 years ago. Our focus initially was to develop a cellular analysis platform that could analyze intracellular activity quantitatively. We subsequently migrated towards development based on our research of a drug that could target the PI3K/AKT/mTOR pathway called PAM. Based on our research, we identified, A, what we thought the required mechanism was to address this pathway, and then we found a drug that corresponded to our supposition, called gedatolisib. As it turns out, gedatolisib was owned by Pfizer. We had a relationship with them because of the other research we were doing. They had decided to out-license it. We raised our hand and said, "Well, if we were ever going to get into the drug development business, it would be for a drug that hits this pathway and this drug." Now we have had two phase III readouts, in the second-line setting in breast cancer. We have two current ongoing phase III studies in the frontline setting for advanced breast cancer. We have an early phase study that's ongoing in prostate metastatic castration-resistant prostate cancer. Great. Yeah, it's a great overview. You just had the mutant top-line data reported at ASCO on Tuesday. Can you give a quick recap of the presentation and the most important efficacy and safety signals you're seeing there? Ultimately, we had a 3-arm trial design. We had geda combined with palbociclib, which is a CDK4/6 inhibitor, and fulvestrant, which is an ER inhibitor, tested against alpelisib. Fulvestrant/alpelisib is a PI3Kα inhibitor. Then we tested gedatolisib by itself, rather just with fulvestrant. This provided the first head-to-head comparison of two drugs hitting the same pathway, this PI3K/AKT/mTOR pathway. The triplet and doublet each showed over 11 months median PFS. Each showed a doubling of the likelihood of survival without disease progression or death. Very, very favorable results. The hazard ratios were 0.5, essentially indicating that we were significantly exceeding our internal estimates. I think the implied estimates out there. I think there's been some misinterpretation of those analyses. What were your internal estimates for HR? We had between 0.56 and 6, based on an assumption of about seven and a half months for the control arm, which is alpelisib, and then about 13 months for the triplet. As it turns out, it was 5.6 and 11.1 for that primary analysis. Right. Yeah. Yes, it's pretty impressive data. One of the observations which you mentioned is that the PFS for the doublet and the triplet were similar. Why do you think adding CDK4/6 isn't driving incremental benefit to this population? I think you have to think of it in two ways. We showed that there's a significant contribution in the wild-type setting with the CDK addition to geda and fulvestrant. We think in the wild-type setting, these three pathways play a more co-equal role. Clearly, in the mutational setting, PIK3CA mutant setting, the PAM pathway plays a more important role. It's probably a primary driver. CDK is less important, we don't see a benefit on a median PFS basis or hazard ratio basis. There was a higher objective response rate, nearly 50% with the triplet, 36% with the doublet. Clearly some additional anti-tumor activity, very strong duration of response. You'd say data's immature for overall survival, but you'd look at the overall survival curves and the data, that's more favorable with the CDK4/6 addition. We think there may be more of a latency effect or, again, more of a passenger effect than a driver effect. Still relevant when you dive into the data. Got it. Yeah, it makes sense. You have the response rate difference, and then it could potentially translate into OS benefit. Do you think the lack of separation could be driven by small numbers in the doublet? It could be a factor. I think sometimes PFS can hinge on a few patients. What was interesting to us was that half the patients had an objective response in the triplet. The duration of response was nearly 16 months. We had half the patients, the duration of response of 16 months, and an overall PFS of 11 months. It indicates that half the patients getting the triplet are getting very, very extended duration of treatment beyond the PFS. Right. Yeah. Interesting. In the mutant cohort, median PFS came in below phase I at the 14.6. Should we attribute this to baseline differences or are there other factors? There were baseline differences. I think a quarter to a third, I forget exactly, of the patients in that early phase study hadn't had prior CDK. That's a big difference. Yeah. You can see a doubling of outcomes in non-CDK-treated patients in this setting. That added some not distortion, but just some difference in the results. We also had a very, I don't want to say tough to treat, but the patients with significant tumor burden, 80% had liver lung mets. We had 15% that were endocrine resistant. Many of the studies done these days, particularly with the SERDs, exclude the endocrine resistant because they're just not responsive at all to estrogen receptor inhibition. We included those patients. They showed a very meaningful benefit. Those factors can result in a divergence of results from an early smaller sample size study relative to a global phase III. Got it. For the endocrine resistance population, are you saying more on what the benefit was in those patients? We have. It's in the subgroup analysis. I forget the hazard ratio, but it's still a very favorable hazard ratio. Got it. Okay. When should we expect final OS data from the mutant data set, and is wild type still on track for first quarter 2027? I don't know if we've said first quarter. I think first half would probably be a better characterization of wild type, and then probably closer to sometime in the second half. Less predictable OS, smaller number of events, and it's event-driven, not time-driven. Got it. Okay. In real-world practice, especially in the community setting, how do you see doublet versus triplet being used, and would PIK3 testing be required to guide CDK4/6? Sure. In the wild type setting, we expect the doublet and triplet to both be approved. We're nearing the PDUFA date, which is July 17th. Our research indicated that there's, based on quantitative assessment from a survey, about an 85/15 split between preference for the triplet versus the doublet. That was consistent with what we saw when patients were crossed over, a similar proportion by doctors deciding that. There'll clearly be a less heavily weighted preference probably for the triplet. We think there's still an argument for the triplet, or rather, I think doctors could still believe that, particularly for patients, let's say you're a premenopausal woman, younger, you probably want to make sure, or at least we've gotten this feedback, that you want to essentially control all these pathways. The benefit to them of making that decision is that it's not a binary decision. They can ultimately decide, look, palbo, my patient's not tolerating it well for whatever reason. I can discontinue that and continue on with this regimen. I think ultimately what we've created with two regimens, to enable physicians to potentially improve the tailoring of an option based on the patient's profile. Oftentimes drugs, most of them are one size fits all. Some patients may not be suitable for it, and so they're not treated. In our case, for instance, you'll have subgroups of patients who have particularly aggressive disease. You absolutely would probably want to have the triplet. You may have other patients, a 75-year-old woman with more indolent disease. You may not include palbociclib. The fact that we're giving the doctor flexibility to tailor, also the flexibility to modify the regimen without affecting the positive benefit that could be induced is actually, we think will accrue in the long term to higher penetration than we might otherwise be able to get. Yeah. It makes sense. Yeah, definitely makes sense. For the wild type data, you saw some differences based on geographic split. You're not seeing that in the mutant setting. I guess what drove the regional differences, and if CDK4/6 use was a factor, what specifically differed across the geographies? Sure. In the wild type setting, these were patients lacking PIK3CA mutations. We saw U.S. patients get 19 months median PFS. U.S., Europe, Western European countries, and Japan, overall, when you aggregated that data, which was about two-thirds of the patients or 60%, was about 16.6 months. As we dug into the data, we saw that there were differences probably associated with CDK4/6 usage, palbociclib usage, in terms of dose reduction or interruption. There was also interruption of gedatolisib or dose reduction that occurred in these countries disproportionately. Geda doesn't, for instance, introduce any myelosuppression like neutropenia. In these countries, you saw significant alignment of reduction of palbociclib and geda for those reasons. Well, they shouldn't have. We think that, in fact, affected the results, because we saw a correlation in the countries with palbociclib penetration. In the mutant population, we didn't see that variance, and it probably relates, again, this is speculation, but I think it's sensical, is that the palbociclib played less important role. Independent of how you may have managed the palbociclib, to the extent you maybe didn't have an optimal management, it would've had less effect. It's just less dispersion potential based on the less significant role CDK4/6 was playing. Got it. Okay. That all helps, makes sense. Given the efficacy and safety profile to date, how do you expect uptake to look in the second-line mutant setting? How do you think the physician decision tree will evolve versus standard of care? Well, we think the reaction, we were at ASCO just this week, so coming from Chicago to here. It was a great meeting for us. It was great data in our view, and also in the view of the KOLs we met with. We presented many of these KOLs. These are important docs doing a lot of research. On a confidential basis, we previewed the data with them just to get their take and see if we were missing anything, and it was universally positive. They just thought, "This is great data. It's unequivocal. You guys should be the standard of care." Some of them actually got more specific and said, "Well, I was hoping you could do 0.65 hazard ratio. If you did that would be a home run." Obviously, we did 0.5, so that's better. Lower is better with this metric. We came out of ASCO just feeling, oh, this is great. The reaction will be very positive. Obviously, some investors had different expectations, but ultimately, I think that'll resolve because this data really is outstanding, and you really rarely see in a study that compares two drugs of the same class head-to-head, a hazard ratio that low, 0.5 hazard ratio, a doubling of efficacy. Yeah. Makes sense. I guess maybe just digging deeper into the question, though, as far as market research goes. What percent of the mutant population do you think you can capture versus standard of care? A lot. No, I do think we'll establish a new standard of care. We don't want to get too specific in our forecasts, but again, it's not often you have such a wide separation. The drugs are also better tolerated, so we win both on the efficacy standpoint, we also think we win on the tolerability, from the tolerability perspective. That would certainly seem to translate to a significant preference for our regimen over the other guys. We think we'll be the ultimate winner. Yeah. In that segment. Yeah. Makes sense. I guess another way to look at it is that our penetration estimates that we have in our model are pretty conservative. Would you agree with that, or disagree? No, your conservative penetration assumptions, no. I think how you model it needs to be thought about, because essentially you have to take into account what is a reasonable penetration assumption and then factor in what's the time to achieving that penetration, peak penetration, and then you assume a linear achievement of those penetration targets. If you work backwards from those assumptions, you'll come up with, assuming your penetration target's accurate, a reasonable estimate of revenue potential over time. Got it. Yeah, makes sense. For the wild type approval coming up soon, July 17th, where does the review stand today, and are you already in labeling discussions? Do you want me to give you the play-by-play here? Absolutely. No. We're approaching the end. We have a priority review. They're still reviewing it. That's good. We haven't been kicked out yet. No, we're very optimistic. Again, we have breakthrough designation, which gave us the opportunity to interact with the agency on a regular basis, asking questions along the way about topics that are relevant for a new drug submission. We came into the process and incorporated in our NDA a resolution to the extent there were any questions upfront. We really, we think, minimized the risk of getting surprised because we asked all the hard questions, tried to get alignment to the extent there was anything else that the FDA would've wanted. We resolved that upfront. Got it. Okay. As far as the label goes, what are the items you're focused on there as far as population, dosing, safety language? Yes, we are focusing on all of those. Okay. Obviously, the FDA's responsibility is to create a label that communicates effectively to doctors what they should expect and how they should think about managing patients on the label. That's totally as expected and appropriate. We expect a label that will be informative, and appropriate. Yeah. Okay. Makes sense. Can you update us on launch readiness, what's been completed so far, and what still needs to be done? Sure. We began the commercialization preparation process a couple of years ago when we brought on our Chief Commercial Officer, subsequently built out senior people in the marketing, commercial operations, market access, sales areas, medical affairs, and then completed build-out of those organizations, except for the sales force last year, last fourth quarter, and then brought on our sales force. Today we have all of the sales reps, they're all trained. They've all been released into the wild. They're not able to market or sell, but they are able to profile, and doctors understand what their practice habits are, their perspective on patients and potential patient subgroups, just to essentially get familiar with the doctor and their general clinical approach to treating these patients, which is a good way to lay some foundation, going into a launch. Right. Yeah, makes sense. For the first 90 days of the launch, what are the key milestones that you want to achieve there? Maybe talk about access, accounts, patient starts. Sure. You've got a lot of different variables that we're focused on. Obviously, from a market access standpoint, we want to have submitted and have moving forward dossiers at all of the national accounts, the payers. From a strategic account standpoint, we've been having these discussions, again, for 18 months with them. We can't move forward specifically, but we essentially are making sure the constituency that is going to ultimately make decisions about the formulary or the pathway preference, et cetera, is fully informed. We want to get all those dossiers submitted. We want to get NCCN submitted immediately. From a strategic account standpoint, similar to the payers. From a sales standpoint, that's all groundwork that's required to support your sales efforts. From a sales standpoint, we want all of our reps to have obviously met with all the doctors in their territory. Each of the reps have roughly 80 docs or so that they're responsible for. We want them to hit certain targets for number of doctors prescribing, number of docs repeating prescription, writing scripts, and overall level of interaction. There's some other activities that are relevant that we expect the reps to perform, and those will also be tracked. What are the targets? I'll get back to you on that, Maury. All makes sense. Based on the market research, how do you expect uptake to differ between community versus academic settings? Given this is IV infusion, what are the initial bottlenecks you expect? Right. There that should ease over time? Our research doesn't indicate wide separation in usage between an academic center and a community setting. Community setting will account, though, for roughly 80% or so or more of the patients treated. Obviously that's where we need to make sure we do a great job. A good chunk of them are affiliated with larger networks like US Oncology, Florida Oncology, Texas Oncology. You work with them because there's a bit of a top-down component to that decision process. Work accordingly. Over at ASCO, one of the doctors we spoke with mentioned that for the community setting, there could be capacity issues. Yeah. It's interesting they mentioned that. Yeah, just one thing to point out. In the breast cancer setting as well as many other settings, infused drugs are a very important component of their practice. In the breast cancer, Herceptin, PERJETA, $10 billion revenue drugs at peak, pembrolizumab and HER2, TRODELVY, all billion-dollar drugs. Every patient ultimately will get chemo, that is often weekly infused. They've built practices around infusing patients with drugs. It's a well-worn operational procedure that they have. We can essentially plug and play into that. I haven't heard one person talk to me about capacity. Yeah. I don't think it's a problem. Okay. For launching in second line, are there subpopulations of patients that you think could be early adopters? Well, I think certainly, patients with aggressive disease. Again, doctors, I think it's more a function of the doctor, and their comfort launching new or rather, prescribing new drugs. Doctors are somewhat like consumers. You have early adopters or experimenters. You have mainstream, and then you have the skeptics. There's a continuum of folks you work with. Certainly, as we're doing work profiling accounts, to the extent we can characterize whether these people seem to fall in the early adopter category versus a skeptical category, you focus on the early adopters. Part of it is understanding, based on their individual preferences, who they think are the first patients to treat. It's really going to be territory by territory, doctor by doctor targeting. Got it. Yeah, makes sense. By patient subgroup. Yep. A lot of it will just be a function of where the doctor's preference is. Yeah. One of the advantages of geda, initially the drug, we expect to be approved in patients lacking mutations, and we've now demonstrated that the drug is very active and superior in the mutant setting. Ultimately, our position will be that we are giving doctors, clinical oncologists, the medical oncologists, the ability to utilize a single treatment. Yeah. That's suitable for all patients who've progressed after their first line of therapy. With having both data sets in the wild-type and mutant, it builds out the value proposition of the drug. Doctors theoretically wouldn't have to test in second-line. It's self-reinforcing because I think as doctors typically get more experience with a drug, they get more comfortable with it. Once they get comfortable with the drug, they tend to stick with it. We saw this in a first-line setting with palbociclib versus ribociclib. Palbociclib was by far the preferred CDK4/6 inhibitor in the first-line setting. Probably had a 60-30-10 split between the three drugs, with palbo being the lead. Now it's probably a 50-40 with ribo in the lead. Ribo flipped the script because they got favorable OS data, but palbo's hung in there because it's a good drug. It's a great drug. Patients tolerate it well, and they're comfortable with it. I think over time, by having this ability to treat essentially an all-comer population, will lead the doctors to use this drug very frequently, and in turn feel very comfortable with it. Our goal for them to consider this their go-to option for all their patients who've progressed after their first treatment. Makes sense. When you think about the value proposition, how does that translate to how you're thinking about pricing for the drug? Could it be more in line with CDK4/6 inhibitors or alpelisib or something different? No, I think the second-line drugs which were launched later, and if you look at the second-line drugs in this setting, you'll see them in this $25,000-$27,000 wholesale acquisition cost range. That certainly sets a benchmark. I'm not saying that's our price, but that gives you a perspective or frame of reference for what to expect for drugs of this type in this setting. Got it. How do you view the competitive landscape and wild type over the next approximate five years? Are there any specific mechanisms you're focused on? Well, I think fundamentally, by addressing all three critical drivers of this disease, ER, PAM pathway, CDK4/6, you're optimizing the potential outcome you can offer patients because these three pathways are linked, and you're going to optimize benefit by hitting all three. PAM pathway plays an outsized role, clearly, in particularly the mutant setting. We think over the next five years, we'll be well-positioned to have essentially the superior regimen that's going to offer the best potential outcomes for their patients. Yeah. Makes sense. Wanted to talk about the frontline setting as well. Sure. Which could be an even bigger market opportunity. You recently updated that you're going to include endocrine sensitive patients. What are your expectations for enrollment timelines in endocrine resistant versus endocrine sensitive cohort? Sure. We've set up VIKTORIA-2. VIKTORIA-2 is essentially two studies in one. Endocrine resistant patients, these are women who progress almost or recur very quickly after they were diagnosed with early breast cancer and develop metastatic disease. Endocrine sensitive, which is two-thirds of the population, are women who recur much later. That's relevant because they respond differently to current standard of care therapy. We're enrolling these patients essentially simultaneously. We're intaking patients. If they are endocrine resistant, they would get assigned to study 1. If they're endocrine resistant, assigned to study 2, we think we'll enroll proportionately. Essentially endocrine sensitive is about two-thirds of the total number of patients. Resistant, about one-third. We've got more patients, though, in the endocrine sensitive population, I think 740 versus 440. It'll just be naturally, that may balance out where they can enroll somewhat proportionately to completion, roughly the same period of time. There'll be a much longer follow-up. The duration of treatment for current standard of care in the endocrine sensitive population, it's around 24 months, two years. Only seven in the other indication. We think today current estimates would be around end of 2028 for data from the resistant population study one, and then in 2030 timeframe for the patients in the sensitive population. Got it. Makes sense. In the endocrine resistant frontline setting, there's no clear benchmark from gedatolisib. How are you framing expectations for activity in this population? Well, again, with the first line, rather the second line data we've shown the data in the wild type population improves outcomes relative to fulvestrant monotherapy by nearly four-fold. It's clearly active 5+ months just on a duration of time off of a two-month delta. If we are able to achieve similar deltas, we'll create a statistically significant, clinically meaningful result for these patients. Got it. Inavolisib is approved in frontline endocrine resistant, but only in the mutant setting. Is that data a relevant benchmark for your frontline endocrine resistant all-comer? I think it's a relevant benchmark. That drug, because of the hypoglycemia it induces, is really only appropriate, as it turns out, for patients who are pre-diabetic or diabetic. Which unfortunately includes about half of women diagnosed with breast cancer. It's a very narrowly targeted drug. Ultimately, we're positioning our drug as an all-comer independent of your HbA1c status, your glucose level status, or your PIK3CA status. Our study endpoints are all-comer, intent to treat. They're not separate endpoints by PIK3CA mutational status. Yeah. You report 48 months median PFS from the gedatolisib phase I-B study in the endocrine sensitive data set. How should we interpret that versus what's achievable in the phase III, given the baseline differences and inclusion/exclusion? No, I would like to think we could do 48 months, I think that's dreaming a little bit. I think the benchmark is pretty clear. All three of these current CDKs deliver similar results, about 25 months for these patients. Clinically meaningful benefit in that setting is probably at least five months. If we've shown in an early phase study 48 months, and to be clinically meaningful, we need to maybe hit 30, we think that that's enough margin for error to have a lot of confidence. We think gives us the kind of margin for error that makes it highly probable, or at least in our view, very high probability of success. Got it. Okay. About a five-month- That would translate to both, we think, statistically significant and clinically meaningful. Got it. Okay. You also announced a subcu formulation for gedatolisib. How materially could that change the commercial uptake, and what do you need to do there? Anything more on timelines for? Sure. Getting there? Subcu formulation is really designed almost and intended to be developed in parallel with the development of the indication for the endocrine sensitive patients. We don't think penetration will really be affected by IV administration. Certainly subcu, when it's available, would be preferred. Absent its availability, it won't, we think, limit our penetration. However, in the endocrine sensitive population, where these patients, if the data bears this out, is on for two and a half to three years, we think subcu would be important to optimize penetration of that. We expect the data and an approval, potentially, if everything shakes out the way we'd like, in the 2031 timeframe. That's when we think we will have, and that's what we're targeting, is to have that subcu available in that 2031 timeframe. It's typically a five-year development cycle. You look at some other programs done in other companies, it's a four- to five-year development process. Yeah. Okay. Makes sense. Also wanted to ask on, there's been discussion around the Relay data, where they showed 11 months data in the phase II. You guys have two phase III readouts. I guess, what are your views on how to contextualize? Well, in general, it's hard to place as much weight on a phase I-B study as a phase III study. I also think in these studies, it's really important is to analyze the baseline characteristics of the patients because there are some variables that can yield significant differences in these characteristics. One characteristic that can significantly affect the results is the inclusion of patients with non-measurable disease. It was interesting, in our study, we saw that patients with non-measurable disease who got fulvestrant responded four times as well if they had bone-only disease versus those who had measurable disease. If you have 40% of your patients, let's say, or a significant proportion, have non-measurable disease, you're going to create an upward lift. Yeah. In your overall number. Typically, in phase III studies, you're enrolling patients who have measurable disease because you need to assess PFS very reliably. Secondly, I think in general, what we've found, certainly with the SERDs, is that if you have a similar mechanism, even if you have different ways of achieving it, which a number of these SERDs are different, you find that there's only so much biological potential out of a target. There's other drugs under development, alpha inhibitors, PI3Kα inhibitors. Our data, as it turns out, is 10 times more potent than a couple of the, like alpelisib, for instance, or the Relay compound. We do a good job of inhibiting alpha. What's more important is that we're shutting down the pathway by inhibiting all the other critical components of this pathway. Ultimately, that's what we think is important. The SERDs really haven't shown very significant differentiation, even though they have different potency, different mechanisms to address your target. Ultimately, the biology will win out, and there's only so much juice you can get out of the squeeze. You can't inhibit more than 100%. Right. Yeah. Makes sense. We're pretty much out of time. Maybe in closing, if you just want to comment on cash runway assumptions and key updates that investors should focus on. Oh, sure. We just closed a convertible note offering. We think that convertible note gives us cash, certainly through 2029. We raised $500 million has a greenshoe, where we could take down another $75 million. We'll pay down about $130 million of debt, we will net, if the greenshoe's exercised, around $420 million. That's a good addition to the coffers. We had $380 million at the end of first quarter. On a pro forma basis, $900 million, we think that'll carry us quite a ways. Got it. Okay. Thanks so much for joining us today, Brian. Oh, you're welcome.