Hi, everyone, and thanks for continuing to join us on the Stifel two-day virtual oncology event. My name is Brad Canino, Senior Analyst here. Very happy to do the next fireside with Celcuity. We've got Brian Sullivan, CEO. Brian, thanks so much for joining us.
You're welcome. Happy to be here.
Perfect. Can we just kick off with an intro to the company and the state of the business as we're sitting here right in front of first pivotal data for the company?
Yes, yes. Sure. I started the company with our Chief Science Officer with the initial focus of developing a platform that could quantify signaling pathway activity in live tumor cells. We started working on the PI3K/AKT/mTOR pathway and have subsequently evolved to now focusing on how to treat cancers that involve the PAM pathway. It is the most important oncogenic pathway, or at least amongst the most important pathways in our view, most highly altered.
Our lead asset, Gedatolisib, is a pan-PI3K/mTOR inhibitor. We have three studies in process: two in breast cancer, one in prostate cancer. We have an ongoing phase III study that will report first data this quarter in one of the cohorts, wild-type cohort. That study is treating second-line patients who have ER-positive breast cancer.
We have a phase III study that we're just getting going now in first-line patients with ER-positive advanced breast cancer. We expect to enroll our first patient this quarter as well. We have an ongoing study in metastatic castration-resistant prostate cancer. That's a phase I B study. Expect to have top-line preliminary data from that towards the end of the Q2 .
How do you like to describe the how and the why is Gedatolisib's pulse and differentiated from the suite of other pathway inhibitors?
Sure. It is important to understand the structure of the pathway because that is really what drives the approach that we have taken, our drug has taken, and that is what differentiates it from the other drugs that have been approved in this space. Most pathways that people are familiar with involve a single target that can lead to downstream inhibition of pathway signaling.
The PAM pathway, PI3K/AKT/mTOR pathway, has multiple targets or nodes that must be targeted because each of them can cross-activate the other in the absence of inhibition. When this pathway was first discovered 20 plus years ago, almost every major pharma had a pan-PI3K/mTOR inhibitor because the biological imperative essentially required you to comprehensively shut this pathway down, hitting these various targets. Otherwise, you would have a suboptimal if or no treatment effect.
Gedatolisib then is able to hit at low nanomolar concentrations all class I PI3K isoforms, as well as mTORC1 and 2, the two mTOR complexes. Whereas the drugs that are approved hit what we refer to as single nodes, let's say PI3K alpha, mTORC1, AKT.
While they've gotten approved, we would argue based on the available data that it's a suboptimal approach because if we look at the non-clinical data, we would see that Gedatolisib is 300-fold more potent than these drugs when you analyze breast cancer cells or prostate cancer cells, tumor cells, relative to these approved single-node inhibitors. We think that just reflects the nature of the pathway and the importance of inhibiting it comprehensively to induce either half-maximal or maximal effect.
What is the rationale for Gedatolisib's pulse to work in patients who don't have oncogenic alterations on the pathway that's being targeted by the mechanism?
Sure. There are two factors I think that need to be considered. One is that the PAM pathway's role is, think of it as a junction, and it regulates various important critical cell functions, including the regulation of key metabolic processes that include glycolysis. That's particularly important in tumor cells, which require a tremendous amount of energy and glucose relative to normal cells to survive.
Essentially, they're hyperdependent on PAM pathway activity and the resulting regulation of glucose. That makes it a fundamental driver, independent of the presence of an activating mutation, particularly in tumor types like HR-positive breast cancer that involve other pathways that are linked to it.
The other factor that makes it possible for Gedatolisib to be effective, independent of activating mutations or makes the pathway relevant, is that it requires this comprehensive inhibition because essentially the critical role of this pathway has led to this essentially redundancy of these other nodes. The pathway can still function and perform these metabolic processes. It can adapt if, in a sense, one node is inhibited. In the absence of an activating mutation, you still need to comprehensively blockade it.
You may induce, and the data suggests that you get better effect even with a single-node inhibitor if you're only hitting a single-node inhibitor in cells that have that activating mutation. Interestingly, though, Gedatolisib, again, with the various cell lines that we've analyzed, has no difference in effect, in fact, the same potency and cytotoxicity, whether or not a mutation is present. We think that just reflects the nature of the pathway's role as a regulator of these key metabolic processes that, if inhibited, essentially disrupt tumor cells' ability to survive.
Yeah. Obviously, with the availability of the single-node inhibitors, patients are segmented between patients with PI3K alpha mutation and not, and the treatment paradigm has evolved based on that. Based on your phase I data, how close is the clinical activity between those two subgroups of patients with and without the pathway mutation?
The non-clinical data is almost identical. The preliminary clinical data that we have suggests they're very comparable, but there's probably a numerical advantage relative to it with patients that have a PI3K mutation versus those that don't.
In our early phase data, the objective response rate in patients that lacked PI3K mutations was 60%, 63%, whereas in patients that had a mutation was 73%. The PFS rate at 12 months, 49% in wild-type patients, 60% in patients that had mutations. A difference, and we would expect, as we report out data, to see a numerical difference. It wouldn't be a statistically significant difference, but I think meaningful potentially clinically in terms of the effect size difference.
The data you're describing are from a triplet. Obviously, fulvestrant is there, an endocrine backbone agent, which pretty much every patient gets across all lines, but also a CDK4/6 inhibitor. The question is, why is it important to continue that CDK4/6 inhibition when these are patients who have progressed on that mechanism in the front line?
Right. One of the most important, I think, findings over the past few years is recognition of the interconnected role that the three pathways play as a promoter of ER-positive breast cancer. The estrogen receptor pathway, CDK4/6 pathway, and the PAM pathway.
It appears, just based on the data we've generated, as well as various non-clinical models, that these pathways are intrinsically involved, and inhibition of one can drive and cross-activate the other pathways. There is a significant amount of non-clinical data that we've generated, as well as others in various labs across the world, that have found that tumor cells that become refractory, no longer sensitive to CDK4/6 inhibitors, that pathway can be reactivated, and then the cells resensitize to CDK4/6 inhibition in the presence of a PAM inhibitor. In effect, you eliminate that escape route.
You drive activity back down to that CDK4/6 pathway. You need to maintain that pressure. Essentially, you've got this circuit that needs to be closed in order to optimize control. That explains why in our first-line phase I B data, we showed very, very, we think, impressive results, 48 months median PFS in treatment naive patients, which we think provides a pretty good demonstration of, again, the intrinsic role these three pathways play at the outset of the disease. They are the driver of the disease.
You can potentially optimize treatment or require to optimize treatment blockade of all three pathways, either upfront on the first line or if you didn't do it upfront in the front line setting, then as a second-line option for patients. That is what we're evaluating in our various studies.
Let's talk about the study, VIKTORIA-1, second-line pivotal trial. What are the key design elements you'd highlight of that trial?
Sure. We're enrolling patients whose disease progressed after treatment with a CDK4/6 inhibitor and an aromatase inhibitor. These are second-line patients. That's standard of care for most patients. They're allowed to have had up to two prior endocrine therapies, but no chemotherapy in the advanced setting. We would expect most of these patients to be pretty much second-line patients. They can't have had, for instance, prior PAM inhibitors.
They're randomized according to PI3CA status. We're essentially enrolling all comers, but then assigning them to a cohort one or cohort two, depending on their PI3CA status. In the wild-type set, in the wild-type cohort, we have three arms. We have a triplet, Gedatolisib, palbociclib, and fulvestrant. We're comparing that to fulvestrant as the control. We also have a doublet, Gedatolisib, fulvestrant, and we're comparing that to the control as well [Inaudible] What the hell? I think I'm back. I'm back.
Yeah. You were paused for a little while?
Everything shut off. I apologize. That was very strange. Let me get my camera back, I guess. Here, hold on one sec.
We've got the wheat background.
Yes. I'm here.
There you are.
I apologize. No, literally, our entire system shut up. I think I was just describing the trial design. Actually, I'm sorry, I've just got to reconfigure my system here. Hold on. I've got to go off camera because I had something screw up.
No worries.
I'm back. Sorry about that.
No worries. Okay. It's a trial design.
It's an interesting trial design. In the mutant cohort, we have a triplet. Again, in this case, though, in the mutant cohort, we're comparing to alpelisib and fulvestrant as a primary endpoint. The two cohorts have independent statistical analysis plans. In the case of the analysis of the triplet versus fulvestrant and the analysis of the doublet versus fulvestrant, those are tested hierarchically. We're sparing alpha and making the design very efficient statistically.
Okay. How has enrollment progressed in this study relative to your assumptions? What do you think that means?
Sure. We have completed enrollment in the wild-type cohort. We announced that in the Q4 . We actually were right on track 100% with our projection from over two years ago for the overall enrollment of the study.
We found that the mix and the ratio of patients who lacked mutations versus those who had mutations was slightly different from what we had projected. We thought it would be a 65-35 ratio, just certainly variance. It turned out to be more of a 60-40 ratio. That meant that the wild-type cohort took slightly longer to enroll than our initial projection, slightly improved enrollment in the mutant cohort, but not substantially really changing things. The overall enrollment, we were very pleased with. Very, very engaged sites from around the world and kind of pretty much met our expectations.
In broad strokes, what effect size do you hope to see in these two different patient cohorts?
I mean, we'd like to see as big an effect size as possible. Certainly, what I can speak to is the available data for fulvestrant as a control. I think the data is fairly clear, at least in our mind. Patients who've received fulvestrant post-CDK4/6, there's been four randomized studies that have a population similar to ours.
That data ranges from 1.9 months up to about 2.8 months. We call it an average of, let's say, 2.75 months. We think it's most likely, just based on the results from those studies, that your control would be in that two and a half to three months range. We obviously reported in our preliminary study much higher PFS than that. We can't really, it's not appropriate for us to project. What we can talk about, though, and I think what's relevant is what's clinically meaningful.
What would be a win?
I think a win, and certainly more is always better. When you talk to KOLs or regulators, they define a clinically meaningful result as an improvement of roughly three months or more relative to your control. Now, there are different ways of measuring activity because I think in this setting now, what you have are studies taking place that are enrolling a mix of populations. You've got some studies enrolling first-line and second-line patients together or with or without prior CDK4/6.
Clearly, those patients have far different prognosis. The results that they would obtain from the same regimen are far different. The hazard ratio improvement will be very important to achieve as well. Again, we hope and expect that the preliminary data we have represents what we'll see in the phase III, but we'll see. The bar relative to what we've reported to date with our preliminary trial, the phase I B trial, suggests we have a pretty good margin of safety, margin for error to achieve a meaningful and important result.
Yeah. I would agree. The KOLs I've spoken with have suggested that three-month delta is being a meaningful metric to think about. A lot of the conversations with investors I have today are around the competitive intensity in this space. Some of that includes the CDK plus oral SERD doublets, which have emerged with, I would say, at least high single-digit absolute PFS results. The question to you, and I'd like to know how you'd like to answer this, is how do you see yourself competing in the space with these new benchmarks out there as well?
Sure. I think, for instance, you're probably referring to the EMBER-3 study. That's an example of a study that, to be frank, because of the way they designed it, makes the data hard to interpret because the primary analyses included patients who were first-line and second-line. In the second-line patients, some had and some did not have prior CDK.
If you were to talk to investigators, they'd say they do not know how to interpret that data. One way to interpret the data is to look at the hazard ratio. The hazard ratio was 0.58, in that range. We do not think it is unreasonable to think that we would beat that if the early phase data that we reported is reported in this setting. We'll see.
I think that that study had some other challenges that, again, may limit the applicability of the data and utility of that going forward. Ultimately, again, people will have to look at the relative contribution of the regimen compared to control because the absolute numbers are hard to use.
If you've enrolled first-line patients, you're going to see a higher absolute top-line median PFS number. That's just the way it is. Fulvestrant as a front-line therapy will offer patients potentially 10-11, 12 months. If you include those patients in your analysis, you've essentially goosed the top line. That doesn't necessarily represent what the second-line patients would get.
That'll be one of our responsibilities if we are commercializing this drug to ensure that we present our data in a way that is very interpretable and helps oncologists understand the relative benefits or potentially ways of assessing the various data sets out there.
All right. Okay. Now, back to VIKTORIA-1. What toxicity profile do you hope to achieve? Is the rate of discontinuation an important metric you're looking at for the study result?
Sure. I do think the rate of discontinuation is probably the best holistic measure of the overall tolerability of a regimen. That is certainly the feedback we have received as we have done our preliminary kind of commercialization preparation work because it just captures whether patients can stay on the drug. In this setting, there are drugs out there, PI3K alpha inhibitor, mTORC inhibitor, that had 26%, 24% discontinuation rates due to adverse events.
In our preliminary data, our phase I B data, with the phase III dose, the discontinuation rate was 4%. Now, if we look at the overall data set from that trial, 138 patients, the discontinuation rate was less than 9%. I think discontinuation rate of 4% to 8% is certainly something we think is reasonable to expect.
Again, based on the feedback we've received from investigators who've worked with and treated patients with some of the other drugs in this class, they see a highly differentiated tolerability profile. Again, that's preliminary data. We'll have to see. It was very encouraging for us as we began activation or selection of sites for our VIKTORIA-2 study, the front-line study.
At every VIKTORIA-1 site, the study we hope to report data out this quarter, who we wanted to participate in our front-line study wanted to participate. We interpret that as favorable. People had a good experience with the drug. The patients had a good experience in terms of general tolerability. They must have seen something that suggests that they wanted to participate in ongoing development of the drug.
Yeah. Now, one of the challenges with some of the competitive pathway inhibitors that we've seen launch recently in terms of interpreting their safety results is they have relatively strict criteria for who they select into the trial. When the drug goes out in the real world, the toxicity rates generally start to creep up. How does your criteria for enrollment around parameters that can affect the safety profile compare to some of these recently launched drugs?
Sure. We have over, or I think, nearly 500 patients of data that's been reported publicly. We haven't reported the phase III data. Referring to the published data for Gedatolisib, we have not had to create, for instance, a limiting HbA1c level, for instance. We have enrolled in those early phase studies patients who are prediabetic or type 2 diabetics, uncontrolled diabetes. Type 1 diabetics, we haven't been enrolled. I think what you're referring to is hypoglycemia.
Given the role this pathway plays in regulating glycolysis, that has been the third rail for drugs in this class. We don't expect to see any difference between what we're enrolling and what we're experiencing in our phase III studies versus what the real world will show. There is enough patients who have been treated with the drug that we think the profile, particularly around hyperglycemia, is well-defined, well-characterized.
Got it. Maybe last question for me. Assuming we get positive results in the next few months here, what could an NDA process for Gedatolisib look like? Given there still is an unmet need for a lot of these breast cancer patients, is there an accelerated timeline available for review?
There is. We have breakthrough designation for this indication. Typically, breakthrough designation is given, A, based on available data you've presented to the agency and the review of it, but also primarily because there's a significant unmet need. That really hasn't changed. There is a pathway called Real-Time Oncology Review, RTOR, that we'll request. If that is not provided or granted, there's the priority review. Given the big unmet need, given the preliminary data, we think it's pretty reasonable to expect we would qualify for some form of accelerated review.
Okay. Wonderful. Unfortunately, we're out of time. Brian, thank you so much for joining us. Look forward to hearing the news updates in the next few months. Exciting time.
Thank you. Yes, it is. Thank you.
All right. Bye now. Thanks everyone for joining in.
You're welcome.