Welcome, everyone. I am Josh Schimmer from the Cantor Biotech Equity Research Team, and very pleased to kick off our annual healthcare conference with Arthur Kuan, President and Chief Executive Officer of CG Oncology. Arthur, welcome. Thanks for joining. You have a lot going on. Maybe just take a few moments to set the stage and frame where CG Oncology is in the cretostimogene program.
Sure. Thanks, Josh. CG Oncology is developing cretostimogene, which is an oncolytic immunotherapy for the treatment of cancer. Our initial focus is in non-muscle invasive bladder cancer. The landscape is evolving rapidly, but really, I think we have built a body of evidence to suggest that creto can work across all of NMIBC, starting in high-risk BCG unresponsive disease, moving into BCG exposed, potentially BCG naive, as well as in intermediate-risk non-muscle invasive bladder cancer.
Just today, happy to share that from the BCG exposed results earlier this year, we are actually planning our third phase III trial, which we can get into as well. This will be in the BCG exposed high-risk setting, comparing cretostimogene plus gemcitabine versus gemcitabine. I am pleased to share that we have had very positive FDA alignment on the trial concept. The full details of the trial design will be revealed later on once we have our 12-month complete response rate in the creto plus gem data that is going to be available later this year.
Is gemcitabine the standard of care in that patient population? If not, how do you make that a control arm?
It's a great question, and one that we asked the FDA directly. It turns out that for patients with BCG-exposed disease, technically, you should give them more BCG, and then they could potentially turn into BCG unresponsive disease. However, in an era of BCG shortage, what we're noticing in the community is many practices are actually just giving single-dose gem. It's based on that feedback that we've gathered from the real world and from sites that we work with, this turns out to be a pathway that is acceptable both from a regulatory perspective as well as from a clinical feasibility perspective. We may potentially also offer up the opportunity, if you go on single-dose gem, to access creto subsequently.
The control arm would be a single-dose gemcitabine?
I wouldn't say single dose. I should say monotherapy gemcitabine.
Okay. Got it. It sounds like a hybrid between PIVOT and BOND.
Yes.
Yeah. In terms of the administration of cretostimogene, would it be more similar to BOND or PIVOT? Because they're slightly different-
Right.
...approaches.
I think when it comes to high-risk disease, we tend to go beyond a year. Which is something that we focus on given the higher risk and stage of that disease. Based on our Cohort CX data, this was combining creto plus gem, we tested it in two different schedules. Currently, we're leaning towards the sequential schedule. Simply, the data looks very comparable both in the high 80s, 90s. It really comes down to tolerability and convenience. The sequential therapy seems to be more tolerable compared to the concurrent administration schedule.
When you say sequential, what would the regimen be? Is it one then the other, then one then the other? Is it three weeks of one, three weeks of the other?
Exactly. Basically we tested creto, gem as the sequential regimen. That forms the basis of it. In the induction course, it's actually creto, gem, creto, gem. You basically take out one of the cretos and replace it with gem. On the maintenance, it simply would turn into creto, creto, gem. Instead of three creto weekly administrations.
Okay. Got it. Why don't we come to PIVOT-006 study? There are certainly a lot of investor questions around that, and in particular, the timing that was first accelerated and then a little delayed, but seemingly for very different reasons. Maybe walk us through what's been going on with the timelines and where we are towards unblinding that data.
Sure. When we designed and launched PIVOT-006 early on, we really didn't have a whole lot of information to go by when it comes down to the control arm and how that control arm might perform. In the literature at the time, there was only the SWOG study that compared a single-dose gem versus saline. That was the information available at the time. Later on, after we started the trial, UroGen's ATLAS trial read out, and there was a control arm component to it that was just TURBT without any therapy. That was roughly around a 50% RFS at the 15-month time point based on the Kaplan-Meier curve. It's with those two information, as well as our own experience enrolling patients in this study. We enrolled across the U.S. in 90 centers.
Based on our own internal projections, it really beat that by about nine months, so timelines were moved forward. Initially on clinicaltrials.gov, we had a projection of initial readout somewhere mid next year. Obviously we think that that is not accurate. There was initial guideline change to first half of this year. It's with that, a lot of it is based on the overall event accrual number that we have access to. I think there's a lot of questions around. We had an AUA abstract placeholder that was withdrawn. The context there, and I've shared this as well, is this is a very high-profile study for the urology community, and people are eagerly awaiting for those results because it'll be the first time where a prospective randomized controlled study was done in the intermediate-risk population.
Everyone's really curious about what the control arm could do, what creto could do. When you have a high-profile study like that, we do get inbound outreaches and collaboration discussions with these medical conferences. That was a case where if we had data, it kind of allowed us to present that information at AUA. Unfortunately, the events weren't hit at the time. As of August 6th, in our last 10-Q, we've said that we've basically hit the majority of target events.
That's important information. I think this is, again, just to remind everyone, it's an event-driven trial. Oftentimes, it's very hard to predict when the final events may come in, so we can only do our best to make that estimation. Today, I would still say that we are definitely anticipating results in the very near term, which is also consistent with what we've said a month ago.
Will you announce when the majority of target events have been hit or answer a question as to have they been hit by now?
Right. I think our approach would be simply just disclosing the top-line results.
Okay.
Yeah.
Got it. So it is possible you've hit the number of events, you're just not going to say one way or another?
Right, yeah.
Okay. So intermediate risk. Well, actually first for PIVOT, what is the powering that the trial is?
Yeah, so we've not disclosed the powering assumptions, but we have said that what would be meaningful clinically would be a 30% relative risk reduction compared to the control arm.
And there's no second year of treatment in PIVOT, whereas there is in BOND, is that evolving or why not give a second year of preventive therapy?
No, that is a great question. When we looked at in an era where BCG was not in shortage, BCG was actually given over a year, right? Many intravesical chemo, for example, were also given over a course of a year. We did not want to overburden this patient population with a three-year maintenance regimen. However, I think when and if we do establish that there is a clear answer here, optimizing scheduling is always something we can work on downstream.
Would that be in a randomized trial setting or not?
I think there are different approaches one can think about that. Could there be reinduction opportunities even in that setting where progression risk is low?
One common investor debate is really trying to understand how similar intermediate risk is to high risk and the extent to which we can borrow the data from high risk because we do not really have any data, at least until PIVOT comes out, to inform how cretostimogene might behave in intermediate risk. One obvious difference between the two, intermediate and high risk, is percent of FGFR3 activating mutations. I am not sure if you have any data to show that cretostimogene works comparably in FGFR3 mutated tumors versus not. Possible that you have been able to generate some of that along the way?
That's a great observation. I would say that, first of all, in PIVOT-006, we do intend to look at FGFR3 alterations. That's going to be an answer that hopefully will provide a clear answer to at some point once we have all the samples analyzed. Prior to PIVOT-006, you're right to point out that in high-risk, high-grade disease, which is predominantly carcinoma in situ. In high-grade disease, they tend not to start with an FGFR3 mutation. FGFR3 tends to start in more low-grade luminal or papillary disease. There is a sizable population of high-grade papillary tumors in high-grade disease. But the prevalence of FGFR3 is not that high in those settings. I think CIS is still enriched, and there's very little FGFR3 mutation in CIS.
In terms of whether we have evidence of this, in BOND-003, we did look at some urine sample analysis on urine ctDNA of FGFR3. Very few were positive. But in those that had a positive urine ctDNA, it seemed like creto was also working in that setting. But again, it's just directional. It's a very small number. I wouldn't read too much into that. I think the main point still is can creto work in an FGFR3-positive tumor? And the way I'll kind of reframe that is I go back to creto's MOA.
All we need is E2F1, the transcription factor that activates creto. And so the question I'll raise is, well, is there E2F1 in intermediate-risk, low-grade, or high-grade tumor types? And I shared this example. Whenever you resect a low-grade tumor, it often comes back at a different location. That points to field cancerization. If it were to keep coming back at the same spot, then I would have a different answer. I do believe there is E2F1 and what's the exact threshold of E2F1 that is necessary for creto activation? We don't know exactly, but I can tell you that all we need is some E2F1 to kickstart that replication cycle.
Okay, got it. Why don't we come to the creto BLA filing now for high risk, because that's another very common topic for discussion amongst investors, and maybe just help frame how it's been evolving, what the final gating steps are to complete the submission, and your confidence that when it is complete, it'll be airtight on CMC.
Yeah. I think many investors and your readers have expressed that, hey, it seems like manufacturing is a black box, right? You do not provide as much timeline, a detailed Gantt chart on what is going on compared to clinical trials. I would just say conceptually, what we are dealing with here is a replication-competent vector. Conceptually, this is something that has a very high yield and compared to something that is non-replication competent, this is off-the-shelf, so it grows in bioreactors. So it is something that is actually very scalable, and we have commented that this process is actually very robust. In fact, we have actually completed all the necessary BLA-enabling studies related to manufacturing. For example, process characterization. This is one thing that when you go from clinical to commercial, you need to do extensive characterization studies.
What I mean by that is you have to stress test all the possible parameters that could possibly break the process. So we are tuning the pH levels, oxygen levels, et c. That data package is available, and it actually points to how robust the process is. So that is one thing that if I were on your side of the aisle, I would definitely focus on that first. The second piece of it is process validation. This is basically once you have locked down your process and you continue to repeat that at basically your commercial scale. The good news is what I have shared initially is we have a very high yield with this process. For our initial commercial scale, we are actually not changing the scale at all. It is still the same scale.
We will eventually change the scale in anticipation of our future demand and the number of patients we can serve. But for the first BLA, we do not intend to change the scale, and we do believe that we are going to have plenty of inventory sufficient for that launch, well into that launch. So really what it comes down to is there is a lot of studies that went into studying the different parameters, all the testing, and we are in this stage where we are diligently compiling all those reports and then putting them into the right formatting for the Module 3 , which is the only module that remains in our BLA.
That work, it definitely takes time, but we are confident that if we get it done correctly, we could avoid possibly their downstream consequences of rushing into it. Because again, as a reminder, this one Module 3 would also support our PIVOT-006 BLA in intermediate risk and possibly other BLAs to come. So this is basically a one for many situation.
What part of it has taken longer than originally expected? Why that?
Yeah, I think the quantum of reports that are needed is definitely one. Basically, the more extensive your characterization is, the more reports and data that are going to be needed. The good news is they are all done. That is the good news. This is now in the stage of documentation and generating these reports that are fit for the BLA submission. Our team is working diligently on this. I think there are always investors obviously who want more clarity and more communication on this. I think our goal is to make sure that once we have any timely information around this, we will always be very quick to share with investors the latest update.
Was there anything unexpected that came up during the process characterization or validation that required additional work that you had not initially anticipated?
I would say that the FDA, through our conversation, given our BTD status, they have been very open in discussing with us what is needed and what is not needed, what is necessary for BLA versus post BLA. We feel like there is very good alignment. Again, as a reminder, the good news is all the assays that we are using is at the same site, and they have all been validated. That always helps when you are doing these report generations.
I would say that the Biovire activity, this is our drug product fill and finish facility, this is something that we've spent a lot of time on. We've invested a lot of capital into it, and we're at a very good place, in my view, getting them ready for inspection readiness, which is the other work stream that is also going to happen once you've filed the BLA. Once it's accepted, we can expect the inspection to come at any time.
Okay, got it. Now, in hindsight, if you had started any of these endeavors a little earlier, because you've been working over manufacturing for years, so it's kind of hard to imagine it was a matter of getting ahead of things when you have been so far ahead. I think we're all still trying to, again, understand the timeline and maybe a little bit of slipping there, and if it was just natural prep, working through processes, or if some of these things in hindsight, no blame, but just-
Yeah.
Again, to clarify where we are in the moment.
I think you've been following the oncolytic virus space for 10 years now, since your 2016 Bible. The way I would describe this is oncolytic virus as a class, there's now two approvals now. Both of those are in-house manufactured. Our shareholder base and investors early on opted in for outsourcing manufacturing. I don't think there's right or wrong on this. It's some investors prefer to build their own day one in anticipation of a potential approval. We took on the outsourced path.
Now, the number of CDMOs that have experience with a commercial oncolytic viruses, obviously there's only two approved ones, and they're both made in-house. So there's relatively a lack of experience on that front. I think over the years, and it's more thematically on a macro level, most companies would face that challenge. We were also kind of. This was years ago. There was a time when cell therapy was very hot. They also require making lentiviruses. You got a lot of capital going into that cell therapy lentivirus manufacturing mode. It could also kind of compete against-
Yeah.
...the attention on just pure-play oncolytic viruses. This is more on a high level. I would say we're at a very good place now, given the clinical conviction that we have and the data that we have to support that conviction and the resources that we have. It's very different if you were a startup right now working with a new CDMO. It's going to be a lot challenging.
Okay, yeah.
We've gone through that phase already.
Thank you for that color. Why don't we move on to market dynamics and another very common question that comes up is for high-risk NMIBC, how large is that unmet need? In reality, I think we're seeing a number of product launches into either high-risk or intermediate-risk, more or less a similar cadence. They're kind of hitting maybe a couple hundred million run rate and still growing. Help us understand the market dynamics, because one question that comes up often is why aren't these products' adoption curve faster? Is it a reflection of the limitations of the product or the limitations of the market? How would you address that?
Yeah, no, that's a great question. I think each of the competitive launches, the branded products, all have their own idiosyncratic challenges. I won't go through each one of them in detail, but what's clear is you can see that pre and post J-code, there's a clear kind of shift in ramp in revenue. Taking the intravesical gemcitabine approach, you'll also notice there's a ramp post J-code. The lesson learned really, I think, our customers definitely, the larger oncology group practices, they definitely are very sensitive about the J-code. That's one key lesson learned. I think that this launch, I think we're going to continue to monitor how the competitors are doing. I think what's most important for us is to profile who might be the early adopters, who might be the mid-cycle adopters, and late adopters.
I think what we've been doing as a company is to continue to invest in clinical research across different indications. It actually helped CG and cretostimogene establish our profile and brand awareness. Some of the early adopters that we're seeing may not entirely come from the large urology community practices before the J-code. It could actually come from the academic centers who might be involved with our clinical trials, for example. We're going to continue to learn this. We have the benefit, being the fourth entrant, to learn from our competitors who have branded products that are in the market. We get to study the pre and post J-code dynamic and which are the accounts that are making those decisions, and what are the trade-offs that they're thinking through.
For the timing around permanent J-code, the time to get one will depend on when exactly you get approved.
Yes, it is about six months from there.
Right. Well, depending on, because the J-code assignments are on a fixed calendar, whereas approval is kind of whenever you submit. To what extent might you consider finalizing the BLA for creto to align with the J-code cycle, which also kind of gets into the question, would you consider even holding off to incorporate PIVOT-006, just to maybe, again, help align with the right permanent J-code cycle?
Those are great questions and one that we continue to discuss internally. I would say as of now, because BOND-003 for BCG unresponsive, we have breakthrough therapy designation. PIVOT-006 does not yet have that, and certainly, we do not want to jeopardize the potential to get priority review on the first BLA. To that question, I would say we definitely have thought about it, but I think just getting into the market first with unresponsive disease for creto, I still think is the right approach to open up the market. There is possibility for us to publish the PIVOT-006 results in a peer-reviewed journal, and should creto be approved at that time, certainly, NCCN inclusion could be a potential option to kind of bridge into that gap that you are describing.
I've just got a question on my mind that just flew out of my head, so we can come back to that. Maybe coming back to the market dynamics and lessons learned from some of the other product launches, what else do you think you can do to perhaps have a swifter adoption curve than the others have had?
Yeah. I would say that, first of all, there's an approach where we just look at the data, Medicare claims data. I think through looking through the Medicare claims data, certainly there are new ideas that have emerged from them. Initial target accounts that we had in mind might not be the most obvious early adopters. Those are details that are emerging. Secondarily, I think, what do the customers really care about? What do the healthcare providers really care about? There are definitely nuances between academic providers versus community providers.
I think that workflow certainly is a very important variable, and the good news is, when we approach a practice, creto's workflow actually, on top of course, our efficacy and safety data, workflow is definitely emerging as something that's quite refreshing because it's the same way how BCG is given, and they're very used to having a medical assistant perform that procedure or installation. Because that's something that typically you would think, hey, if you have the best efficacy and best safety and best durability, it should just sell itself. I think for the larger urology group practices, they are very focused on the workflow. Not disrupting their day-to-day, not adding additional burden or hassle to what they have to deal with. They certainly don't want to get a phone call on a weekend about a certain type of AE.
Right. Okay. My forgotten question came back just around the filing and the application review times. Would you consider purchasing a priority review voucher to address that one question you had at one time?
I think internally we did, I personally did ask that question to our CFO, actually.
He was like, no way-
No.
too expensive.
I think obviously there's a mathematical answer to that, and of course, there's also the availability of such a-
Fair. Yeah.
...a voucher.
Coming back to market dynamics, another question that comes up often is related to where are the patients, because there's the incidence of high-risk NMIBC and the prevalence, and given the multiyear time until muscle-invasive bladder cancer, we'd envision that there are a bunch of patients out there who've been treated with BCG and failed and don't really have many options, like the warehoused or pent-up demand. Yet we don't necessarily hear about them when we talk to practitioners. Is the math wrong around what the prevalence should be relative to the incidence? Are we not looking for the patients in the right location? What do you think?
Yeah, no, this is a great question. I think when it comes to BCG unresponsive disease, you got to fail 5+2 doses of BCG, recur within 12 months of that. When you just look at that, certainly you tend to have to think more about prevalence because it's less likely that the same patient gets treated January 1st, 5+2 with recurrence within 12 months. So we do think a little broadly. I think that the line is sort of artificially cut when FDA created this BCG unresponsive definition. It really left all these BCG-exposed patients who have a recurrence after 12 months left without something that's on label.
Our view is that the exposed population is actually a lot larger from a prevalence standpoint, and the unresponsive patients, if you think about the context, they have already failed BCG, and they really want access to potentially more clinical trials, assuming their goal is to prevent and/or avoid radical cystectomy. But then they might need to get a radical cystectomy. So they potentially could get more concentrated into key academic centers that are high volume with great access to BCG and clinical trials.
In the community setting, I think there would be more BCG-exposed high-risk patients. Because they may have some BCG, but they may not have received the 5+2 and all recur within 12 months. We are still studying this pattern. When it comes to intermediate risk, these are all BCG-naive patients. There is definitely a lot of them in the community setting. That is something that we will continue to study over the next couple of months.
Coming back on this topic to the gem, creto phase III that you are planning. The first data set was encouraging, but I think many of us felt that the real signal would be in the durability of responses. The fact that you are moving towards the phase III would suggest that perhaps you are liking what you are seeing as those patients are followed longer. Is that a reasonable conclusion?
I think given creto monotherapy's experience, we see this very long tail from 12- 24 months. I think our plan is to move as fast as we can. We do not know yet the 12-month results, but I think we are definitely encouraged by just a fundamental MOA of creto and what we have seen in the monotherapy results.
Okay. What else should we be looking for over the next 12- 18 months from CG?
Yeah, I think all eyes on our PIVOT-006 top line readout, and we do anticipate hopefully presenting the information at a medical conference as well. Our team is laser focused on getting the most robust CMC package ready for submission, which will be used to support multiple filings, not just the first one. I think the 12-month creto and gem data that will be available end of this year is also something that I would pay attention to.
Outstanding. Looking forward to a lot of important updates from CG in the months to come. Arthur, thanks for joining.
Thanks, everyone. Thank you.