Can you please provide a brief overview of the company, its pipeline, as well as its platform technology, which I'm particularly interested in?
Sure. As you all know, we just launched IBTROZI , our ROS1 TKI. Some of the notable things about this drug are that it has, in our published JCO publication, an 89% response rate and 46-month PFS in the first-line setting, which is actually the highest ORR PFS seen with any cancer drug in any cancer to date. We're very excited about that. In just our first seven weeks of launch, we did 70 prescriptions, about 10 a week, which is about 5x in seven weeks what repotrectinib did in their first 10 weeks of launch. We'll be announcing our first full quarter earnings in November, so excited about that.
We think that with this profile of this incredibly long duration of response, it makes it a very unique commercial opportunity because if you just take the number of patients, new patients a year, about 3,000 a year, multiplied times the drug price, about $350,000 a year, that's about $1 billion of new patient starts every year. Because you have a nearly four-year PFS or DOR, you're talking about stacking that leads to almost $4 billion a year by year four. That's just based on DNA testing. RNA testing will detect about 30% more ROS1 fusions than DNA. We think this is probably close to a $5 billion a year market. We think that IBTROZI is a really attractive agent there.
You know, you recently just, you literally just presented updated data from the pooled TRUST-I and TRUST-II trials. Maybe elaborate a little bit more on the survival benefit of IBTROZI here, because I know it's been continually being pushed out further and further out. To me, it really is record-setting what IBTROZI has demonstrated so far.
In the first-line setting, the data we just put out at a world-long show again, close to a four-year duration of response, which is really exciting. Response rates are very similar to what I just mentioned, approaching 90%. In the second-line setting, in the TRUST-II data, we just presented a roll-along this weekend. We have a 62% overall response rate. Importantly, ROS1 lung cancer is a disease that starts 36% of the time in the brain. In 50% more cases, when they do progress, they progress in the brain as the first site of disease progression. If you look at IBTROZI 's second-line intracranial response rate, it's 66%, which I would say is really, really robust. Our overall response rate in the second line, we just presented a roll-along of 62%. Again, puts it at the top of the pack of agents in that space.
I know that one of your competitors, Nuvalent, also presented data at WCLC. I think before we actually dive into the Nuvalent data, one of the interesting things that was recently pointed out to me was their eligibility criteria. It seems like they're hand-selecting patients versus what repotrectinib's trials did and what your trials did. Can you elaborate a little more on that point?
Yeah, I think that's something that even we only learned relatively recently about. If you look at the protocol, which we've had a chance to review, they exclude any secondary oncogenic driver other than ROS1. MET amplification, BRAF, NRAS, KRAS. I think that the issue there is that if you look at concomitant mutation rate in almost all lung cancer, it's pretty significant. The literature spans a wide gap from close to 10% to up to 60%. We believe that number is closer to at least the 20%-3 0% range. If you look at all, the rate of concomitant mutations is above 30%. We think that's a pretty significant exclusion because, number one, it's not a real-world population. Number two, you have to wonder what kind of label that's going to get. In repotrectinib's study and in our study, we made no such exclusion.
We took all comers or a truly real-world population.
Any idea, because obviously you probably do have the data about the patient populations in TRUST-I and TRUST-II, what was the rate of concomitant mutations in that, if you have them?
You know, since we decided to take everyone, we didn't even look at that. We'll look back and see what that is. We would expect that, since this is a very large database, we enrolled over 400 patients over the last 10 years in the study. We would expect it to be pretty close to a real-world population. If you look at just the second-line response rates, just to make an apples-to-apples comparison, following one prior TKI, as I mentioned, in TRUST-II, we had a 62% response rate. Over the weekend, Nuvalent published a 51% response rate as an overall response rate in second line after one TKI. We think that we compare favorably. If you look importantly at intracranial response rate, Nuvalent's presented intracranial response rate was 48%. That included two patients with unconfirmed partial responses. We only present confirmed responses.
If you take those two out, the response rate's 45%. As I mentioned, our intracranial response rate confirmed in the second-line setting is 66%. We would say that that's pretty competitive.
Yeah, no, you know, it sort of baffles me why the Nuvalent market capitalization is where it is versus, especially when you look at the profiles of IBTROZI versus their drug. Any thoughts as to where the disconnect is coming from, from an investor standpoint?
I think that most investors, and probably even most physicians, don't actually really understand what a Kaplan-Meier curve is and what censoring means. If you look at the new Nuvalent presentation over the weekend, they're suggesting that their progression-free survival is a 22-month duration of response of 22 months and progression-free survival of 23.8 months. That would look really long. For those who aren't familiar with Kaplan-Meier curves and statistics, the way to think about it is that you have, OK, let's say 100 little robots that you charge anywhere from five seconds to an hour. You have no idea how far they're going to walk. You put a quarter of them out on the football field, and there are three robots that have made it to 22 yd. You can't say that the median duration of response or PFS is going to be anywhere close to 22 yd.
Yeah.
That's three robots. When you put all your robots on there, you'll kind of see how many yards they'll get to. What's interesting, if you look at the Nuvalent data that was presented over the weekend, the number of patients who crossed the six-month mark was 26. The number who crossed the 12-month mark is 6. The number who crossed the 18-month mark is 3. The number who crossed the 24-month mark is 0. It's very hard to understand how you can suggest that you have a 22-month or 24-month DOR PFS based on three patients. That's an extrapolation. I would say it's extremely aggressive. We don't make those kinds of suggestions because we think that's not rigorous.
Yeah.
The numbers we posted are with far, far more patients and far, far more follow-up time. I would say that investors may not appreciate the subtlety of how Kaplan-Meier's work and what censoring means. I think that perhaps they have misinterpreted how robust those numbers are and are trying to compare a number based on three patients compared to our data, which has far longer follow-up and far more patients.
Obviously, one of the things that Nuvalent continually sort of touts is its potential for improved safety. I want to get your thoughts on the safety profile of their drug relative to yours and how big of a differentiator is it?
OK, so Nuvalent's primary claim on safety is that they don't hit [NTRK], so they have very low dizziness rates. We have some [NTRK] activity. We're about 20x more selective for ROS1 over [NTRK] than repotrectinib. If you look at our dizziness rate, it's 22%. crizotinib's dizziness rate, so crizotinib doesn't even cross the blood-brain barrier. crizotinib's dizziness rate is 20%. For our drug, it does cross the blood-brain barrier. We're 2% higher than a drug that doesn't cross the blood-brain barrier. Of that 22%, 90% was grade 1 and resolved within 72 hours. We just don't consider that to be a significant clinical AE. Our most common clinical AE is diarrhea. 80% of that is grade 1, starts within 48 hours, and resolves within 24 hours. We don't consider that to be a problematic AE. Our most common AE of any sort is not clinical.
It's actually a laboratory finding. It's elevated liver function tests. That's by far the most common AE we have. If you look at our safety database of 337 patients, the number of patients who discontinued IBTROZI for LFT elevation was 1 out of 337, so 0.3% discontinuation rate. We don't think that's a problematic safety profile.
Yeah.
As I said, we have from our phase I study a quarter of our patients approaching a decade out of therapy. They're taking drug every day for nine years, 10 years. We don't think that suggests that we have a tolerability issue. I would say that we think that given our efficacy, which I imagine is the best profile that's been seen, those numbers have not been matched by any drug in any path. We think that combined with our tolerability profile gives us a very attractive drug in this space.
Yeah, you know, when I looked at the ROS1 data, you know, there was only a 2% discontinuation rate due to TEA. You're at a 2.3% discontinuation rate due to a treatment-related adverse event. It doesn't seem like there's really that much big of a difference here. You didn't have any discontinuation.
If you look at, clearly safety events, adverse events accumulate with time.
Yeah.
What's notable is that Nuvalent's 2% discontinuation rate for TEAs is at five months. If you look at our discontinuation rate for TEAs, exactly apples to apples, at five months, it was also 2%.
Yeah.
It is actually identical. What's interesting is that since the data in the Nuvalent trials is accumulating rapidly, since their earlier presentations, now suddenly there's a 36% edema rate that hasn't been talked about for 15%, shortness of breath, 3% of which is grade 3. These are, you know, we always say that the AEs that patients care about are the ones they feel. We would put shortness of breath right up there as something that, you know, people get a little concerned about.
Yeah.
We don't think that invisible AEs like LFT elevations where they're asymptomatic and only one in 337 patients discontinuing is a problem.
Yeah. You mentioned before the initial experience with commercialization of IBTROZI . I believe you said it was 70 patients over. How many weeks was that?
Seven.
Seven weeks, right? 10 patients a week, which was more than what repotrectinib did in the first several months of its launch. Why do you think that is? Is it in part due to the change in those NCCN guidelines? Do you think that when people are modeling the launch trajectory of IBTROZI , it is going to be significantly different than what we saw from repotrectinib?
I do think it's very different. If you look at the IQVIA script data, about 10% of repo scripts are falling off every month. By five months, there's about a 50% discontinuation rate. We know that the main issue with repotrectinib in continued use is their high dizziness rate, about 65% in their published study. It's very difficult, you know, when you have problems with ataxia and have more difficulty walking or driving. I think that's an issue. On top of that, there are a couple of new tailwinds that didn't exist at the beginning of the repo launch. On January 7th of this year, NCCN guidelines changed for the first time to contraindicate IO chemo. IO, which was commonly used in ROS1, no longer can you give IO. It's actually contraindicated.
Now they require a ROS1 agent for a ROS1 mutation, which is what it should be because that's the right thing for patients. That just changed on January 7th. That's a nice tailwind. We were incidental beneficiaries of that, so happy about that. The second thing is that, of the three other approved drugs, crizotinib doesn't cross the blood-brain barrier.
Yeah.
For a drug that starts in the brain a third of the time and progresses in the brain half the time, you just can't give a non-CNS penetrating drug. That's just not, that's not right. repotrectinib has a 16-month PFS. You know, when we're talking about a 46-month PFS, it's hard to justify getting anything. repotrectinib is really the only real choice. The tolerability has presented challenges. I think, as you pointed out, in our first seven weeks of launch in a summer, we did about five times the repotrectinib sales numbers in their first couple of months of launch. We think that it's what's borne out in the marketplace, which is what we would expect based on what we know about the profiles. We also did have the tailwinds of the NCCN guideline change.
I think as genetic mutations in general lung cancer have converted lung cancer to one of the most treatable cancers in oncology, suddenly everybody wants to know what NGS testing shows. We're seeing a, that's a rising tide that's floating every boat. I think we're lucky to be on the tailwind side of that. I think our data are really robust and durable and tolerable. I think that that's all helping.
Yeah, no, I would certainly agree with you. You know, why don't we shift gears now to a few of your other assets within the pipeline, first one being your IDH1 inhibitor?
That's pretty interesting. I think the most notable thing that we've seen in the last couple of months is that if you look at Royalty Pharma's royalties from vorasidenib, which is the only IDH1 inhibitor approved, and it's only approved in low-grade glioma, those royalties from Royalty Pharma suggest a run rate of at least $750 million a year for vorasidenib already in its first six months of launch. That drug's going to be a billion-dollar-plus drug. That's pretty interesting. I think that if you look at the pivotal study for vorasidenib in low-grade glioma, the INDIGO study, the response rate was 11%. In our first IHE study, our response rate was triple that, 33%. Now, vorasidenib did show an improvement in PFS from 11 months to 27 months, a 16-month improvement in PFS, which is really robust.
We're going to present for the first time our PFS number later this year in a second IHE study along with another response rate. I think that's what should be noted. That's going to be very interesting. Based on what we're seeing, we've already kicked off a pivotal study in high-grade glioma. In high-grade glioma, vorasidenib is not approved, and their response rate is zero. In high-grade glioma, our response rate is 17%, but a third of those have been complete responses. I think I mentioned previously that we have one glioblastoma grade 4 astrocytoma that's been in a complete remission and no tumor left for three and a half years. We have a second high-grade oligodendroglioma whose tumor has been gone for about two years. We think that we've already kicked off a pivotal study in high-grade glioma, which vorasidenib is not approved in.
In low grade, we think our data are so compelling that we're going to be initiating a head-to-head study against vorasidenib in low-grade glioma. We're just in our final discussions with FDA to finalize that clinical trial plan.
You know, one of the interesting things about your IDH1 inhibitor, at least from my perspective, is the durability of its effect. It seems to be like when I look at the AE profile, it seems like there's some immune adverse events in there.
Yeah, funny you should say that. We were struck by that too. We tried to figure out for ourselves, like, why are we seeing such different data between safusidenib and vorasidenib?
Yeah.
We did a number of things. We looked at a model of immune competent versus immune compromised xenografts. What we found is that safusidenib and cytinib work way better when there was an immune system to activate. When we did, we showed a dramatic improvement in activity and mortality where vorasidenib, even at five times a clinical trial dose, did not. We're seeing a much better ability to activate immunity. If you look at all our responses, they take 6 months- 12 months to kick in.
Yeah.
Just like you've seen in IO. If you look at durability, as I mentioned, the ones we've mentioned, they last for years. That sounds like IO. We looked at our top eight adverse events, and out of our top eight adverse events, five of them are immune related. We see skin hyperpigmentation like vitiligo. We see arthralgia like rheumatoid arthritis. We see rash like lupus. We see alopecia like alopecia areata, a well-known autoimmune disease. Vorasidenib doesn't have any of those adverse events. It mainly sees liver function tests abnormality. We have a very different AE profile that looks immune. We have these really long and slow to kick in, but long durability responses that sound like IO. In our clinical models, we show immune activation, which vorasidenib doesn't show. We think that safusidenib is possibly a new oral IO agent.
We're actually in the process now of testing safusidenib in non-IDH1 mutated cancers to see whether or not this is a more generalized immuno-oncology agent. Right now, we do know that our potency against IDH1 is very high. We know we're going to kick these two trials off in IDH1 mutated gliomas, high and low grade. We think there's a possibility, depending on what we see, that we could expand this into perhaps other tumors that are not even IDH1.
Yeah.
That would be really different.
You know, one of the things that I've always been fascinated about with Nuvation i s your DDC platform. For those who might not be aware of what that platform is, to give them just a broad overview of it and talk to us a little bit about NUV-1511. When are we going to see initial data for that? We don't even know what components there are in this compound yet.
That's true.
I'm curious.
I started Nuvation in 2018. At the time, ADCs were all the rage. ADCs have done incredibly well, and some of them are just some of the best drugs in cancer. An ADC is an antibody coupled to a small molecule warhead. Antibodies are really big, and they're harder to make, and they don't often traverse cell membranes very well. We decided to see if we could recreate the same thing with two small molecules, both a small molecule targeter and a small molecule warhead. That would be significantly smaller than an ADC, allow you to target many things inside the cell, and probably have better bystander effects. We created a molecule called NUV-1511. It was our first DDC that we developed that had really striking preclinical activity. One end of that molecule is a well-known chemo molecule.
The other end we haven't talked about, but it's an FDA-approved drug that is actually not even used for cancer, but it binds a target that's seen in many cancers. We tested that. This is our first such molecule. We are now developing many agents behind it. As we learn a little bit more every day about how DDCs work, including our DDCs, we can mix and match warheads very easily. We figured out now how to make these things so that we can have orally bioavailable molecules. We've learned a lot about what warheads pair well with what targeting agents, and it's been a learning process. It's taken us about four years to figure out the chemistry for this. Now, with our experience, we are looking to broaden this platform. We're going to have a number of other indications and warheads and targeters that we're contemplating.
We're still in the process of enrolling this NUV-1511 trial, which is being tested in five different indications where nothing works, and hope to have some data out later in the year.
OK. I guess the last question for me before I turn over to the audience is, what's your current war chest look like? What's the operational runway? How much dry powder do you have to deploy?
We added at the end of the last quarter. In Q2, we had about $600 million to make it simple. We're expecting at least a payment from Nippon Kayaku when they get approved, which will add $25 million. We also have $50 million we can take from our deal with SeaGen. You remember the early year with SeaGen where we had like $150 million of royalty, which valued taletrectinib at a very high level because it was 5% royalty, $150 million, so $2 million there, and $50 million + $50 million debt. Total $250 million only to $200 million.
Wow.
Yes, we have a pretty significant war chest. We always said that even prior to the SeaGen financing, we had enough to get to profitability. That's even stronger now, honestly, and to deploy eventually to interesting things, lots of things to be done. In terms of burn, what we said without getting too much detail, because there's still variability, is that obviously our Q2 is the first quarter in which we have our sales force and our kind of commercial spend. It's a decent proxy for where we're going to be in terms of spending in the years to come and quarters to come.
OK. That's really all my questions. We'll now take any questions from the audience if there are any. All right. David, Philippe, thank you so much for joining us today.