Good day, and thank you for standing by. Welcome to the ORIC Enozertinib Program Update Call. At this time, all participants are in a listen-only mode. Please be advised that today's conference is being recorded. After the speaker's presentation, there will be a question and answer session. To ask a question, please press star one one on your telephone and wait for your name to be announced. To withdraw your question, please press star one one again. I would now like to hand the conference over to your speaker today, Dominic Piscitelli, CFO.
Good afternoon, and welcome to the ORIC Pharmaceuticals enozertinib Program Update conference call. Earlier today, we issued a press release highlighting updated data from our ongoing phase I-B study of enozertinib in first-line non-small cell lung cancer patients with EGFR atypical mutations. You can find the press release on the investor page at oricpharma.com. We have pre-recorded our prepared remarks, after which we will host a live Q&A session. Before we begin, starting on slide two, during this conference call, we will be making forward-looking statements, including those based on our current expectations and projections about future events and trends that may affect our business, utilizing data available to us as of the cutoff date set forth on the applicable slide. ORIC's actual results may differ materially from those expressed in or indicated by such forward-looking statements.
For a description of risk factors associated with investing in ORIC, please refer to our recent filings with the SEC. ORIC specifically disclaims any obligation to update any forward-looking statements except as required by law. This presentation contains interim results based on the data from enozertinib clinical trial as of the cutoff date set forth in the applicable slide. During this presentation, we will not be speaking to any additional data subsequent to such date. Now turning to slide three. During today's call, we'll discuss preclinical differentiation, clinical data highlights, and next steps, and finally, the commercial opportunity, followed by Q&A. Joining me on the call today, we have several members of the ORIC management team and our guest KOL, Dr. Joshua Sabari. Now, let me turn the call over to Jacob.
Thank you, Dominic. Turning to slide five. We're excited today to review the updated global phase I-B data from our first-line EGFR atypical cohort, which are being presented at ESMO. This slide summarizes the key takeaways, which we believe further reinforce enozertinib's potential best-in-class profile in this population and support our plan to move rapidly into phase III. First, the opportunity. First-line EGFR atypical non-small cell lung cancer has a market opportunity of more than $2.5 billion in the U.S. alone, and it lacks a best-in-class therapy that combines meaningful systemic efficacy, compelling CNS activity, and favorable tolerability. Second, today's enozertinib data set a new benchmark for best in class within EGFR atypicals. To level set, we enrolled a population with more than twice the rate of baseline CNS disease as seen in competitor datasets, and most of these patients had no prior treatment for their CNS metastases.
Even with that context, enozertinib delivered an 84% confirmed systemic response rate, a 70% confirmed CNS response rate with multiple complete responses, and an 88% six-month landmark progression-free survival. Third, safety. Enozertinib continues to show no significant off-target toxicity, and there were no treatment-related discontinuations. Finally, supported by these stellar data, we plan to initiate REDWOOD-1, a global phase III trial in first-line EGFR atypical non-small cell lung cancer in the first half of 2027. Turning to slide six. We are developing enozertinib for two sizable patient populations within non-small cell lung cancer that have thus far been woefully underserved by the currently approved and late-stage development therapeutics, all of which suffer from a lack of compelling CNS activity and come with meaningful toxicity challenges.
Collectively, there are over 11,000 patients annually in the U.S. with these mutations, representing a combined commercial opportunity of more than $3.5 billion annually in the U.S. alone, with more than $2.5 billion in atypical and more than $1 billion in exon 20. We believe each of these first-line opportunities has blockbuster potential in the U.S. on its own, and we see a comparable ex-U.S. opportunity. An oral, well-tolerated CNS-active agent should garner substantial share in these settings. Keith will walk through the market assumptions later in the call. I have already mentioned the importance of CNS activity, and slide seven helps explain the rationale. Approximately 35% of patients with EGFR mutated non-small cell lung cancer present at diagnosis with CNS disease. That figure does not include the patients who have micrometastases that are not yet detectable on conventional imaging. Over time, about half of patients will develop brain metastases.
The bottom half of the slide shows why this matters. With drugs that are not CNS active, outcomes are dramatically different for patients with versus without brain metastases, and we show two examples here from EGFR exon 20. For both sunvozertinib and the combination of amivantamab plus chemotherapy, you can see the dramatically different hazard ratios achieved in patients with versus without baseline brain metastases. This is perhaps most noticeable in the sunvozertinib example, with only a 4% risk reduction or 0.96 hazard ratio in the patients with baseline brain metastases. An effective brain penetrant therapy, many of which exist for other targets in non-small cell lung cancer, can drive long-term outcomes by treating existing CNS metastases and preventing the development of new CNS lesions. Turning to slide eight. This is not a new concept in targeted therapy across multiple lung drivers, including classical EGFR, ALK, RET, and ROS1.
There are best-in-class TKIs that combine meaningful systemic efficacy, compelling CNS activity, and good tolerability. Those drugs were often third, fourth, or even fifth to market within their respective categories. But what most prominently differentiated them from their predecessors was profound CNS activity that led to superior PFS. Ultimately, those brain-penetrant drugs became the market leader within a few years of launch. For EGFR atypical and EGFR exon 20 mutations, no agent, either approved or in late-stage development, combines all of these attributes. Slide nine provides a snapshot of the enozertinib data that underlie our excitement for this opportunity in first-line EGFR atypicals. The slide also lays out the comparator data within atypicals for afatinib, osimertinib, which is in the NCCN guidelines, and the two late-stage investigational therapies, amivantamab with lazertinib and firmonertinib. It is worth spending a moment on the enrollment criteria.
While afatinib did not allow patients with untreated CNS metastases and did not report any CNS data, the other comparators all enrolled those patients at or near the rate seen in the general population, around 30%. In the case of enozertinib, an astounding 65% of patients had CNS metastases at baseline. We weren't trying to enrich for these patients. Physicians in the trial believe enozertinib is profoundly CNS active, and so they disproportionately enrolled their most challenging patients onto the trial. What's even more impressive than this exceptionally high rate of baseline CNS metastases is that from day one, we have allowed patients with untreated brain metastases onto the trial. Again, as you can see from the chart, it is relatively uncommon for trials to allow the enrollment of patients with brain metastases if they haven't already been treated with radiation or surgery.
In the enozertinib cohort you will hear about today, nearly half of all the patients had untreated brain metastases upon study entry. It is with that backdrop that enozertinib still managed to achieve an impressive confirmed systemic response rate of 84%, confirmed intracranial response rate of 70%, and six-month landmark PFS of 88%, with each of these efficacy measures being the highest ever seen in EGFR atypicals based on published data. It did this while demonstrating a well-tolerated safety profile with no significant off-target toxicity and no treatment-related discontinuations. What is clear from these data is that enozertinib is emerging as the potential best-in-class therapeutic option for frontline EGFR atypical mutations. With that, let me hand it over to Lori to remind you of the preclinical profile before Pratik covers the detailed phase I-B results. Lori?
Thank you, Jacob. Slide 11 highlights our key aims in identifying a clinical candidate that would support a best-in-class profile for EGFR atypical mutations. Enozertinib was designed to have multiple areas of differentiation. First, enozertinib has strong cell potency against a wide variety of EGFR atypical mutations and demonstrates tumor regressions in vivo. Second, enozertinib is notable for its selectivity towards EGFR while sparing all other kinases. Third, a crucial feature of enozertinib is brain penetrance. This high brain exposure drives regressions in xenograft tumors grown in the brains of mice. As shown on the right, these three areas of preclinical differentiation have all played out in the clinic, with potency translating into clinical responses across a breadth of EGFR mutations, selectivity translating into minimizing off-target tox liabilities, and brain penetrance resulting in compelling intracranial responses in patients.
Taking each area of preclinical differentiation in turn, starting first with potency on slide 12. A key differentiating feature of enozertinib is strong cell potency across a variety of EGFR mutations. In this figure, cell potency in EGFR PACC and classical-like mutations is compared to EGFR wild-type potency. For any given compound, you want to see as many mutant symbols as possible falling below the gray horizontal bars that depict EGFR wild-type potency for that compound. For enozertinib, nearly all the mutant potencies are below its wild-type potency, indicating a promising predicted therapeutic index. But this clearly isn't the case for the competitor compounds. Enozertinib stands out with excellent potency in mutants and a favorable potency ratio versus wild type. Moving to slide 13, another key differentiating feature of enozertinib is selectivity. In these data, profiling multiple compounds across the kinome, enozertinib is exquisitely selective, a key differentiation versus competitors.
The table at the bottom of the slide summarizes the off-target kinase results, with enozertinib having no off targets, while competitor EGFR compounds have multiple off targets. Off-target binding can lead to adverse events not related to EGFR, such as hematologic, cardiac, or liver toxicities. Finally, turning to brain penetrance on slide 14. Brain penetrance stands out as one of the most distinguishing features of enozertinib, especially given that best-in-class molecules for lung cancer must be able to tackle and prevent brain metastases. On the left, in head-to-head studies in rodents, enozertinib has a high free brain to plasma ratio similar to osimertinib, which is a brain penetrant drug that has demonstrated strong CNS activity in the clinic. Shown on the right are results of an in vivo study measuring anti-tumor activity in the brain.
This EGFR classical mutant lung cancer model was inoculated into the brain of mice, and tumor shrinkage was measured by imaging. The log scale on the Y-axis shows enozertinib achieving deep regressions in tumors grown in the brain compared to the competitor compound with proven brain penetrance. Next, I'll hand it over to Pratik, who will discuss clinical results that show that these preclinical best-in-class properties have translated well in patients.
Thank you, Lori. Before I review our new ESMO data, one quick comment. These data all come from our global phase I-B trial, which is being conducted at 39 major academic centers in 10 countries in North America, Europe, and Asia Pacific. We've enrolled over 350 patients across the various parts of this study, starting with dose finding, followed by dose optimization, through to the dose expansion cohorts. Now turning to slide 16. Last year at ESMO Asia, we presented data from dose optimization, which we used to select the RP3D. These data largely focused on previously treated patients with EGFR atypical or EGFR exon 20 insertion mutations. For this latter group, exon 20 mutations, alongside enozertinib monotherapy, we are also evaluating combinations with chemotherapy and with amivantamab, which are still in dose exploration.
Today's update focuses on the key efficacy endpoints that are listed here at the bottom, as well as safety for the EGFR atypical cohort, by which we mean patients with PACC mutations or the classical like L861X mutations, either singleton or compound. These patients are all treatment naive with broad latitude with respect to baseline CNS involvement, including active CNS metastases, and all received enozertinib monotherapy at the RP3D of 80 mg once daily. Slide 17 shows the demographics and baseline characteristics for the 26 patients in this update based on a data cutoff of September 4th, 2026. 88% had a PACC mutation, 12% had the classical-like L861Q mutation, and of these mutations, 69% were compound and 31% singleton. The most notable aspect, which is highlighted here, is the baseline CNS disease burden. 65% of patients, 17 of 26, had brain metastases at baseline.
Of these 17, 12 patients or almost half of today's dataset had previously untreated brain metastases. In other words, they had active CNS disease at trial entry for which no prior CNS-directed therapy had been administered. If you look at the literature for first-line atypical, in general, about 30%-35% of patients will present with brain metastases at diagnosis, which is also generally in line with competitor datasets. We have 65% or more than double the rate versus competitors, and this enrichment isn't because we're trying to recruit more of these types of patients. It's because our study physicians believe enozertinib is profoundly active against CNS disease. Also, in contrast to many of the competitors, which require brain metastases to have been treated prior to enrolling, we have never had such a requirement.
We have always allowed untreated brain metastases into the trial, and so 12 of our patients, or 46% of today's total dataset, came into the trial with previously untreated brain metastases, making this population overall more difficult to treat versus other competitor datasets where most patients, if they have CNS metastases, have already been treated. I'm spending time on this important context because it makes the efficacy results I'm about to show you all the more remarkable. First, turning to slide 18, we have the safety profile. Treatment-related adverse events were predominantly Grade 1 or 2. Grade 3 events occurred in six patients, or 23%, and all resolved with dose modification. There were no Grade 4 or five events, and importantly, there were no discontinuations due to treatment-related adverse events.
On the right, the most common treatment-related events were diarrhea and paronychia, with only single-digit rates of Grade 3 toxicity. The next most common events were rash, acneiform dermatitis, and stomatitis, again, with very low rates of Grade 3 and all resolved with dose modification. These events are all consistent with EGFR wild type inhibition, with no significant off-target toxicities such as cardiac, hematologic, or hepatic. Overall, the safety data are in keeping with what we've seen and presented previously with enozertinib, showing that enozertinib is associated with a manageable and predictable adverse event profile. Now to efficacy on slide 19. We'll start with systemic ORR in the response evaluable population, which consists of all patients with the opportunity for at least two post-baseline systemic tumor assessments. That's 25 of the 26 treated patients.
By investigator assessment, the best ORR was 88%, and the confirmed ORR was 84%, with 21 confirmed partial responses. The remaining four patients had stable disease for a disease control rate of 100%. We also subjected these scans to blinded independent central review, which is the more stringent regulatory standard and is generally reserved for phase III trials. The BICR-confirmed ORR is 80%, very consistent with the investigator read. On the waterfall plot, you can see responses across a diverse set of singleton and compound mutations, including L861Q, with many deep regressions. The asterisks mark patients with CNS disease at baseline, and they are distributed throughout the plot with no apparent difference in response, which is what you'd expect from a CNS active drug.
Thus, despite enrolling a population with more than double the rate of baseline CNS metastases versus competitor data sets, enozertinib demonstrated a meaningfully higher best-in-class ORR, which was validated by a BICR review. Slide 20 shows the molecular response from circulating tumor DNA. Among the 22 evaluable patients, all achieved at least a 90% reduction in EGFR mutant ctDNA and 82% achieved full clearance, in each case within the first two treatment cycles. These exceptional ctDNA data corroborate the high systemic ORR that I just showed. Turning to the CNS data, starting with slide 21, we have intracranial response by investigator assessment. This analysis consists of the 10 patients with measurable target CNS lesions as assessed by investigator using modified RECIST 1.1 criteria, which focuses only on target lesions in the brain. The confirmed intracranial ORR was 70% with two complete responses.
The remaining three patients had stable disease with tumor shrinkage for a CNS disease control rate of 100%. Of note, 10 of the 10 patients had previously untreated brain metastases, as indicated by the hashtags, making any observed tumor regression in these patients solely attributable to enozertinib. Importantly, as a further indicator of CNS activity, we saw no new CNS lesions arise in patients regardless of whether they had brain metastases at baseline. Slide 22 shows CNS response as assessed by BICR using the more stringent RANO-BM criteria. As with BICR assessment for systemic ORR, BICR for CNS disease is also conducted centrally, but in this case involves expert neuroradiologists who are experienced in evaluating all brain lesions, target and non-target. They also employed RANO-BM criteria, which builds on RECIST by adding CNS-specific rules such as clinical features to finalize response determination.
This central review identified 16 patients with CNS metastases at baseline, of which one patient was deemed not evaluable, resulting in 15 evaluable patients by BICR. These patients were a mix of those with measurable target lesions and those with only non-target lesions, and many had previously untreated brain metastases, again, as indicated by the hashtags on the waterfall plot. Through this analysis, the confirmed BICR intracranial ORR by RANO-BM was 67%, including an impressive six patients, or 40%, who achieved a complete response. Five of these CRs were in patients with untreated brain mets at study entry. Also, three of these CRs were in patients with non-measurable CNS disease, which means all of their non-target lesions disappeared. W ith this more stringent assessment via BICR using RANO-BM criteria, the CNS activity of enozertinib remains very consistent with what we saw by investigator.
Slide 23 depicts the swimmer plot showing time on treatment for the 26 patients. As you can see, responses generally occurred by four weeks, which is the time point for the first on-study scan. But tumor regressions continued over time, with some responses occurring as late as four to six months on treatment. At a median follow-up of 10.5 months, median time on treatment has not been reached, and the majority of patients remain on treatment. Median PFS and median duration of response are immature, but the Kaplan-Meier estimate of six-month landmark PFS is 88%. Follow-up is ongoing, but clearly we're happy with how enozertinib is performing so far in this population with such a high prevalence of CNS disease. It sets us up well for further development. The next two slides show two patient vignettes, which highlight the profound systemic and CNS activity of enozertinib.
The first patient on slide 24 is a 59-year-old man with treatment-naive lung cancer harboring a compound PACC mutation. At study entry, he had five brain metastases, three target and two non-target lesions, none of which had received prior therapy. The patient achieved a deep, confirmed partial response after cycle 4. This response was accompanied by an intracranial partial response after cycle 3, which evolved to a complete response after cycle 14, all with only low-grade adverse events that largely resolved. The patient remains on treatment at cycle 18, so this response has now been ongoing for about a year and a half. The second vignette on slide 25 is of another patient with treatment-naive lung cancer. In this case, a 67-year-old man. His cancer had a PACC mutation, and on presentation, he had a previously untreated single brain metastasis in the frontal lobe.
This patient achieved a systemic partial response after cycle 1 and an intracranial complete response after cycle 4. This patient remains on treatment at cycle 18, so almost out to a year and a half so far. We highlight these cases because they illustrate what a brain-penetrant compound can do for patients who would otherwise have been at risk for CNS progression had they been treated with a non-CNS active compound, and these are only two of the six complete responses we saw. Now that I've had a chance to present our data in detail, let's take a step back on slide 27 to look at the benchmark table Jacob showed earlier.
You can clearly see that in comparison to not just the currently available therapies of afatinib and osimertinib, but also to the data with other late-stage investigational therapies, enozertinib is demonstrating systemic and CNS response rates that are higher than all these competitor data sets in a far more challenging patient population with more than double the rate of baseline brain metastases. And although it is early, the 88% six-month landmark PFS appears to be outdistancing the others. These enozertinib efficacy data, coupled with a safety profile that is very predictable and consists of EGFR wild-type related adverse events without the significant off-target toxicities seen with each of the other therapies, really puts us in a position of having a potential best-in-class molecule. Slide 28 builds on the point I made earlier about why CNS active therapies matter, even in patients without baseline CNS disease.
The top half of this slide shows non-CNS active therapies, where the benefit, as expected, is limited in patients who have brain metastases. The examples here are sunvozertinib and amivantamab plus chemotherapy, which Jacob covered earlier. The bottom half shows the value of CNS active therapies, where benefit is seen across all patients, those with or without baseline CNS disease. Alectinib, for example, confers a risk reduction of 49% in patients without baseline CNS metastases and 60% in patients with them. And for osimertinib, similarly, it was 54% and 53%, respectively. In other words, with a truly CNS active agent, the presence or absence of brain metastases does not change the outcome. That's the profile we're aiming to deliver with enozertinib in EGFR atypical non-small cell lung cancer.
Slide 29 shows two historical examples of the impact of CNS activity on long-term clinical benefit in the form of progression-free survival. In the FLAURA Trial in first-line EGFR mutant non-small cell lung cancer, osimertinib, which is brain penetrant, had a modestly higher systemic response rate than gefitinib or erlotinib across all patients, 80% versus 76%. But as expected, a much higher CNS response rate, 66% versus 43%. This translated into a median PFS of 18.9 months with osimertinib versus 10.2 months with the non-CNS active control agents. Similarly, in the ALEX Trial in ALK positive non-small cell lung cancer, alectinib versus crizotinib showed very comparable systemic response rates of 83% versus 75%, respectively. But alectinib's higher CNS response rate of 59% versus 26% with crizotinib translated into a tripling of the median PFS to 34.8 months versus 10.9 months with crizotinib.
These are different drugs against different targets, but the pattern is consistent. Modest improvements in systemic response, but with major improvements in CNS response, translating into dramatically improved PFS. That is the pattern we believe could play out with enozertinib. Turning to slide 30, I would now like to provide an overview of our phase III plan. We have already completed our end of phase I interactions with the FDA, which has allowed us to receive alignment on a number of critical factors to our late-stage development plan. Specifically, the recommended phase III dose of enozertinib of 80 mg once daily has been confirmed. With respect to the phase III trial design, we have discussed and reached an alignment on the overall study design, including patient population, the trial endpoints, the control arm therapy, and the statistical methodology of the trial.
With this in mind, REDWOOD-1 will be a global randomized phase III trial in treatment-naive patients with advanced non-small cell lung cancer harboring EGFR atypical mutations, PACC or L861X. As with all our clinical work, patients with treated and stable brain metastases or untreated but asymptomatic brain metastases will be eligible. Patients will be randomized one to one to enozertinib 80 mg once daily versus physician's choice of osimertinib or afatinib. The study will have dual primary endpoints of ORR by BICR, which is intended to support accelerated approval and PFS by BICR to serve as the confirmatory trial to support conversion to full approval. Key secondary endpoints are overall survival and intracranial ORR by BICR.
These are the main design elements of the study. We intend to have additional regulatory interactions with other global agencies. After these are completed, we expect to initiate this trial in the first half of 2027. We are excited about the potential benefit enozertinib may offer in this patient population of non-small cell lung cancer with PACC or L861X mutations and look forward to progressing to phase III in the near future. I would now like to introduce our guest, Dr. Josh Sabari, who has graciously agreed to join us to provide his own perspective on these data and on the therapeutic landscape in general. Dr. Sabari is a thoracic medical oncologist at NYU Langone Health and Assistant Professor of Medicine at NYU Grossman.
He is an investigator across multiple targeted therapies and immunotherapies with extensive experience in EGFR targeted therapies, including being a PI with enozertinib. Josh, thanks for joining us today here in Madrid at the ESMO Conference.
Pratik, thanks for having me, and very exciting data, and very excited to see all the data this weekend at this conference.
Before we discuss the enozertinib clinical data, I would first like to take the opportunity for you to review the atypical patient population and the current therapeutic landscape for these patients. And just for clarity, by atypical, I am referring to PACC mutations that form the core of our data here. P erhaps first you could spend a few minutes discussing the spectrum of the presentation of these patients with EGFR atypical mutations. Is it like the rest of the EGFR patient population or are there some unique aspects?
Yeah. T his is a unique population. A PACC mutation or P-loop and αC-helix compressing [lesion] group makes up about 15% of my EGFR mutant population. In order to put that in perspective, you have common activating mutations, exon 19 DEL and L858R. There you have approvals, osimertinib and chemo or obviously amivantamab and lazertinib. And on the other end of the spectrum are the rare mutations, the EGFR exon 20 mutations. But for PACC mutant populations, there is a significant unmet need. We don't yet have standard approved agents, in my opinion. Afatinib we'll discuss and then also the NCCN recommendation for osimertinib. But this is a unique population. These are older patients in my clinic, more male, heavy prevalence of CNS metastases, and we also see a higher rate of smoking in this patient population compared to the exon 19 or even L858R population.
Do you see a higher risk CNS in this population?
I would say about a third of my patients at initial presentation have CNS metastases. But over the course of their treatment, likely because we have ineffective CNS penetrant drugs, upwards of 60% of my patients will develop CNS metastases.
Now if we move to the currently available therapies, as you said, the NCCN guidelines include both afatinib and osimertinib. Let's start with afatinib. What are your thoughts about that?
Remember, afatinib is a second generation EGFR inhibitor, and this was really approved based on a small subset analysis from the LUX-Lung trials, and this is before we differentiated PACC mutations. Really the label is in atypical mutations and includes L861Q. The response rates in some specific mutations are in that 50%-60% range, but in others in that 40%-50% range. For afatinib, really we don't have good CNS penetration. Most of my progression, my primary progressive site of disease is in the brain in the afatinib treated population. The other drug that we can use that does not have an FDA label but is NCCN recommended is osimertinib. Osimertinib, as you know, is a third generation EGFR TKI. In L861Q, which is not a PACC mutation, the response rates are quite good.
But in G719 or S768, response rates are quite low in that 40%, maybe 50% range, and the durability is not there. Osimertinib does have good CNS penetration, but it doesn't have good activity in these PACC mutant populations. I n clinic, we're often making a decision between two ineffective therapies. One that has activity, afatinib, but no CNS penetration, and one, osimertinib, that doesn't have great activity but has CNS penetration. So we need better options for our patients.
So another option or a regimen that's not in the NCCN guidelines but does have data is the ami-laz combination.
Yes, I participated in that study. This is the CHRYSALIS-2 single arm study of about 65 patients that were enrolled, and we did see impressive response rates as well as a PFS around 11 months. We also showed a median overall survival there of about 41 months in this patient population. But as you know, amivantamab and lazertinib is a tough regimen to give to a patient, and it's not really truly selective per se to the PACC mutant group. This is a regimen that may have broader activity across multiple EGFR mutations, and it comes at a cost.
And chemotherapy is notably absent, so does it get used much?
It's interesting. Chemotherapy is not commonly used in the PACC mutant population in the community. In the academic settings, we started to use FLAURA2 off label somewhat in this patient population. We really do need a CNS penetrant, EGFR TKI that will overcome the steric hindrance that you see with some of these other third-gen and second-gen EGFR inhibitors.
Great. Now let's shift to the agents that are currently in development. Notably, firmonertinib, they're running a phase III. Silevertinib has put out some data as well. What's your overall impression on their systemic efficacy?
Yeah. For firmonertinib, we've seen some early data in the PACC mutant population. Clearly, we saw more data in the exon 20 space, which was unfortunately negative, and I'm sure we'll discuss. For silevertinib or BDTX-1535, we did see decent activity, but again, it came at a dramatic cost. I think about 84% of patients require dose reduction or dose interruption due to significant Grade 3 or greater adverse events. We know that that's not the current dose as they'll move forward. I think we need a better tolerated as well as better CNS penetrant therapy in this patient population.
Both of those agents have generated data, as you've said, but they've had some restrictions in terms of the types of patients with CNS disease. Can you comment on the relative eligibility versus in the enozertinib trial?
Yeah. I think from a clinical standpoint as an investigator, what I like to provide to patients is a therapy in active, untreated CNS disease. That's really the gold standard, and we'll talk about how we assess response in this patient population. In a patient who has treated CNS metastases, maybe we're measuring the response to radiation, for example, or resection. I think it's important that when we talk about the CNS activity, we're talking apples to apples, and active untreated CNS metastases really is the gold standard when developing a TKI.
Great. Now you brought it up. You recently there was a disclosure about the firmonertinib phase III in exon 20. Maybe we just take a little departure there. What's your interpretation of the study results as we know them so far?
Yeah. I was very hopeful for that study and was somewhat disappointed. This is the FURVENT study, a randomized phase III study of firmonertinib at 240. They also did look at a 160 mg dose versus standard of care in this population, which is chemotherapy. There, the PFS was around 11 months. This is from a press release in the treatment arm. In the control arm, about 8.5 months or so. It didn't meet stat sig, and this was a negative trial based on press release. We look forward to seeing the full data disclosure.
Sure.
What's interesting is if you look at the 160 dose, the response rate was lower and the PFS rate was in that eight-month range as well. I was somewhat disappointed by the data in the EGFR exon 20. As you know, there is a randomized phase III study, the ALPACA trial in the PACC mutant population, and I don't think we could read through, but it is somewhat concerning.
Just sticking to exon 20 for a second, do you think chemo combination then is the way to go for that set of targets?
It's interesting. For exon 20, the standard of care is amivantamab plus chemotherapy, the PAPILLON regimen. We just saw a data disclosure at WCLC 2026 of zipalertinib plus chemotherapy, and there the PFS looks to be sort of best in class. But again, it comes at a unique toxicity, neutropenia, hematologic toxicity, somewhat driven by zipalertinib, but also compounded with the use of chemotherapy. I think the gold standard in the clinic is if you have a TKI that is active and CNS penetrant that can deliver response rates north of 60%-70% that has intracranial activity as well as durability, that's really the gold standard. That's what we want for our patients.
Great. L et's now get back to atypical mutations and the enozertinib data that we're presenting here and that you've certainly been a part of generating. F irst, let's talk about the patient population in the trial in this data set. 65% of patients with CNS metastases at baseline, including 46% of patients who had untreated brain mets. How would you position this enrolled patient population versus the other PACC data sets that are out there?
Yeah. Deep, this is a very highly selected population by academic clinicians who are enrolling patients on these studies because of the preclinical data, because of the preclinical activity in the brain. We are putting a sicker patient population onto the study. We are putting patients with higher rates of CNS disease. My typical practice is about a third of my patients with PACC mutations have CNS metastases. Here, 65%, that's double what we would expect to see in the clinical setting. I think despite that, seeing robust CNS activity in a population with a heavy rate of CNS metastases is quite reassuring to me. I think as clinicians, investigators, we've selected this population based on the preclinical and clinical intracranial activity that exists with enozertinib.
Now seeing these data, do you have further conviction about CNS activity?
I was quite impressed when we saw this data set. It is a small population, 26 patients, but to see response rates north of 80% in the brain, I mean, that's what we expect for drugs like alectinib or [inaudible] , for example. I am very hopeful and very excited about this CNS activity because there is a significant unmet need in my patient population.
We have been talking obviously a lot about CNS response and our data set, other data sets. Can you perhaps step back a bit and talk just about the sort of methods of assessing CNS response? Certainly, there is RECIST for systemic response, but CNS response involves, you mentioned prior therapy is an issue. The standards that you use, BICR versus not. What are you-?
Yeah. I t is such a gray zone in sort of drug development right now, and I think the gold standard is modified RANO-BM criteria, which is the real gold standard of strict criteria for identifying intracranial activity. Also whether that lesion was treated in the past or not is important. Who selects the target lesion? Is it the investigator? Is it blinded independent central review? I was quite impressed to see your data, not only from investigator assessment, but to a blinded independent central review using the gold standard, the modified RANO-BM criteria that is the gold standard in 2026.
Great. Now let's shift to safety. You've had experience with enozertinib. How would you position enozertinib versus other agents in this space in terms of safety profile?
Yeah. This is a drug with expected EGFR wild type toxicity. We see things like rash, for example, and diarrhea, but relatively well-tolerated. If you look at the data generated, it's what I've experienced with the compound. At 80 mg, the recommended phase II dose, 23% had a Grade 3 toxicity. When you look at the rates of rash and diarrhea, they're all sub 10%. These are side effects that we know well and know how to treat and manage in the clinical practice setting. If you compare and contrast the safety profile with the efficacy profile, I think this is a very manageable safety profile.
We did not see significant cardiac tox, AST and ALT elevations, so transaminitis. These are things that are more concerning in the clinical practice setting. Pneumonitis, we did not see that toxicity with this agent. I think these are toxicities that we know well and know how to manage in clinical practice.
Can you comment, you work at a top-tier academic facility, versus the community oncologist, does the safety profile play well into the community oncology space?
I think it does. Any oncologist with experience with first- and second-generation EGFR inhibitors knows how to manage some of these cutaneous adverse events. I think what community oncologists are not well-versed at managing are maybe transaminitis, AST, ALT in the 100s or 200s, or cardiac toxicity, QTc elevations. I feel more comfortable in the academic setting. But it's common to see dose reduction and discontinuation in the community with those toxicities. With enozertinib, we did not see any dramatic hepatic toxicity or cardiac toxicity in the study.
Great. As you saw, we are, and we've spoken to you previously about the design of our phase III, about how to position the phase III with FDA. Can you comment on the design as we presented?
I think this is a really interestingly designed trial. I'm excited to get it off the ground and enroll patients, who remember this is a significant unmet need in my population. I like that you're studying it versus the control of osimertinib, which is what is commonly used in clinical practice, as well as afatinib, which is the FDA recommendation or guideline recommendation. I think it's a very well-designed trial. I like that you included L861Q, and we'll look at that as a separate subset of patients. Because again, PACC mutant populations and the L861Q population is a group where we don't have tremendous therapies that are approved and are CNS-penetrant for my patient population. I think you're going to see a whole lot of CNS active disease enrolled in this trial, and it's reassuring to see the data from this experience so far.
What advice would you give us on the proportion of patients with CNS disease that we enroll into phase III?
Well-
Should we try to throttle it down?
I would open the study wide open, but you clearly are going to have to stratify by CNS disease. I think, as the data you've shown so far, there's activity, and I would enroll patients with active intracranial untreated disease. That may play out well for you in the control arm, especially if patients are receiving afatinib, where we know afatinib does not have significant CNS penetration. But vice versa, in those patients with L861Q, osimertinib may have better activity. I think it'll be important that you stratify and that we look at that data in more detail. I would enroll this broadly because that's where the patients are, and that's what the patients need.
Josh, thanks. I know you're fresh off a plane, and I know you had to make time in your busy conference schedule to be with us today. Any parting thoughts you want to leave?
Yeah. First off, Pratik, thanks for having me, and it's exciting to get here to ESMO, not at ASCO, at ESMO 2026. I think for this patient population with an active disease, there's still a lot unknown in the community practice. In academia, we've clearly defined this based off the [AMAC 2021] [inaudible]. We're now starting to see this patient population emerge because on osimertinib or afatinib, they're not deriving durable responses. We're not seeing CNS activity. This is a population where I think is one of the highest unmet needs, particularly those patients with active CNS disease. Even in the patient without active CNS disease, primary site of progression is typically the brain in our current sort of regimen. Having a drug that prevents CNS metastases and treats CNS metastases will lead to more durability, but overall better quality of life for our patients.
Thanks, Josh. I'll now turn it over to Keith to discuss the commercial potential of enozertinib. Keith?
Thanks, Pratik and Dr. Sabari. Turning to slide 34, this shows the total addressable market in first-line EGFR atypical non-small cell lung cancer. In the U.S., we estimate an annual incidence of approximately 7,000 patients and assume a treatment duration of 15 months. Then factoring in the current pricing of EGFR TKI therapies, we estimate an addressable market in the U.S. of over $2.5 billion and a total global addressable market of more than $5 billion for this first indication alone. Enozertinib's compelling systemic and CNS activity, combined with its well-tolerated safety profile, has strong potential to deliver durable clinical benefit for these patients. Achieving this highly differentiated best-in-class profile would position enozertinib to potentially capture a substantial share of the more than $5 billion global EGFR atypical market.
Turning to slide 35, a question we sometimes get is whether a best-in-class molecule can still be commercially viable if there are other competitors approved ahead of it. The classical EGFR and ALK markets provide excellent case studies. In the case of classical EGFR, osimertinib is considered best in class based upon its profound CNS activity that has led to best-in-class PFS. Despite it being the fourth EGFR inhibitor to market, you can see how rapidly osimertinib gained share after its first-line approval to quickly become the market leader. Similarly, for ALK, alectinib and lorlatinib have both achieved best-in-class status based upon their CNS activity and long durability. They were third and fifth to market, and since alectinib's first-line approval, these therapies have rapidly become the most prescribed ALK inhibitors.
Turning now to slide 36, to pressure test enozertinib's market potential in front-line atypicals, we had an outside firm conduct independent market research with 75 U.S.-based academic and community oncologists. The physicians assessed blinded target product profiles of enozertinib, noted in the research as product Y, against a competitor next-gen EGFR TKI, noted as product X. The competitor profile reflected published efficacy and safety in EGFR atypical patients, while product Y reflected improved CNS and systemic efficacy and a safety profile with no significant off-target toxicity as compared to the competitor. The improved product Y target profile is one that we believe enozertinib could deliver in the REDWOOD-1 phase III trial based on the compelling CNS and systemic efficacy data thus far.
The key insight from the research was that oncologists indicated they would ascribe the blinded product Y profile approximately 60% share of the EGFR atypical market, reinforcing the notion that a differentiated, best-in-class, highly CNS active therapy can drive prescribing decisions and significant market share, especially considering the high proportion of patients who present with or will develop brain metastases over the course of their disease. This prescribing preference of multiple oncologists expressing strong preference for product Y based on its promising CNS activity is reflected in their verbatims. I'll now hand it back to Jacob for wrap up and next steps. Jacob?
Thanks, Keith. Turning to slide 38, as mentioned today, we believe enozertinib's profile within first line EGFR atypicals is clearly differentiated. In a frontline population with far more challenging disease than competitor datasets, as evidenced by the exceptionally high rate of patients with baseline brain metastases, enozertinib demonstrated exactly the characteristics that produced best-in-class agents time and time again against other oncogenic drivers in lung cancer, namely strong systemic efficacy paired with robust intracranial activity, including in patients with previously untreated brain metastases, plus a well-tolerated safety profile. Slide 39 lays out our roadmap for enozertinib. Having successfully completed our end of phase I interactions with FDA, over the coming months, we plan to complete other global regulatory interactions in anticipation of an expected first half 2027 initiation for REDWOOD-1.
Given the number of patients with lung cancer driven by EGFR atypical mutations and the lack of compelling therapeutic options, we feel it's imperative for us to move enozertinib as rapidly as possible through this final stage of development. Meanwhile, we continue to also work diligently on developing enozertinib for patients with EGFR exon 20 mutations. Given recent competitor developments in that space, it's critical that we thoroughly evaluate all three of our potential phase III treatment regimens to understand whether a monotherapy approach or one of our two combination approaches will yield the best overall profile. We look forward to sharing data on all three of these cohorts concurrently in mid 2027, after which we would look to potentially initiate REDWOOD-2 within that population.
In closing, we're very proud of the team and pipeline that we've assembled here at ORIC, and we're looking forward to providing additional updates for both enozertinib and rinzimetostat over the coming quarters. Before we open it up to Q&A, I'd like to thank our investigators, as well as the entire ORIC team who've worked diligently to tackle our mission on behalf of patients. Most importantly, I'd like to thank our patients and their families whom we hope to help overcome resistance in cancer. With that, let's open it up for Q&A.
Thank you. As a reminder to ask a question, please press star one one on your telephone and wait for your name to be announced. To withdraw your question, please press star one one again. One moment for questions. Our first question comes from Anupam Rama with JP Morgan. You may proceed.
Hi, this is Duke Pelz for Anupam Rama at JP Morgan. Thank you so much for taking our question. You've talked about the higher proportion of patients with baseline brain metastasis than compared datasets. Was there a particular focus on enrolling these patients? It seems like it was physician driven. What does this mean for REDWOOD-1?
Sir, can you repeat your second question?
What does this mean for REDWOOD-1?
Got it. Okay. Let me have Pratik take your first question. Maybe Josh can weigh in on that as well, and then Pratik will take your second question.
Sure. Thanks for the question. I think as you saw from some of the remarks and from the discussion Dr. Sabari and I had, from the beginning, we've been very confident about enozertinib's brain penetrant properties. It was originally designed to be brain penetrant, and that was borne out through all the preclinical profiling. Now I think through the clinical data, we're seeing strong evidence of that. That's why we have, from the very beginning, allowed patients to be enrolled in our study who have active or untreated brain metastases. This contrasts with the narrower enrollment criteria used by all the other competitors. In fact, if you look back at our phase I data disclosure, where we talked about some initial experience in exon 20, we had 86% of patients with brain mets at study entry.
Since then, actually, we've had the rare chance to collect CSF from several patients and saw good penetration of enozertinib into the brain and was able to achieve therapeutic levels. The shorter answer is no. We have not made a specific effort to enroll these patients, but we have shared the preclinical data, the clinical data as it's developed with our investigators. I think they've seen by themselves the potential of enozertinib in CNS disease. Josh, why don't you?
Yeah, I think it's simple. When you're sitting with a patient in the clinic and they have brain metastases and they're asymptomatic from their brain metastases, so less than 1 cm, no vasogenic edema, they're not on any steroids, you have two options. You have an option for systemic therapy, or you have an option for local therapy, typically stereotactic radiosurgery. The norm in this patient population, the PACC mutant population, has been to give radiation upfront because none of the therapies that we've had have CNS activity. Once a study opens that allows patients to enroll with active untreated CNS metastases, we, the physicians and patients, vote with their feet. This is a selected population by the investigators because there are no other options for this population. Put yourself in the patient's shoes.
I'd much prefer to give a systemic therapy that has both thoracic activity as well as intracranial activity than I would for starting radiation upfront. Really, it's a clinical paradigm that investigators were selecting these patients to go on to this study because there were no other options available for our patients.
Thanks, Josh. I think the second question was around REDWOOD-1. As I referenced during my remarks, we did have our end of phase I meeting with FDA and a big chunk of that meeting was spent on the phase III trial design. We also obviously expect to gather ex-US regulatory guidance since REDWOOD-1 will be a global study. The good news is, though, that we're already conducting our phase I in 10 countries, so the program has a sizable global footprint as it is. The other activities to get to phase III include the typical operational activities, onboarding a global CRO that we would engage, and then they would help us define the final countries and sites for the trial.
With that in hand, we would make the country level and site level regulatory and ethics submissions, and all of that would lead to final activation of the trial, which we expect to happen in the first half of 2027.
Okay. Thank you so much.
Thank you.
Thank you.
Our next question comes from Prakhar Agrawal with Cantor Fitzgerald. You may proceed.
Hi. Thank you so much for taking my questions and congratulations on these excellent data sets. Maybe firstly recognizing that it is too early for a mature PFS, but what does a six-month landmark analysis tell you about the durability here that you expect to see with longer follow-up and continued differentiation? Second question, how do you think about the translation of higher systemic and CNS response rates to longer PFS benefit and your confidence on PFS differentiation versus other investigational therapies in this population? Thank you, and congrats again.
Yeah, thanks Prakhar for the questions. I'll ask Pratik to weigh in on your first question and then maybe Pratik and also Josh can weigh in on your second question.
We have presented here six-month landmark PFS, which is 88%, and it's tracking very favorably to both the approved agent and other investigational therapies. But you also need to put this number into context. This number is from a population of patients that has significantly more CNS involvement, twice of what you see in other data sets. In fact, the proportion of patients we have just with untreated CNS mets is higher than others have with any CNS involvement, treated or untreated. A direct side-by-side would ignore these important differences.
That said, we haven't reached a median time on treatment, and the majority of patients are still on treatment, so there's a high censoring rate, and therefore, the median PFS and median duration of response are still immature. But I think we're tracking very well, especially given the patient population that we have enrolled. What are our expectations for how what we're seeing now will translate into longer PFS? We've shared many of the precedents in this space, TKIs against potent oncodrivers in lung cancer, and you see the same pattern playing out repeatedly. Earlier drugs may have had similar systemic ORRs to the CNS penetrant drugs, but they fall short specifically when you look at CNS response in each of these cases. That translates into a much longer PFS for the CNS penetrant drug in the overall population.
The reason being that the brain is the point of first progression in so many of these patients, as Dr. Sabari mentioned during his remarks. If you go back to the alectinib and crizotinib example that we referenced in the presentation, one fact that is buried in the supplemental data is a figure showing the non-CNS progression rates of alectinib and crizotinib, and they're identical. The whole differentiation between those two drugs is on the CNS progression rate. We think we're very well positioned for this to play out with enozertinib.
Josh, you want to add anything?
Yeah, I think in 2026, for any TKI, it's critical to have robust CNS activity. Because without CNS activity, Pratik, like you said, you're going to see site of first progression being the brain. Not only is it a progression event, but patients can often be symptomatic from CNS disease. T here's a high morbidity with regard to CNS progression. I think having CNS activity is crucial. The other thing for not meeting median time on treatment or median PFS, I always educate my residents, my fellows, that's a good thing. That's a patient on therapy remaining on therapy. I hope that we'll see long-term durable results with this therapy. The fact that it has CNS penetration, I think, for me, will improve upon the PFS compared to historical agents that do not have CNS penetration.
Thank you. Our next question comes from Maury Raycroft with Jefferies. You may proceed.
Hi, this is [Farzin] on for Maury. Thank you for taking our questions. For the first one is, for your patients with brain mets, you are very explicit about which of them had measurable lesions versus not, and previously untreated. Why do these distinctions matter, and what are your expectations for phase III on that front? Secondly, you also talked about the CNS activity in this setting. How do you think about the performance of the control arm, like the physician's choice of osimertinib and afatinib? Thank you so much.
Yeah. Thanks, [Farzin]. I think we have touched on both topics at a high level during the call, but maybe I will ask Pratik and also Josh to weigh in on your first question, and then Pratik can talk about your second one.
Measurable lesions in the CNS are more amenable to investigator assessment since they can follow the same rules that they use for RECIST assessment systemically and just apply them to the brain. But when you have a BICR, an independent array of radiologists, so you have two independent radiologists reading each of the scans, and if there is a disagreement, you have a third radiologist who adjudicates. This is a very impartial and well-executed response assessment. Here, you also have expert neuroradiologists. T hey are trained to look at lesions that are of all sorts of sizes, including non-target lesions, smaller lesions. And so you just get a better, more complete assessment of the patient's CNS disease.
I think the measurable versus non-measurable is important when you're looking at investigator assessment, but then the BICR allows you to look at the entire spectrum of disease. Now, we touched earlier on the importance of calling out active untreated mets. FDA has expressed their views through a couple of their published guidances, and we confirmed that through our recent FDA interactions. Specifically when you want to demonstrate CNS activity, FDA is particularly interested in CNS lesions that have not been treated or were treated at least several months before instituting new therapy to isolate the effect of the investigational therapy versus continuing tumor response from prior radiation or a compromised blood-brain barrier from prior therapy that could confound interpretation of brain penetrance.
Because if you have previous weeks of therapy, you can impair the blood-brain barrier, and then even non or poorly brain penetrant compounds can result in some degree of tumor regression in the brain. But that's not truly a demonstration of CNS activity. T hat's why we're calling out these patients with untreated or active brain mets. I think it forms the basis of a very strong demonstration of CNS activity without the confounding that prior treatment can introduce. Now, actually, Josh, do you want to?
Yeah. Pratik, I agree. I think in the clinic, we want to understand if the drug is benefiting the patient in the brain, and if they've had prior radiation or prior systemic therapy with CNS activity or even chemotherapy for that matter, you can't differentiate whether the therapy that you're on, like for example, your enozertinib, is causing the response.
And the FDA is very clear with this, right? W hen you look at modified Brain Metastasis criteria, there's strict rules on how you select target lesions. Lesions have to be progressing on prior therapy. They can't be regressing. They can't be previously radiated, right? Or if they were radiated, they have to be growing. And I think that's really important, right? T he fact that we enrolled patients on this study with active untreated CNS metastases, there's no question in my mind that the regression is related to the activity of the TKI enozertinib. And that's really what we want to define in the clinical practice setting is this drug actively penetrating the brain, and is it causing response in the brain?
In a lot of other studies with TKI or even ADCs, you'll see 30%, 40% CNS activity that just may be due to a disrupted blood-brain barrier. Here, we're seeing responses intracranially north of 80%. And again, most importantly, many of these patients have active untreated CNS disease.
And then your second question was around the CNS activity in the phase III compared with the control arm drug. I think we touched on some of that when Josh and I were speaking. But as he said, and I think is well known, osimertinib does have well-described CNS activity. But the data to support this is largely in the EGFR classical mutation settings where it has its indications
And here, osimertinib has high potency and gets into the CNS at sufficient levels to maintain that potency against classical EGFR mutations. But in our case, we are talking about EGFR atypical mutations. And here, osimertinib has much less potency. W hile it does get into the brain, you are not going to get the level sufficient for it to be as active as it is against the classical EGFR mutations. And so even though osimertinib is conventionally considered a CNS active drug, its activity is still impaired in the PACC atypical mutant population. And so I think we have every ability to differentiate versus that in our phase III. Now, the other drug, afatinib, as Dr. Sabari said earlier, really does not get into the brain very well. It is actually the better PACC drug.
Systemically, it is more active than osimertinib, but it is associated with significant toxicity, which really plays into the risk-benefit of using that drug and why, as Dr. Sabari said, more therapeutic options are needed. W e think enozertinib has a definite edge against both of the control agents that we are going to be employing in our phase III study.
Thank you. Our next question comes from Cory Kasimov with Evercore ISI. You may proceed.
It is [inaudible], not Cory. Thanks for taking the question. Given the relatively limited experience in patients with the L861Q mutation, what informed your decision here to include this subgroup in the phase III REDWOOD trial, and how do you think about the impact to the control arm here?
Yeah. Pratik?
Sure. Well, again, L861Q right now is afatinib's the only approved drug for this. Osimertinib is in the NCCN guidelines for this target. But as you saw from our data, enozertinib appears to be very active against this mutation. Just to dwell on the patients we enrolled, in our evaluable patient population, we had three patients with singleton L861Q mutations, and all three had a partial response. One of these patients had CNS lesions at baseline, and they've had a confirmed complete response per RANO-BM assessment, and all three patients are still ongoing on treatment. We think we have an exceptional drug for these patients. If we were to take them out, they aren't driving our overall numbers. If you take out those three responders, you're left with a pure PACC population, which would be 18 responders out of 22 patients now.
And there now the confirmed ORR is 82%, so still exceptionally high in PACC alone and compares very favorably to the other agents out there, like the firmo number of 68%. But we've included the L861 mutations into our development plan because one, we see that we have exceptional activity, although osimertinib is quite active in these patients, we feel we have the edge here, particularly still with CNS activity. And we will be stratifying them in the REDWOOD-1 design so that they're equally represented on the experimental and control arms. Then finally, it's a sizable patient population. It's 3.5% of EGFR mutations. So we think if we can benefit these patients, we should try to demonstrate that.
Got it. Just a quick follow-up on here. You focused on the benchmarking here on firmonertinib and the amivantamab combo. What's your perspective on the other TKIs that are kind of starting to play in this space, and how does that impact the global commercial opportunities that you estimated there in the slides?
Yeah, this is Jacob. I can take that, and I'll ask Keith to talk about the market opportunity. In terms of the benchmarking, we included any of the competitor programs that are in late-stage development and have enough data that you kind of understand what the profile looks like for a potential phase III. I n other words, a phase III dose. And the only programs that are both in late-stage development and have enough data that you understand what their profile is are really the two you mentioned. T he combo of ami-laz, and then firmonertinib. T here's some other programs that folks are probably familiar with, some of which Josh mentioned in his comments today as well. Silevertinib which to our knowledge, basically doesn't have any data yet at their go-forward phase III dose.
There's zipalertinib that has just over half a dozen patients worth of data. I n other words, too few to understand what the profile is. Sunvozertinib has got about 15 patients worth of data that's pooled across two different ISTs. A gain, I think it's just hard to know what the profile of those compounds is at this point, although obviously we'll watch those competitors quite closely. Then maybe Keith can talk about what the commercial opportunity might look like in the context of both our results as well as the competition.
Sure. Thanks, Jacob. Yes. Based on the EGFR classical and ALK therapy commercial case studies we shared, it's evident that drugs like osimertinib, alectinib, and lorlatinib with best-in-class CNS and PFS profiles achieve market leadership status pretty rapidly, despite them being fourth, third, and fifth TKI to the market respectively. While enozertinib may be a next-in-class therapy to the market, we believe its compelling systemic and CNS activity and potential best-in-class profile could really rapidly achieve market share that's substantial and then become the preferred TKI in frontline EGFR atypical, similar to those drugs that we highlighted in EGFR classical and in ALK.
That prescribed preference came through loud and clear across the oncologist feedback in our blinded product profile market research that I shared, where they described approximately 60% share of the EGFR atypical market to a therapy consistent with enozertinib's target profile. This was across all patients, including those with and without brain metastases. Remember, we tested this with academics and community oncologists and based on Dr. Sabari's notes as well, this would be a profile that would be attractive to both of those settings.
Thank you. Our next question comes from Derek Archila with Wells Fargo. You may proceed.
Hey there. Thanks for taking the questions and appreciate the update today. Congrats on the data. First, just with no treatment-related discontinuations, but dose reductions in 39% of patients, I guess how should we be thinking about the tolerability over a longer treatment period? Has this been consistent with the broader experience at this dose? I guess maybe the bigger question is, what's the ability for these responses to hold after these dose reductions? That's question number one.
Yeah, Pratik.
Sure. As I said during my remarks, we've now treated over 350 patients in our phase I study. Amongst those patients, 150, a little over 150, have received the 80 mg recommended phase II dose that we're moving forward with as a single agent. Some of these patients had other targets, not atypical, but say exon 20 or may have been in the second line or later setting, but they all received the 80 mg dose. Using this larger safety data set as a basis for comparison, a profile we showed today in the first line, atypical patient population is right in line with the overall 80 mg safety profile. The most common AEs remain the most common, so diarrhea, paronychia, dermatitis, acneiform, stomatitis, and rash, and their rates of Grade 3 are also just about the same.
We also, in the larger data set, don't see significant off-target toxicity. I think this bodes well for long-term tolerance of the drug. You can see on the swimmer plot we have patients well past a year, many patients well past a year. Some of them do have dose reductions, but they're maintaining their response. We think 60 is actually a very active dose and is sufficient for those patients who need to dose reduce down to 60.
Very helpful, And there-[crosstalk]
Oh, go ahead.
This is Jacob. I'm just going to add in the other thing there that you'll recall from the phase I dose escalation experience that our first confirmed response is in the 40s in terms of 40 mg once daily. O bviously with 80 mg being the dose in the phase III, we have a lot of room to negotiate, even if there do need to be some dose reductions.
Yep. Helpful. Thank you.
Sorry, second question.
Yeah, second question. I was going to ask from the waterfall on slide 19, it seems there was a slightly lower response in the singleton PACC mutations versus the compound. I do not know, any biological explanation, or would you just attribute that to small numbers? I guess what is your view on the mix we should expect for REDWOOD-1? Thanks.
I think the mix for REDWOOD-1 is probably the mix you see overall. We are pulling from a lot of sites from multiple countries. I think some of the difference that you might be seeing is an issue of small numbers. I would not read too much into that at this point.
Got it. That does.
Thanks, sir.
Thank you. Our next question comes from Laura Prendergast with Stifel. You may proceed.
Hi, this is Charles Kirby on the line for Laura. We were wondering, are you guys articulating any expectations for timing to top line data for ORR for a potential accelerated approval and then PFS for full approval?
Yeah, Pratik, you want to start on this?
Yes. W e do have these dual primary endpoints for the study that is after our discussion with FDA. This is built around an initiative that they have called Project FrontRunner. The concept is to encourage development of new drugs in earlier clinical lines, earlier clinical settings by designing a single trial, randomized trial that has the elements for both an accelerated approval, in our case, ORR, and then follow up for those same patients for conversion to full approval, in our case, PFS. T hat's the design of the study, and we have two endpoints that we'll read out at different time points.
I'll just say in terms of actual specific guidance on timing of top line data, it's probably too early for us to give you that guidance at this point. We'll look to start the trial first half of next year, first half of 2027. After that, we may at that point choose to provide some refined guidance in terms of when to expect top line data.
Got it. Thanks, guys. Appreciate it.
Thank you.
Thanks.
Our next question comes from Asthika Goonewardene with Leerink. You may proceed.
Hi, this is [Lena] for Asthika. Thanks for taking the questions, and congrats on the data. First one is with the recent Firmo update, how has that changed your thinking about atypical? If you can specify both in terms of the market share you can achieve and also how that might structure a clinical trial to address the variability with the tumor.
Yeah, I can take one. I think we covered a little bit and Josh touched on it in his comments as well today. I think the recent events with firmonertinib and their phase III outcome just further highlights the challenges in finding an ideal approach for those patients. I think you now have, in that space, a couple different demonstrations that chemotherapy does pretty well as a control arm in those phase III studies. You've got two examples of small molecules that have failed to beat chemotherapy, a single agent in firmonertinib and mobocertinib. You do have a positive study from sunvozertinib head-to-head with chemotherapy, but you have to keep in mind that their patient population was far from representative. It was just about 12% of patients that had CNS disease in that population, and they had to be previously treated and stable.
And so again, back to the comments Josh made on the call today, it's not in any way representative of a truly CNS active compound, and quite underrepresented in terms of the patients that had CNS mets to begin with. Even after all that pre-selection, they wound up with a hazard ratio that was 0.96 in patients with a prior history of CNS mets. I wouldn't in any way characterize that as a validation of the single agent hypothesis here in terms of going head-to-head with chemo. Now you do have, and we referenced also on the call today, a couple different combinations with chemotherapy that seem to be more successful, at least in terms of the efficacy. Y ou have amivantamab combined with chemo, and then you also saw zipalertinib combined with chemo.
But obviously in both cases, they came at the cost of significant toxicity, of pretty sizable increase in toxicity over monotherapy alone. I think taking it all together, we're looking at all the options with enozertinib. We're looking at enozertinib mono in exon 20. We're looking at enozertinib monotherapy. We're also looking at, obviously, two different combinations. One is with amivantamab and one's with chemo. We'll wait until we have more mature data to decide on what's the best path forward within exon 20.
That's helpful. Thank you.
Thank you.
Great. Thank you.
Our next question comes from Yigal Nochomovitz with Citigroup. You may proceed.
Hi, this is Joohwan Kim on for Ygal. Thanks for taking our question. Maybe just a quick follow-up from us on Johnny's question. Regarding stratification, can you clarify whether the intended use of osi or afatinib will be specified before randomization and incorporated into the stratification? Given that treatment patterns could vary by geography and mutation subtype, how are you thinking about the eventual mix of the two agents within the control arm? Thanks so much.
You are correct that we would stratify a priori by the physician's choice of osimertinib and afatinib. As Josh said during the call, physicians will factor in a number of considerations with respect to their patient when they choose which control arm therapy that they would select for their patients. That selection would be a stratification before randomization in the phase III study. Yes, you are right that there may be geographical differences in the sort of preference for afatinib versus osimertinib from country to country. We expect the stratification to account for those differences. But again, I think we will have good representation of both control arm agents in the aggregate control arm experience.
Go ahead. Thank you.
I am not sure the second half, if there was more to your question than that.
No, that answered it. Thank you very much.
Thank you. Our next question comes from Paul Jeng with Guggenheim. You may proceed.
Hey, guys. Congrats on the great data, and thanks for taking the question. First, on the CNS activity, I am curious if you have done any early subgroup analysis on landmark PFS in patients with CNS mets at baseline and whether you are also seeing strong concordance with the broader population. Then my second question is just what do you view as the bar for ORR, just for accelerated approval for REDWOOD-1 as well as the bar for PFS for full approval? Thank you.
Sure. In terms of the CNS PFS landmark, actually if you subgroup the patients by CNS disease at baseline versus not, they are actually doing equally well. We are not seeing a decrement in PFS projection for the patients who have CNS disease at baseline. V ery well playing out with the profile that we are establishing for enozertinib. With respect to the ORR and PFS, I do not think there is honestly good benchmarks out there. You have the osi afatinib benchmarks, which have a PFS of about 12 months across the PACC patient population, ORRs in the 30%-40% range. We expect to beat that significantly, both in terms of ORR and PFS in our phase III study. But we have not yet disclosed the specific parameters we are designing that study around.
Great. Thanks very much.
Thank you. Our next question comes from Kelsey Goodwin with Piper Sandler. You may proceed.
Hi, this is [Kirsten] on for Kelsey. Congratulations. Two quick questions from us. First, on the control arm of the REDWOOD-1 trial, will you implement a cap similar to what you did for Himalayas to ensure there's no option that can go above a certain rate? Then can you please remind us what assumptions are baked into your cash runway guidance and how many phase III trials are included across rinzi and enozertinib? Thank you.
Yeah. Pratik will take the first and I can take the second.
Yeah. W e still haven't decided about whether we would cap. Our preference is actually not to cap. I think there are, as in the previous question, there will be geographic preferences of afatnib versus osimertinib. There will be patient-specific preferences, so I don't think we want to force the hand of investigators by putting a cap in place for these two agents. It really would then reflect the real-world use of these two agents in the control arm.
Yeah. Then in terms of the cash runway guidance that Dominic has laid out for all of you, it's always been from a conservative perspective. What I mean by that is when we talk about having cash in the second half of 2028, we already accounted for one phase III trial for rinzi.
Obviously, Himalayas-1, which we just kicked off a few months ago, but we also accounted for REDWOOD-1 and plans to initiate that trial in the first half of next year. So exactly in line with the base case that we articulated today for enozertinib. T he announcement today of our plans to initiate REDWOOD-1 first half 2027 does not change the cash runway guidance that we've previously provided, still in the second half of 2028. Obviously, most importantly from a corporate perspective, that cash runway guidance gives us ample runway past the top-line data readout for rinzimetostat in CRPC.
Great. Thank you so much. Just one more. What other activities do you need to complete before you can initiate the phase III study in EGFR atypical?
Sure. As I mentioned during remarks, we've had already a phase I meeting with FDA, but we want to have ex-U.S. regulatory interactions as well, because we plan to do a global study. Then there's all the operational activities, engaging a global CRO, and then generating a final site and country list, and then going to all of those sites for both ethics and regulatory approval. All of that just takes bureaucratic time, and so that still, though, puts us in good position to start the study in the first half of next year.
Fantastic. Thank you so much. Congratulations again.
Thank you.
Thank you. This concludes the conference. Thank you for your participation. You may now disconnect.