Good day, and thank you for standing by. Welcome to the Black Diamond Therapeutics Phase II Update webcast. At this time, all participants are in a listen-only mode. After the speaker's presentation, there will be a question-and-answer session. To ask a question during the session, you will need to press star one one on your telephone. You will then hear an automated message advising your hand is raised, and to withdraw your question, please press star one one again. Please be advised that today's webcast is being recorded. I would now like to hand the presentation over to your first speaker today, Mark Velleca, CEO of Black Diamond Therapeutics. Mark, please go ahead.
Good morning, and welcome to our webcast. Before we begin, I'd like to remind you that we will be making forward-looking statements. Today, we are very pleased to present encouraging initial results from our phase II trial of silevertinib in frontline patients with non-small cell lung cancer. We will also describe our plan to conduct a randomized phase II trial of silevertinib in patients newly diagnosed with glioblastoma, or GBM. Each section of the presentation will include commentary from a clinical expert in these two areas of oncology. We believe that silevertinib has broad potential to benefit non-small cell lung cancer patients across multiple lines of therapy. Today, we will focus on data in frontline patients presenting with non-classical EGFR mutations, a large market with significant unmet medical need. Here are the key takeaways from today's webcast.
One, silevertinib is clearly active in frontline lung cancer patients with a robust response rate across a broad range of distinct non-classical EGFR mutations. Two, with durability of response continuing to mature, we expect that PFS data in the second quarter of next year will enable us to chart the course for a registrational study. Three, we continue to engage potential partners for pivotal development, as we believe that a partner can help maximize the opportunity for silevertinib to benefit patients globally. Four, the anti-tumor activity for silevertinib in the brain is remarkable. As you will see in the lung data presentation, there are multiple examples of patients with extensive CNS disease who experience striking responses and ongoing clinical benefit. Five, given this robust CNS activity and promising clinical results for silevertinib in patients with GBM, we plan to initiate a randomized phase II trial in newly diagnosed GBM patients.
We believe that the properties of silevertinib make it uniquely positioned to overcome prior shortcomings of EGFR-targeted therapy for GBM patients. And finally, given our capital-efficient operating model, we expect to execute this randomized phase II trial and deliver data with the cash that we have on hand today. Now, I'd like to turn the presentation over to Sergey Yurasov, Chief Medical Officer at Black Diamond. Sergey?
Good morning, everyone. I'm very pleased to share the promising initial data from our phase II trial evaluating silevertinib in frontline non-small cell lung cancer patients with non-classical EGFR mutations. To remind you, silevertinib is the most advanced fourth-generation EGFR TKI in development. Based on encouraging data in over 200 patients, we believe that silevertinib has the potential to benefit both patients with non-small cell lung cancer and patients with glioblastoma. We all know that osimertinib/Tagrisso has provided clinical benefit to more than one million non-small cell lung cancer patients with classical EGFR mutations. However, as we have recently learned from the widespread use of the next-generation sequencing, there is a subset of patients presenting with non-classical mutations, or NCMs, for which there remains significant unmet medical need.
NCMs represent a group of over 100 distinct oncogenic mutations, which collectively comprise approximately 25% of the EGFR mutation space for newly diagnosed patients. As shown by numerous published studies over the past several years, referenced here in the center panel of the slide, the EGFR TKIs of gefitinib, osimertinib deliver limited benefit to NCM patients with substantially shorter progression-free survival in patients with non-classical mutations versus patients with classical mutations. PFS is also significantly shorter in patients with baseline brain metastasis, and the afatinib shows a PFS of only 6.1 months. silevertinib has the potential to address the needs of patients presenting with the full spectrum of non-classical EGFR mutations. We have demonstrated that silevertinib potently inhibits NCMs that are not covered by osimertinib. Moreover, unlike afatinib, silevertinib is highly brain-penetrant.
Patients with EGFR NCMs are especially prone to brain metastasis, with approximately a third of patients presenting with CNS disease at diagnosis. As shown in recent publications evaluating afatinib and osimertinib, there is a higher rate of CNS progression for NCM patients as compared to patients with classical mutations. And it is expected that the emergence of CNS disease limits overall survival for these NCM patients treated with second and third-generation TKIs. silevertinib delivers potentially the best brain penetrance among all EGFR TKIs, as demonstrated by superior preclinical Kp,uu evaluation and by clinical data demonstrating pharmacologically relevant drug levels in brain tumor tissue. Moreover, we have previously presented data for silevertinib demonstrating CNS responses in non-small cell lung cancer patients who had previously progressed on osimertinib. And today, I will present compelling CNS response data for frontline NCM patients. This phase II trial enrolled 126 patients across three cohorts.
We reported last fall initial data in patients with recurrent disease, cohort one and two, and will present final data next year. Today, we're reporting the initial phase II data of silevertinib in frontline patients. We enrolled 43 patients presenting with 35 unique non-classical mutations. More than a third of these patients presented with baseline brain metastasis, including a subset with measurable brain lesions, which enabled us to assess CNS response. Patients were treated with oral silevertinib at a daily dose of 200 milligrams, a dose selected based on our clinical experience with silevertinib in the recurrent setting. As of the data cut on November 3rd, patients had a median follow-up of 7.2 months, with 29 patients remaining on therapy. All patients were treated at sites in the U.S. and Canada. 74% of patients were Caucasian and 16% Asian.
Patients presented with 35 unique non-classical mutations, including PACC and classical-like NCMs. A subset of 16 patients presented with compound NCMs. Of the 43 patients in our ITT population, 40 patients received at least one post-baseline scan and can be visualized on this waterfall plot. 26 patients of 43 patients achieved a confirmed radiographic response by RECIST 1.1 criteria for an ORR of 60%. 25 patients experienced a confirmed partial response, and one patient experienced a confirmed complete response. This patient population presented with a diverse spectrum of NCMs, which is what is expected in a real-world setting. Just over half of patients presented with PACC mutations, and just under half presented with classical-like or other mutations. A subset of patients presented with compound NCMs. Responses were observed across all mutation subclasses, with the ORR ranging from 56% - 73%. The disease control rate is over 90%.
Robust CNS activity is important for NCM patients, as they frequently present with CNS disease at baseline. Seven patients in our study presented with target CNS lesions at baseline, and we observed a confirmed CNS response by RECIST in six of these patients. The seventh patient also had a CNS response, which is awaiting confirmation. We believe that the CNS activity of silevertinib is critical for driving prolonged progression-free survival. Consistent with the radiographic responses seen across patients with a wide range of NCMs, we observed striking molecular responses, as evidenced by decreased variant allele frequency of VAF across mutations. Among a group of 26 patients evaluable for VAF ctDNA changes, we observed decreased VAF for all mutations in all patients, and with complete clearance in 81% of patients.
With a median follow-up of 7.2 months, it is too early to calculate median PFS, and duration of response is not yet reached. However, as can be seen on this swimmer plot, preliminary durability trends are encouraging, especially in those patients with CNS disease. While the majority of patients experienced a response on the first scan, a subset of patients presenting initially with stable disease experienced their responses at subsequent scans, second to fifth. All patients who had an opportunity for confirmation of their response did ultimately confirm. This waterfall plot captures the depth and preliminary duration of response we observe across patients. The median time to first response is 42 days, and the majority of patients experience deepening of responses over time, with a median time to deepest response of 126 days.
It is noteworthy that while the majority of patients underwent dose reduction, with median time to dose reduction of 74 days, patient response continued to deepen after dose reduction. Indeed, several patients experienced their first response after a dose reduction. The CNS activity achieved with silevertinib is striking. We show here a case example of a patient who presented with a large target brain lesion at baseline. This patient experienced a rapid and confirmed CNS response, as shown in the radiographs on the right. Multiple other target lesions in the lung and liver also responded, and this patient remains on study today. silevertinib exhibits a tolerability profile consistent with the EGFR TKI class, with the most common adverse events being rash, diarrhea, paronychia, and stomatitis.
Consistent with other EGFR TKIs, we found that adverse events were higher when patients are treated with silevertinib in the frontline setting, as compared to our prior experience with silevertinib at the 200 milligram dose in the recurrent setting. Adverse events were managed by supportive care, dose interruption, and/or dose reduction to 150 milligram or 100 milligram daily, with minimal impact on response depth or durability to date. Indeed, several patients experienced their first response after dose reduction, and nearly all patients with a response showed maintenance or deepening of response after dose reduction, including all patients with CNS responses. In this phase II study, we demonstrate robust efficacy for silevertinib in frontline patients presenting with 35 unique NCMs, a mutational profile that reflects the real-world setting.
Patients presenting with distinct subclasses of EGFR NCMs, including PAC and classical-like mutations, demonstrated confirmed ORR rates of 56%-73%, with an overall rate of 60%. The tolerability profile for silevertinib is typical for an EGFR TKI, and these were managed with supportive care and dose interruptions or reductions without compromising efficacy. Lung cancer patients with EGFR NCMs are prone to CNS metastasis, and previous studies with second- and third-generation EGFR TKIs have shown erosion of efficacy in patients who present with CNS disease. Strikingly, silevertinib has achieved responses in all seven patients with target CNS lesions, six confirmed and one awaiting confirmation. Initial evidence of durability, together with robust CNS activity, bode well for prolonged PFS in the NCM population, where CNS progression is historically high. These data support the continued development of silevertinib in the frontline setting for patients presenting with a broad range of EGFR NCMs.
I'd now like to introduce Dr. Balazs Halmos, an investigator who participated in this frontline study of silevertinib. Dr. Halmos is Associate Director for Clinical Science and Director of the Multidisciplinary Thoracic Oncology Program at Montefiore Einstein Comprehensive Cancer Center. He's also a member of the ASCO Non-Small Cell Lung Cancer Guideline Committee. Dr. Halmos, thank you for being with us today. I would like to start with the first question around the unmet medical need for patients with EGFR non-classical mutations. We could see it over the last several years that this has become a more important topic at many scientific meetings, and we'd very much appreciate your understanding of where the unmet medical need for these patients.
Absolutely. It's a very important chapter of precision medicine in a way. We uncovered about 20 years ago that patients with non-small cell lung cancer fall into multiple different categories based on the mutational subtype of their particular lung cancers, with EGFR-mutated lung cancer becoming kind of, you know, this high-profile subset of precision oncology. But identifying these EGFR mutations can yield us, you know, being able to offer excellent targeted therapy for the majority of these patients with what's called classical EGFR-mutation positive disease. Unfortunately, that leaves a smaller subset of patients behind, patients with so-called uncommon or atypical EGFR-mutation positive disease, where the targeted therapies have some efficacy, but the efficacy is modest. Really, really leaves a lot, a lot, lot behind. You know, we need better treatments for those patients. We currently can offer second or third-generation molecules such as afatinib or osimertinib.
We'll just have modest activity for this particular class of patients, and afatinib also lacks CNS penetration, a very important element, since these patients many times either present or will ultimately develop brain metastasis. So there's a very strong need for new medicines that can be more efficacious and that can have brain penetration and manageable toxicity for patients with uncommon EGFR-mutation positive disease. Thank you. Next, I would like to touch on CNS metastasis in non-small cell lung cancer and the challenges that you face when managing these patients. What are the typical treatments that they get, and how we can improve on the current practice? So it remains a huge challenge, both in terms of the number, but also the impact on the individual level.
About 25% of the patients will present at the outset with brain metastasis when they have metastatic EGFR-mutated lung cancer, and another 25% will develop it over the course of the disease. So one out of two patients ultimately will need to face, you know, brain metastasis, a very frightening term just on its own. Just think about cancer penetrating into the most critical organ that makes us human, and the treatment of it traditionally has not been easy. It might have been surgery, it might have been radiation to the entire brain. Just imagine the potential side effects of either of those approaches as to functional deficits that they can leave behind. We've improved in terms of local treatment options with tailored radiation courses, but still the toxicity is significant.
So having medical treatments that can effectively penetrate the brain and treat the disease without the need of radiation is tremendous. We're able to offer it now for many types of patients with biomarker-positive metastatic lung cancer, such as classical EGFR-mutation positive lung cancer. But the treatments could improve, and we certainly do not have effective treatments for uncommon EGFR-mutation positive lung cancer when they have brain metastasis. So the need, you know, continues to exist, and we're very hopeful that we'll have better treatments for such patients. Thank you. You had patients treated on our silevertinib clinical trial. Can you please share some of your clinical observations from this study? Absolutely. You know, I'm just so fortunate as a clinical investigator.
It's always just a thrill to participate in trials with new agents, you know, that have the hope of delivering, but not each and every agent will deliver, and not each and every one of our patients will benefit. And we've had the good luck at Montefiore participating in the silevertinib trials, you know, where we had patients both with uncommon EGFR mutation disease with nice response, you know, to the drug, as well as patients with brain metastasis, in fact, bulky brain metastasis with tremendous shrinkage and durable benefit from this excellent agent. And lastly, we've also had a patient who had developed resistance to the earlier generation of compounds with a special mutation that the earlier compound, you know, was, you know, kind of resistant to, and this was overcome, you know, by silevertinib as well.
So again, we've had three sets of, you know, excellent, you know, favorable responses that we've witnessed with this molecule. So feel very fortunate that we've had a chance to participate and see our patients, you know, get great benefits out of it. Thank you. Last but not least, I would like to ask you to give us your perspective on how to best manage EGFR TKI-associated toxicities that may emerge during treatment, including silevertinib. Definitely. You know, by now we have 20 years of experience with targeted therapies that we use to manage patients with lung cancer, including multiple generations of EGFR targeted agents such as NAB, the next generation drug silevertinib.
You know, there's an expectation that many of these medicines will have some very specific side effects, both gastrointestinal, such as diarrhea, as well as skin, both related to the normal activity of EGFR in these organs. We've also learned, you know, we have developed tools whereby we can manage these toxicities, both, you know, Imodium, of course, you know, for diarrhea, as well as multiple ways of managing the skin rash that can develop, that can include topical agents, steroids, antibiotics that can include oral antibiotics. We have nice regimens, ways of stepping it up, preventative prophylactic ways of, you know, managing it as well. Lastly, of course, some patients will need those holds, you know, for some time until, you know, the toxicity improves or dose reductions over time.
We've witnessed now through a number of generations of drugs that many times those dose reductions ultimately do not compromise the outcome of patients. We can maintain the efficacy by maintaining, you know, the quality of life that our patients deserve on these excellent, you know, targeted medicines that patients might many times need to take for a long, long time. So quality of time is, quality of life is absolutely critical, and I believe that with these strategies, we're able to achieve that for silevertinib just as well. Thank you very much. Last, I want to thank you for participating in our clinical trial and for, you know, great care that you took of our patients. Thank you very much. And thank you so much, Sergey. It was a great, great pleasure and great honor to be with you today. Thank you.
Thank you, Dr. Halmos. My name is Elizabeth Buck, Chief Scientific Officer at Black Diamond. I will be providing an introduction to our planned phase II trial for treating newly diagnosed glioblastoma patients with silevertinib. While EGFR inhibitors have transformed the treatment paradigm in non-small cell lung cancer, all past attempts to position EGFR inhibitors in the GBM setting have failed. We believe silevertinib overcomes the challenges that contributed to these past failures and has the right properties to effectively treat EGFR-altered GBM patients. We now know that GBM tumors are characterized by the expression of multiple EGFR variants. silevertinib demonstrates highly potent preclinical activity across the full spectrum of EGFR alterations found in GBM patients. The pharmacology for EGFR alterations in GBM has also been misunderstood in the past. These EGFR variants, including EGFR variant 3, can actually be paradoxically activated by reversible EGFR inhibitors like Erlotinib and Dafitinib.
We designed silevertinib with a covalent binding mechanism that does not cause paradoxical activation, and finally, CNS penetrance is critical to the success of any therapeutic to treat GBM. silevertinib delivers potentially the best CNS profile for an EGFR inhibitor. This has been demonstrated by clinical PK data in GBM tumors showing pharmacologically relevant drug levels in brain tumor tissue, and is further supported by the exceptional CNS anti-tumor activity that we presented today in lung cancer patients. Oncogenic EGFR variants are found in approximately 50% of all GBM patients. These variants are often co-expressed with other EGFR alterations and EGFR amplification. EGFR variant 3 is one of the most common oncogenic mutations in GBM. Patients with this variant represent over 4,000 patients in the U.S. alone. EGFR variant 3 is one of the best understood oncogenic mutations in GBM. This mutation is highly oncogenic through a unique activation mechanism.
It is also present as a compound mutation in nearly all cases. Compound EGFR mutations are highly actionable in non-small cell lung cancer patients, and this may also be the case in patients with GBM. We have demonstrated in both cell-based and in vivo models that silevertinib inhibits EGFR variant 3 and all other EGFR alterations found in GBM. silevertinib also improves survival in an intracranial tumor model derived from a newly diagnosed GBM patient carrying the variant 3 mutation. In summary, we believe silevertinib addresses the key factors contributing to past failures of other EGFR-targeted therapies. As we compare silevertinib's profile to data for these other therapies, only silevertinib has been shown to deliver potent broad-spectrum activity together with robust CNS penetrance. Now I'd like to turn the call over to my colleague at Black Diamond, Chief Medical Officer Sergey Yurasov.
Sergey will discuss the clinical experience with silevertinib in GBM patients to date, as well as our plans for a randomized phase II trial in newly diagnosed patients.
Thank you, Liz. As shown in the left two panels of this slide, silevertinib has demonstrated encouraging tolerability and preliminary efficacy across two trials in GBM patients with recurrent disease. On the next slide, I'll summarize the data for silevertinib in GBM patients with such disease. First, tolerability. As shown in the table to the left, silevertinib is generally well tolerated in 38 patients dosed at 200 milligrams daily. To date, more than 60 GBM patients with recurrent disease have been treated with silevertinib, with the longest on therapy for more than 16 months. silevertinib has demonstrated encouraging efficacy, including a range of responses in this heavily pretreated GBM patient population.
Moreover, based on our own preclinical data and clinical data from others, we expect that the combination of silevertinib plus temozolomide or TMZ can be safely administered to patients. Most importantly, the phase 0/I trial, also called the window of opportunity study, has clearly demonstrated that pharmacologically relevant exposure of silevertinib can be achieved in GBM tumor tissue, as shown on the next slide. In the window of opportunity study, patients received silevertinib at 200 milligrams daily for five days prior to their scheduled surgical resection. After resection, PK analyses are performed to measure silevertinib levels in plasma, CSF, and tumor tissue. Importantly, silevertinib achieves a mean exposure level in brain tumor tissue that well exceeds the IC50 of silevertinib for EGFR variant 3 and other EGFR alterations.
Exposure in non-enhanced brain tumor regions is especially important because these are the regions where the blood-brain barrier is intact and highlight the strong brain penetrant properties for silevertinib. Unfortunately, GBM patients today have limited treatment options. Standard of care is surgery followed by radiation plus Temozolomide and then maintenance with TMZ. The overall survival benefit for patients receiving surgery, radiation plus TMZ, then TMZ maintenance therapy is two and a half months. In the unmethylated or MGMT negative patient population, TMZ provides only marginal benefit. Those patients who are methylated or MGMT positive have better survival outcomes. We believe silevertinib has the potential to benefit GBM patients expressing EGFR alterations. To rigorously test that hypothesis, we have designed a robust randomized phase II trial evaluating the combination of silevertinib plus TMZ versus TMZ alone. The design principles for that trial are shown on the next slide.
The design of this trial integrates features which we believe are critical for success. First, we will focus on those patients with the highest unmet medical need, the unmethylated or MGMT negative population, which represents approximately two-thirds of all GBM patients. We will further focus on patients presenting with EGFR variant 3, the most common and best understood oncogenic mutation present in GBM patients. EGFR variant 3 expression is routinely measured in standard diagnostic tests already in place. We will be measuring both PFS followed by OS, as it is clear that response rate is historically unreliable in glioblastoma. Finally, as you will see on the next slide, we have a robust statistical analysis plan that will be governed by an Independent Data Monitoring Committee, or IDMC. Part one of this trial is a safety run-in to establish that silevertinib can be safely administered with TMZ.
Part two is the randomized portion of the trial where we will randomize 150 patients after they have completed chemoradiation. Patients will either receive TMZ plus silevertinib, the experimental arm, or TMZ alone, the control arm. IDMC will monitor the study for safety and perform both an interim and final analysis for efficacy. The primary endpoint of this trial is PFS by blinded independent committee review, followed by overall survival as the secondary endpoint. We are already gearing up for this trial to initiate in the first half of next year and expect to deliver data in 2028. I'd now like to introduce Dr. Minesh Mehta to comment on the considerations in designing the trial to targeted EGFR-altered GBM and the potential for silevertinib to benefit this patient population. Dr. Mehta is the John and Mary Lou Dasburg Endowed Chair in Radiation Oncology at Baptist Health Miami Cancer Institute.
He's also the chair of the Brain Tumor Committee for NRG Oncology. Dr. Mehta, thank you for being with us today.
It's a pleasure, Sergey.
I would like to start with a question that probably on many's minds today. With only two successful phase III trials in newly diagnosed glioblastoma, what are the key lessons that the FDA and oncologists working in this field have learned when designing studies for patients with such severe and dismal brain tumor?
Sergey, that's a critically important question, and there are at least four lessons that we have learned that explain many of the failures of the past. We've seen, for example, trial design and endpoint selections being a crucial element. Many times we see people using small, non-randomized phase II trials that show promising results that then subsequently fail to show promise.
So this issue of using non-randomized trial designs has been a historic problem of the past. More importantly, if you do a small trial, and even if it's randomized, if you don't power it appropriately, then you're going to run into power calculation problems and statistical analysis problems, and that'll serve to cause the trial to fail. The second key factor is endpoint selection. If one does not very carefully select an endpoint, you could end up with an endpoint that, like a mirage, potentially looks positive, but in the end does not give you sufficient data for positivity. And a very good example of this is radiographic response. Many people have used radiographic response only to show that this often has minimal to no impact on either progression-free survival or overall survival, which are really the key endpoints that patients care about and that the FDA cares about.
Making sure that these critical endpoints of progression-free survival and overall survival become the drivers of any clinical design is a crucial element for future success. One of the issues with progression-free survival is to ensure that progression-free survival is not just simply the opinion of one individual, but it's truly validated. The way you validate that is by really using an independent central review committee, and this approach of using such a committee to validate progression-free survival is crucial. That can then give you a go-no-go signal to move into a phase III trial. The final fourth point is patient selection. Glioblastoma is not just one disease. It's a disease composed of many molecular subcomponents, and these have different prognostic and predictive implications.
And if you do not homogenize the population that you're selecting, it is very likely that you'll have a trial that will not succeed. So these are at least four reasons. There are probably many more, but these are the key reasons that explain some of the failures of the past.
Well, thank you for that perspective. You have participated in many very large, very well-designed clinical trials that investigated EGFR-targeted agents for patients with newly diagnosed glioblastoma, and they were unsuccessful. So when you look back on your experience with these clinical trials and targeting EGFR agents, why do you think those agents were not successful?
So like with many other drugs, one of the key factors that we often do not pay sufficient attention to is the drug, does it actually get to the tumor?
This issue of getting the drug into the tumor is a critical question that is frequently inadequately addressed. Part of the challenge in terms of answering this question is that one can think of glioblastoma as a disease with two compartments. There is the central enhancing compartment where the blood-brain barrier is mostly broken down, it's imperfect, and many things can penetrate that part of the tumor and enter the tumor. That gives sometimes false hope that a drug is actually getting in because it gets into the enhancing component of the tumor. The key challenge is the component of the tumor beyond the enhancing component, the so-called non-enhancing component of the tumor. This is a component that does not typically enhance with classic gadolinium agents on MR imaging, but we know categorically that microscopically, if you resect this part of the tumor, there is disease there.
If the drug does not penetrate that portion of the tumor, it doesn't matter how much drug gets into the enhancing portion, you have a large component of the tumor that remains untargeted. So I think a key reason for failure of many, many agents is this poor penetrance of the drug into the non-enhancing compartment of glioblastoma. The second reason explaining the failure of some of the prior trials that targeted EGFR alterations is that the term EGFR alteration is used as a large sort of description of one entity. In reality, there are many pathways of EGFR oncogenic drivers of glioblastoma. You can get EGFRvIII mutations, you can get EGFR amplification, you can get other pathway activations that stimulate EGFR, and all of these contribute to oncogenesis.
If only one component of this EGFR targeting is addressed by an agent, it is unlikely that that strategy will be successful. One really needs to shut down all EGFR-driven oncogenesis in order for an anti-EGFR agent to be successful, and that is where drugs in the past have not been successful. I'm glad you mentioned the complex biology of glioblastomas, and I want to ask you, if you were to pick an EGFR target, what would be the pros and cons of going after a variant 3 in this disease? I think there are a few key factors that an ideal or a classic drug would require in order to demonstrate success against an EGFRvIII mutated glioblastoma. First and foremost is the blood-brain barrier penetration that is critical, that is crucial.
Secondly, it must have specific on-target activity with minimal off-target activity to minimize the toxicity of the agent. Thirdly, it should have pan-activity across the EGFR oncogenesis pathway and not just limited to the EGFRvIII mutation. And fourthly, there is the phenomenon of paradoxical activation. When an inhibitor is used to target a specific target, such as, for example, EGFRvIII, there is actually the possibility that after target binding, instead of inhibition, one can actually get paradoxical stimulation. And should such a thing happen, the activity of the agent would be dramatically diminished. And so ensuring that no paradoxical activation is occurring is also a key design element of the appropriate targeting molecule.
Well, thank you. You've been in this field for many years. My last question to you is, why are you still an optimist in this field? Why do you still continue working on opening new trials and looking for new therapies for these patients despite prior failures? What makes you an optimist?
Sergey, that is a very important question. As an oncologist, we can approach life in one of two ways. One is target the diseases where we have made great progress and we have very favorable outcomes, and we will have patients for whom we can do a lot in terms of prolonging their survival. But that only serves one part of the universe of oncology because we have the other part of the universe of oncology where we have challenging diseases. These patients have needs. In fact, they probably have more needs because their therapeutic choices are highly limited. These are the patients that need a lot of guidance, they need a lot of direction, they need a lot of clinical trial approaches.
In all honesty, although progress has been slow, the science has progressed dramatically, and that has led to some small incremental steps in survival improvement, but with the promise that we will probably turn the corner on many of these very difficult-to-treat diseases with better improvements in therapeutic targeting.
Thank you very much for your work, and thank you for sharing your thoughts today with us. Thank you.
It's a pleasure, Sergey.
Thank you.
Thank you, Dr. Mehta, for your thoughtful comments. I'd like to make a few closing remarks and then open the call for Q&A. We now have a clear path forward to advance silevertinib. While we await PFS data in our lung cancer trial and continue our partnering discussions, we will initiate a phase II trial in GBM.
Our balance sheet enables us to execute the GBM trial independently, and we expect to deliver data in 2028. Based on everything we have learned from the more than 200 patients who have received silevertinib, we are confident of its potential to bring transformative benefit to patients with lung cancer and glioblastoma who need better treatment options. In closing, we would like to thank our investigators for your collaboration. We are also grateful to all the patients who have placed their trust in us. And to the 21 employees of Black Diamond, thank you for working tirelessly to advance silevertinib. Operator, please open the line for questions.
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. And to withdraw your question, please press star one one again. Please stand by for our first question. The first question comes from Farzin Haque with Jefferies. Your line is open.
Hi, congratulations on the update, and thank you for taking my question. You have 29 patients of 43 patients still on therapy and have very high CNS response rates. What is the anticipated threshold for median PFS to justify a relatively smaller size pivotal versus what your competitor ArriVent has alignment for? And then what are your expectations for the control arm given that you have a broad mutational coverage and potential accelerated approval path?
Thank you, Farzin, for the question. Mark here. Sergey, can you take that?
Thank you, Farzin. That's a very important question about CNS activity and what we know, and I think our investigator highlighted that outcomes in patients with CNS disease are dismal. So we expect approximately 2/3 of the PFS that typically is seen for overall patient population. These patients with CNS disease would run to 50%-60% of that number. Now, what is really important to remember is that it's a larger proportion of patients who have non-classical mutations who develop CNS disease. And so the fact that we have patients with these ongoing CNS responses bodes really well for our projected PFS.
And then the second part regarding potential control design or trial design?
Correct. So it's certainly subject to discussion with the FDA when we have PFS data in hand, but I can say that right now when we look at what is accepted as standard of care, we certainly have a substantial leg up against both osimertinib and particular afatinib on CNS activity.
Got it. Thank you so much.
Thank you. And the next question will come from Kelsey Goodwin with Piper Sandler. Your line is open.
Hey, good morning. I guess maybe to build on the last question, given the evolving landscape, what in this dataset stands out most to you? I don't know if the KOLs are still on the line, but in terms of differentiation for silevertinib. And then secondly, in terms of upcoming regulatory discussions and the potential for a partner, I guess how important is a partner in your view to have ahead of the regulatory discussions to kind of have them side by side with you when you're talking to the FDA in order to get a trial design that's optimal in your view? Thank you.
Yeah, thanks, Kelsey, for the question. I'll take the second part, and then maybe Liz can comment on what strikes us with the data versus some of the molecules that are already approved and from osimertinib. We do think a partner can maximize the value to both patients and shareholders for this drug, but we need a partner, the right partner who shares our vision of how much benefit this drug can do for patients not only with lung cancer, but with GBM. We also need the right structure to maximize value for shareholders, and then to your point, it needs to be done at the right time. It could be that that partner is by our side at the FDA meeting. We will be prepared to go if that partner is not by our side.
What we've done in the GBM study is be able to generate a robust dataset with capital that we currently have without a partner. So that, again, maximizes the value of the drug across two indications. As far as kind of what strikes us about differentiation of the drug, Liz can comment, and then again, Sergey can reiterate what he said about expectations for a pivotal trial.
Yeah. So there are really two major aspects to comment on first about what differentiates silevertinib from other EGFR TKIs. First and foremost is the broad spectrum activity. We now know that non-classical mutations comprise dozens of unique variants. In order to be effective in that space, the drug has to target all of those variants. In the study that we presented here today, we're describing a real-world population.
This population presented with 35 unique non-classical mutations, and we show quite impressive clinical activity throughout that spectrum of mutations. The other aspect that truly has silevertinib standing out from other EGFR TKIs is brain penetrant properties. And this is really demonstrated in this patient population where CNS metastases are so common. A third of patients presenting with CNS metastases at baseline, and the vast majority of patients who progress on afatinib and osimertinib develop CNS metastases. We demonstrated activity in all of these patients with CNS target lesions, and so it really bodes well for what we would expect to see for durability of the agent in the setting.
Perfect. Thank you. I think we can get that answer.[crosstalk]
All right. Thanks, Kelsey.
Thank you. And the next question will come from Sean McCutchen with Raymond James. Your line is open.
Hi, guys. Contrasting the data, and thanks for the questions. There's a couple from us. First, how are you framing the efficacy in the classical-like patient population, perhaps patients where OSI would be relatively effective, at least for L861Q, and how does that color your view on the breadth of patients you'd like to include in a pivotal program? And then secondly, you do have a relatively high grade three cutaneous tox. Any novel approaches you're looking to implement on mitigation and prophylaxis as you think ahead to a phase III protocol like the COCOON regimen for ami Labs, for instance? Thanks.
Sure, Sean. Thank you. Liz will take the first part of that question, and Sergey the second part. Liz?
Yeah. So you're absolutely right. In this broad patient population, we evaluate patients both with PACC mutations and groups of classical-like mutations. Now, we recognize that osimertinib actually performs quite well within a limited number of these classical-like mutations, and you specifically pointed out the L861Q mutation where historically OSI has done quite well in that setting.
We recognize that, and where we also have shown just with a couple of patients in this population who presented with L861Q, and we also did quite well in that patient population, but I'd also point out that our population included many of these non-classical-like mutations that historically osimertinib has not done as well in. I would even point out some of the atypical exon 19 deletion insertion mutations where it is known that osimertinib does not perform as well in that patient population. There is an unmet need in that setting.
These are a structurally diverse group of mutations versus classical mutations, and where we were quite impressed with the activity in that population as well.
Sorry. And I can follow up on the management of EGFR TKI-related adverse events. I would refer to what our investigator described, over 20 years of experience and learning experience how to successfully manage the side effects of EGFR TKI. So those protocols are very well established, and the first step is patient education and provided supportive care along the established guidelines. The second important piece in managing the toxicity is the fact that we have a wide therapeutic index, and that allows us for successful dose reduction without losing efficacy, without compromising the responses. So with those things in hand, as you can see, the discontinuation rate has been quite low.
Last but not least, if you carefully look at our swimmer plot, you can see that we have now five patients who continue treatment beyond progression, which means those decisions are made based on patients' quality of life, ability to tolerate the drug, but also when they face subsequent therapeutic choices, and they have to make that decision.
That answers your question, Sean? [crosstalk]
Thank you. Our next question will come from Robert Driscoll with Wedbush. Your line is open.
Thanks, guys, and congrats on the data. Just maybe on the GBM study, have you discussed this phase II design with the FDA at all, just maybe with a view to thinking about support for pivotal design in the future?
I can tell you the FDA is aware of our intent to run this phase II trial. Can't comment beyond that.
Okay. Then maybe a question for Liz. Do we know if the EGFR variant 3 mutation confers a negative prognosis in GBM for patients? And maybe if so, how are you thinking about that in the study?
Yeah. So we know that EGFR variant 3 is in almost all cases coincident with EGFR amplification. So we may not have that data specifically for EGFR variant 3. We do know that patients who carry EGFR amplification, that this is a negative prognostic factor. And so really focusing on patients with EGFR variant 3 allows us to also include patients with high amplification of that receptor as well.
Got it. Thank you very much, guys.
Congrats on the data again.
Thank you. As a reminder, to ask a question, please press star one one on your telephone. And the next question comes from Ellen Horste with TD Cowen, and your line is open.
Hi, guys. Congrats on the data. Thanks for taking our question. We're just wondering how should we compare the EGFR-specific tox like the GI and the skin rash to the first afatinib dataset that was more biased toward Asian patients? And in general, how much better is EGFR tox tolerated in Asian populations, and to what extent do you think this contributes to the discrepancies that we've seen in the safety profiles for these two drugs? Thank you.
Sure. Sergey?
Thank you for this question. There are numerous studies of first and second-generation EGFR TKIs that were conducted in the Western world versus Asia. And we can go back all the way to two LUX-Lung 6 studies that were well analyzed, and the outcome of such meta-analysis clearly shows that the rate of EGFR TKI on target toxicity is substantially lower in studies that were conducted in Asia.
Thank you. And the next question will come from Laura Prendergast with Stifel. Your line is open.
Hey, guys. Congrats on the great data. Can you provide a little bit more color regarding dose selection? In pretreated patients in the phase II, you said that you had investigated 100 milligrams and 200 milligrams to satisfy a . And based on still having deepening responses despite dose reductions, it seems like you do have pretty active doses at 100 milligrams and 150 milligrams . Could you maybe elaborate a little bit more on the profile that you observed in those patients you treated with 100 milligrams and how it compared to the 200 milligrams , and just generally any other color on dose you can provide?
Maybe I'll just take the first part to remind folks that we will have the recurrent setting data being presented in the second quarter of next year. That will be a full dataset, and we'll be able to show all of the 100-milligram patients that were in that dose optimization. As far as our approach to dose for the pivotal study, Sergey, perhaps you can comment.
Yes. So, Laura, very good question. The important thing to remember is that we have established firmly that doses of 100 milligram and above provide very good therapeutic coverage, and we've seen responses at doses at 100 milligrams and above. So with that in mind, we will discuss the choice of a dose for the pivotal study, and it is possible that we may run a small lead-in phase as part of a phase of a pivotal study design. But that, of course, is subject to PFS data that we're waiting for and discussion with the agency.
Got it. Thank you. And just one more question for me. Regarding the glioblastoma study, is there any possibility that this could have any registrational intent given the unmet need for unmethylated patients?
The study is not meant to be registrational. Obviously, it all depends on data. It's well-powered to be a very compelling proof of concept phase II.
Got it. Thank you. Congrats again.
Thank you. Our next. Oh, I do apologize. One moment, please. The next question will come from Brad Canino with Guggenheim. Your line is open.
Hey, good morning. One for me for the three PACC non-evaluable patients at the bottom of the swim lane that fell off within under a month. Was that because of toxicity or symptomatic progression withdrawal or something else? I have a follow-up.
Those patients discontinued due to adverse events that occurred early before scans.
And then you mentioned the potential of looking at a safety run-in for a lower dose. Would you prefer 150 mg or 100 mg? And then I don't know if you disclosed within the slides for the reductions, for the 77% reductions, what proportion of those went to 150 milligrams versus went down to 100 milligrams ? Thank you.
Yes. Go ahead, Sergey.
Yeah. So I can provide details. So all patients that required a dose reduction were dose-reduced to 150 milligrams . A small subset of those patients went down to 100 milligrams , and all of those patients maintained the response.
Great. Thanks.
Thank you. This does conclude today's conference call. Thank you for your participation, and you may now disconnect and have a good day.