Q4 2026 Silevertinib Program Update Webcast - BLACK DIAMOND THERAPEUTICS INC

Good morning and thank you for standing by. Welcome to the Celebratinib program update webcast. At this time, all participants are in 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. To withdraw your question, please press star one-one again. Please revise that today's webcast is being recorded. I'd like to hand the presentation over to your first speaker today, Mark Meleka, CEO of Black Diamond Therapeutics. Mark, please go ahead. Thank you, Operator, and thank you all for joining today's webcast. We are excited to provide an update on silivertinib, which we believe is a best-in-class EGFR TKI for patients with non-small cell lung cancer and glioblastoma. After my introductory remarks, Sergei Urasov, Chief Medical Officer, will present updated Phase II results in frontline patients with non-small cell lung cancer, key takeaways from our end-of-Phase II meeting with the FDA, and our pivotal development plans. He will also provide a brief update on our GBM program. I'll make some closing comments, and then our Chief Scientific Officer, Liz Buck, will join us for Q&A. Before I go further, I would like to remind you that during today's webcast, we will be making forward-looking statements and actual results could differ materially. Please refer to the disclaimer on this slide and the risk factors in our recent SEC filings. Our webcast will focus on a key determinant of progression in patients with EGFR mutant non-small cell lung cancer, brain metastases. We know that approximately 40% of patients with non-classical mutations, or NCMs, present with brain metastases at diagnosis, and approximately 80% of patient with NCMs will eventually progress in the brain. Because of this fact, overall response rate is often not a strong predictor of progression-free survival. Simply put, if a therapy can delay progression in the brain, patients have an opportunity to live longer. Based on our recent end-of-phase-2 meeting with the FDA, it is clear that they recognize the importance of controlling brain metastases to deliver meaningful survival benefit to patients. Today, we are excited to share with you new analyses of our phase-2 data in patients with and without brain metastices. As you will hear from Sergei, these data are highly compelling and demonstrate progression-free survival that is twice that of available TKIs. These data drove a highly productive and collaborative meeting with the agency last month regarding our Phase III plans. We came away with a clear definition of our patient population, patients with all non-classical mutations, and a primary endpoint of CNS overall response rate for potential accelerated approval. Both of these outcomes make silibertinib's path to potential approval differentiated from any other TKI in development. Finally, the FDA acknowledged our justification for our pivotal dose of 150 milligrams. We have been enrolling patients at that dose for the past three months and we'll submit data from that cohort to the FDA prior to our planned Phase III trial, which we expect to initiate in the second quarter of 2027. Before I highlight the data from patients with and without brain metastases, the results from the overall ITT population of 43 patients has not changed since ASCO. Systemic ORR and median PFS remain at 60% and 15 months, respectively. The data from patients with and without brain metastases are unprecedented. In patients with untreated brain metástases, Silibertinib generated an 86% CNS-ORR, which compares to a 35 to 36% CNR-OR seen with Afatinib and Osimertinib from published data. In patients with untreated brain metastases, silibertinib delivered 12.2 months of median progression-free survival, doubling the median progression- free survival of afatinib and osimertinib from published data. And in patients without brain metástases, silivertinib delivered 23.5 months median PFS, compared with 9.8 months seen with osimertenib historically. At ASCO, we reported that none of these 24 patients had progressed in the brain. Now, we are thrilled to report that with a median follow-up time of nearly 16 months, that is still the case. Not a single patient has developed a brain metastasis. We're focusing on patients with non-classical EGFR mutations. They represent about a quarter of newly diagnosed cases of EGF-R mutated lung cancer, and you You can see that in the pie chart on the left. There are over 50 mutations that make up the group of non-classical mutations, and Siliberton potently targets all of them. This is one of the big wins from our FDA meeting. They also agree with this definition of mutations for our patient population. We're not limited to pack mutations, which make up only half of these non- classical mutations. Our patient population is this entire piece of the pie. The reason why the FDA acknowledges this full group for silibertinib is because we've already demonstrated confirmed radiographical responses in so many of these mutations. When you think about trial enrollment or even an eventual label, this is a major win because it's so easy to understand as a physician. A patient can come onto our trial if they have an oncogenic EGFR mutation other than a classical mutation or an exon 20 insertion. Of note, silivertinib is also very active against classical mutations and in preclinical studies is 10 times more potent on L858R as compared to osomertinib. We believe this, combined with silivertonib's robust CNS activity, affords multiple potential opportunities to develop silivertinib in other patient population and lines of therapy, especially in the adjuvant setting. With that, I'll turn it over to Sergei to discuss our recent data, FDA feedback, and Phase 3 plan. Thank you, Mark. First, a quick reminder of the Phase 2 study we'll be discussing today. We enrolled 43 patients in the frontline Phase 2 cohort, all in the United States and Canada. This is critical because, as we all know, EGFR-TKI studies conducted primarily in Asia can produce early data that do not always stand up in global trials. Our trial also enrolled a real-world patient population with over 33 unique non-classical mutations, unlike other trials which enrolled a much smaller number of mutations. The data we presented at ASCO this past May had 11 months of median follow-up and now, with almost 16 months of median follow-up, we see that the systemic aura and median PFS are mature and maintained. We're also seeing striking CNS activity and evidence that silivertinib is delaying progression in the brain in both patients with and without brain metastasis. Well, let's talk about this striking CNС activity now because it is the key driver of the durable benefit we're seeing in patients both with or without baseline brain metastasis. These robust data drove our FDA conversation and laid the foundation for our unique phase 3 design. First, silivertinib has shown the strongest brain penetrance of any EGFI TKI out there right now. As seen in the panel on the left, silvertinib is the only EGFi TKI to demonstrate high levels of unbound drug directly measured in brain tumors from treated patients. As shown in the histograms, silovertinib achieves high concentrations in both non-contrast enhancing and contrast enhancing tumor regions well above the threshold of 5x IC50. This means that silobertinib crosses an intact blood-brain barrier to deeply penetrate all areas of the tumor. Inability to cross the blood-brain barrier is a significant limitation of most other drugs. The next panel is our Phase II results in frontline patients showing an 86% CNS ORR per RAIN or BM criteria in patients with baseline brain metastasis who received no prior radiation treatment to the brain. That's a critical piece because many other studies report CNS responses from patients who have been pretreated with radiation. Furthermore, Rayna BM criteria are more stringent than resist. FDA only accepts CNS-OR per Rayna-BM to evaluate CNS activity. The 86% CNS OR for silivertinib is more than double the historical CNS or R data for osimertinib and efatinib. Of course, it's not just the response rate that matters. Durability is also critical. Our median PFS is more than 12 months, while you can see benchmarks with osimertinib and afatinib are in the 5 to 6-month range. So, silivertinib is doubling the median PHS of currently available treatments. Now, turning to the benefit we're seeing in the 24 patients in our study without baseline brain metastasis, the results are remarkable. With nearly 16 months of median follow-up, none of these 24 patients have progressed in the brain. And I should remind you that every patient in our study gets a brain MRI every six weeks. So even if a patient developed an asymptomatic brain metastasis, it would be detected by an MRI. These surveillance scans are often missing in other trials. Our patients without baseline brain metastasis demonstrate a median PFS of 23.5 months, almost two years. For context, oesometinib is at 9.8 months in these patients. Finally, our overall incidence of CNS progression is one quarter of what is seen in patients who receive afatinib. This impressive data allowed us to align with the FDA on a differentiated path to approval that no one else can take on right now because they lack the robust CNS data that we have. Our end-of-Phase II meeting with the USDA was the most collaborative meeting I've had in my 20 years in the industry, and we got everything we hoped for out of the meeting. The first critical win is the definition of our patient population, because it is simply all oncogenic EGFR mutations, excluding the two classical mutations, exon 19-del and L858R, and exon 20 insertions. There was no packed discussion. There was not an alphabet soup of mutations. The FDA recognized our robust data showing activity across a broad group of non-classical mutations far more than any of our competitors. So, our patient population is the entire group of non-classical EGFR mutations. Next, the feedback we got from the FDA on our study endpoint is, I think, what makes this trial such an exciting opportunity for us. The FDA recognized the fact that for a drug to be successful in frontline non-small cell lung cancer, showing brain activity is key. If you win in the brain, you win overall. That's why our endpoint for accelerated approval is CNS ORR per Raynaud BM criteria, not systemic ORR for resist. Systemic ORR, and this is direct FDA language, just needs to be similar to the control with OS showing no detriment versus the control. And that is a huge fundamental difference from what anyone else must show. And then, of course, full approval is based on systemic PSS per resist, as is common practice in frontline lung cancer trials. Finally, on dose, this was a simple and short conversation. The FDA reviewed all the data we provided from those in over 200 patients with this drug, and they acknowledged our rationale for the 150-milligram dose. All they asked us to do is provide safety data from a sufficient number of patients dosed at 150-millimeter dose before enrollment starts in our Phase 3. We expect to deliver those data in the second quarter of 2027, as we have been enrolling patients at 150 milligram in cohort four of our ongoing phase two trial since August. This FDA meeting was a home run for us, and together with our data allowed us to design a differentiated phase three trial that positions filivertinib, if approved, to be the drug of choice for all frontline patients with EGFR non-classical mutations. Our planned phase 3 trial will enroll patients with non-classical mutations which are routinely detected by standard NGS testing. When a patient is diagnosed, they all get an MRI as part of standard of care, and they either have brain metastasis or they don't. If they have brain Metastasis, they enroll in the randomized part of the trial, which is the top section of this slide. This 300-patient trial will look at cilovertinib versus investigators' choice of afatinib or osimertinib. The accelerated approval endpoint, as we discussed previously, is CNS-ORR per rena BM criteria. As you can see, we currently sit at 86% CNS ORR versus the comparators in the mid-30s, and the trial is well-powered to detect an even smaller difference. The endpoint for full approval is median progression-free survival, and we're running now at 12.2 months versus historical control of five to six months. So once again, the trial is well-powered to show such benefit. In parallel to the registrational trial, everyone who does not have brain metastasis, which is approximately 60% of patients, goes into a single-arm open-label cohort that will evaluate response rate and PFS and is expected to be completed at the same time we're filing for accelerated approval. This creates a potential opportunity for us to engage with the agency at the time of label negotiation using that single- arm data from patients without brain metastasis. At the minimum, we would see conclusion in NCCM guidelines exactly how osimertinib is currently used today for patients with non-classical mutations. The key is that silivertinib is sitting at almost two years progression for survival right now in these patients without brain metastasis versus 9.8 months for osimertenib. To put it in further perspective, silivertinib's performance so far in this patient population is better than what osinertinib achieved in the FLORA study for patients with classical EGFR mutations, which was 19 months. Finally, it's not on this slide, but you will see it in our upcoming catalyst slide a little later, the FDA also suggested we run a cohort for patients with leptomeningeal disease. These patients currently have a very poor prognosis and no good treatment options. I think that really speaks to the potential the agency sees in our data set and desire to work with us to bring better treatment options to patients who have significant unmet medical need. A quick note on safety. We have seen no new safety signals and median dose intensity in our phase 2 study is at 126 mg per day. As shown on the PK curves on the right, Siliverdinib shows dose linear PK, and we're confident that 150 mg dose will provide more than adequate exposure for efficacy while improving tolerability. Now let's quickly turn to our phase 2 trial in patients with glioblastoma, or GBM. GBM is a highly aggressive cancer with a very poor prognosis. Approximately half of patients carry oncogenic EGF alterations, representing almost 50,000 newly diagnosed patients globally. Unfortunately, there are very few treatment options for patients with GBM because of silyvertinib's high brain penetrance and potent coverage of all oncogenic EGFI alterations, we believe that we have the right molecule to finally bring a precision medicine to patients with GBM. Our current phase two is a randomized study of silivertinib plus temozolomide versus temozолomide alone in a defined patient population, patients with EGFR variant 3 and MGMT unmethylated status. Our phase two initiated this past spring, starting with the safety run-in, and we are pleased to report that we have completed that part one, and the trial is now enrolling patients in the randomized part two at 150 milligram dose. The primary endpoint is PFS by independent assessment, with overall survival as a secondary endpoint. All survival analysis will be conducted by an Independent Data Monitoring Committee, or IDMC. Interim PFS data are expected in the first half of 2028. And if the interim PFS are positive, there is potential to convert the trial to a Registrational Phase III trial with an overall survival endpoint. End point. And now I'll turn it back to Mark. Thank you, Sergey. In terms of catalysts, we have many over the next 24 months. Namely, for the lung program, we expect cohort four data in the second quarter of 2027. Assuming we initiate our phase three trial in the the second quarter of 2027, we expect preliminary data from the two open-label cohorts of the Phase 3 in 2028, and CNS-ORR data, the endpoint for potential accelerated approval, in 2029. And for GBM, interim PFS data are expected in the first half of 2028. So, in summary, we believe that silivertinib is a best-in-class EGFR TKI that has the potential to deliver substantial clinical benefit to patients with non-small cell lung cancer and glioblastoma. With the support of the FDA, we are excited to advance this novel therapy to patients who are in need of better treatment options. With that, we will now open the call for questions. Operator? As a reminder to ask a question, please press star 1-1 on your telephone and wait for your name to be announced. To withdraw your question, press star 11 again. Please stand by while we compile the Q&A roster. And our first question comes from the line of Farzan Haq of Jefferies, the line is now open. Hi. Good morning. Thank you for taking my question. I had a question on the phase three. Like across the U.S., Canada, Asia, and the European sites that you're planning, how much does the NCM subtype mix and standard of care differ? And could that affect like phase three enrollment or control and performance? What are you hearing feedback from the site PIs on the proposed design and willingness to randomize the patients as they come in? Yeah, so I'll turn that over to Sergei, but I just want to make sure I understand the question. So it's choice of Afatinib versus Osamertinib in these different countries that you mentioned, and I think Sergei will mention also South America and Mexico that we're also planning. Sergei? So thank you for your question, a very important consideration, and we're confident there will be no trouble in enrolling this patient exactly because we have an investigator's choice there in the control arm. So there are areas like United States where semertinib is often used off-label in this patient, and that can be an option. Now, there are other countries in South America, particularly in Asia, huge markets in Asia where afatinib is the dominant choice, and most investigators are very comfortable with using that drug. So the trial can accommodate all the spectrum. And so I think we believe that it will be very representative of the actual real-world practice that exists for patients with non-classical mutations. Got it. And then the follow-up is, I can understand why FDA was comfortable with CNS ORR for accelerated approval, but there are two competitors. They're basically pursuing approval based on systemic ORR. And you said that FDA only expects the BDT-AXA systemic ORRR to be comparable to co-control and they're sufficient. Does that imply that competitors have risks in actually using the tissue superiority versus the same control that you're using? We think it does. If you look at a FATNF, it delivers a 66% systemic ORR. That is the comparator arm in both of our competitor trials. That has a very high bar to overcome. Thank you. Yep, makes sense. Thank you. One moment for our next question. Our next question comes from the line of Kelsey Goodwin of Piper Sandler. Your line's now open. Oh, hey. Good morning, and congrats on the update and all the progress here. I think two quick ones from me. In terms of the endpoint CNS-ORR, what kind of follow-up did the FDA ask for with that? I think you had said data in 2029, and then maybe building on that last question in terms of kind of how this will maybe allow you to make up some time versus competitors. How do you think about timing now and are you able to catch up a bit to the two competitors mentioned previously? Thanks so much. Maybe I'll take the competitor piece and then I'll turn it over to Sergey to talk about, you know, readout and follow-up. We do believe we can catch up. We think this will take about 18 months to enroll. We expect to start in the second quarter of next year, so that's why we're confident of data in 2029. And Sergey, you can comment based on your in-depth discussion with the agency on what they're expecting to see when we deliver that CNS ORR data. Yeah, exactly. So the sufficient follow-up is built in as part of that CNR ORR analysis. It's a very important aspect. And so Our expectation is that around the time we complete enrollment in the trial, there will be sufficient data set, not just to describe CNS or I and do the comparator that FDA requires us to do, but also provide sufficient data on the durability of that benefit. So that all comes as one single package, and it should be in early time enough to drive the accelerated approval filing. Okay, great. Thank you so much. And then maybe just one more, if you don't mind. In terms of the safety cohort, cohort four, do you need any follow-up there? And is there anything else that needs to be demonstrated in terms of efficacy, or is that solely safety-based? Thanks. The way we see it, it's a checkbox that has to be done before a phase three trial opens up. So a minimal follow-op, simply to describe, if we will, the rate of dose reductions. As you may recall, it was at 82% when we showed, when we presented ASCO. And it's very simple to demonstrate that it will be, there will be substantial improvement in tolerability early on. Okay, perfect. All right. Thank you so much. Congrats again. Thanks, Kelsey. Thank You. One moment for our next question. Our next question comes from the line of Shawn McCutcheon of Raymond James. Your line is now open. Hey, guys. Thanks for the questions, and congrats on the Phase 2 data holding up with longer follow-up. A couple from us, could you speak to your powering assumptions on CNS RRR by Rano BM for the Phase 3 study, how that informed your trial sizing, and your confidence in that lower bound based on the fact that you're using a lower starting dose than you did in phase two? And then perhaps, you know, the obvious question, but maybe we can get some pushback to this end, but can you speak to, you Know, a bit more on the market sizing, you know, patients with brain meds and can you kind of give a sense of the anticipated population with non-classically UFR mutations untreated with active or stable brain med and do you expect a meaningful enrollment of patients with asymptomatic or stable CNS metastasis? Thanks. I think I'll turn the first question over to Sergei regarding the powering of the study, and I can handle the second one on market. So, Sergei? So, it's more than that. So, let me just go to the benchmarks where we are today, 86% versus 35%. So currently, it's more than well-powered to show that difference. If anything, if the number goes down for the experimental arm, there's still sufficient power to demonstrate a difference the way the FDA expects based on the confidence interval. And now, of course, for progression-free survival, again, we make appropriate adjustments to make sure that the study remains well-power beyond the phase 2 estimates that we currently have at hand. So in regards to the second question, I think it was two parts to the second question. One was just, you know, can you enroll these patients? Can you find them? We're confident we can. We were able to do that in the phase two. I'll remind you, we had 19 patients with brain mets. Sixteen were untreated. Patients do not want to receive radiation to their brain. There's significant sequelae. And in our conversations with investigators who will be part of this phase three program, it's very clear that this is a big opportunity for them to offer a pill for a patient with a brain met rather than radiation. As far as the market, yes, the registrational trial is focused on patients with brain mets, but that single arm cohort is going to read out at the same time. Should it deliver the same magnitude of benefit that we're currently showing two-year PFS, we are confident that this drug will be used in all patients with NCM, which is 25% of the EGFR market, three times the size of the Exxon 20 insertion market, double the sizeof the PAC market. Excellent. Thanks, guys. Thank you. One moment for our next question. Our next question comes from the line of Laura Prendergast of Steve Fogelman is now open. Hey, guys. Congrats on the update. I wanted to follow up on Kelsey's question a little bit. Can you provide more detail on your dose selection discussions with the FDA, exactly what you need to show from cohort four, and how fast you could be able to turn around the submission of data from cohort 4 into initiation of this trial? Yeah, I think, Sergey, maybe comment on when dose reductions occur, because that obviously affects timing of delivering these data. We're highly confident we're going to be able to do that given that we've been enrolling since August. Go ahead. Exactly as Mark opened it up, it's a very straightforward exercise. We know that the median time to dose reduction is approximately six weeks. So within the six weeks, we'll know whether the patient requires it or not. Now, that's really the focus of FDA request. There is no specific requirement to the number of patients that we need to submit. So even a small number of patience may be sufficient. Got it. Thank you. Can you also speak to the screening for EGFR mutations involved in the strategy? Basically, can any patient with any EGFR mutation that's not classical or an exon 20 enroll? And then as far as how rigorous or simplistic this could be, how does this compare to your competitors who need to identify a PAC mutation? Maybe I'll take the first part. Yeah, it is that simple. If you don't have a classical or an Exon 20 insertion, you're in our trial, which was a big win from the FDA meeting. I think Liz can comment on, you know, how these are determined and differences from the competition, which is a tougher exercise. Yeah, that's right. So, these non-classical mutations are readily detected using standard next-generation sequencing tests already in effect for this patient population. This is a much broader group of mutations than is described for PAC, where you really have to go to a specific specified list of these non-classical mutations in a much smaller patient population. Got it. Thanks very much, guys, and congrats on the update today. Thanks, Laura. Thank you. One moment for our next question. Our next question comes from the line of Robert Burns of HC Wainwright. The line is now open. Hi, guys. Thanks for taking the question today, and congratulations on the data. It's quite impressive, especially with the longer-term follow-up here. First question for me, maybe going back to the powering for the Phase 3, you know, can you provide a little more granularity with regard to the CNSOR delta that would, you know, trigger statistical significance here? And then the second question, obviously, you know, I see that you're pursuing NCN addition and compendia listings. Maybe you could talk talk a little bit about, you know, what you expect the uptake to be should that actually occur rather than a full FDA approval. Maybe, Sergey, you want to take the first one? I'm powering out. So powering, you Know, the study is sufficiently powered to show, let's say, 35-point difference, right? So it's right now we're at double. So even if there was, you know, if there was a change in a real world trial, we have enough power. The important thing to remember is that the benchmark that we're showing there for afatinib and osimertinib is pretty solid. We've talked to a lot of people around the world who has a very extensive experience with afatinab and osimetinib. And there is a strong consensus that those are not very CNS active drugs. So that's why getting off the CNS OR gives us the best chance of showing the difference. And to your question about, you know, will compendia listing be sufficient? We absolutely think it will be because I point everyone to a recent paper, Meow It All, Journal of Thoracic Oncology, which describes in hundreds of patients treated with osomertinib how prevalent CNS progression is. 75% of patients treated with osomertinib who have non-classical mutations progress in the brain. So, you know, that is a big issue. And if we can now show in our single-arm cohort that these patients are not progressing, the drug will be widely used, even if it is only included in the NCCN guidance. Awesome. Thanks, guys. And congrats again thank you thank you for joining us today this concludes today's conference call you may disconnect have a good day