Morning. Thank you for standing by, and welcome to the Cullinan Therapeutics zipalertinib REZILIENT3 Results Conference Call. As a reminder, this call is being recorded. A slide deck that you may find helpful while listening to this call is available on the Events section of Cullinan's Investor Relations website at investors.cullinantherapeutics.com. It is now my pleasure to turn the call over to Nick Smith, Head of Investor Relations at Cullinan Therapeutics.
Hello, everyone, and thank you for joining us on today's call to discuss the results of zipalertinib + chemotherapy from the phase III REZILIENT3 trial in first-line EGFR exon 20 insertion mutation non-small cell lung cancer. My name is Nick Smith, and I'm the Head of Investor Relations at Cullinan Therapeutics. Before we begin, I would like to remind you of the safe harbor provisions outlined on slide number two. During today's presentation, management will be making certain forward-looking statements which are based on current information, assumptions, and expectations that are subject to change and involve risks and uncertainties, which may cause actual results to differ materially from those contained in such forward-looking statements. These risks are fully described in the company's filings made with the Securities and Exchange Commission, including our annual report on Form 10-K. You are cautioned to not place any undue reliance on these forward-looking statements.
In addition, this call contains time-sensitive information accurate only as of the date of the live broadcast, September 14th, 2026. Cullinan undertakes no obligation to revise or update any forward-looking statements to reflect events or circumstances after the date of this call. As shown on slide three, joining me on the call today are Nadim Ahmed, Cullinan's Chief Executive Officer, and Dr. Jeff Jones, Chief Medical Officer. It is my pleasure to now turn the call over to Cullinan CEO, Nadim Ahmed. Nadim?
Thank you, Nick, and welcome everyone to our call today to discuss the remarkable and potentially practice-changing results for zipalertinib in combination with chemotherapy from the phase III REZILIENT3 trial in first-line EGFR exon 20 insertion mutation non-small cell lung cancer. Today, we're very much looking forward to providing you with an overview of the results and also our perspective of the exon 20 commercial opportunity, where we firmly believe zipalertinib combination therapy will become a first-line standard of care treatment. These results were showcased last night as a late-breaking oral presentation during the Presidential Symposium at the World Conference on Lung Cancer in Seoul, South Korea. The fact that the study was selected for this plenary session is a clear testament to the significance of these results.
I will also point out that the Presidential Symposium is focused on featuring notable advances in lung cancer research and treatments with the potential to change clinical practice. Before I ask Dr. Jeff Jones, our Chief Medical Officer, to dive into the REZILIENT3 results, I would like to take a moment to emphasize key elements of the compelling profile of zipalertinib for patients with exon 20 non-small cell lung cancer. Zipalertinib, which we are co-developing with Taiho Oncology, is an oral EGFR inhibitor with unique design characteristics that enable a potentially best-in-class efficacy and safety profile. It was just a few weeks ago when we first announced that the REZILIENT3 trial met its primary endpoint of median progression-free survival early at a planned interim analysis. This is the first and only study in first-line exon 20 non-small cell lung cancer to achieve its primary endpoint at the interim analysis.
The six-month improvement in median PFS demonstrated by zipalertinib combination therapy is the best PFS result observed to date in the front-line setting. In fact, the first treatment that has shown a median PFS longer than 12 months in this patient population. In addition, phase II studies have already demonstrated the differentiated and broad clinical profile of zipalertinib across multiple patient subsets. Zipalertinib monotherapy has demonstrated clinically meaningful results in relapsed patients, in patients with brain metastases, and patients harboring uncommon mutations. With our comprehensive clinical development program, we have the opportunity to establish zipalertinib as a future standard of care across multiple EGFR non-small cell lung cancer indications. Firstly, in the second-line setting, our NDA for treatment of relapsed exon 20 non-small cell lung cancer based on the results of the phase II REZILIENT1 trial is under active review by the U.S. FDA.
The results of the phase III REZILIENT3 trial, the subject of today's presentation, enable potential indication expansion into the front-line setting. Our parallel REZILIENT2 phase II study is enrolling a cohort of patients with uncommon mutations in the metastatic disease setting, where we showed promising response rates already. Finally, Taiho has the phase III REZILIENT4 trial, an ongoing adjuvant study in patients with exon 20 and uncommon mutations. We look forward to sharing these data in the future. With that, let me turn it over to Jeff to dig into the zipalertinib REZILIENT3 pivotal results. Jeff?
Thank you, Nadim. As Nadim said, we are excited to share results from the randomized phase III portion of the REZILIENT3 study. To begin, let's review the study design. REZILIENT3 is a phase III clinical trial to evaluate the efficacy and safety of zipalertinib in combination with chemotherapy in adult patients with previously untreated, locally advanced or metastatic non-small cell lung cancer harboring EGFR exon 20 insertion mutations. Patients were randomized 1:1 to 1:2 treatment arms. In the investigational arm, patients were treated with oral zipalertinib 100 mg twice daily in combination with platinum-based chemotherapy. In the control arm, patients received only platinum-based chemotherapy. As of the May 2026 data cutoff, the study enrolled 279 patients, of which 140 were treated with zipalertinib + chemotherapy and 139 were treated with chemotherapy alone.
The primary endpoint for the study was progression-free survival, as assessed by blinded independent central review per RECIST criteria. While the primary analysis was expected after 162 progression-free survival events, a pre-planned interim analysis was to be conducted after 122 progression-free survival events. Of the 140 patients randomized to zipalertinib + chemotherapy, an impressive 62% of patients were still receiving treatment as of the data cutoff. Slide seven summarizes key attributes of the enrolled patient population. While EGFR exon 20 insertion mutations occur with higher frequency in Asian populations, nearly 60% of patients were enrolled in centers outside of Asia, primarily in the U.S. and Europe. More than half, nearly 60% of all enrolled patients had an ECOG performance status of one.
Most notably, approximately 31% of patients enrolled had a history of brain metastases, a rate far higher than the previous PAPILLON and WU-KONG28 studies and a rate more representative of real-world patient populations. As noted previously, the primary endpoint of this phase III trial was progression-free survival, depicted by treatment group on slide number eight. Zipalertinib + chemotherapy delivered a remarkable, statistically significant 50% reduction in the risk of disease progression or death. Zipalertinib plus chemotherapy demonstrated an unprecedented median progression-free survival of 14.5 months, compared to a median of 8.5 months for chemotherapy alone, a statistically significant and, more importantly, clinically meaningful difference of six months. Extending beyond the previous ceiling of less than 12 months, the progression-free survival of 14.5 months and an absolute benefit of six months over chemotherapy alone is truly without precedent and underscores the potentially practice-changing strength of these data.
Other efficacy variables are summarized on slide number nine. Zipalertinib + chemotherapy demonstrated a 65% overall response rate, comparable to the results previously reported for the PAPILLON and WU-KONG28 regimens employing amivantamab and sunvozertinib, respectively. What stands out, however, is the durability of response and clinical benefit. Zipalertinib + chemotherapy demonstrated a median duration of response of 14.2 months. Again, unprecedented in this patient population. Turning now to slide number 10. While the results were still maturing at the May 2026 data cutoff, initial overall survival outcomes are promising. While not yet statistically significant, the hazard ratio of 0.72 favoring zipalertinib + chemotherapy is very encouraging. The management of CNS metastases in patients with non-small cell lung cancer remains a critical unmet need. The subgroup analysis of patients with CNS metastases depicted on slide number 11 is very important.
Survival outcomes in this patient group were favorably impacted by treatment with zipalertinib + chemotherapy. Patients with CNS metastases experienced a notable improvement in progression-free survival when treated with the combination regimen, a PFS hazard ratio of 0.38. Moving on to slide number 12. While patients who received zipalertinib + chemotherapy did experience a higher incidence of adverse events and treatment discontinuations, dose reductions and supportive care may favorably impact patient management. Importantly, higher-grade adverse events were most frequent during the first four cycles of platinum combination chemotherapy, particularly hematologic adverse events. Higher-grade hematologic adverse events were notably less frequent and less severe thereafter. Adverse events occurring in more than 25% of patients are detailed on slide number 13. In general, the safety profile of zipalertinib + chemotherapy was consistent with the safety profile of the individual agents and consistent with the previously reported zipalertinib monotherapy data.
The most common treatment-emergent adverse events in this study were low-grade EGFR-related dermatologic toxicities and chemotherapy-related cytopenias. While direct cross-trial comparisons cannot be made, treatment with zipalertinib plus chemotherapy demonstrated a differentiated profile with regard to grade three or higher diarrhea and rash. Both adverse events are well-known on-target toxicities associated with EGFR tyrosine kinase inhibitors. The overall incidence of rash and diarrhea was also lower. To summarize, for patients with EGFR exon 20 insertion mutations, these data represent some of the best results yet reported in the first-line setting. The combination of zipalertinib and platinum-based chemotherapy reduced the risk of disease progression or death by 50% and delivered a statistically significant and clinically meaningful six-month median PFS improvement as first-line treatment for patients with EGFR exon 20 insertion mutation non-small cell lung cancer.
While the survival data are not yet mature and the observations not yet statistically significant, at the interim analysis, there was already a favorable trend toward improved overall survival among patients with zipalertinib + chemotherapy. Importantly, the observed safety profile of the zipalertinib combination regimen was manageable and largely consistent with the safety profile of the individual components of the regimen. Lastly, the independent data monitoring committee recommended unblinding the study, and crossover to zipalertinib monotherapy will be allowed. Continued follow-up is ongoing to further characterize the overall survival benefit and other exploratory endpoints. I'll turn the call back over to Nadim, who will highlight how zipalertinib fits into the exon 20 treatment landscape and the commercial opportunity in that population.
Thank you, Jeff. Despite the data sets from the PAPILLON and WU-KONG28 studies, there is still a persisting unmet need for patients with exon 20 non-small cell lung cancer. The REZILIENT3 study results for zipalertinib combination therapy clearly demonstrate the regimen's potential to be a best-in-class treatment for patients with first-line exon 20 non-small cell lung cancer. Zipalertinib combination therapy demonstrated a median duration of response of 14.2 months and median PFS of 14.5 months. While direct comparisons across studies can't be made, the six-month difference in median PFS is an unprecedented result relative to the results observed today in the front-line exon 20 setting, where previously, a three to 4.5-month median PFS benefit has been observed for monotherapy and combination therapy respectively. These data also support a combination approach over monotherapy treatment with a tyrosine kinase inhibitor for improved clinical outcomes in the front-line setting.
REZILIENT3 also enrolled the highest percentage of patients with brain metastases as compared to other first-line trials in this patient population. The results of zipalertinib combination therapy in this specific patient population with difficult-to-treat disease are also very encouraging given the unmet need and growing population of patients with brain metastases. We also previously shared the promising results of zipalertinib monotherapy in patients with relapsed exon 20 non-small cell lung cancer from the REZILIENT1 study, where again, there was a higher proportion of patients enrolled with brain metastases. The progression-free survival results in the relapse setting are also encouraging, indicating that both responding patients and those with stable disease derived clinical benefit from zipalertinib treatment. Relative to amivantamab + chemotherapy, as well as sunvozertinib monotherapy, we're also encouraged by the safety profile of zipalertinib combination therapy.
Overall, zipalertinib combination therapy demonstrated a differentiated safety profile for adverse events associated with EGFR inhibition, including skin toxicity and diarrhea, which supports the proposed mechanism of action being more selective for mutant versus wild type EGFR. The observed safety profile, taken together with the unprecedented efficacy results discussed today and robust clinical activity across different segments of patients with non-small cell lung cancer, provides a clear opportunity for zipalertinib to become a new standard of care across multiple disease settings, which has always been our focus for the program. On slide 18, looking at the potential opportunity for zipalertinib in the U.S., approximately 16% of non-small cell lung cancer cases harbor EGFR mutations, with insertions at exon 20 accounting for 12% of these mutations. This results in a relatively conservative annual U.S. incidence of approximately 3,000 first-line patients and 1,500 second-line patients.
Importantly, represents an important billion-dollar-plus total addressable market opportunity. With the clinical data delivered by zipalertinib regimens in the front-line and relapse settings, we would expect significant uptake and large market share across patient segments. With our partner, Taiho, we also have future opportunity for continued revenue expansion by generating important data in our ongoing REZILIENT2 and REZILIENT4 studies, where Taiho is exploring zipalertinib in uncommon EGFR mutations in the metastatic setting, and as an adjuvant therapy in both exon 20 and uncommon EGFR mutations. In addition to the previously received upfront payment of $275 million, zipalertinib offers Cullinan a very attractive strategic and economic opportunity as a potential source of non-dilutive capital. Our collaboration with Taiho offers a very cost-effective commercialization approach to unlock the near-term billion-dollar-plus exon 20 total addressable market opportunity.
Recall that with our partner, Taiho, we have a co-development and co-commercialization agreement in which Cullinan retains a 50/50 development cost share and a 50/50 commercial profit share in the U.S. Importantly, in addition to the profit share, Cullinan is also eligible to receive a total of $130 million in additional near-term payments for U.S. regulatory approvals of zipalertinib in second-line and front-line exon 20 non-small cell lung cancer. Importantly, our partner, Taiho, is leading the commercialization activities, and Taiho already has three commercially approved products with an established and efficient infrastructure for the commercialization of zipalertinib. To conclude, on slide 20, the compelling and potentially practice-changing results from the REZILIENT3 study established zipalertinib combination therapy as a highly competitive regimen with the potential to become a new standard of care in the front-line setting for patients with exon 20 non-small cell lung cancer.
Importantly, these data also support combination therapy as the optimal approach in the front-line setting compared to single-agent tyrosine kinase inhibitor treatment by demonstrating the best observed median PFS, the best observed median duration of response, significant clinical activity in patients with brain metastases. At interim analysis, an early positive overall survival trend has already been observed, and manageable rates of EGFR-associated adverse events. From a regulatory perspective, in April 2026, we announced our NDA in relapsed exon 20 non-small cell lung cancer was accepted for review by the U.S. FDA, and our next step, together with Taiho, will be to discuss these first-line results with the FDA. Lastly, as we approach these important regulatory approvals, our collaboration with Taiho offers significant near-term non-dilutive capital, which also allows continued investment in our promising T-cell engager pipeline across immunology and oncology.
Thank you for your attention. I'll now turn the call over to the operator to take your questions.
Thank you. 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. Please stand by while we compile the Q&A roster. Our first question comes from the line of Brad Canino with Guggenheim Securities. Your line is now open.
Hey, good morning. Congratulations on the great data this weekend. Two questions from me. One is just, as you speak to physicians with these data, what are you hearing about potential market share in their practices for this regimen versus the monotherapy or amivantamab? Then second, anything that you saw in the study around rates of AEs around the hem tox and management of it that was different across regions? There was a pretty big difference in the hazard ratio in the forest plot for the different regions, and I'm wondering if that was a cause of that or if that's just something that happened in the numbers. Thank you.
Thanks, Brad. I'll take the first half of your first question, and then I'll pass it over to Jeff for the rest of the question one and two, since he is on the ground in Seoul. I think from an uptake perspective, something we've heard very consistently, for example, in the combination therapy setting with the PAPILLON study and amivantamab, is that the EGFR-related toxicity, especially related to skin toxicity, is quite difficult to manage. We have seen in the frontline setting, even post the PAPILLON results, that some physicians have still opted for platinum-based chemotherapy in the face of some of these adverse events, and we've been told that physicians have been waiting for our results. I think certainly from an EGFR-associated toxicity, I think we have a very strong profile, and we would expect large market share in that frontline setting.
The second thing I think that's important to note as well is that between both the PAPILLON study and now our study, it's clear that the combination therapy approach is going to be more optimal than a TKI monotherapy approach, just based on the progression-free survival alone across both of those studies, certainly compared to sunvozertinib data to date, and the fact that we're starting to see an OS benefit as well, which was shared in that same symposium with combination therapy approaches. I think our view is from a positioning perspective, especially in a disease like exon 20 non-small cell lung cancer, which has particularly poor prognosis, that most patients are going to opt, and physicians, for combination therapy. And then within that combination therapy, I think we do believe we have a potential best-in-class profile.
Let me turn it over to Jeff to address both what he's hearing on the ground from physicians as well as your question around regional AE rates. Jeff?
Yeah. I would just underscore that in almost all situations in oncology, patients do best when you lead with your best efficacy. I think far and away, achievement of a 14.5-month progression-free survival in the zip + chemo arm cannot be discounted in any way. It's remarkable. It's the first time that a randomized study has demonstrated a progression-free survival in excess of a year. Every way you cut it, across all relevant subpopulations, patients are benefiting. And I think that's the overwhelming impression that people walk away from these data. Now, in terms of the adverse event distribution, there are a number of factors that could explain some of the adverse events, some particularly hematologic adverse events, as well as management of adverse events. Taiho are continuing to look at the data to understand regional variation in AE management.
The issue that you raise, Brad, which could potentially explain some of the observations. Some of the hematologic toxicity could be different depending on the platinum agent used, with carboplatin being much more likely to induce higher grade cytopenias. A lot of that analysis is yet to come since what was presented here at World Conference on Lung Cancer was based on top line data from the interim analysis. Still a lot of data to mine to better understand the pattern of adverse events and their management.
Yeah, I would just add to what Jeff said before moving on, Brad, is look, any way you look at the PFS Kaplan-Meier curves, we see the separation very early, and it stays consistent. So even in the face of any additional AEs or any sorts of discontinuations, the benefit accrued to patients is very early and very persistent. As Jeff says, we haven't seen a median PFS of longer than 12 months in this setting previously. So obviously, we're very encouraged by the results. Thanks, Brad.
Thank you. Our next question comes from the line of Matthew Phipps with William Blair. Your line is now open.
Morning. Thanks. Thanks for taking my question. Congrats on the very strong data here. I was wondering if you saw any difference in activity between near loop versus far loop mutations. That was brought up by the discussant as an outstanding question. Then can you remind us maybe how many patients would present with brain mets at diagnosis in your expectations, given the very strong results there? Thank you.
Yeah, maybe the last part of your question I'll take, and then I'll pass it over to Jeff. Qualitatively, at least, Matt, I would say that patients do tend to accrue brain mets over time through lines of therapy. So you do tend to see it grow between the frontline setting as you go through and the relapse setting, just based on the biology of this disease. Let me pass it over to Jeff now.
Yeah. So, Matt, while that was brought up by the discussant, I would emphasize that right now, and with the discussant, no one is making treatment decisions based on near loop, far loop, helical, location of the mutation in exon 20. The data is insufficient from any study to guide practice in that respect. In the REZILIENT3 study, that data is being collected, but it is not yet been analyzed. It is something that we will ultimately be able to answer, but can't today.
Yeah. Then just to close out the brain met question, Matt, I think this is one where now consistently across both the frontline and the relapse setting, our studies have accrued more patients with brain mets, relatively speaking, in both of those settings. And we clearly saw very encouraging response rates in brain mets in the relapse setting, active brain mets I'm talking about now. And in the frontline setting, we are obviously pleased with the subgroup analysis showing the robust PFS in the brain met population, which we haven't seen with other molecules in this setting. Thanks for your question.
Thank you. Our next question comes from the line of Andrew Berens with Leerink. Your line is now open.
Hi. Thanks for taking the questions. Congrats on the data, and appreciate you guys doing the event. Yeah, another one for me on the brain mets. Just can we get some additional color on the activity you are seeing? And I did notice that there was an imbalance in the brain mets patients in the arms of the trial. Any reason for that, and do you think that there is any implications given that the upper boundary does approach one, although it is below one? Thanks.
Thanks, Andy. Jeff, do you want to take that question?
Yeah. So any imbalance in the arms is simply a sort of artifact of randomization. There is no other reason for it than that. But what we know from the REZILIENT2 study, this is a phase II study where Taiho is very carefully characterizing the intracranial response rate of zipalertinib, is that the drug has about a 30% intracranial response rate in patients with active brain metastases, including from patients with leptomeningeal disease, the hardest to treat subset of patients with CNS metastases. So the drug does have documented, demonstrable evidence of intracranial activity. And I think in this study, as the discussant pointed out, not only was it that patients could have previously treated brain metastases, patients in this study could have untreated brain metastases, some of which were up to 2 cm in size.
Not only is there a higher frequency of patients with brain metastases, these are not the same patients with brain metastases as in the WU-KONG or PAPILLON comparator trials. So I think that makes the outcomes here all the more remarkable in that patient population.
Okay. No, thanks, Jeff. Can I just clarify, the reason that previously treated versus non-previously treated is important, I think, is because some patients come into the trial after getting radiation and there could be a delay in the actual activity that you see in shrinking the tumors. Is that why it is such an important
Well, it's because those patients will have had actual treatment for brain metastases, yes. So their brain metastases are more likely to be under control than a patient with a larger brain metastasis or one who hasn't had It's less a delay in, I'm not sure if you're meaning a delay in starting systemic therapy, or you're saying that it could potentially delay progression in the CNS from having prior treatment?
Yeah, no. I've heard from some KOLs that sometimes patients come into the trial, they might have gotten radiation therapy a few weeks before the trial, a month before, and that some of the shrinkage then is seen from the prior therapy.
Yeah. No, but in this.
It's been a long while.
Yeah, but in this instance, we are not assessing the patients. The outcome is progression-free survival. So we are not assessing shrinkage of the patient's tumor because of prior radiation or any other prior therapy. This is where patients often fail in the CNS. But I think this is a really robust outcome in those patients.
Yep. Thanks.
Thanks, Andy. Just to clarify on the question of imbalance, the brain mets in both arms was around 31%, so they were very similar. Thanks, Andy.
Thank you. Our next question comes from the line of Alex Thompson with Stifel. Your line is now open.
Hey, great. Good morning, and thanks for taking our question, and my congrats on the data. Maybe two from us. I guess on the sort of progression of the data and maturation of the data over time, can you speak a little bit to how you would expect discontinuation rates to evolve over time from here? Secondly, can you talk a little about the nuts and bolts of sort of the profit sharing here, and how you plan to recognize revenue? Thank you.
Yeah, I'll take the second one first, Alex, and pass the first one back to Jeff. Taiho will book the sales, and we take in the 50% of the net profits. That's fairly straightforward from that perspective. Jeff, do you want to take the other question about what we expect to see with discontinuation rates over time, given most patients have now been treated?
Yeah. As Dr. Tan pointed out in the presentation, the higher grade adverse events were largely confined to the first four cycles of treatment when patients were receiving zipalertinib in combination with platinum chemotherapy. The higher grade adverse events, and particularly hematologic adverse events, really fell off dramatically in the subsequent four months and beyond. So I expect, as we saw with zipalertinib monotherapy in the relapse refractory setting, that the drug is well-tolerated for longer term administration when it's not being given in combination with platinum. In terms of what we continue to look for, we'll of course continue to look for the adverse event profile over time, but I would also reemphasize the really positive trend in overall survival that was already emerging at only 30% information for overall survival.
Again, not yet statistically significant, but an early separation in the curves that is very encouraging given the magnitude of the progression-free survival outcome.
Thank you.
Thank you. Our next question comes from the line of Samantha Semenkow with Citi. Your line is now open.
Hi, good morning. Thanks very much for taking the questions, and congratulations on this data as well from me. Just another one on the brain mets. I am wondering how we should think about adoption in terms of how physicians are assessing whether the patient has brain mets or not at the start of treatment. How much emphasis do physicians tend to put on these subgroup analyses for driving treatment decisions? Then just from the commercial opportunity perspective for zipalertinib, how should we think about time to profitability for the collaboration for Cullinan? Thanks very much.
Thanks, Sam, for your question. I will take the second one first, and Jeff can take your first question around brain mets and the importance of treatment decisions up front in the face of brain mets. From a commercial perspective, I think, look, we are very, very pleased with the results. I think painting the broader picture of how do we think ultimately patients with frontline exon 20 will be treated. I think two things I will point out. It has become very clear between both the PAPILLON and REZILIENT3 results that you get the best efficacy results when you use combination therapy over TKI monotherapy. If you compare the results relative to what we have seen, at least so far with sunvozertinib. As Jeff said earlier, in oncology, the approach is use your best treatments first. We see the vast majority of patients receiving combination therapy up front.
There may be a group that are contraindicated for chemotherapy or that are frail, that they may be candidates for monotherapy. We expect that to be the minority of patients. I'm also going to address your first question by what I'm about to say in some ways, from a commercial perspective, at least. I think the fact that we now see clear activity in patients with brain mets in both the frontline and the second-line setting, and that we have studies that have over-enrolled patients with brain mets, I think that's a very strong commercial position. When clinicians are going to choose between combination therapies, I think they're going to look at the fact that we've had the longest PFS reported to date. We've broken that ceiling of median PFS of 12 months, as Jeff said earlier.
The fact that we have very robust activity in patient brain mets. I think our view is that within the combination patient segments, which will be the majority of patients, that the first choice will be zipalertinib + chemotherapy, in other words, a zipalertinib combination therapy regimen. Let me hand it over to Jeff now to address the brain met question.
Yeah. I think, Sam, it's an important question since for EGFR patients in general, exon 20, perhaps more so, more than 50% of them will ultimately have CNS metastases. So it is literally an important and growing problem as survival continues to improve for this group. Being able to address CNS disease is an important consideration in improving outcomes for the group as a whole. So I think it's a very important consideration in selecting therapy. Again, patients with most cancers do best when you lead with your best therapy, and I think that's just one of the characteristics that really make the data that we presented yesterday morning quite compelling.
Thanks, Sam.
Thank you. Our next question comes from the line of Phil Nadeau with TD Cowen. Your line is now open.
Good morning, that is, and our congratulations on the impressive data. A few questions from us. First, on submitting the first-line data to the FDA, can you talk a little bit more about your strategy? Is there a chance that you could get a simultaneous approval for first line and second line? Second question on overall survival. Can you remind us of the statistical analysis plan? When will the next analyses looking at overall survival take place? Then last, just in terms of the PFS, it did look like the control arm PFS in this trial was a bit longer than in some of the precedent studies. Any thoughts as to why that could be? Thank you.
Great. Thanks, Phil. I'll take your first question around FDA interactions, then I'll pass over the other two questions on OS analysis timing and also your PFS question. I think, look, I don't want to get ahead of our partners, Taiho, who are the IND and NDA holders. I think our view is, we're very excited about these results, and we want to continue to have discussions with the FDA. I'm not going to speculate about what those sorts of approvals look like. But I do think these results are compelling, and I think the FDA will find these results compelling, and we'll update you in the future on that question. Jeff, you want to take the question around subsequent OS analysis timings and the other question on the outperformance on PFS on the control arm in this study?
Yeah. Let me take the last one first because I can't explain it. I don't think we have a good answer for why chemotherapy would've done better in the patients that were enrolled to REZILIENT3. But in general, it would tend to bias the outcome of the study against us. Not only did it not, but we were able to read out this really important progression-free survival outcome at the time of the interim analysis. Even with fewer events, the magnitude of benefits emerged, as Nadim mentioned, very early as progression-free survival curves separated. In terms of the overall survival analysis, it was timed again to be conducted when we reached the number of events that would have been required for the primary analysis, which is expected later this year.
Great. Thanks, Phil.
Perfect. Thank you for taking our questions.
The thing I think I will add to what Jeff Jones just said also is the fact that we have a six-month improvement in progression-free survival at interim analysis. We are seeing that early. We are obviously very excited about what the OS results could become. But obviously, we will follow those patients up for survival. Thanks, Phil.
Thank you.
Thank you. Our next question comes from the line of Kaveri Pohlman with Clear Street. Your line is now open.
Yes. Good morning. Congratulations on great results and thanks for taking my question. On the CNS benefit, is there any literature or clinical experience or emerging evidence suggesting that stronger CNS activity during initial treatment could translate into reduced risk of developing brain metastases as the disease progresses, supporting first-line use of zipalertinib? Just like an additional follow-up, as the data matures, which secondary endpoints or longer-term efficacy measures do you view as most important in influencing physician adoption and treatment preference? From the regulatory perspective also, do you expect you would be comfortable moving forward with filing based on current data set, or you want to wait for the mature data? Thank you.
Sure. Thanks, Kaveri. Let me take questions three and two in that order, and then I'll pass on the question to Jeff around brain mets and risk of subsequent relapse. From an FDA perspective, I will point out that our view, and if you follow the precedent, for example, in the PAPILLON study, is that we don't need OS for regulatory approval. The base of approval typically has been progression-free survival. Our view, Kaveri, is we don't need any additional data for a frontline approval. Of course, OS is clinically meaningful, and we will continue to follow those patients. Again, the fact that we see such a large PFS benefit early at the interim analysis, makes us feel very encouraged about what the OS benefit could be in the future. But our view is that PFS is what you need for regulatory approval.
Secondly, in terms of endpoints, it links back to the answer to this question. Obviously, OS is going to be important. We're going to continue to follow that. PFS2 is looked at these days as well. You want to see from randomization to, for example, the second relapse, are you continuing to see the benefits? That will also be an important endpoint to continue to follow. But again, I'll emphasize that it's PFS that's required for regulatory approval, and that's what we're going to be moving forward with. Jeff, the question on brain mets and subsequent risk of CNS relapse.
Yeah, Kaveri. I think the most robust data supporting that CNS control is in part what drives improvements in progression-free and overall survival in EGFR mutated non-small cell lung cancer comes from osimertinib in the FLAURA study, demonstrated greater intracranial control, and that in part drove improvements in progression-free survival. When you add chemo to osimertinib, you get further improvement in intracranial control. You saw further improvement in both progression-free and overall survival. I think that's the precedent.
Thank you.
Thanks, Kaveri.
Our next question comes from the line of Julian Harrison with BTIG. Your line is now open.
Hi. Let me add my congratulations on these impressive results, and thank you for taking the questions. On the topic of brain mets driving relapses in this setting, to what extent do you expect zipalertinib's CNS activity to perhaps translate to a longer-term overall survival benefit? Second, I am wondering what the physician feedback has been on the higher grade cytopenias. Can you give us a sense for how manageable these adverse events are and how they are typically navigated?
Great. Thanks very much. Jeff, do you want to take those two questions?
Yeah. Oncologists know how to manage cytopenias. We learn from the day we start prescribing chemotherapy. To the extent that these adverse events occurred during the chemotherapy combination part, the platinum combination part of the trial, physicians have a number of ways of managing this. One can be giving a different platinum, cisplatin, having a lower rate of hematologic adverse events, dose reductions, as well as growth factor support. All of those things are a part of the armamentarium all practicing oncologists are used to mobilizing to manage hematologic adverse events. So in general, when we say that the adverse event profile in that respect is manageable, I think that is quite true. Notwithstanding the Heme adverse events, patients benefited from therapy despite dose reductions. So I think that is generally the case in oncology, and I think that is demonstrated here.
You had another part of your question, I believe.
Yes, brain mets, and do we expect an impact on OS by managing to address patients with brain mets, Jeff?
Yeah. Absolutely. As we were talking about just a moment ago in response to Kaveri's question, that's kind of the experience that's been demonstrated in the FLAURA trials for osimertinib, and I think the same logic applies here. Very early, only 30% information for overall survival, already impacted by a significant number of crossover patients in the chemo arm at the time of disease progression, and yet a hazard ratio of 0.72 for overall survival. Not yet statistically significant, but with the magnitude of progression-free survival, I think it's quite encouraging for what we expect to see in the future.
Great. Thanks, Jeff.
Thank you. Our next question comes from the line of Sean Laaman with Morgan Stanley. Your line is now open.
Hi, good morning. This is Katherine on for Sean. Thanks so much for taking our question and adding my congrats on the data. Just on the adverse event profile, you discussed the drop-off in hem tox after cycle four. We were just wondering if you could separate how much of that reflects completion of platinum versus dose reductions, other treatment modifications. Can you clarify whether you observed efficacy being maintained in patients who required those modifications? Thank you.
Jeff?
Yeah. For the first part of your question, I think it's largely discontinuation of platinum. A lot of the hem tox there is driven by the platinum in combination. In terms of your second question, I'm not sure I completely heard all of the question.
Jeff, it was.
Can you repeat it?
Yeah. Patients who had dose modifications, were they able to achieve efficacy?
While we haven't broken out a specific efficacy analysis for patients who had dose reductions, the efficacy reported reflects the patients as they were treated. Notwithstanding any dose modifications, the patients in the zipalertinib plus chemotherapy arm still experienced a median progression-free survival of 14.5 months, a six-month improvement over chemotherapy alone, the first time that magnitude of benefit has been demonstrated in the frontline for zipalertinib. I don't think, in this instance, that the dose modifications adversely impacted outcomes for the patients treated with zipalertinib + chemotheraphy.
Right. Thanks.
Thank you. Our next question comes from the line of David Dai with UBS. Your line is now open.
Great. Thanks for taking my questions. Also, my congrats on the great data here. A couple of questions from me. During the discussion session, there is a step that highlighted 9% rates of adverse event-related deaths, compared with the lower rates recorded in competing phase III studies, including three patients with treatment-related septic shock. Can you provide some color on what gave rise to the deaths? Are these related to mainly patient characteristics, and how do you think the physician would view the risk-reward here? My second question on assuming potential approval for zipalertinib, which physician or patient segment do you believe represent the greatest early adoption opportunity?
I think, David, it was hard to hear fully, but I think your first part of the question was around the AE profile, including deaths. I will come back to you on the adoption question, but why don't we have the first question to Jeff. Jeff, are you there?
Oh, sorry. Managing from Seoul, so it is a little late in the evening here. In terms of the adverse events leading to death, those are detailed in the presentation, and many of them were respiratory tract-related events. These are things that commonly occur in patients with lung cancer. When you looked at the adverse events that were adjudicated to be treatment-related by the investigators, that was only three patients in the zipalertinib and chemotherapy arm, the three patients with sepsis septic shock, as you mentioned.
I would add, David, in terms of adoption, again, I would go back to the fact that we see the PFS benefits so early and sustained. Remember, and we tend to forget this, but a reminder that exon 20 patients have a much poorer prognosis than many of the other EGFR mutations. The point that Jeff made earlier, this is why it is important to address these patients with combination therapy upfront, because this is a poor prognosis disease, so you need to be aggressive in your treatment. Any time we had any sort of dose modification or dose adjustment, as physicians should do, especially in the first four cycles, that risk or tolerability risk was far outweighed by the PFS benefit we saw of over six months. Again, going beyond the previous 12-month median PFS ceiling we have seen in frontline studies.
I think, again, the overall benefit completely outweighs the competing risk from a tolerability profile perspective. I think our view still is that we are going to get very large adoption across patient segments, starting with the frontline setting with zipalertinib combination therapy.
Thank you. Our last question comes from the line of Boris Peaker with Jones Trading. Your line is now open.
Great. Thanks for squeezing me in, and let me add my congratulations on the data. Just want to probe a little more on the brain mets. Can you comment on what proportion of patients in each arm, the combo or the chemo arm, experience CNS as their primary progression versus systemic outside of CNS progression? And a financial question, there is about $100 million in milestone upon frontline approval. I just wanted to kind of get a sense of, is that one payment? Are there several conditions that add up to $100 million? And what could the timing of recognizing that milestone be?
Sure, Boris. I will take that second question. That is straightforward. It is a one-time payment following the U.S. approval in the frontline setting of $100 million. And then for the second line, it is $30 million, to give a total of $130 million. Jeff, the question of site of relapse.
Yeah. Right. Yeah, in the REZILIENT3 trial, we've not yet analyzed the patients by pattern of disease progression. We're not sure. I can't answer that question yet. In terms of patients as a whole, we know that the rate of CNS metastases is in excess of 50% ultimately in the lifespan of most patients with exon 20. The risk increases as survival improves if they are not treated with CNS active therapy.
Thank you. At this time, I'm showing no further questions. I would now like to turn the call back over to Nadim Ahmed, Cullinan's Chief Executive Officer, for closing remarks.
Thanks, Shannon. Thanks, everyone, for dialing in today. Our view is that these results are yet another example of Cullinan's strategy to match the right modality to the right target, as well as our executional strength. It was a few years back, we entered into a partnership with Taiho, with significant economic upside for Cullinan. Together with Taiho, we're very excited about the potential to address the significant unmet need in exon 20 lung cancer, and also being able to offer zipalertinib combination therapy as a best-in-class first-line treatment option for patients in need of efficacious, durable, and tolerable therapies. Thanks again, everyone, for dialing in, and we look forward to keeping you updated in the future.
This concludes today's conference. Thank you for your participation. You may now disconnect.