Hello, welcome to the Cullinan Pearl Clinical Update webinar hosted by Cullinan Oncology. Please be aware that this call is being recorded. I'm Juliet Labadorf, of [Investor Relations] here at Cullinan, I'm pleased to introduce today's call, which will focus on the ongoing phase I-IIa clinical trial evaluating CLN-081 in non-small cell lung cancer patients with EGFR or Epidermal Growth Factor Receptor exon 20 insertion mutations. The data sets presented are available as a poster presentation in the 2021 American Society of Clinical Oncology annual meeting. Before we proceed, we may be making certain forward-looking statements during this program. This may include statements regarding our preclinical and clinical development plans, clinical trial designs, the strategy of our product candidates, among other topics. We encourage you to review our forward-looking statements in this slide deck, which will be accessible on our website this week until the call. Providing prepared remarks from the call today will be Owen Hughes, Cullinan's Chief Executive Officer, Jon Wigginton, Senior Advisor to Cullinan and Chairman of the Scientific Advisory Board, and Leigh Zawel, Chief Scientific Officer of Small Molecules. Cullinan CFO, Jeffrey Trigilio, will be available during the Q&A. Following remarks, there'll be a live Zoom Q&A session moderated by our CEO. To indicate that you'd like to ask a question, you must click Raise Hand. When it's your turn, I will introduce you and then unmute your line. Once the call has concluded, an archive recording of the webinar, along with all materials presented, will be available on our website, cullinantherapeutics.com. I will now hand the call over to Cullinan's CEO, Owen Hughes. Thank you, Juliet. On behalf of my colleagues here at Cullinan and our partners overseas, namely Taiho Pharmaceutical in Japan and Zai Lab in China, I'm pleased to share with you our latest data set for Cullinan Pearl's CLN-081, a selective EGFR kinase inhibitor seeking to address exon 20 insertion mutations. Overall, we are very pleased with the emerging profile, both in terms of safety and efficacy. We recognize there's much work ahead. I'll be remiss, however, in not thanking our patients first and foremost, the Exon 20 Foundation, as well as our investigators and various collaborators across the globe. As you'll see, many patients are benefiting from CLN-081. It is our goal to make sure that continues to transpire. With that, allow me to turn it over to Leigh Zawel, our CSO of Small Molecules, to provide a bit of background on 081 before we jump into the clinical data. Leigh? Thanks, Owen, and good morning, everyone. Let me just take a couple of minutes to set the table for Jon's clinical update. EGFR is an extremely well-validated oncogenic driver in non-small cell lung cancer. The effectiveness of TKIs deployed in the exon 20 insertion mutant population has been limited by challenges in establishing the requisite mutant to wild-type selectivity index. It is critical in this space to have a drug that's very potent against exon 20 insertion variants whilst also sparing wild-type EGFR, and a failure to have that requisite selectivity index results in wild-type EGFR toxicities like skin rash and diarrhea. CLN-081, the structure of which is shown in the upper right corner of slide five, is based on a unique pyrrolopyrimidine scaffold. The preclinical data package that we were first exposed to indicated to us that CLN-081 had the potential to meet the challenges in the exon 20 insertion space. I want to walk you through two pieces of preclinical data, first focusing on the lower right-hand corner. What you're looking at in the dashed oval is a representation of CLN-081's IC50 in cell lines bearing wild-type EGFR divided by the IC50 in cell lines bearing six different exon 20 insertion mutants. What you can appreciate is that regardless of the colored shape, i.e., which mutant you focus on, the selectivity index is greater than one, indicating a sufficient delta between mutant and wild-type EGFR. We also see that CLN-081 compares quite favorably to a field of EGFR TKIs, regardless of which variant you focus on. In addition to having this differentiating wild-type to mutant selectivity index, CLN-081 is also differentiated in that it is sparing of wild-type HER2. It also doesn't have activity against exon 20 HER2. Finally, in preclinical models of exon 20 insertion non-small cell lung cancer, as exemplified in the lower left, where we're looking at an ASV variant model, we see potent single-agent activity and durable tumor regressions, and this was seen in multiple different models representing different variants of exon 20 insertion mutant lung cancer. With this profile, we were quite excited to test the compound in the clinic. With that, I'll turn things over to Jon to walk you through the clinical update. Great. Thank you, Leigh. Good morning to everyone. As Leigh highlighted, the features of the molecule suggested, based on its selectivity, that it could have a more favorable risk-benefit profile in the clinic. As I'll walk you through in the next few slides, we're encouraged by the data that's accumulating on that front to date. Before going further, let me just highlight that both the data shown and the statements made relate to an April 1, 2021 data cut. There are additional details regarding both the safety and efficacy in our ASCO poster and the presentation by Dr. Zofia Piotrowska, and I refer you to that as well. This first slide here highlights for you the architecture of the trial. In this phase I/IIa trial, we've tested doses ranging from a starting dose of 30 mg given orally twice daily on up to 150 mg orally twice daily, the highest dose tested to date. Dose escalation began with a single patient accelerated titration design that transitioned to a conventional 3+3 design on the first occurrence of a Grade 2 or greater drug-related adverse event that occurred at 100 mg. In addition, the protocol includes a number of nice features that allow us, guided by pre-specified safety and efficacy criteria, to further expand our cohorts to as high as 13 and 36 patients guided by dose criteria. You'll see here that we've expanded at dose levels including 30 mg, 65 mg, and 100 mg in particular. We've noted recently that we're expanding our 100 mg cohort all the way out to 36 total patients in our phase II cohort expansion. The trial is being executed across a geographic footprint in the U.S., E.U., and Asia-Pac. We also hope to begin enrollment in partnership with Zai Lab in China as well. This slide highlights some key demographic characteristics for the population to date. Key things that I'd like to call out indicated in the upper green box are that over 70% of patients had at least two prior lines of therapy. Also, 40% of the patients had prior EGFR TKI, including poziotinib or mobocertinib, and also 56% of the patients had prior checkpoint inhibitor therapy. All in all, a very heavily pretreated patient population, with 70% of them being third line or greater. As I mentioned, there are more details in the poster regarding the safety profile of CLN-081. I'm going to focus on some key high-level characteristics of the molecule here. Let me orient you. The columns indicate dose levels from left to right, starting at 30 mg, the beginning dose, over to 150 mg, the top dose tested to date. In total, we have 45 patients evaluable for safety and their enrollment and respective dose levels is annotated there for you. Let's turn now to some key EGFR-related toxicities, in particular rash and diarrhea. 76% of patients have some form of rash, although as you will see here, it's Grade 1 and Grade 2. No Grade 3 events have been seen, and it's mostly Grade 1 at lower doses and then as you get into higher doses, you can see Grade 1 or Grade 2, manageable with conventional supportive care. Diarrhea, as you know, has been and can be a particularly challenging toxicity for patients being treated with EGFR TKI. Importantly, we've only seen 22% of the patients that have seen diarrhea of any grade and only one case of Grade 3 diarrhea at our top dose tested, 150 mg. That contrasts quite significantly, as you're probably aware, with some other molecules in the class that have seen rates of all grade diarrhea and Grade 3 diarrhea that are several fold higher than those numbers. Something that we think may help us to differentiate in this space. Importantly, also, it's worth noting, as you can see annotated here, that we've begun to see some laboratory abnormalities, including transaminase elevations and anemia. We're continuing to follow those. In a few of the patients, they have been Grade 3 or greater. I also would highlight for you another issue with molecules in this space have been very high rates of treatment-related dose reduction or discontinuation. Down below in the bottom row, we've annotated those by dose level. As you can see so far, the relatively moderate rate of dose reduction or discontinuations. Importantly, I should mention also only one dose limiting toxicity, that case of Grade 3 diarrhea that I mentioned to you at our top dose tested of 150 mg. I'll turn things over to Leigh, who's going to walk you through encouraging PK data that has been accumulated to date. Thanks, Jon. The safety data suggests that we've been able to avoid significant wild-type EGFR toxicities so far, and this squares well with the PK profile. Just a couple of features that I want to point out on slide nine. You're looking at a plot of the unbound plasma exposures of CLN-081 from 30 minutes out to eight hours. Each curve summarizes that relationship for the 30 mg cohort in blue all the way through to the 150 mg cohort shown in yellow. First point to make is we're seeing a nice dose proportional increases in exposure as we increase the dose, suggesting that the PK is well-behaved so far. With regard to safety, I want to point your attention to the horizontal red line, which indicates the GI50 for wild-type EGFR. Importantly, for much of the dose cohort, we're below or just at the red line for wild-type EGFR inhibition. It's really only at the 150 mg cohort that we creep above that line for a few hours. Potentially explaining the safety profile that we're seeing so far. Finally, the last point is with regard to the potential for efficacy. The gray bar and the green bars, the horizontal lines rather at the bottom indicate the GI50 for inhibition of exon 20 mutant cell lines for ASV in gray and SVD in green. We were quite pleased to see that right out of the gate, even at the 30 mg of BID cohort, our plasma exposures were above the GI50 for the entire eight-hour period here, required for inhibition of those mutant EGFR cell lines, suggesting broad potential for efficacy. With that, I'll turn things back to Jon. Thank you. I'm going to walk you through now several slides summarizing our efficacy data. Again, I'd refer you to the excellent presentation by Dr. Zofia Piotrowska, a poster presentation at ASCO, Dr. Piotrowska from the Massachusetts General Hospital. Shown here, we have our swimmer's plot, and there's a lot going on in this slide, so let me orient you before we dive into the data. Annotated over on the left by patient is the dose level at which they were treated, low doses at the top, high doses at the bottom. Annotated over on the right are the specific exon 20 mutational subtypes. As you can see in the swimmer's plot, as indicated by the dark green bars, a number of patients achieved confirmed partial responses, the light green bars indicating patients with unconfirmed partial responses, several of which are quite durable and continuing to build in durability. In addition, I'd highlight the dark blue bars indicating patients with stable disease, again, several which are highly durable and building. Also highlight for you the gray bars indicating patients that are not yet evaluable, two of whom had just not yet reached their first on-treatment scan, and then the red bar indicating a single patient that had progressive disease as their best response. The red squares that you'll see in the bars indicate points of progression for the treated patients. What you should take away from this slide, in our view, is that we see activity, high rates of activity, durable activity that extends across the spectrum of doses tested to date, that extends across a variety of different exon 20 mutational subtypes, and is of building durability in this maturing data set. I'd also like to call out a couple anecdotes. Two patients at the very top are treated at 30. They had seven prior and five prior regimens, respectively. Both received prior poziotinib and responded but later progressed. Both received subsequent mobocertinib, best responses of stable disease, but later progressed and came on to this trial and responded to CLN-081. With this, the encouraging what appears to be differentiated safety profile and the spectrum of activity that we're seeing here, including the ability of patients to respond after progressing on prior exon 20 EGFR inhibitors, we're very encouraged by this clinical profile and the opportunity to develop this molecule further. This spider plot here gives you a sense of the kinetics of response. I think the key takeaway here is that in most patients, this drug appears to work pretty rapidly. The first scan that patients go through on treatment is at day 42 or six weeks, and subsequent scans are every nine weeks thereafter. What you can see, each line indicates an individual patient. If you rolled all these up, 76% of the patients have tumor regression at their first post-baseline scan. The drug looks to work quite rapidly in many of the patients, something that can be important in patients that have bulk disease in particular. In this waterfall plot here, we're looking at the percent change in the dimensions of the target lesions from baseline. What you can see as indicated by the green bars, indicating a high proportion of patients that achieved partial responses. You also see the blue bars indicating patients with stable disease, and a single patient indicated in the gray bar that had progressive disease. In totality, out of 42 response evaluable patients, 98% of them achieved either a partial response or stable disease as their best response. Several of whom, as you'll see as indicated by the E's, had prior EGFR TKI. I'd just like to call your attention to even the stable disease patients. You'll see that a number of those folks had substantive disease regression. This slide here rolls up some of the important numbers looking at the populations. Again, we're looking at dose levels in the columns ranging from the starting dose of 30 mg on the left over to 150 mg. The far right column, rolling up all patients, all 42 response evaluable patients across all dose levels. I'd like to focus in particular at the 100 mg dose level, where we've enrolled 13 patients as of the data cutoff. I mentioned to you, and I think you're aware that we've announced that we're expanding that cohort on up with another 23 patients to take us out to the full 36 patients per protocol, encouraged by both the safety and efficacy data that we're seeing to date. What we have seen is seven out of 13 patients had a partial response, yielding an objective response rate of 54%. Of those seven, six of the responses were confirmed for a 46% confirmed objective response rate. one of the patients will remain unconfirmed. Importantly, we've seen good disease control. Out of the 13 patients, nine achieved disease control, as indicated by either a partial response or stable disease lasting for six months or more, yielding a disease control rate of 69%. It's important to note that out of those 13 patients, there's another three that have ongoing stable disease and have not yet reached a six-month cutoff point. I'd also like to turn very briefly over to the total population. I mentioned this when we were looking at the swimmers. Again, in the 42 response evaluable patients across the totality of dose levels, 21 of those patients achieved an objective response. Of those, 13 were confirmed, yielding a confirmed objective response rate of 31%. Of the eight that are unconfirmed, five are pending a confirmatory scan, and three will remain unconfirmed. In addition, you'll see a substantive disease control rate, as indicated in the lower right-hand corner. Once again, the drug's showing activity across the dose range tested, and we think particularly important activity at 100 mg as well. You'll note that we've seen activity at 150 mg, and we will consider evaluating that dose level further in the future. I'd like to pause there and turn things back over to Leigh Zawel, who's going to talk a little bit further about some of the other characteristics of CLN-081 and how those may serve up other opportunities for the molecule. Great. Thanks, Jon. Honestly, we've been laser-focused on the post-chemo exon 20 setting. Yeah, just wanted to highlight some CLN-081 attributes that speak to the potential to get outside of the exon 20 space. In the first slide, I presented a wild type mutant selectivity index that was specific to half a dozen exon 20 variants, and we saw there that the index ranged from five-fold to over 100-fold, depending on the variant in question. What you're looking at is a similar sort of output here comparing CLN-081 in green to osimertinib in blue. Here we're focused on the traditional sensitizing mutations, L858R and exon 19, either with or without a T790M. Actually, the wild type mutant selectivity indexes in this population are even greater than in the exon 20 population and compare very favorably to osimertinib, particularly as you can appreciate in the exon 19-alone setting. I'd also point out that we've identified a couple of TAGRISSO resistance variants where TAGRISSO obviously doesn't work, but CLN-081 does. Again, we're very pleased with the safety profile. We're pleased with the activity in the exon 20 population, and this data suggests potential utility in the sensitizing population, either early in treatment or post-TAGRISSO. With that, I'll turn things over to Owen for concluding remarks. Thank you, Leigh. Based on the data to date, we are very encouraged with CLN-081's emerging profile. There are two topics that I'd like to discuss in more detail now. One, with respect to exon 20. We will continue to prosecute the existing phase IIa trial and will look to collect additional data that will enable us to initiate a regulatory discussion with the agency sometime in the coming months. In addition, upon confirmation of the recommended phase II dose, which could very well be the 100 mg BID dose that we're studying now in the expansion cohort, we'll look to initiate a first-line pivotal study in exon 20 patients. We believe CLN-081 is well-poised to address this indication, given the need to maintain dose intensity over a prolonged period of time. Secondly, given the safety profile to date and the preclinical data that Leigh just touched on, we believe there's an opportunity to extend CLN-081 beyond exon 20 non-small cell lung cancer, either as a monotherapy or in combination. For that matter, the combination hypothesis may very well bear fruit in exon 20 as well. We are working diligently to bring CLN-081 to as many patients as appropriate, and we'll look to update folks as our plans progress. I just want to touch on three things, safety, efficacy, and next steps overall. From a safety perspective, we hope we've been able to show you the data that is very encouraging to us. From a safety profile, we believe that the reduced frequency and severity of GI events has the potential to differentiate CLN-081 relative to other molecules. All AEs to date have been manageable and reversible. From a PK perspective, CLN-081 has been well-behaved, with dose proportional increases in Cmax and AUC and no evidence of accumulation. While preliminary, the initial efficacy data remains encouraging, showing anti-tumor activity in a heavily pretreated patient population, including efficacy across a range of dose levels, across a spectrum of EGFR exon 20 variants. After progression on prior TKIs, including small molecules addressed specifically for exon 20 variants. After progression on checkpoint inhibitors with high rates of response with solid disease control in a maturing data set. As it relates to next steps, we've experienced a very nice enrollment cadence in the expansion cohort and look forward to updating folks as we progress throughout the year, likely at a major medical meeting in the fall. In addition, assuming the data continues to unfold in a similar fashion, we anticipate initiating a development plan discussion with the agency in the second half of this year. With that, I want to thank you all for your time today, and we'll revert back to Juliet for questions. Thank you, Owen, Jon, and Leigh. It's now time for the live Q&A portion of the program. As a reminder, if you'd like to ask a question, please indicate so by clicking raise hand. Once you're selected for a question, I'll introduce you and then ask you to unmute your line. The first question is from Andrew Berens at Leerink. Can you hear me? We can indeed. Oh, thanks. I guess I just wanted to get some clarity on the potential development in exon 19, given what you just showed us. Would it be a combination strategy with a drug like a MET inhibitor or some other agent? Any idea what type of resistance mechanism you would expect to see in exon 20? Yeah. We're evaluating a number of possibilities. Certainly, the importance of MET in resistance to TAGRISSO is not lost on us. A MET inhibitor is certainly one combination that we're keen to explore, among others. In terms of the TAGRISSO resistance mutations, I'm going to present some data on that at the NSCLC Summit next month. We'll talk specifically about which of those resistance mutations CLN-081 works on. That's also part of our development plans going forward. Okay. In terms of exon 20, what do you think the resistance mechanism would be there? We anticipate MET being important. We anticipate C797S alteration being important. For CLN-081, it's early days, and we're keen to track the duration of the response and also what the mechanisms of resistance will be. Okay. Thank you. Thanks, Andrew. The next question is from Jeff Hung at Morgan Stanley. Go on, Jeff. Jeff, are you there? Sorry about that. We'll go to Josh Schimmer now. Actually, we'll go to Ed White at H.C. Wainwright. Hi, can you hear me now? We can. Great. Thanks. Congratulations on the data. Just wanted to ask a question about the 100% tumor reduction best responses and why those aren't considered to be CRs. Sure. Jon? Sure, yeah. The waterfall plot that you're looking at reflects the change in the dimensions of the target lesions, right? Those are patients that may not have had a complete resolution of non-target lesions, which means they wouldn't be a complete response. Okay. Do you have any data on where these non-target lesions were or how relevant they are? Well, across the population, they would represent the diversity of different places that people can have non-target lesions sitting in, liver, bone, brain, those sorts of things. Okay, thanks. Just a question for Owen and on the next steps that you're going to take. If you have a discussion with the FDA in the second half of this year, how are you thinking about running a pivotal trial? If you can give us any thoughts on timing or how you think about the size of the trial. Sure. I think there are some predicates that have been outlined by our competitors in terms of the size of the trials, the duration, and the endpoints. Our assumption at this point in time is that although some of those agents will be approved, is that we can follow in their footsteps. I think the most important thing for us is to continue to collect the data and achieve what we believe is necessary in order to actually initiate that discussion. At this point in time, durability is just the one thing that we will require. That's simply just additional follow-up on the patients that are on therapy. Our hope is that we can initiate those discussions fairly soon, being in the coming months, and be in a position to do what many other companies have done in a similar position, which is make or attempt to amend the existing protocol so that the 36 patients that we have in the 100 mg cohort by the time that we go to the agency, can act as the basis for the initiation of a pivotal study, i.e., those patients would serve as the first patients in, and let's assume that we would need 100-125 patients. Plus or minus, just based on predicate. We simply need to enroll another 70 patients. The other thing I would just say there, as it relates to the accrual, obviously exon 20 is a competitive space. We're certainly aware of that. I think one of the benefits of doing the deals that we've done thus far, A, keeping Taiho involved with the Japanese rights and bringing on Zai with the Chinese rights, is that you will now have three companies actively enrolling all at the same time across the globe, all working on the same protocol. Our hope is that if that can be affected seamlessly, is that we would be able to enroll the pivotal study, quote-unquote, "fairly quickly." I don't know what that timeframe is exactly, but I know that we're working quite well with our partners in Japan and China, and I know that Zai is very anxious to get started, as are we. Great. Well, thanks for the update, Owen, and again, congratulations. I appreciate it, but fortunately, it's the drug that's doing the work, not us. Thanks to the folks at Taiho for developing a fine drug. Thanks. Next question is from Jeffrey Hung at Morgan Stanley, who's having some IT difficulties, so I'll just read it for you. Can you provide additional details on the 100 mg patient that will remain unconfirmed? A follow-up question, it appears that there were some patients in the 30 mg cohort, who had previously seen progression, and now they're recorded as responses. Can you please give some details on what changed over time? Sure. Jon, you want to talk to that one? Yeah, sure. The patient at 100 that I believe you're referring to, you said that is an unconfirmed partial response that is not confirmed. Is that your question? Why they did not confirm? That's right. Yeah, that's right. That patient had a non-drug related adverse event, worsening of dyspnea, and that came off study, possibly complicated by aspiration. Came off study pre- confirmatory scan. The second question was color on the patients at 30 mg that were previously recorded as progressors who now have a partial response. Yeah, I guess maybe can you be more specific about which patient you're referring to? Maybe it's more, Jon, just in terms of, given what we've experienced from a COVID perspective, given what we've seen with the lack of CRAs being able to get into sites to do site monitoring, just some of the changes that happened in the database in the last six-12 months. Yeah. That happens sometimes where the coding will change and things get adjusted during monitoring. Again, we don't see anything that changes the story here. Yep. Thanks. Perfect. Now going to Josh Schimmer at Evercore. Thanks for taking the question. You've talked about registration paths and plans, both in exon 20 insertion settings and beyond. Are you prepared to be a little bit more granular in terms of what kind of studies, particularly combination studies, you think are going to be critical to move into earlier line settings? In exon 20, Josh, or outside exon 20? Both, really. Sure. I think, in the exon 20 space, we're intrigued with the data from those that are developing bispecific antibodies that are targeting EGFR and cMET As Leigh mentioned earlier in response to a question, it is very likely one of the resistance mechanisms that emerges over time. I think given the safety profile that we have to date and the efficacy that we've shown across the dose range, I think an interesting combination approach would be taking the antibody and 081 and combining those and really trying to extend durability in the second line. As it relates to areas outside of exon 20, once again, Leigh mentioned the C797S, cMET. Those are two areas that we're looking at diligently from a business development perspective. We do think that they are of interest with CLN-081 as the backbone. Once again, I think it's a function of what we're seeing from a safety perspective and what we're seeing from a preclinical perspective. Then even beyond that, certainly one of our competitors has shown us very intriguing data in the earlier lines of the traditional sensitizing mutations as a combination approach. Obviously, TAGRISSO is a phenomenal drug and has established a very large beachhead. I'm not quite sure we would go there just yet, but in certain TAGRISSO failures, as well as in second and third line, I think there are opportunities combined with both small molecules and bispecific antibodies in order to address essentially unmet medical needs for patients. Leigh, anything else you'd talk about? No, I think you covered everything well, Owen. Nothing to add. What about first-line exon 20? Is that on the table? Yeah. I should have mentioned that as well, but yeah, by all means. I think the premise holds that if you can extend durability to the second line through a combination approach, by all means, you want to do it in the front line as well. I do think that if you can combine two small molecules and have an efficacy and safety profile that's commensurate with a single molecule, that would be potentially a gold standard, but obviously that's conjecture at this point in time. Rest assured, we're thinking about it, and we recognize that we have a very good start here, but we're certainly not resting on our laurels. We do think that there are opportunities to combine various agents in order to improve the outcomes for patients. All right. Thanks very much. Thanks, Josh. Next, we'll hear from Chad Messer with Needham. Great. Thanks. Can you hear me okay? We can. Awesome. Great. Never 100% sure till I get the affirmative. Let me add my congratulations on the great data. Maybe we could just talk about dose a little bit more. I think I get and agree with your rationale for going forward at 100 mg with this next study, safety tolerability very paramount here. That said, I think it's really hard to rule out based on what you know about 150 mg, whether that might not be perfectly safe or safe enough to merit what might be more efficacy i guess would be the conclusion there. You had mentioned it might be of interest to explore that dose in the future, but I was wondering if you could maybe talk a little bit more about what an appropriate setting for that might be. Sure. Jon, do you want to touch on that? Yeah. I think, as you've seen from the data and we've talked about before, we'll make a decision in terms of a final dose designation based on a matrix of things, including the safety profile, of course, also the response rate, response duration, PK profile, et cetera. On that front, 100 mg looks very good, very encouraging based on what we've shown you. Your point is well taken about 150 mg. That is not off the table at this point to take a look at some point. It does have some step-up in the safety signal there that we're mindful of. Whereas, as we've shown you, there's no Grade 3 diarrhea at 100 mg, and the safety profile otherwise looks favorable. That's why we've started emphasizing 100 mg, and we'll reconsider 150 mg over time. Great. Thank you. Congrats again. Sure thing. Thank you all for joining. Any other questions? I think at this time there's no other questions. Just so everyone is aware, a recording of this webinar will be posted to our website and be available for 30 days after the webinar is complete. All the materials presented will also be available on the website, cullinantherapeutics.com. The last thing is just that in July, Leigh will be presenting at an ASCO Lung Conference. It'll be much of the data that you've seen here, as well as some additional new preclinical data that we have not shared to date on some atypical EGFR mutations, as well as some of the traditional sensitizing mutations. We'll make sure that we get that into your hands as well. With that, thank you very much for your time today and enjoy the weekend. Thanks. Bye-bye