Hi, welcome to the Atara Biotherapeutics ATA188 Interim Analysis Disclosure conference call. At this time, all participants are in a listen-only mode. A question-and-answer session will follow the formal presentation. If anyone should require operator assistance during the conference, please press star zero on your telephone keypad. Please be advised that today's call is being recorded. I'd now like to hand the call over to Eric Hyllengren, Vice President of Investor Relations and Finance at Atara Biotherapeutics. Please go ahead, sir. Thank you, Laura. Good afternoon, everyone, and welcome to Atara's ATA188 EMBOLD study interim analysis disclosure conference call. Earlier today, we issued a press release announcing completion of the ATA188 phase II EMBOLD interim analysis in patients with progressive multiple sclerosis. This press release is available in the Investors and Media section at atarabio.com. On today's call, members from the Atara executive team will provide an update on the phase II EMBOLD study for ATA188 and also answer any questions you may have. Joining me on today's call are Dr. Pascal Touchon, President and Chief Executive Officer, Dr. Jakob Dupont, Executive Vice President and Global Head of Research and Development, Dr. Ajay Joshi, Chief Medical Officer, and Utpal Koppikar, Chief Financial Officer. We will begin with prepared comments from Pascal, then open up the call for your questions. We would like to remind listeners that during the call, the company's management will be making forward-looking statements. Actual results could differ materially from those stated or implied by our forward-looking statements due to risks and uncertainties associated with the company's business. These forward-looking statements are qualified in their entirety by the cautionary statements contained in today's press release and the company's SEC filings. These statements are made as of today's date, and the company undertakes no obligation to update these statements. Now, I'd like to turn the call over to Pascal. Pascal? Thank you, Eric, and thank you all for joining us this afternoon. As we announced earlier via press release, we have completed at the end of June, as planned, the interim analysis of the ATA188 phase II EMBOLD study in patients with progressive forms of multiple sclerosis. The IA included review by the Independent Data and Safety Monitoring Committee, or IDSMC, of the efficacy, safety, and biomarker data available at the data cutoff time. The IDSMC review identified no safety concern with ATA188, and they recommended the study continue without sample size adjustment or any other modifications. The primary focus of the interim analysis was on Expanded Disability Status Scale, or EDSS, improvement at six months and overall safety, and also included evaluation of EDSS beyond 6 months for patients with longer treatment duration, other clinical data and magnetization transfer ratio or MTR biomarker data. Recall, the IA was designed to assess if a sample size adjustment was needed and was planned to occur before enrolling patients 80, so as to have the possibility to increase sample size if needed. The main assessment in the IA was EDSS improvement at six months, which in the 24 patient open label phase I study appeared to be predictive of confirmed EDSS improvement at 12 months. As a reminder, confirmed disability improvement by EDSS at 12 months is a primary endpoint for the EMBOLD study, which had been recommended by the FDA. At the time of the IA, there were EDSS data on 34 patients evaluable at six months and on 15 patients at 12 months. This was across both placebo and active groups with 1:1 randomization in the trial. Based on the analysis of this data set available at the time of the IA, there was insufficient data to draw a conclusion about the predictive value of EDSS improvement at 6 months for EDSS improvement at 12 months. The IDSMC believed that the six-month interim endpoint may be an inaccurate measure of the potential of this intervention in this condition. The IDSMC did not observe any safety concern that precludes the completion of the study. After reviewing the IDSMC assessment and recommendation, we are proceeding with their recommendation that the study continue to completion without a sample size adjustment or any other modification. Now, I would like to give more details on our enrollment in the study. After completing the IA at the end of June, the target enrollment of 80 patients was achieved. In addition, following the recent landmark publication in Science and Nature earlier this year, momentum around the program continued to build with physicians and patients. As a result, there has been a recent acceleration in enrollment for the EMBOLD study. Total final enrollment will be higher than 80 patients as patients we were already in screening through the end of June will be allowed to continue the enrollment process. The final readout of the study primary endpoint of confirmed disability improvement by EDSS at 12 months and communication of this data is planned to occur in October 2023. In order to maximize the potential benefits to patients and value creation for Atara, we'll continue to evaluate partnering options for ATA188 as the EMBOLD data mature. At this time, while remaining opportunistic, we do not intend to share the data from the IA with potential partners due to insufficient data to draw a conclusion about the predictive value of EDSS improvement at 6 months for EDSS improvement at 12 months. Meanwhile, we plan to continue preparing accordingly for phase III readiness, including interacting with the FDA as part of our Fast Track status and further developing our scaled manufacturing process in stirred-tank bioreactors. Finally, as we have said before, our goal for the phase II EMBOLD study remains to demonstrate a potentially transformative profile for ATA188 versus placebo in order to bring new therapeutic options to patients in need. While focusing significant efforts on ATA188 development, we are continuing to make progress with our other assets. We will give a full update at our next quarterly call, but we would like to highlight recent progress with tabelecleucel as we have received cGMP certificates from EMA inspectors, and we have sent answers to all EMA questions received. We therefore continue to anticipate a decision regarding EMA approval for tabelecleucel in Q4 2022. Meanwhile, we are continuing to have a constructive dialogue with the FDA to find a potential path to BLA submission that will not require conducting an additional phase III study. I would like to extend my gratitude to the Atara staff, our collaborators, and the patients involved in our studies. Together, we hope to bring transformative therapies to patients in need. I'll now turn the call back to the operator to begin the Q&A portion of the call. Operator? At this time, we'll be conducting a question and answer session. If you would like to ask a question, please press star one on your telephone keypad. A confirmation tone will indicate your line is in the question queue. You may press star two to remove your question from the queue. For participants using speaker equipment, it may be necessary for you to pick up your handset before pressing the star keys. One moment while we poll for questions. Our first question comes from the line of Salim Syed with Mizuho. You may proceed with your question. Great. Good afternoon, guys. Thanks for the color thus far. Couple of questions from me, if I can. Pascal, I'm curious if you could just give us a little bit more color. Why do you believe the six-month data is not predictive? Obviously, it was, you know, 85% predictive based on the phase I data. Now that you have the full enrollment here, are there any baseline characteristics potentially, or weighting PPMS versus SPMS patients that may have made this not predictive at six months? Also, second question, just on the no sample size adjustment, is another way of saying this is that you're seeing just greater variability in the data, greater standard deviation, and therefore there's no sample size adjustment? Is it no sample size adjustment because the data thus far is acting as you would expect according to your statistical protocol that you had set out at the beginning of the trial? Thank you. Thank you, Salim, for your question. AJ, do you want to take the first one? Sure. Thanks, Salim. You know, kind of as you noted, in the phase I open label study, we had over 85% predictive value for six months to twelve months. You know, for the part I wanna kind of emphasize for this part isn't so much it's not predictive, it's just there's not enough data to see whether it's predictive or not. That's the kind of difference between what you were stating and what we're trying to articulate. I think that's part of what that is based upon is we had 34 patients, as noted, at the six-month time point, and we had 15 at the twelve-month time point. When you know, kind of now in this study, it's not all patients getting therapy. There's half getting therapy and half getting placebo, essentially. When you look at all of that, it turned out that at least there's just not enough information at this point to confirm or deny that greater than 85%. Jakob, do you want to take the second question? Certainly. Thank you, Salim, for the question. This is regarding variability of EDSS. What I'll say is, as we previously stated, and to protect the integrity of this placebo-controlled study and to ensure that we complete it without introducing potential biases by public communications of the IA results, we're only disclosing today the IDSMC recommendations and our other decisions, regarding next steps for the program and our rationale for that decision. Pascal? Oh, thank you, Jakob. Salim. Thanks, guys. Thank you. Our next question comes from the line of Tessa Romero with JP Morgan. You may proceed with your question. Hey, guys. Thanks so much for taking our question. So I think just two for me. The first is. Can you talk directionally a little bit more about the treatment effect that was observed at the 12-month versus the six-month time point and if it's improving or not over time? I have one more if I could. Yeah. Thank you for the first question, Jakob. Yeah, absolutely. Hi, Tessa. Thank you for the question. Once again, we're trying to protect here the integrity of the placebo-controlled study and to assure that we don't introduce potential bias by disclosing publicly the interim results. Thus, we're only disclosing the IDSMC recommendations and the decisions regarding next steps, which are namely to keep the sample size as is and to not make any further modifications to the study design. Tessa, you had a second question. Yeah. I mean, I guess, yeah, the second question was just if you could explain a little bit further what the conditional power modeling approach that you used at the time of the IA and how that informed either the sample size staying the same or being modified. AJ, do you wanna take that one? Thanks, Tessa. The way we've never really articulated the actual conditional power that we're going for, but the way it generally works is that when you take a look. Remember the couple things. The 12-month time point is of course a critical time point. Since we weren't gonna have that many patients at the 12-month time point for the interim analysis, we were looking to see was there a way that we could do a 6-month interim and kinda and predict what would happen at the 12-month time point. That's where the conditional power comes in, because then you can take a look at whatever statistical assumptions you use at that point for drug versus placebo, would then inform the probability for achieving a certain conditional power and whatever that conditional power target that you have. That's how the general model will work. You set a conditional power target, and then you have the assumptions for drug versus placebo at that earlier time point, which will then give you kind of the probability of hitting that conditional power. You know, generally the more data that you have, the better chance you have of trying to, you know, get a good idea of what that conditional power looks like. Does that help? Yeah, that helps. Then just to confirm, guys, like, so it sounds like you won't disclose to us how the treatment effect at six or 12 months compares to what observed in the IA, right? It sounds like that's something we need to wait for in the full data set. Thanks so much. Yes, you're right, Tessa. We cannot disclose anything related to the treatment effect, active or placebo. What we can say is really, as we said since the beginning, we never say that we'll disclose any data. We say we'll disclose a decision, regarding sample size adjustment as well as the rationale for that decisions. That's what we are doing today. Thanks for taking our questions. Our next question comes from the line of John Newman with Canaccord. You may proceed with your question. Hi, guys. Thank you for taking my question. Question I have is, in the phase two study, was the dose used consistent across all patients, or did it change at some point for cohort three versus cohort four? Thanks. Thank you, John, for your question. AJ, do you want to address that one? Sure. Thanks, John. Yeah. Good memory. We, you know, when we initially started the study, let's just go back to the phase I for just a quick moment. Remember, we had, you know, both cohort three and cohort four dosing looked really good. Cohort four dosing ended up looking a little bit better once we had that data set. However, when we started the EMBOLD study, the cohort three dose had actually triggered these predefined rules that we had to move on to establish the phase II dose. We started the study with the cohort three dose. When we got the data from the cohort four, where it showed a little bit better activity on the EDSS portion of it, but just marginally better, but, you know, somewhat better, then we decided to move the cohort four dose. The overall perspective was, this was kind of a threshold dose effect. Once you got above that dosing for cohort three and four, they both worked, and cohort four was maybe just a touch better. That was the rationale behind it. In terms of how that translated into the study itself, the first 18 patients in the study were cohort three dosing, and then everyone else after that is cohort four dosing. Remember, that's randomized in placebo. Randomized between placebo and active. Does it answer your question, John? Yes. Thank you. Our next question comes from the line of Phil Nadeau with Cowen and Company. You may proceed with your question. Good afternoon. Thanks for taking our questions. It sounds like what you're saying is that the 6-month data, your initial analysis suggests it's not predictive of 12 months, what's likely to happen at 12 months, and therefore no sample size adjustment can be recommended from the 6-month data. Is that an accurate interpretation? Yes, it is. You guys obviously knew how many patients were gonna be up to 12 months when you designed the interim analysis. It's probably not a surprise how many 6-month patients you have versus 12-month patients. I guess if you took the six-month data at face value and assumed the predictive nature of it from the phase 1 trial, what would have been the recommendation? What would have happened to the trial? Would it have been increasing sample size, stay the same futility? Kind of if you forget about what you're seeing at 12 months, but just take what you're seeing at six months, assume it's predictive, what would have happened to the trial in that circumstance? Yeah. Phil, I think just to be clear, we cannot comment about data at six months or data at 12 months. What we can comment upon is the rationale for the decision, which was really on the aspect of the ability for the six months data to be able to be predictive to the 12 months and whether or not there was enough data to make that assessment. Now, maybe, AJ, you can clarify a little bit why we thought that this IA with that number of patients at six months and that number of patients at 12 months makes sense and was to be conducted before the enrollment of patient 80. Sure. Yeah. You know, although the sample size at the time of the IA, you know, was small, as you, as we mentioned, 34 at six months and 15 patients at twelve months, with that 1-to-1 randomization, that number of patients was sufficient. If you look at the statistical modeling that we put in and the assumptions that went into that, including some of the predictive capacity we talked about, that was sufficient to have an adequate interim analysis. Again, if all of those assumptions held. That's kind of the first piece of it. Why would we still do the timing that we did and maybe not wait a little bit longer for more? Well, a couple things. One is, of course, this was still, it was still important to do because this is the first time you get kind of placebo-controlled safety information, and as you noted that there were no safety concerns that came out. The second piece is that remember that we needed to, you know, whatever sample size adjustment we needed to make, we needed to make before we got to that 80th patient. As you can tell, we, you know, rapidly enrolled that 80th patient after the interim analysis. If we had waited longer, we actually wouldn't have that proper opportunity to adjust the sample size. It was kind of both of those different factors went into the specific timing. Again, from the numbers that we had and the numbers that we knew, if all of the assumptions had held, that would have been a, you know, that we would have been at least able to perform an adequate interim at that time point. Yeah. In fact, just to confirm on that, given the current limited size and duration of observation in the dataset at the IA time, these assumptions cannot be fully assessed. It's not that they are wrong or right, it just can't be fully assessed. That's why the data needs to mature to the primary endpoint analysis at 12 months. Got it. In the past, you said one scenario today was that the trial stopped for futility. Can you remind us what the futility criteria were? Given that the trial is completing, can we assume that the data therefore did not trip that futility criteria? Yeah. Thank you. Jakob, do you want to take that one? Phil, just in terms of the futility analysis, as part of the interim analysis, we did a full analysis covering all possibilities, including futility. We have not disclosed boundaries for futility in this particular study. Again, as stated, to protect the integrity of the study, particularly the placebo-controlled nature of this and to not introduce bias, because we really wanna get to that final result on the study when we have the primary endpoint data at the 12 months confirmed EDSS, we're not able to disclose these aspects today. Got it. Just last question from us. To be clear, management all knows the data. This isn't one of those circumstances where management doesn't know the data. IDMC does its analysis and comes up with a conclusion without management having understanding of the data. Like, you guys have actually seen and analyzed the results yourself. No, not the management. Only a very few required individuals within Atara research and development team, and of course, the IDSMC have had access to the unblinded data. Management have reviewed the IDSMC recommendation and their rationale to continue study and maintain sample size. That's the management didn't have access to the unblinded data. Got it. That's very helpful. Thanks for taking all of our questions. Our next question comes from the line of Ben Burnett with Stifel. You may proceed with your question. Hey, great. Thank you so much. Apologies if I missed this, but I just wanted to kind of go back to the statistical analysis. Any color you can provide on sort of the alpha spend on this interim analysis, specifically, on the efficacy portion. I guess, does that in any way impact, in your belief, the IDSMC's interpretation of the six-month interim endpoint? AJ, do you wanna take that one? Sure. Thanks for the question. In terms of I'll start with the second part. Does it impact their assessment? That's at least from the conversation with IDSMC this has almost no role. I don't think they brought it up once in the conversation. From the standpoint of the, The alpha spend itself, it's actually extremely small because it's, you know, in the past we've articulated that we did not intend to stop the study at the time of this interim analysis. Whenever you do an interim where you're not intending to stop, your alpha spend is extremely small. Here it's an extremely small spend. What's nice about that is that, you know, it doesn't really have a major, or hasn't, doesn't have much at all of an impact in terms of the successful outcome from the 12-month perspective. Very small alpha spend, and as far as I can tell, there's no impact from what the IDMC said. Excellent. Okay, that's super helpful. I guess also just a cash question. It's understanding that I think at your last earnings cash guidance into Q4 of next year, I guess kind of put that into context for when this study is expected to read out and kind of just frame sort of the cash runway. Thank you. Sure. The cash runway is into Q4 of 2023, and we think cash runway is beyond when this study is gonna read out. Does that answer your question, Ben? Okay, thank you. That's great. Thank you very much. Our next question comes from the line of Michael DiFiore with Evercore ISI. You may proceed with your question. Hey, guys. Thanks so much for taking my question, for hosting the call. A few from me. For one, I just wanna drill down a little bit more on the futility criteria. Has the futility criteria changed or has it been withdrawn entirely in light of the fact that the six months EDSS improvement data is no longer predictive? Are you willing to confirm that the study isn't futile in that sense? Second question is, you know, I think you mentioned, and please correct me if I'm wrong, that in your prepared remarks you said you did not intend to share data from the interim analysis with potential partners. Should we read anything into that in terms of the success or potential success, you know, of the program? One housekeeping question is that, and again, correct me if I'm wrong, you said that I think you may have said that more than 80 patients will be enrolled because of some accelerated recruiting dynamics at the beginning of the month. If that's so, what would be the final number of enrolled patients in the trial? Thank you. Thank you, Mike. Do you want to address the first one, Jakob, and I will take the last two? Yeah. As noted, as part of the interim analysis, we did do this full analysis covering all of the possibilities. Again, the IDMC also had futility as one of the considerations there. We can't disclose the nature of that data, obviously, for the reasons stated in terms of maintaining the integrity of the study and really getting to the end of the study and the primary endpoint. Because recall that the endpoint that was assessed at the time of the interim was not the primary endpoint of the study. That's what's gonna come in October of 2023, as Pascal mentioned. I also wanna confirm that we have not made any changes to the futility criteria whatsoever during the conduct of the trial. That's been in place, and that's been reviewed by the IDMC, and they obviously came to their determination to continue the study, not change the sample size, and to continue the conduct of the trial's design. On your question about partnering, I'm not sure there is any particular readout of, or comment on partnering. As I said, Atara will continue to evaluate and discuss partnering opportunities for ATU and ATA as the data mature in order to maximize potential benefit to patients and value creation for Atara. We've been very clear from a strategic point of view, corporate strategy point of view, it makes sense to partner that program to move to phase III and other type of studies there. Now, at this time, again, we remain opportunistic, but we do not intend to share broadly the data from the IA with potential partners, mainly due to the insufficient data to draw a conclusion about the predictive value of the EDSS improvement at 6 months for EDSS improvement at 12 months. Again, we remain open. We're going to continue discussion. We are not planning to share broadly the data from the IA with potential partners, but we'll remain opportunistic on that front. On your questions related to the enrollment, as we said, we're going to continue to enroll those that have been into the screening process till the end of June, until we took the decision related to the sample size post the IA. We cannot give a precise number, but certainly you can expect at our next quarterly call that we can clarify where we are in terms of enrollment in a study to make sure that you have a clear data there. Great. Thanks so much. Our next question comes from the line of Salveen Richter with Goldman Sachs. You may proceed with your question. Hi. Thanks for taking our question. This is Tommy on for Salveen. We just had a question on the path forward from here. If you could give us any color on the timing of regulatory discussions and how the updates here affect your thinking on the design of the phase III studies, maybe in terms of target enrollment or the 12-month timeframe for EDSS. Like how are you now thinking about running the phase III studies? Thank you. Thank you. As we said, we're going to continue to work on the phase III readiness. We have a Fast Track status that allows us to interact with the FDA, and we are planning some interaction with the FDA to discuss about some aspects related to that phase III readiness, especially aspects related to the product to be used, the manufacturing process to be used. As you know from previous communication, we are working on a new manufacturing process that allows scale-up type of manufacturing with potentially up to 20,000 doses per leukopak per donation. We need to work out with the FDA on different aspects to be ready for the phase III. No change in terms of our plan from that point of view, in terms of getting ready for phase III readiness. In terms of the particular protocol of the phase III, as we disclosed previously, we had had some discussion with the FDA related to the Fast Track designation, and we communicated the type of phase III planning that we envisage at this stage. We are not planning any particular change right now, but we'll of course continue to discuss that with the FDA and other regulatory authorities as applicable, to be able to finalize a plan for phase III after the phase 2. Thank you. Our next question comes from the line of Yigal Nochomovitz. Sir, you may proceed with your question. Hi, Pascal and team. Thanks for taking the question. Just on a factual question, can you just remind us factually what was the exact criterion or criteria that triggered the IA? Given that you had 34 patients at 6 months and 15 at 12 months, was this sample size for the IA smaller than what you had modeled or expected? Thank you. Thank you. Yigal. AJ, do you want to take that one? Yeah. I think the best way to look at this, Yigal, thanks for the question, is it's a timed IA, so, you know, didn't need a particular trigger. You know, for the minimum number of patients we needed to have that kind of, you know, adequate power, the adequate interim analysis of the data, again, if all the assumptions held, was 32. So we had 34, so we had passed that number. Then, of course, you know what happened with that. You know, we've talked about, you know, given the limited size and duration of this dataset, we just couldn't fully assess those assumptions, and therefore we need to let the data mature to the primary endpoint. Okay. Just going back again to this no sample size determination by the IDSMC, just so I understand. Is the IDSMC saying no sample size adjustment solely because they don't have enough data at six and 12 months to render an opinion one way or the other on whether to change the sample size, or was part of their decision not to change the sample size also a function of having seen the initial efficacy delta between ATA188 and placebo, whatever that may or may not have been? Jacob? Yeah, sure. Yigal, thanks for that. The IDSMC had standard options at the time of the interim analysis, and one was to maintain the sample size. The next was to increase sample size, discontinue the study or modify the study design and so forth. Those are standard options, and that was the case also for our IDSMC. They did recommend that the study should continue without modification or sample size adjustment, to really get a readout, as AJ says, of the primary endpoint at the timing of the final primary endpoint, which is gonna be here in October of 2023, which is really that main endpoint that's also been recommended by the FDA as well for the phase III registrational program. Also to add to that answering your other part of your question, that the IDSMC had reviewed all unblinded data on efficacy, safety, and MTR biomarker data to arrive at their recommendation that the study continue with sample size adjustment. They had access to all the unblinded data, efficacy, safety, and MTR in particular. Okay. Thank you. Our next question comes from the line of Tony Butler with Roth Capital. You may proceed with your question. Thanks very much. Pascal, you just mentioned MTR, and I wanted to ask if, in fact, what was the pre-specified timing for an MRI to determine the quantitation? Are there multiple time points for subsequent MTR? That is, for example, is one at three months, six months, or and/or at twelve months? The second question, pretty simply is this. You've made some comments about the safety aspect of the trial to date. Can you make any statements about dropouts? Thank you. MTR. AJ, do you want to clarify the MRI timings? Sure. From the MTR perspective, we have, you know, baseline, 6 months and 12 months. That's what we had reported out in some of our prior datasets, is the 6-month and 12-month changes, and we have that certainly in this study as well. On safety and dropouts? I don't think we're commenting on the particular dropout piece. Obviously, you saw, you heard from the IDSMC perspective that there's no safety concerns, but we haven't commented on dropouts at this point. Thanks. Just one additional on the MTR. Did all the roughly 15 patients at 12 months have MTR done for that last or the last MTR done at 12 months? Yeah. I mean, again, as far as I can recall, you know, from a protocol perspective, these people did have all the proper protocol assessments at the right time. You know, every once in a while you have a COVID issue that changes timings here and there. As far as I recall, yeah, they were per protocol. I appreciate that. Thanks very much. Ladies and gentlemen, we have reached the end of today's question and answer session. This concludes today's conference. You may disconnect your lines at this time. Thank you for your participation. Enjoy the rest of your day.