Good day, ladies and gentlemen, and welcome to the Atara Biotherapeutics second quarter 2019 financial results conference call. At this time, all participants are listen-only mode. Later we will conduct a question and answer session. Instruction will follow at that time. [Operator Instructions] I will now like to turn the conference over to your host, Dr. John Craighead, Vice President of Investor Relations and Corporate Communications. You may begin.

Thank you, operator. Good morning, everyone, and welcome to the Atara second quarter 2019 financial results and corporate update conference call. Earlier this morning we issued a press release providing an overview of the company's second quarter 2019 financial results and recent operational progress. This press release, as well as an updated investor presentation, are available in the investor and media section of atarabio.com. I'm joined on the call today by Dr. Pascal Touchon, President and Chief Executive Officer; Utpal Koppikar, Chief Financial Officer; Dr. Chris Haqq, Chief Scientific Officer; and Dr. AJ Joshi, Chief Medical Officer. We'll begin with prepared comments from Pascal and then open the call for your questions. I'd like to remind listeners that 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 the 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 would turn the call over to Atara's President and Chief Executive Officer Pascal Touchon. Pascal?

Thank you, John, and thank you everyone for joining us this morning. Today's my first conference call as the Chief Executive Officer of Atara Biotherapeutics. It has only been a few weeks since I joined Atara in late June, but I've already been impressed by the expertise and commitment of our teams in developing T-cell immunotherapies to transform the lives of patients with serious diseases. Indeed, I am confident that we are now in a strong position to execute on our commitment and create value for patients, physicians, and shareholders because of tab-cel, multiple sclerosis, and next generation CAR-T programs. What I would like to do today is first provide a brief overview on my background and tell you why I'm so excited about Atara's potential. I joined from Novartis, where I serve as global head cell and gene and member of the oncology executive committee. In that role, my responsibilities included the regulatory approval, pricing and reimbursement, and global launch of Kymriah, the first ever CAR-T approved globally in two indications. I was also writing about this global CAR-T strategy, clinical development, manufacturing and technical operations, and the financial performance of the oncology cell and gene activities. Prior to that whole was global head strategy, business development, and licensing oncology at Novartis, and beforehand executive vice president of Servier, where I initiated the partnership with Cellectis and Pfizer on allogeneic CAR-T. I believe my experience makes me uniquely suited to carry out Atara's mission to transform the lives of patients with cellular diseases. I think work over the last five years in the field of auto boost CAR-T as well as third generation genetic allogeneic CAR-T. I believe cell therapy is the next therapeutic frontier in oncology and immunology, following its transformative impact on patients with P-cell malignancies. What…

Thank you. [Operator Instructions] And our first question comes from Anupam Rama from JP Morgan. Your line is now open.

Good morning. This is Tessa filling in for Anupam this morning. Thank you for taking our question, and it's very nice to meet you, Pascal. So my question is on MS. With first efficacy data for ATA-188 coming at ECTRIMS in the coming weeks, can you walk us through scope of data here and what the points of differentiation are that we should be looking for in the phase 1 readout? Specifically, what are your expectations in the context of the disease for the T secondary endpoints that you will be outlining for us? Thanks so much, guys.

Thank you very much for your question. I will ask AJ Joshi to start answering that question.

Sure. Thank you, Pascal. So just to put things in perspective. This is, of course, a study that is focusing on the progressive MS population, and really the focus in progressive MS is to delay deterioration. That's really the end goal. Now in order to assess that particular parameter, most parameters actually you need about one to two years of assessment to really prove anything. So our challenge in setting this study up was how do we get an earlier read than a couple of years? Because, as you know, these data are going to be around six months data on a couple of cohorts of patients. We work with our leaders to essentially define multiple parameters that are well defined, clinically meaningful, and essentially assess those parameters across multiple time points. What you're looking for across those parameters is basically similar change across multiple parameters at a specific time point, and what, I should step back for just a second. So those parameters assess both clinical function, physical function, cognitive function. It's a variety of MS issues. If you see a similar movement in multiple parameters at a specific time point, that gives you confidence that there's a real signal there at an earlier time point than that one to two years that I was talking about. So success for us here would look like movement across multiple parameters at the time points that we describe when the data are presented.

And just to add on that, ECTRIMS is an important step in building up evidence in terms of safety and efficacy for ATA-188, so then we can move to at the latest date when we have completed the phase 1a study into phase 1b. We expect to present initial efficacy results from the lower cohort as well as additional safety result from the higher cohort. But the fourth and final planned cohort is almost fully enrolled, but we do not expect to have efficacy data available at the time on the fourth cohort at ECTRIMS in September. And this is just, as I say, a step, and we plan of course to progressively present additional data at future congress as they continue to mature. Other question?

No, that's great. Thank you very much. I appreciate the color.

Thank you. And our next question comes from John Newman from Canaccord Genuity. Your line is now open.

Hi. This is Chris on for John. Thanks for taking my question. We were just wondering do you have any incremental color for us for how fast new clinical sites can be opened? And if you have any details, additional details, on what you're doing besides that to help enrollment?

Thank you for your question. I think AJ, you might want to start answering that question.

Sure. So in terms of speed of getting sites up and running, for PTLD you're really looking at larger scale academic centers. So those, you typically have a little bit of a longer lead time to open those sites. The reality for us is though that we've been building toward this for a while, so much of that lead time is already built in. So we're starting to really see the benefits of getting those sites coming in because, as you know, academic centers often take anywhere between six and nine months to really get up and running. But again we've done a lot of that lead-in already so we do expect to open up several additional sites in the U.S. later this year, and in addition we are opening up additional geographies because we just got, we recently got a "no objection" letter from Health Canada. So we'll be opening up our first Canadian site later this year, and then looking forward to opening up European sites next year. In terms of any additional work that we're doing, much of the focus for us now is optimizing recruitment at the existing centers, and this has been really, we've been leveraging a variety of resources, including medical science liaisons and other activities that I implemented in the past, in previous lives and ultra-rare indications. So multiple activities we've used in ultra-rare indications so that way, so I think again, Let me step back. Multiple activities from ultra-rare indications that have been used in the past, including MSLs and including a variety of outreaches to both the patient advocacy community as well as the established physician societies.

And I'd just say that one of my first priority when I joined was to look in details at the way clinical operation were run to make sure that we optimize or chance to accelerate recruitment in that very important study. I have been satisfied by what I've seen, and that's why I am confident regarding the guidance we've given, which is of course linked with the enrollment in these studies. What we should say as well that some of the challenges that we faced here are linked with the disease itself. It's a notorious disease happily progressing where sometimes the patient cannot wait for the administrative burden of being recorded in a study, and that's we're using a EAP program or SPUs as well to be able to treat these patients. Additionally, we have to recognize that there have been a number of competitive trials, either directed at PTLD or basket clinical trials in this space that have been started over the last two years, and I think there are about 12 clinical trials right now in that space. So that's also something to take into account into enrollment, but it also tells a lot about the attractiveness of that particular medical need in terms of new innovative therapies.

Got it. Thank you very much. That was very helpful.

Other question?

Thank you. And our next question comes from Salim Syed from Mizohu. Your line is now open.

Thanks so much, and welcome, Pascal. A few from me, if that's OK. One on the MS program. Curious how you guys are thinking about the risk that these T-cells, they're gonna be in MS patients. They will have immune systems. So when you're looking at efficacy, like how do you guys think about the risk here that you're giving enough T-cells so that they last, right? I know you can compare to PTLD, but that was immune system knockout patient. How are you guys thinking about here that these T-cells will be able to last enough to provide the efficacy that you're looking for? Number two, just on allo program versus auto program, 188 versus 190, if there's any theoretical risk here of one potentially getting past the blood-brain barrier more than the other or if there's a mechanism in place that that the human body essentially would reject 188 because it's not perfectly HLA matched? And then the last question is just then could you provide more interim details on the tab-cel program. How many patients do you need for the interim and what would the OR hurdle be? Thank you.

Thank you very much. AJ, do you want to start?

Sure. So in terms of the persistence question, we have actually experience with EBV-derived T-cells, EBV-targeted T-cells in an immunocompetent population with the nasopharyngeal carcinoma program. So as you know in the phase 1 studies at MSK we had about a 20% response rate in essentially an immunocompetent population. So we should be very well able to apply that concept to what's happening in MS because you've got a similar scenario, immunocompetent population using allogeneic T-cells. So from that perspective I think there's relative confidence that we should have enough persistence to get an efficacy signal and maintain efficacy appropriately.

Okay, do you want to comment on the blood-brain barrier passage?

Sure. I think here the tab-cel experience results were a good guide in that our experience that the EBV-specific T-cells have the ability to traffic. They cross the brood-brain barrier, and in our phase 2 experience we have observed several patients who've had CNS-located PTLD to have objective responses. We expect that to continue as well in the setting of MS, and in previous American Society of Hematology presentations we've presented on the activity of the antiviral T-cell platform with essentially the same response rate observed in patients with CNS disease compared to those whose disease presented systemically.

And on the auto and allo program in MS, AJ?

Can I pause it? Would you mind repeating the specific question on auto versus allo?

Oh, no, I think you answered that question. Then the last question was just around the interim details, how many patients and what the OR hurdle would be.

Sure. So, as you might imagine, any time you're doing an interim analysis and you're working on that with authorities, you're gonna have to show substantial benefit and with some adequate level of certainty, which is essentially ORR. Our current RR, when you look at the total patient population, we have 33 patients, the null hypothesis is 20%, so OR would have to be 37% to meet the statistical hurdle. And you might imagine any interim analysis that we do would have to have a higher ORR statistical hurdle to meet the requirements.

Okay. Great. Thanks. Thanks so much guys. Thank you.

Thank you. And our next question comes from Matt Phipps from William Blair. Your line is now open.

Hi. Thanks for taking my questions, and, Pascal, welcome. Good to have you. You guys mentioned in the press release that 34 sites are now enrolling in the pivotal tab-cel trial. So I'm wondering how many of those enrolled both solid transplant and bone marrow transplant patients as opposed to maybe just one or the other. When it was two separate studies there was not, I mean a lot of them, Some of them were overlapping, some of them weren't. And then also just when can we get some disclosures on the enrollment of the trial? Do we have to wait until you guys say that you submitted a BLA, or will we get some info? I mean, I think some disclosure on how it's going would be important given how long it's taken.

Thank you. Thank you for your questions and for your welcome remark there. So on the number side we 34 that are unique sites, but of course some of them are doing both so we have about 23 sites for 301 and 27 sites from 302 that are open for enrollment right now. And I think as we said, we are going to increase this number of sites in the U.S., and also with the recent good news from Canada open by the end of the year Canadian site before moving to Europe following the CTA submission there. In terms of your questions on the communication related to the enrollment, in such an open study we have decided not to communicate on the number of patients being enrolled, and we will communicate on that at the time of the initiation for submission when we go into that phase of the development of the product there. What is important to have in mind is that we want to recruit the full population in both cohorts, even though we do the interim analysis on a smaller number of patients.

Again thank you.

And just to add on that, just to be clear, we plan to communicate on our guidance. So when we say we will initiate BLA submission in second half of 2020, we plan to communicate on that initiation at that time.

Thank you. And our next question comes from Philip Nadeau from Cowen and Company. Your line is now open.

Good morning. Pascal, let me add my congratulations on your new role. First question is on tab-cel and the European regulatory discussions. Can you give us some sense of what elements still need agreement between the company and the European regulators? Is it end points, enrollment criteria? Where is the debate?

I mean, as you know, we've adapted, amended our protocol for SOT, and we have merged the studies together, the SOT and the STT study into one study with two cohorts, but that's something that has been based on our discussion and constructive dialogue with the AVA. Now we need to discuss with the MEA about these changes and see how do they react to that and then whether that fit in terms of the timing of potential submission of a CMA. We are in the time process, which is a very unique type of process for advanced therapies where you can have regular type of interactions with the reporter and then of course you may benefit for an accelerated review then. So we are actively engaged with reporter and other members of the committees to be able now to clarify when is the timing for the MEA submission of a CMA. But it's really around this change in the protocol that I've linked with permission of the AVA that now we need to discuss with the EMA.

And do you have some sense when your discussions with the EMA will conclude?

It's moving rapidly. I cannot say exactly when it will be conquered, but what I can say that as soon as we have clarity there we will make that a matter for public communication.

Perfect, and then next question on the next generation CAR-T. Can you remind us what is the difference between your autologous ATA-2271 and the mesothelin CAR-T that was developed by Memorial Sloan Kettering and from which we have data?

Yes. Chris, do you want to start on that one? Then I will explain our strategy there.

Sure. So we're very excited at the opportunity to incorporate next-generation co-stimulation 1xx, for example, as well as the T-cell-intrinsic checkpoint inhibition through the use of the PD-1 dominant negative, and we'll be using the identical binding domain as has been used in the presentations given at ACR and ASCO. So we're using the clinically validated mesothelin binding domain, together with the best available technologies as we advance the mesothelin CAR-T program.

I think the key here is that the binding domain is really unique and in a sense that the data that we have presented, the early initial data we have presented at ASCO and ACR, clearly show a level of safety first of all and efficacy that is not very common in mesothelin targeted type of therapies. So we have that ACAV. That's something that we are learning from the first generation CAR-T development of our collaborators at MSK, and that same ACAV is being part of the construct of the next generation with the new customary domain that is aiming at having the right balance between expansion and persistence of the CAR-T's. And then, of course, we have added the PD-1 DNR to have really a more physical way of addressing the need to target PD-1 and make sure PD-1 and make sure that we have enough activity of the cells when they have penetrated into the tumor.

Perfect. And then my last question is on the earlier pipeline. Is there any update on ATA-3219, the anti-CD-19 CAR-T, or ATA-2431, the CD-19, CD-20, CD-22 CAR-T?

Yes. We are continuing these program, of course. We have our collaborators there for the one you mention, which is look at three targets there, and then that's with Moffatt of course. And then for the CD-19, which is an entire development, this development is progressing well, and as you know objective here is really in very well-known type of CAR-T target to be able to show efficacy and safety that will allow to prove the concept of allogeneic CAR-T based on EBV-specific cells. So both program are progressing. We don't have any special comment to make on specific type of achievements there, but they are progressing very well.

Great. Thanks for taking my questions.

Thank you. And our next question comes from Tony Butler from Roth Capital. Your line is now open.

Yes, good morning. Good morning, Pascal, and welcome, as well. My question is around 188 and MS and ECTRIMS. There are three questions. Number one is, you state in the release that you were looking at progressive forms of MS, but I think since trials it's actually both relapse-remitting and also progressive. So I'm curious what we get at ECTRIMS directly? That's number one. Number two is while you clearly won't have any, or there will not have been enough time for patients to have been on drug to see probably any changes in MRI, I'm simply trying to understand what could you present that would at least provide enough information that the T-cells do have activity and more importantly are affecting the MS directly. I would appreciate some specificity there if possible. And then finally did I hear correctly that there were patients or at least a patient that had PTLD MS that did clear? If that's true could you please describe clinically how that was actually determined? Thanks very much.

Sure. So this is AJ. Let me start with the third question, and apologies if there was any miscommunication or misunderstanding. There is no PTLD patient with MS that we are describing. I think what we were describing earlier is the concept that we have proof that the T-cells do get into the CNS. They are able to cross the blood-brain barrier and access the CNS compartment and exert function. Because we've treated patients with PTLD, CNS PTLD, we've also treated patients in a slightly different setting with CNV retinitis and a variety of other CNS conditions related to CNV with CNV-targeted T-cells. So there's very good proof that these T-cells that we create, virally targeted, access the CNS compartment well and exert their function. So that was the proof source for why we believe all of these cells are getting in to exert function for MS. So it's not really PTLD. Hope that answers question three.

It does, AJ, and thank you. My apologies for misunderstanding.

No. No worries. No worries. Getting back to the first question on progressive versus relapse-remitting. You're right. Our original thought process was to take a look at both. As we got deeper and deeper into the understanding of the autologous ATA-190 program and what we wanted to focus in on this program, we specifically decided to stay targeted on progressive MS, and also it's really a progressive MS population that really doesn't have a lot of inflammatory disease. So when you mentioned MRIs, you're right. There's a lot of times you can get relatively early reads in active disease by tracking MRI status, but this is not active disease. We really want to focus in on the progressive component, which is where really most other therapies have failed to, have failed. And we think that's the differentiator here. So for us the win is, The early win, There's an early win and a later win. The later wins are sustained responses on various disability measures like EDSS scores and timed 25 foot walk tests and whole brain volume production, but again those are one to two year parameters. These are not six month parameters. So the six month parameters, what you're looking for or at least what we're looking for working with our thought leaders is there's about six or seven different measures of physical function, cognitive function, a variety of different things, well-established measures. And what you want to see is stability across those measures or no decline. Obviously if you have improvement that's fantastic, but it's way too early quite frankly to even assess an improvement. But if you're able to show stability and no decline in function, that is the win, because these progressive patients will decline in function. So that's really what we're looking for is can you show stability or no decline in function across those various parameters, across multiple parameters in a single patient?

And again, I mean, this study is really there to determine a phase 2 dose that we'll use, by the way, in a phase 1b, and that's mainly based on safety and tolerability. But of course all these efficacy parameters will inform our choice of the dose for the next portion of the study.

May I ask one follow up? And that is, in the progressive patients, will all have had or will they still be on, and I'm sorry I did not look for the exclusion criteria, will they have had or still be on Ocrelizumab? Thank you.

Yeah, they would all need to wash out of Ocrelizumab before they enter the study. Ocrelizumab or any other disease-modifying agent would have to be washed out before they enter the study.

Thank you, AJ. Thank you, Pascal.

Sure.

Thank you. And our next question comes from Yigal Nochomovitz from Citi. Your line is now open.

Hi, thanks for taking the question. This is Samantha for Yigal. I wanted to build on an earlier question on your next gen CAR-T for mesothelin, specifically in respect to your dominant negative PD-1. Can you just go through all the advantages you see with including this into your CAR-T versus just dosing a PD-1? Because the MSK data suggest that you're already getting a really high response right with pembrolizumab. I'm just curious on your thought there.

No, that's a great question. Thank you. Chris, do you want to start? And then I'll give some feedback as well there.

Sure. So the inclusion of the dominant negative for PD-1 in the preclinical models had additional antitumor efficacy compared to the preclinical combination of CAR-T cells plus an antibody-based checkpoint inhibitor. Specifically a higher portion of the animals studied had complete response of their tumor, and in addition there was great durability. So we see a couple of important advantages in that way. In addition, of course, there's a factor of patient convenience as the combination being present in the T-cell itself and able to essentially carry that checkpoint inhibitory function into the tumor in a manner like a Trojan horse allows them to then have a more convenient treatment potentially as developed.

Yeah, and I think that the belief there is that to have the joined approach on the construct will really allow more physical way of targeting PD-1, and that is, as Chris has said, backed up by some animal data. But of course we need now to go into patients and to prove that this is a better approach than combining with regular administration of PD-1 blocker.

Thanks for that additional detail. And then in the MSK data at ASCO, 11 out of those 16 patients were PD-1 negative, but you still saw really great response rates in that population. I'm just curious whether you think expressing PD-01 is important for this therapy, and do you plan to screen in your potential, the planned phase 1 for in 2020?

So there was no correlation with PD-01 expression or status on the outcome in the data presented by our MSK collaborators, Dr. Adusumilli and colleagues, at ASCO. You know, one thing to bear in mind is that when T-cells entered the tumor environment, there's a release of cytokines that will dynamically change the expression of PDL-1, and that may be why it's so important to have the combination present. So I think the baseline PDL-1 expression is generally low in mesothelioma, but that may not be the most important factor in a combination therapy that works through an immune mechanism because the cytokines would be expected to dynamically change that axis.

And again, I think the DNR here could be an advantage because then in situ that's what is really important about the activity of the T-cells when they have homed to the tumor there is the in situ immunosuppressive environment that is laid by the PDL-1 expression there. And I think that for the DNR makes a lot of sense.

Great. Thanks for taking our question.

Thank you. And our next question comes from Salveen Richter from Goldman Sachs. Your line is now open.

Yes, hi. Thank you for taking my question. It's Mariana Brightman for Salveen. I have a couple. One of them, following up on the relevance of rationale for the changes. I was wanting to get a better sense for the timeline, and I was just trying to understand what are the manufacturing stages that the process is in. What other vectors have been made and how far along are we basically in preparation for the 2020 study?

So when you say the change of the timeline for the meso…

No, no. Just let's, Understand the timeline. What is…

Okay, OK. Sorry. Sorry. Okay. Oh, no. Sorry for that. So we're working with our collaborators at the MSK to be able to be ready for the I&D, and this is progressing very well. Our guidance has been that we will be, I mean, our collaborators will be filing an I&D in 2020 on this next generation CAR-T, and we are very confident in this guidance. Meanwhile, as you know, we are also progressing in supporting the first-generation study that is continuing. There's an important medical need, and as you can see the efficacy and safety results are encouraging there. And we believe we are learning a lot from that study, and that learnings could be applied to our first study with the next generation meso CAR-T there. So timelines are in line with the guidance, and I think are progressing very well there for 2020. And manufacturing at this stage is to be done initially at MSK and will be transferred at our Atom unit in due time.

Got it. Thank you.

Thank you. And our next question comes from Maury Raycroft from Jefferies. Your line is now open.

Hi. Thank you for taking my questions. This is David on for Maury. First question is regarding tab-cel. Just can you share with us any updates on the expanded axis program such as updates on the number patients included? How about patients enrolling? And if you will report any new data from this program?

Thank you for your question. So we are not giving any number of patient at this stage, but the program is really recruiting a lot of patients that either cannot participate to the study itself for the phase 3 study, cannot be enrolled in the phase 3 study, or are different type of patients there. So we're very pleased with the enrollment in that program, and we will communicate at future congresses new data on this program.

Got it. That's very helpful. And the second question is with the tab-cel, can you tell about the status of your nasopharyngeal carcinoma trial? And how is enrollment currently going and what's the sort of the plan for the disclosure?

Thank you for your question. AJ, do you want to answer that one?

Sure. So the NPC study, as you know, is ongoing. We're in the first phase of the program. We're looking to enroll anywhere between 12 and 24 patients in the study. As you know, it's the first time we've ever really combined cell therapy with PD-1 inhibition from, When I say cell therapy, the first time we've combined tab-cel with PD-1 inhibitor. So the major focus for us here is going to be really assessing the safety of the combination initially. From a timing of data, we're not expecting to present data this year on that program just yet.

Ah. Got it. That's very helpful. And then last question is on the mesothelin CAR-T. Can you just share with us any preliminary thoughts around your potential trial design for the CAR-T program in 2020?

So again, the I&D is planned for filing in 2020. We are not right now giving any details on the protocol because we're using it right now, and they will give details on that protocol in due time. Again that is important that we will take into account all the learnings from the first generation and what we're doing there, and I think that's an important aspect to optimize the I&D and the continuing study for these next generation CAR-T. We're very confident that we will be able to start that in 2020 as planned with an appropriate protocol that will really allow us to have proof of concept of the clinical efficacy and safety of these next generation CAR-T as rapidly as possible.

Got it. Thank you for taking my question.

Thank you. And our last question comes from Ben Burnett from Stifel. Your line is now open.

Hey, great. Thanks so much for taking my questions, and, Pascal, great to see you on board and congrats. Two questions from me. Just one first just to clarify on ATA-188. What cohorts or dose levels specifically will we see efficacy data for at ECTRIMS? And then also have patients so far in the phase 1 progressed or made it to the point where they can enroll in the open lable extension? And if they have, I guess, what's sort of been the enrollment rate or retention rate into that study?

Thank you, Ben, for your question. AJ?

Sure, and I may ask for clarification on part two, the question. So in terms of the cohorts, we'll be presenting data on two cohorts out to the first two cohorts, out to about six months' worth of data. We may have a little bit more on one of the initial cohorts, but again it's really just a timing issue in terms of how long the patients have been in study at this point. In terms of enrollment into extension, I apologize, but can you clarify that? Because for what we have right now, just as a plan is, as we finish this fourth dosing cohort, and we're close to finishing the fourth cohort now, we'll make kind of a database decision as to what the dosing will be for the randomized portion of this study. And is that what you're talking about as extension?

Well, I'm referring to a study skim off that you guys presented at EAN wherein it looks like cohorts 1, 2, 3, and 4 had the opportunity to go into an open label extension after one year of treatment. I was just asking if anyone's made it to that point yet and what the sort of retention rate's been into that section.

That's, okay, now I, I apologize. Thank you for the clarity. Yes, we do have a few patients from the first cohort that have gotten out past a year and are going into the extension. I'm not really going to be able to comment on percentages or anything like that because it's a little bit too early, but we do have patients who have gotten out to that time point and will be…

Okay. Okay, understood. Thank you for that. And then just one last one on tab-cel and PTLD. I guess can you just talk in a little more detail about the logistics of getting patients enrolled into this study? I guess really why has it been so challenging to forecast enroll rates? And I guess what gives you the confidence now that enrollment can be accurately forecast going forward?

Sure. It's a great question. So probably the prime difficulty, you know this is an ultra-rare disease, so clearly ultra-rare presents challenges, and this is a different challenge than some of the other ultra-rare diseases out there because this is one where the patients really do come up randomly. There's not major centers of excellence where you're going to have a high concentration of patients. That's step one. Step two is the timing because when you look at this you're really looking at patients, catching them right as they fail therapy. Now normally that's not that big of a problem, but here these patients progress so rapidly to death that there's a short window of opportunity to really help them. So catching them at exactly that window is a second challenge. We've done really well actually. In terms of the operational efficiencies we've created to make sure we can catch these patients, that act has gone extremely well, and we've gotten a lot better over the course of the study in achieving that. So that's what gives us a bit more confidence that we'll be able to enroll correctly or at least on the pace that we're targeting. The third thing just to keep in mind, and I'm going to roll this back into the EAP program a little bit, part of the challenge that you have is, as I mentioned, these patients kind of come up everywhere. And when we find a patient who, for example, is not at a clinical trial site, we'll actually move him over to wherever the clinical trial site is, irrespective of where that is in the country. But you also have to remember that these patients are often so sick that you can't actively transfer them. Or they have a medical, they're able to be transferred medically, but they have a specific issue that actively prevents the transfer process. And in those cases, those are patients that get onto our expanded access program, which is what Pascal had alluded to where we actually have a fair number of patients on our expanded access program, both for PTLD and for other EBV conditions. But those are the major factors into why I think the enrollment has been a little bit difficult to work through, but since we have focused in on our existing centers we've created these operational efficiencies to make sure that that timing component, which is really critical, is not as much of a factor as it has been in the past.

And to build on that, I think our confidence in the guidance is linked to the fact that we've learned so much onto the study that new we're confident that we can deliver what we said regarding the initiation of BLA submissions second half 2020. Also to have in mind that in the commercial setting, it will be vastly different there because we'll have the product available within three days to any site, any physician that is in need of that product for his patients across the country. And in the US, which is a very attractive market for this type untreatable disease, that means that this possibility to treat the patient extremely rapidly, offered by our top sale as allogeneic T-cell that is available within three days, is going to be immensely useful for patient and physicians.

Okay. Great. Thanks so much for the additional color there, and congrats again, Pascal. Thank you.

Thank you very much for that.

Thank you. And I'm not showing any further questions at this time. I will now like to turn the call back to Pascal Touchon, President and CEO, for any further remarks.

Thank you very much, and thank you, everybody, for having joined that call. I want to think really you for taking the time to join us today. We look forward to providing a date for the remainder of the year on a regular basis as we continue to advance our position as a leader in the auto-shell allogeneic T-cell immunotherapy. Thank you very much.

Ladies and gentlemen, thank you for your participation in today's conference. This does conclude today's program. You may all disconnect. Everyone have a great day.