Hello, ladies and gentlemen and welcome to the Autolus Therapeutics 3Q 2019 Financial Results Conference Call. At this time, all participants are in a listen-only mode. Later, we will conduct a question-and-answer session and instructions will follow at that time. [Operator Instructions] As a reminder, this conference call is being recorded. I would now like to turn the conference over to your host Ms. Silvia Taylor, Vice President Global Corporate Affairs and Communications. Ma'am please go ahead.

Thank you, Jonah. Good morning or good afternoon, everyone, and thank you for taking part in today's call on the financial results and operational highlights for the third quarter of 2019. I am Sylvia Taylor, Vice President of Global Corporate Affairs and Communications, as Jonah just introduced me. With me today are Dr. Christian Itin, our Chairman and Chief Executive Officer; and Andrew Oakley, our Chief Financial Officer. Before we begin, I would like to remind you that during this call we will be making forward-looking statements. All statements, other than statements of historical facts, contained in this presentation are forward-looking statements. Our actual results performance or achievements may be materially different from those expressed or implied by the forward-looking statements. For a discussion of the risks and uncertainties relating to our business and other important factors, any of which could cause our actual results to differ from those contained in the forward-looking statements, please see the section titled Risk Factors in our annual report on Form 20-F filed on November 23, 2018, as well as discussions of potential risks, uncertainties and other important factors in our other periodic filings with the SEC. The forward-looking statements contained in this presentation reflect the company's views as of the date of this presentation regarding future events and the company does not assume any obligation to update any forward-looking statements. You should therefore not rely on these forward-looking statements as representing the company's views as of any date subsequent to the date of this presentation. On slide three you will see the agenda for today and it is as follows. Christian will provide a brief introduction and that will be followed by our operational highlights for the third quarter of 2019. Andrew will next discuss the company's financial results and then Christian will conclude with upcoming milestones and other concluding comments. And, of course, we will welcome your questions following our remarks. So with that, I'd now like to turn the call over to Christian.

Thank you, Sylvia, and good morning to all of you and thank you for joining us. I'm pleased to review our progress in the third quarter of 2019, as well as some recent company highlights. On slide five and as reported yesterday, we will have four oral and two poster presentations at ASH in December in Orlando. We are pleased that we will be presenting data across the majority of our clinical programs. The focus will be on AUTO1 with three oral presentations alone. This coming weekend also, we will have a poster presentation with preclinical data on our first solid tumor program AUTO6NG at SITC. Turning to slide six. Let's start the discussion of our Q3 operational highlights with our highest priority program AUTO1 in adult ALL. We are pleased that earlier this week the U.S. FDA granted AUTO1 orphan drug designation for treatment of acute lymphoblastic leukemia. Relapse and refractory B-cell acute lymphoblastic leukemia represents a significant commercial opportunity, both in terms of the potential market size, as well as the high level of unmet need in the management of the disease. Worldwide approximately 8,400 patients are diagnosed every year with about 6,000 of those patients coming from the U.S. and the EU5 European countries. While response to initial combination chemotherapy regimen is encouraging, only 30% to 40% of adult ALL patients will achieve long-term remissions and the median survival for adult patients with relapsed/refractory ALL is less than one year. While Kymriah a CD19-targeting CAR-T therapy was approved for pediatric ALL patients in 2017, no CAR-T therapy has been approved for adult ALL patients to date. The only redirected T-cell therapy approved for adult ALL is blinatumomab or BLINCYTO, a bispecific CD19-targeting T-cell engager. Blinatumomab has a 42 response rate -- 42% response rate, yet the durability…

Thanks Christian and good morning or good afternoon to everyone. It's my pleasure to review our financial results for the three-month period July through September of 2019. On Slide 18 net total operating expenses for the three months ended 30th September, 2019 were $35.6 million. That was net of grant income of $0.3 million and that compares to net operating expenses of $17.1 million also net of grant income of $0.3 million, but that's the same period in 2018. The increase was due in general to the increase in development activity increased head count primarily in our development and manufacturing functions and the cost of being a public company. Research and development expenses increased to $27.3 million for the three months ended 30 September, 2019 from $10.1 million for the three months ended 30 September, 2018. Cash costs which exclude depreciation as well as share-based compensation increased to $21.6 million from $9 million. The increase in research and development cash costs of $12.6 million consisted primarily of an increase of compensation-related costs of $5.2 million due to an increase in employee headcount to support the advancement of our product candidates and clinical development an increase of $3.6 million in research and development program. Expenses related to the activities necessary to prepare, activate, and monitor clinical trial programs including the manufacturing and technical transfer activities required for AUTO1 to enable the commencement of the registration study in adult ALL and an increase of $2.6 million in facilities costs supporting the expansion of our research development translational science capability and investment in manufacturing facilities and equipment. And lastly, an increase of $0.7 million in telecom software costs as well as an increase of $0.5 million elsewhere. General and administrative expenses increased to $8.6 million for the three months ended 30 September, 2019 from $7.3 million for the three months ended September 30, 2018. Cash costs which again exclude the depreciation expense and share-based compensation decreased to $5.6 million from $5.7 million. Compensation-related expenses decreased by $0.6 million. IT, communication, general office expenses decreased by $0.7 million and that was offset by legal and professional fees of $0.9 million and an increase of $0.3 million in very preliminary commercial expenses. Net loss attributable to ordinary shareholders was $27.2 million for the three-month period compared to $12.9 million for the same period in 2018. The basic and diluted net loss per ordinary share for the three months ended 30 September, 2019 totaled $0.61 or $0.61 compared to a basic and diluted net loss per ordinary share of $0.33 for the three months ended 30 September, 2018. Cash and cash equivalents at the end of the period totaled $229.4 million and that compares with $247.1 million at the end of September in 2018 and we anticipate that cash on hand provides us with the runway into the second half of 2021. With that I will now hand the call back to Christian to give you a brief outlook on our expected upcoming milestones. Christian?

Thank you, Andrew. Let me conclude this part of the management discussion with a review of the upcoming milestones and news flow through 2020. Let's move to Slide 20. The upcoming 15 months will be an eventful period for us with multiple clinical milestones and opportunities for value creation. Our chief operational focus will be our moving AUTO1 in adult ALL into registration trial in the U.K. and U.S. We also expect to report data across various programs and to progress a number of our other clinical candidates, specifically updates on our ongoing clinical trials initiation of a Phase I study of AUTO1NG in pediatric ALL in the first half of next year a go/no-go decision on Phase II initiation of AUTO3 in DLBCL middle of 2020, initiation of a Phase I study of AUTO6NG in neuroblastoma in the second half of 2020, and initiation of a Phase I study in the next-generation program in multiple myeloma also in the second half of 2020. In conclusion on Slide 23, I'd like to recap the major messages from today's call. First AUTO1 is our foundational program and the first Autolus program expected to move into pivotal stage. Given the positive safety and efficacy profile today, we believe that AUTO1 has the potential to be a best-in-class CD19 CAR T in ALL. Secondly our next priority is on AUTO3 in DLBCL with AUTO3NG as a next-generation opportunity. We expect to report full Phase I data for AUTO3 in middle of 2020 to reach a decision point on Phase II trial initiation thereafter. Third, in the pediatric ALL we have transitioned our focus on AUTO1 and AUTO1NG. While AUTO3 data confirmed the dual-targeting hypothesis, we believe the excellent persistence with AUTO1 is likely to drive long-term remissions. Looking ahead to 2020, we see opportunity for additional value steps for multiple myeloma T-cell lymphoma and the GD2-positive tumor programs. The company has a strong balance sheet with $230 million in cash which provides a run rate to the second half of 2021. And finally, we're looking forward to seeing many of you at the upcoming SITC and ASH Annual Meetings. We'd now like to take your questions. Operator please open the line.

[Operator Instructions] Your first question comes from the line of Gil Blum from Needham & Company. Your line is open.

Hi everyone and thank you for taking my question. Just a quick one about -- so AUTO6NG data that's coming out at SITC, we know that this is kind of a mix of T cells that were transected with two different vectors. What kind of analysis, would you have to do in a product that's this complex before using it in human? Like how would the product be defined?

Well first of all thanks for joining and thanks for your question. Obviously, what we're doing with AUTO6 is, we're introducing a substantial amount of genetic information into a single cell. And that is actually a level of genetic information that you cannot deliver with a single vector. So you have to use two vectors to do that. And this is now actually an approach that's been used in a number of programs that have gone through regulatory review and are actually currently the clinic for other types of indications and obviously have gone through the normal regulatory process and are active in development. Ultimately, what you have to show is you have to demonstrate the activity of the product as is and you design your safety studies etcetera to really understand the activity of the product as a whole. We have to understand obviously also that, even when you look at a product that is transduced to the single vector that we have multiple types of differentiation state of T cells in there which gives you quite a wide range of properties of these cells just based on the differentiation state. So the products are complex to begin with. And the programming on itself we don't believe will add a significant element on top by adding the two vectors in of themselves. Vectors are designed to -- all of them actually recognize the target antigen so that the basic activity is actually shared amongst all transduced cells.

All right. Maybe a bit of an odd question, but if Kymriah or ever used off-label in adults ALL? Is that -- anyone does that?

Well what we do know is that the products obviously are not part of the normal payment process that you can actually get -- actually have them used in. If they're used off label or not that's difficult for us to tell. There's certainly a possibility within oncology for products to be used off label, but it's something we can't judge. And I don't think there's any information out there on what it might be and how many patients might actually be impacted.

Got you. And just the last one, I know we're getting an update on AUTO2, but when could we expect updates on the new program in multiple myeloma?

Yes. So the next-gen version for the multiple myeloma program, we expect to update when we're actually entering into clinical trials and obviously during the course of next year have opportunity to provide an update on the design of the program and we'll do it at that point in time.

Thank you for taking my question and congrats on the quarter.

Thank you. Thanks for joining.

Your next question comes from the line of Jim Birchenough from Wells Fargo. Your line is open.

Hi guys. Thanks for the call and congrats on the progress towards the AUTO1 pivotal. Just on that pivotal question, could you maybe speak to; number one, are there going to be any entry criteria, whether it's tumor burden or other features that will reduce the risk of severe CRS or neurotoxicity? And then maybe if you could discuss what the efficacy hurdle is there if there's some lower bound of the confidence interval you need to shoot for at some point estimate of response just so we have a frame of reference? And then I've got a follow-up.

Yes. First of all, thanks for joining Jim. With regards to the pivotal study that we're planning to do here, obviously when you look at the inclusion criteria, we're including patients that actually are in formal relapse. And in other words, these are patients that have more than 5% blasts in the marrow which is kind of when you have a morphological relapse. So all these patients have manifest disease and obviously given the speed at which the disease moves, can actually have quite a range. And as you see with the data that we have shown at AACR and we'll update at ASH and approximately half the patients have a massive level of tumor burden, but all patients are in formal relapse.

And just in terms of the efficacy hurdle?

In terms of the efficacy hurdle, when we look at the current programs you'd expect and you'd see kind of well from our own programs that we've shown today is that you want to show a robust molecular complete remission rate because that's your entry ticket for a transformational activity. And then obviously you want to have durability of effect. So we would clearly want to improve substantially over and above the durability of effect that we've seen with blinatumomab in this patient population. Remember that was 31% at 6 months. We believe that we want to see somewhere in the range of the double of that at that point in time.

Terrific. And then just on AUTO3 and the go/no-go decision in DLBCL in mid-2020 is durability of response going to be the key there? Or maybe you could speak to what the criteria are for go/no-go? And I guess, if there was a durability that was adequate in ALL is there something different about DLBCL where you might still get away with a win there? Just trying to understand the go/no-go?

Yes. The disease settings are quite different between leukemia, acute leukemia and DLBCL. What we need to be able to do in leukemia is, we need to be able to put pressure on the tumor for very long periods of time. We're talking 18 months 24 months to get to transformational activity and we're actually going to be showing kids that are now really long-term -- have long-term observation on the AUTO1 experience. When you look at DLBCL, it is quite different. What you need to induce is, you need to induce a complete remission. And typically, the complete remission that you can induce and observe at three months or at the latest at 6 months depending on the program is a very good measure for long-term benefit in that patient population. And the effect that it can induce is obviously one that happens relatively quickly usually within 3 months' period of time. Most of those patients have accomplished or have achieved their CR. And at that point most of the patients, if you look at the Yescarta data, the JCAR-17 data, the Kymriah data this can actually sustain that activity. And so what we want to see is, we want to see a robust CR rate as well as obviously have a good sense for the durability of those CRs as well at that point in time.

Just a final question Christian, reimbursement -- inpatient reimbursement for Kymriah and Yescarta has been difficult from what we've heard and we've heard from a number of consultants that the move to outpatient is going to be really important in advance of getting some CAR-T specific reimbursement code for the inpatient setting. And so how important is it to you to have features in your cell that lend themselves to outpatient delivery? And is there some way to incorporate that into a trial design?

I think it is important when you think about the overall cost of therapy, obviously, there is the actual cost of the therapeutic itself of the drug itself, but there's also a significant cost associated with managing the patients. And obviously, the more severe your adverse events are the higher the costs are for the management of the patients. And what cause challenges particularly in DLBCL reimbursement in the U.S. and elsewhere is that that portion of patient management cost was initially not properly covered. And that caused the major issue for the hospitals who were treating these patients. That is now actually being resolved. It's also in part resolved for the Medicare patient, but it remains a significant driver of the overall cost of therapy. So actually having products that have no high-grade CRS cytokine release syndrome that have limited or minimal neurotoxicity is important, because it allows you to actually consider giving the patient the therapy and then actually have the patient in an outpatient setting from there on forward. And as you see with all the programs it is a progression in terms of the information and the experience you have with the product. What you want to make sure is that it captures much information related to that intensity of patient management during the course of your pivotal study even if your patients are initially mostly treated as inpatients and then obviously with increased experience of the products will move more towards an outpatient setting. But it's absolutely crucial to collect that information also when you have conversations with payers, because it is a key element of the value assessment as well.

Great. Well, thanks for taking the questions.

Thank you very much.

Your next question comes from the line of Matt Phipps from William Blair. Your line is open.

Hi, guys. Thanks for taking my question. A few here. I guess first on the AUTO3 abstracts. In the ALEXANDER study five patient discontinued before dosing and four were screen failures. I know this can happen, but I guess what were some of those reasons? And maybe this is related to the AUTO3 manufacturing in the site of the Catapult site. So what is your current turnaround rate you think for these programs through Catapult? And will that help with this enrollment?

Yeah. So, obviously, with the particularly the initial patients that we have been treating, those are patients who are extremely advanced and are at high risk of developing infections as well as some of them had highly progressive or rampant disease frankly. And so that was clearly one of the key challenges that we had in the very first patients that we were had in the program. In general, the turnaround time that we're looking at is comparable to what is -- what we see across the industry. And it is clearly an area you want to focus on very consistently to minimize the time from patient identification to patient dosing. It's really every step along that chain, including obviously the manufacturing elements themselves but equally important literally every step up along also on the hospital side that you want to minimize. And that's a continuous process that we're working on and continue to improve on. And the manufacturing at the new manufacturing site obviously is now open and fully operational. And obviously, it gives us a good environment to keep on making progress on the turnaround time.

Got it. Thanks. What -- and then kind of again comparing AUTO1 to AUTO3 and the next-generation steps, why do you think it is that AUTO3 or shows less persistence or I guess AUTO1 shows more at this point? Do you really think it's all driven by the SCFE, a novel SCFE or something else with kind of the co-stems? Obviously, AUTO3 does incorporate an OX40. And then again, with the AUTO1 next Going, are you using the same CD22 construct for AUTO3? Or is it unique? And how is it different?

Right. So when we look at the various programs particularly comparing AUTO1 to AUTO3, there's sort of two key drivers that we believe will matter a lot particularly in leukemia. One is persistence, which is foundational. Without persistence you do not have long-term benefit. You have no -- you cannot induce actually a transformational effect in the kids. And the level of persistence that is required in leukemia is extremely long. You need substantially beyond a year. We've seen persistence up to two years and onwards with AUTO1. Now the reason why AUTO1 has such an excellent persistency is because of the in part because of the vector system and because in part of the way that we designed the receptor. The vector system we're using is a lenti-based vector, which is known to actually be giving you typically very good overall levels of persistence. This was the key breakthrough when you think about it that Kymriah accomplished versus the earlier programs in pediatric ALL. What we have done with the AUTO3 program is not to build initially on -- or not to build on a lenti system, but actually put it in a retro system. And the reason for that is that we're expressing multiple genes in that particular product. And that is what retro systems are actually much better suited technically to actually do that. So when you have multiple components, you want to deliver them from a single vector. That is where retro systems are particularly good at. What we see with the AUTO3 program actually is even in feeds, overall and you'll see the data at -- presented at ASH is overall, actually very good persistence. What we don't see is exceptional persistence. And that is where there's the difference between the two. The next-gen program that we're looking at is really focused to go a step further and allow the cells to adapt to a complex microenvironment. So it is really focused on adding additional modules to deal with those questions similar to what I described before for AUTO6NG. The chimeric antigen receptors there's also elements there that we're looking at and that may slightly be different compared to AUTO3 going forward.

Okay. If I can ask one more on the multiple myeloma program, the AUTO2 abstract I guess makes it clear why that is being stopped. But going I'm surprised to see on the slide that the NG version doesn't disclose the target. Do you think there is a problem with the April CAR that's maybe driving the suboptimal response and that's kind of being evaluated in the NG product?

Well, we've also made a very thorough evaluation of the performance of the product molecular design, et cetera. And obviously that is flowing into the new product. We haven't guided on the target antigens that we're going after with the program. But obviously, we're doing -- we're literally working on optimizations on every step within that particular product.

All right. Thank you, Chris.

Okay. Thanks a lot, Matt.

Your next question comes from the line of the Debjit Chattopadhyay from H.C. Wainwright. Your line is open.

Hi, guys. Good morning. This is Aaron on for Debjit. So I just had a question directed at the AUTO4/5 program. So I saw a ASH presentation or abstract from group at Baylor College of Medicine. They're using a CD5-directed CAR-T therapy for T-cell leukemia and lymphomas and they got three out of nine elite responses. They had limited durability and they saw some viral reactivation. So are you thinking that this is like -- I don't know if you guys have seen this but do you think it would be a viable competitor? And also what are you thinking on the intent to use AUTO4/5 either as a bridge to stem cell transplant? Or do you think this would be curative in and of itself?

Well, Aaron, thanks a lot for joining. Yes we did we know of the CD4-targeting and CD5-targeting programs. The general I think observation is obviously one of the challenges when you deal with T-cell lymphoma is that you really want to make sure that you're not wiping out critical parts of the T-cell compartment, because of the risk of rendering the patient immune deficient. And so that is sort of the fundamental kind of challenge that I think we have with that type of disease -- with that particular set of diseases because the structures, the antigens this include CD4, it include CD5, it includes many of the other target antigens you find in a T cell are just shared across a vast range of T cells including some of the T-cell clones. CD5 is an active -- often considered also in activation markets it goes up when you sort of have activated T cells. And in that sense obviously you can actually hit the significant number of T cells. It's difficult for us to tell at this point in time based on the published data of what the profile really is and whether it's a sustainable profile and whether it's a safe enough profile. I think we need to see kind of what those -- how these programs shape up to actually get a sense for what they might be able or capable of. As you point out if you run the risk of wiping out sort of essential parts of the T-cell compartment you may need to actually rescue the patients with a transplant and sort of a rescue strategy. That is a possibility. I don't think there's any data to that that we have seen in any other programs. So I think it is not a point where we can actually even opine how they're actually going to fare going forward.

Okay. And about the viral reactivation would you guys anticipate seeing any sort of similar rates? So they saw about 11% do you think that's comparable? Or are you guys shooting for -- since I think you guys were saying it was almost a 50-50 split of the TRC A and B so do you think you would see any sort of viral reactivation or anything on that?

Well I think it's important that when you go after TRBC1 or TRBC2 you're sort of -- you're not focused on a subpopulation of CD4 CD8 as an example. You have a mix of helper’s cells. You have a mix of effector cells in your TRBC1 population or in your TRBC2 population. So you tend to have a full complement of T cells that can actually take care of challenges -- infectious challenges. Now if you wipe out a substantial proportion of the CD4 population that may have or can have an effect obviously on your ability fight viral infection as we all know from HIV viruses, which are obviously predominantly have attacking CD4s and wiping out CD4 cells. So this is not a one - it's not actually a sort of a like-for-like kind of situation. But clearly if you have particular subsets of particular cells that you run a higher risk that you might actually lose a function. I think the approach that we're using actually leaves -- does not actually create a bias towards a subset but it gives you the range of T-cell phenotypes within that type of -- with the same subtype of T-cell receptor. But that shouldn't shift your overall immunological activity.

Okay. Great. Thank you. Thanks guys.

All right. Thank you very much.

Your question comes from the line of Graig Suvannavejh from Goldman Sachs. Your line is open.

Good morning or good afternoon and congrats on the progress and thanks for taking my questions. I've got two for you Christian and one for Andrew. My first question is just on the pivotal clinical trial design for AUTO1. I was wondering if you could walk us through that trial design and specifically in terms of its similarities or the differences between the pivotal trial for blinatumomab. And then my second question Christian just has to do with your AUTO6NG program. I unfortunately won't be able to come down to SITC but in terms of where you see the current state-of-the-art in terms of CAR T and its applicability or utility in solid tumors in your opinion as you assess the landscape, I wanted to get your sense of -- if you had looked at all the ASH abstracts kind of where we are and perhaps what you see as promising and perhaps what you see may not be promising. And then my follow-up question for then Andrew is, Andrew just looking at the third quarter it looks like your G&A was a little lower than we expected and just maybe there's a lumpiness to that but just maybe if you could give us a sense of how we should think about your OpEx in the fourth quarter and on a go-forward basis. Thanks.

All right. Great. Thanks a lot for joining. Thanks for the questions. So the first question was related to the pivotal design. Actually the pivotal design is very similar to the pivotal study that we conducted at the time with blinatumomab. With one key exception and the one key exception is that we allow patients that have previously been exposed to blinatumomab or inotuzumab, which is this change in the standard of care since we conducted the pivotal studies with blina. But it is the same patient population in the sense that's a relapsed/refractory population there obviously including patients that have experienced blina or inotuzumab and failed on either one or both of them. But the basic outlook of the patients at the time that patients have the speed at which the disease progresses is highly comparable. The inclusion criteria with regards to tumor load as answered before on Jim's question, we look at patients that are in morphological relapse. So these are proper relapses and obviously that can be a range. It's either slightly above 5% or it's all the way up to about 90% 90-plus percent. As I indicated in the current study you expect about half of those patients to be in the range of about 50% plus in terms of tumor load so that's an enormous amount of tumor burden. That's sort of the population. It's very comparable very similar in that sense to the original study that we conducted with blina also in terms of the size of the study. And in essence also to -- similar to the answers given before the size of the study that we did with blina which also is about 100 patients was in essence a 2x improvement of the standard of care. And what we're looking to do here is also again now a 2x improvement over and above blina, which is sort of the ballpark that you're aiming for which gives you the same type of size of the study but also a slightly more advanced patient population but the same level of tumor load and highly comparable durations with regards to the time lines the patients have at that stage of their disease.

Maybe I can just quickly follow up there. Does the inclusion of previously treated patients with blina, how do you think that may or may not impact your outcome?

We don't think it will have a significant impact. I mean, there's a limited amount of data from other programs that are sort of including patients that have previously been on blina. There didn't seem to be a major difference between patients plus or minus the experience. And also when you look at the patients at the time of relapse that actually were treated with blina on relapse there's an extremely small proportion of patients we're talking very low single-digit percentages that have lost antigen. So also on that level it doesn't really change the profile neither of the disease nor does it really change the profile from a target antigen perspective.

Okay. Thank you. Sorry about that, interrupt you.

So the second question was related to kind of the solid tumor programs and what we're doing there. What you may remember I mean the approach we've chosen with AUTO6 programming with GD targeting GD2 was to go after a tumor that is homogeneous -- where the target is homogeneously expressed so that reduces one level of variability and then focus on first of all targeting the tumor while -- with an appropriate therapeutic window and that's what we've demonstrated with the original AUTO6 data, where we could show that we actually get partial remissions in the kids without inducing neurological toxicity which was the fear and the adverse event that's particularly known for the first generation of GD2 monoclonal antibodies in the field. We then did as a second step is this is what the AUTO6NG is about is to make sure that the product is resilient and can withstand various lines of defense that the tumors actually do develop. They are obviously, defenses in the category of checkpoints defenses in terms of soluble agents like TGF-beta and then can also create an environment that is depleted of nutrients et cetera. And so you need to have an element that sort of keeps the cell well activated. And that's basically a JAK STAT type of signal that's providing into the cell like all of these signals are intracellular. They don't change -- don't create a toxicity on the outside. They actually are all contained within the cell and don't change the susceptibility of the product to these negative influences. So when we take a step back and look broader at the field, I think that is clearly a theme that you'll see develop over the upcoming periods because it's sort of the second stage. Once you have an ability to target and target with an appropriate therapeutic window the next thing you need to do is you have to make sure that the therapeutic approach actually can be sustained throughout the tumor and to crack the tumor that way. And that is sort of the core to really focus on. And we see a number of approaches with various types of elements that are in exploration at this point in time, but I think that is going to be a key component to give us a much more substantial activity in the solid tumor setting. So, with that I'm happy to hand it over to Andrew.

Graig, thanks very much for the question. I always appreciate the opportunity to answer a financial question. There were some sort of one-off sort of issues that -- not issues, but items that resulted in the G&A being a little bit lower than we would have thought. You could probably expect that in Q4, we'd have an uptick in G&A compared to the third quarter, but that's as far as we're probably going to go in terms of guidance at this point.

Okay. Great. Thanks again for taking my questions.

All right. Thank you, Graig.

Your next question comes from the line of Biren Amin from Jefferies. Your line is open.

Hi guys. Thanks for taking my questions. Christian, I guess on the AUTO1 pivotal trial in adult ALL starting first half 2020, I think previously, you were guiding to Q4 2019 initiation, so I just wanted to kind of understand the reason for the pushout?

Hi there. Thanks for joining. As pointed out in the remarks before, we're filing the CTA in the U.K. this month and we're obviously going through the FDA process beginning of next year and that gives us an ability to obviously open up centers in the U.K. and then in sequence in the U.S. during the first half of next year and get the study off the ground in that period. So, we're slightly later than I think the guidance that you referred to, but it's well under way and with the appropriate feedback from the agencies, which is obviously the key thing.

And then so, I guess for the U.S. patients that would be treated in that study, can you talk a little bit about the manufacturing cells at Catapult? How much additional time will be required because of that shipping of the cells across the Atlantic? Thanks.

Well, the logistics actually is not that much increased due to the actual shipment because, if you're in a center in the U.S., you collect the cells and you have to ship it. It also goes through the U.S. hub, gets transferred from one hub to another, gets to your manufacturing site et cetera. So, it may add a day, maybe two days max, but we don't expect to see more than that in terms of added time. The flight time from the East Coast to London is five hours. This is pretty much the same time you have if you commute from the West Coast to the East Coast or back. So, it's very comparable in that sense. The distances -- there is a bit more water in between, but the distances are quite comfortable.

Got it. And then I guess on AUTO3 for DLBCL, how many more patients can we expect at ASH at the 150 billion cell dose?

As we indicated in the Q2, obviously we had a delay. This is why the decision point moved out to second half -- sorry to the middle of next year. So, it's going to be a limited number of additional patients because of the delay that we have and reported on in Q2.

Got it. And then you talked a little bit about screen fails -- screen failures in that trial due to progressive disease. Are you incorporating bridging chemotherapy to try to I guess minimize screen fails?

To the extent, appropriate for the respective patients, you would include some holding chemotherapy. But again, particularly in kind of the early patients where we're treating, obviously their infections were probably the key issue which required us to sort of actually manufacture to put the patient -- we have the patient cleared first before you could actually get on treatment and that obviously has a significant impact on time.

Got it. And then maybe just the last question. So, on AUTO2, I think there's some prior questions asked about the response rate that you've seen and you're clearly making efforts with the NG program. I guess, I would want to kind of see if you've dissected the AUTO2 data a bit further to see what led to the nonresponses. We're seeing a number of approaches in BCMA, specifically, I think there's an approach out there that Fred Hutch is looking at you're probably aware of this where they're combining with gamma secretase to try to reduce the levels of soluble BCMA. Have you looked at that in terms of as a reason for nonresponse or for a relapse?

So we did obviously do a very thorough analysis and there are clearly three major areas that you -- we felt that we wanted to improve the design to sort of achieve and improve the properties. The first area is that, we wanted to get a deeper response with the product and that is very much around optimizing target engagement and the core potency of the chimeric antigen receptor itself. The second element is to make sure that it can boost the persistence and that obviously has a few other components technically that you can look into and you should look into. And those are the second kind of key places to look at. And then obviously, we need to consider the -- also the impact the microenvironment can have on the overall activity of the product and that's the third area, we've been looking into. And we're considering obviously -- considered in the context of the design for the next-generation program.

Okay. Great. Thank you.

But it's not a single factorial element. There are several components that we believe requires some additional improvements.

Got it. Thanks.

Thank you.

Your next question comes from the line of Miles Dixon from Peel Hunt. Your line is open.

Hi there. Thank you for taking the call. Christian, Andrew, the question is really on the wider CAR-T area. I just wondered with respect to your apparent better durability of AUTO1 versus some of the reported relapse rate out there, whether do you want to comment on any of the -- or changing mood music if you like with payers on reimbursement structure and/or phasing agnostic geographies really. Is there anything you want to comment on that?

Well, I think one of the things that is important to do as you're sort of designing your pivotal studies is not only to think about the data you want to generate or you have to generate to fulfill the regulatory requirements for approval. But you also want to make sure you collect the necessary data that actually are helpful for payers to evaluate the value of the therapy. And that is sort of the key element, that we also spend time on with various bodies on the payer side, to understand the nature of the data that they needed to sort -- to be able to value and evaluate the benefit of that kind of -- of this kind of a therapy and to make sure that that type of data actually is being collected through the conduct of the trial. Often the data sets, there are not the same types of data sets that you would actually use or need for the regulatory approval. So that -- I think that's probably the most important bit because in the end, this is all a question on data and that data needs to be the type of data that can be fitted into the respective models that the payers are using. And that differs quite a bit from geography different jurisdictions and different nature of payers.

Thank you.

I'm showing no further questions at this time. I would now like to turn the conference back to you Christian.

Excellent. Well, thank you very much. Thanks for all of you for joining for the good discussion and we're looking forward to seeing you either at SITC or at ASH in a few weeks' time. Thank you very much and have a great day.

Ladies and gentlemen, this concludes today's conference. Thank you for your participation and have a wonderful day. You may all disconnect.