Cellectis S.A. (CLLS) Q4 2019 Earnings Report, Transcript and Summary

Good morning, everyone, and welcome to the Cellectis' Fourth Quarter and Full Year 2019 Results Conference Call. [Operator Instructions]. I'd now like to introduce the first speaker, Simon Harnest, Cellectis' Vice President of Corporate Strategy and Finance. You may begin, sir.

Thank you, and welcome, everyone, to Cellectis' Fiscal Year and Fourth Quarter 2019 Corporate Update and Financial Results Conference Call. Joining me on the call today with prepared remarks are Dr. André Choulika, our Chairman and Chief Executive Officer; Eric Dutang, our Chief Financial Officer; and Dr. Francisco Esteva, Vice President, Clinical Development. Yesterday evening, Cellectis filed its annual report and issued a press release reporting our financial results for the fourth quarter and year ended December 31, 2019. The reports and press release are available on our website at cellectis.com. As a reminder, we will make forward-looking statements regarding financial outlook in addition to regulatory and product development plans. These statements are subject to risks and uncertainties that may cause actual results to differ from those forecasted. A description of these risks can be found in our most recent Form 20-F on file with the SEC. I would now like to turn the call over to André. André Choulika: Thank you, Simon. Good morning, and thank you, everyone, for joining us today. 2019 has been a year of strong execution for Cellectis. Today, I am proud to say that we are on track with the clinical development of the 3 wholly controlled allogeneic CAR-T cell product candidates as well as 3 partnered product candidates, UCARTCS1 in relapsed/refractory multiple myeloma, UCART22 in relapsed/refractory B-ALL and UCART123 in relapsed/refractory AML, our own wholly controlled product candidates and concurrently in Phase I dose escalation trials called, respectively, MELANI-01, BALLI-01 and AMELI-01. These studies are designed to assess the safety and optimal dose for each product candidate in 3 to 4 different dose cohorts. And when available, we are planning to share the interim data set on these programs in the second half of this year. An additional near-term value driver for Cellectis is the two licensing agreement we have in place with our allogeneic CAR-T cell platform with two world-class partners. We announced yesterday with Servier, the execution of the amendment, confirming the terms of the term sheet signed on February 18, 2020. Under this amendment, Cellectis grants to Servier an expanded exclusive worldwide license to develop and commercialize an allogeneic CAR-T cell product targeting CD19, including ALLO-501 and ALLO-501A product candidates. We're excited about the increased focus by our partner, Servier, and it's U.S. sublicensee, Allogene Therapeutics, on the pioneering UCART19 program in ALL and the expansion into DLBCL patients. As part of the amendment, Cellectis is entitled to receive an additional $27 million upfront payment, up to $410 million in pre-commercial and commercial milestones as well as a flat, low double-digit royalty on sales of the licensed product. While providing Cellectis an attractive economic upside to the product candidates targeting CD19, this amendment significantly enriches our proprietary portfolio of targets with five additional differentiated CAR-T cell targets in both solid and liquid tumors. Our proprietary gene editing platform provides us with a vast amount of new opportunities in the ever-expanding cell therapy arena. We have recently made a selective decision to partner with Iovance Biotherapeutics to use our TALEN technology to develop gene-edited tumor infiltrating lymphocytes for cancer therapeutics, an area where we see tremendous promises and opportunity. We're looking forward to sharing more of this exciting new development program in the future. Hand-in-hand with our increased clinical development, we are seeing strong expansion in the manufacturing effort in the cell therapy space. Here, there are two main trends: First, know-how is the key proprietary asset in this industry. And second, we have seen a strong trend for consolidation. As a result, Cellectis has taken the forward-looking decision to fully integrate its manufacturing expertise as a key strategic asset for the company. In March 2019, we announced our lease agreement for an 82,000 square foot commercial scale manufacturing facility in Raleigh, North Carolina, this new site being designed to develop GMP manufacturing for clinical supplies and commercial manufacturing upon regulatory approval. We are on track with the build-out process, and the facility is planned to be operational next year. In addition to Raleigh, we have built a 14,000 square foot manufacturing facility in Paris, which is designed to produce critical raw and starting material supplies for the ongoing CAR-T clinical studies and our future commercial products. This facility is designed to help us save a significant amount of time and money in the manufacturing process currently done at our CMO partners, MolMed and Lonza. We believe this strategy will enable a smooth transition to target full manufacturing independence. With that, I would like to hand the call over to Dr. Francisco Esteva, Vice President of Clinical Development, who will give you a quick overview of our wholly controlled clinical CAR-T cell program. As a brief introduction, Francisco is a medical oncologist with an outstanding track record of peer-reviewed publications with more than 20 years of experience in drug development. Francisco spent 16 years at MD Anderson Cancer Center and 6 years at NYU, leading clinical research programs in breast cancer and solid tumors. Francisco joined Cellectis in October 2019 as the Head of Clinical Development. We are honored to have Francisco as part of our team, and I'm excited about the positive impact he will have on our clinical development. Francisco, please go ahead.

Thank you very much, André. As André mentioned, our first 3 proprietary allogeneic CAR-T cell development programs, UCARTCS1, UCART22 and UCART123 are all in the first dose cohort. Before speaking about our clinical programs in more detail, I'd like to explain our thought process behind choosing to pursue CS1, CD22 and CD123 as targets. We strongly believe that the potential addressable targets for allogeneic CAR-T goes far beyond the currently known clinical targets, CD19 and BCMA. In an independent study published in Nature Biotechnology in January 2020, Drs. Monica Guzman and Chris Mason and colleagues evaluated the therapeutic landscape for cells engineered with chimeric antigen receptors. They created an interactive CAR clinical trial database, spanning 64 targets deployed in T-cells, natural killer cells or mixtures, from over 500 clinical trials in 20 countries encompassing more than 20,000 patients. In this study, the targets CD22, CD123 as well as CS1, which we chose at Cellectis as our first proprietary clinical targets are standing out as having the highest therapeutic potential among all the targets evaluated, especially CD22, which may have a higher potential as a target compared to CD19 in B-cell malignancies. As we progress through the various dose cohorts, we are closely monitoring the on-target on-tumor as well as off-tumor CAR-T cell interaction. Starting with UCARTCS1, following the IND approval in April 2019, we dosed our first patient in the MELANI-01 Phase I dose escalation clinical trial in October 2019. The patients we are enrolling present with very advanced disease or patients present with a relapsed/refractory multiple myeloma and have progressed on multiple lines of prior treatment, leaving these patients with little to no alternative options. Some patients had a prior BCMA-targeted CAR-T therapy, which either fail them or led the patients to relapse. Our experience so far, so is that with our therapeutic approach, we could potentially help a lot of patients with such a high unmet need. We are starting with a dose of 1 million cells per kilogram. The second dose cohort would explore 3 million cells per kilogram and the third dose would explore 9 million cells per kilogram. For UCART22, we dosed our first ALL patient in the BALLI-01 clinical trial in December 2019. The BALLI-01 study is a Phase I dose escalation trial, evaluating UCART22 for the treatment of CD22-positive relapsed or refractory B-cell acute lymphoblastic leukemia, or B-ALL. Enrollment in the Phase I dose escalation clinical studies ongoing as planned, and we are currently in the first dose cohort. Again, the patients we're enrolling in this trial are very sick. They present with progressive disease, who have failed all other available treatment options, including prior CD19-directed CAR-T cell therapies. Our enrollment criteria select for patients with very high expression of CD22, and these patients can be CD19 negative. If successful, UCART22 could be an extremely valuable treatment option for relapsed, refractory ALL patients, as we do see that there is a relatively high relapse rate after one year of treatment with other therapies undergoing clinical trials in that indication. These patients are not considered eligible for bone marrow transplantation. Our goal is to find a safe and effective dose of UCART22 that can control the disease and ideally make patients eligible for transplant. We are also planning on filing a protocol amendment with the FDA to evaluate the addition of alemtuzumab to the lymphodepletion regimen. Cellectis invented the CD52 knockout concept that is already incorporated in the current UCART123, UCART19 and UCART22 constructs to make them compatible with alemtuzumab treatment. We have the intention to evaluate the safety and efficacy of alemtuzumab-based lymphodepletion regimen in a separate cohort of patients to guide the future development of UCART22 therapy in CD22-positive B-ALL. This comparative evaluation might not start before the completion of the second cohort of the current BALLI-01 study. We are currently dosing patients at a dose of 100,000 cells per kilogram, the second dose cohort would explore 1 million cells per kilogram. Now coming to UCART123. In 2019, we introduced a completely reworked UCART123 construct with an optimized manufacturing process. This new product was approved for testing by the FDA as part of a new IND filing. In January 2020, we announced the first patient dosing in AMELI-01, the Phase I dose escalation clinical trial, evaluating our new UCART123 product candidates in relapsed/refractory acute myeloid leukemia. This new AMELI-01 study replaces the first U.S. clinical trial assessing the first version of UCART123 product candidates. Here too, patient enrollment in this Phase I dose escalation clinical trial is ongoing. We are currently in the first dose cohort of 250,000 cells per kilogram. I would like to highlight the significant unmet medical need for this patient population. Relapsed/refractory AML patients have a very limited life expectancy and currently, there is no treatment -- no effective treatment option for these patients. These patients do not currently qualify for a bridge to transplant approach due to the refractory nature of their disease. If successful, UCART123 could provide an opportunity for this bridge to transplant, which still is the goal for any cell therapy in relapsed and refractory AML. All three Phase I dose escalation studies for our proprietary products are designed to find a safe and optimal therapeutic cell dose. Our primary endpoint is to identify the recommended Phase II dose. With our secondary endpoints, we are exploring the products' expansion, persistence and antitumor activity at different dose levels. We're excited about our clinical progress, and we plan to share data at upcoming scientific companies. With that, I'd like to hand the call over to Eric Dutang, for an overview of the company's financials.

Thank you, Francisco. I will provide a brief summary of the consolidated financials of Cellectis. As a reminder, Cellectis is a 69% shareholder of Calyxt, our publicly traded subsidiary. You can get the breakdown of the consolidated financials prepared under IFRS between Cellectis and Calyxt in the appendices of fourth quarter and full year 2019 financial results press release. I would like to focus on the Cellectis' financial performance since the Calyxt management will present their Q4 2019 earnings right after this call. First of all, I would like to reiterate that we focus our cash spending at Cellectis on the following objectives. Firstly, developing our deep pipeline of product candidates, including the manufacturing and clinical trials of UCART123, UCART22 and UCARTCS1. Secondly, building our state-of-the-art manufacturing capabilities in Paris and in Raleigh. And finally, strengthening our manufacturing and clinical departments, including hiring talented personnel. As of December 31, 2019, Cellectis, without Calyxt, had $304 million in cash, cash equivalents, current financial assets and restricted cash. This cash position will be sufficient to fund Cellectis' stand-alone operations into 2022. The net cash balance which corresponds to net cash flows used by operating activities, capital expenditure and [indiscernible] payments was $51 million for the year 2019, of which $2 million net inflows in Q4 2019, which was notably due to the receipt of a $5 million milestone payment from Allogene and $15 million of R&D tax credits. Revenues and other income were $16 million for the year 2019, of which $3 million relate to Q4 2019. R&D expenses increased by $12 million to $18 million for the year 2019 compared to 2018. And SG&A expenses decreased by $9 million to $17 million for the same period. Operating losses were $83 million for the year 2019, of which $27 million is attributable to Q4 2019. Finally, financial gains were $8 million in the year 2019. To conclude, I would like to say a few words about the consolidated financials. The consolidated cash, cash equivalents, current financial assets and restricted cash position was $364 million as of December 31, 2019. The consolidated net loss attributable to shareholders of Cellectis was $102 million, or $2.41 per share, for the year 2019. This represents an increase of $23 million over 2018, which was mostly explained by the increase in consolidated operating losses. The consolidated adjusted net loss attributable to shareholders of Cellectis, which excludes noncash stock-based compensation expenses, was $79 million, or $1.86 per share, for the year 2019. I will now turn the presentation back over to André for closing remarks. André? André Choulika: Thank you, Eric. 2020 is a pivotal year for Cellectis, as we progress through the clinical development of our wholly controlled or partnered product candidates. We believe we will start to show proof-of-concept on a series of target for our allogenic CAR-T cell product platform. Hand-in-hand with our clinical advancements, we are on track with the construction of our in-house manufacturing site, which are designed to deliver full independence and internal know-how on the forefront of science in gene editing and cell therapy. In addition to our lead clinical programs, we have made significant advances on the R&D side to further establish Cellectis' position as a leader in gene-edited cell therapy field. I would like to highlight 2 published major journal articles by our team, which continues to show our pioneering position in the field of allogeneic CAR-T cell engineering based on our proprietary TALEN gene-editing platform. In addition, we continue to be granted new patents and uphold current issued patent that had been challenged in the gene-editing space, including those using our proprietary gene-editing technology, TALEN, as well as CRISPR-related technologies. These patents further strengthen our leadership and competitive advantage by an investment into the future of these technologies. We are thrilled to see our programs come to fruition as every day gives us new hopes for patient in clinical need. With this, I would like to open the call for any questions you may have. Operator, please go ahead.

[Operator Instructions]. Our first question comes from the line of Gena Wang with Barclays.

And congrats on the -- all the program progress. So I have two questions regarding the UCART22, alemtuzumab cohort. So the first question is more the biology question for André. So just wondering what is your view on the immune system? How much residue immune system we need to maintain in order to have a good balance of, say, toxicity and the persistence of a foreign CAR-T cells? And the second question is regarding the alemtuzumab cohort. Why this additional cohort? And why now? Just wondering do you see Allogene's data? Will you use the lower dose? André Choulika: Well, thank you very much, Gena, for these excellent questions, by the way. So to answer your first question, it's extremely difficult to know exactly what is the ideal window for a better duration for a CAR-T treatment today. It's very interesting because there are like a series of different way to tackle this. And within 2020, we'll have plenty of data that will be coming, either from us or a partner, Servier, or Allogene, but also from other companies, such as Precision Biosciences or others, using, for example, an approach where [indiscernible] beta-2 microglobulin, et cetera. So there are like series of way to do this, and all these data are going to flow in. What we can look at today is essentially the data that has been published, either by autologous CAR-Ts and that shows that there is like clearly no correlation between the duration of the therapy and the persistence of the CAR itself. So there is no clear correlation between the two. What is the best outcome, most of the time, is the deepness and of the activity of the CAR itself in a short period of time or more like the killing potential and the expansion potential of the CAR that is injected, that would be directly correlated to the duration of the therapy itself more than the persistence of the CAR for a long period of time. Nevertheless, also, it's something that we'd like to investigate. We've seen the data of -- that has been obtained with UCART19 that we started, but has been pursued by Servier and Allogene currently, that are using alemtuzumab, was very interesting data, and we think that this approach that we created is very interesting. We decided to have a very deep comparative study in there and our plan during this year is trying to tune up our way to administer this type of therapy to the patient in the best way in order to confirm the way to administer this during the expansion phase that would happen next year. And of course, the window of lymphodepletion, we believe, should be around like over 20 days and should be around a month. But if you want to implement, for example, a re-dosing strategy, it could be interesting eventually to expand this maybe to a couple of weeks more. And we're excited about that. On the other hand, maintaining also the lymphodepletion during this period could be done essentially by, of course, like classical cyclo/fluda, but also potentially by administering other type of drugs or maybe also modifying the CAR itself. And answering totally, this question is not totally very easy. It depends on the targets. So 22, for example, or 19, just hit B cells, 123, go after myeloid progenitors. So could expand the time of the window larger than we expected because the myeloid progenitors won't be able to rebuild, for example, an immune response. CS1 is present on T-cells, B-cells, NK cells, et cetera. So they might deepen the lymphodepletion while expanding. So all these questions we need to answer, and we're still in the first like cohort for 22, and it's currently like we would like to move forward with this. And as we would like to move forward with this, we decided to file this amendment right now, but it should take place probably second half this year in order to be able to make a comparative study that could give us more insights on the way to precondition the patient, the best way to do it on the safety and efficacy side. So there is a leveling behind this that should be done. And as we have all the potential to do this because we, like CD22 is -- like UCART22 is CD52-enabled, so it's like resistant to alemtuzumab. This study -- parallel study is between like a cohort with or cohort without would give us very, very insightful data and would allow us to move forward with probably the best dosing strategy for the expansion.

Our next question comes from the line of Michael Schmidt with Guggenheim Securities.

This is Kelsey on for Michael. First, I guess, could you just provide us the implications of the patent that was upheld by the European Patent Office and maybe next steps there? And then also -- I guess, also following up on the lymphodepletion. Is there a potential to evaluate just alemtuzumab and removing Cy/Flu altogether? And would you maybe consider filing a similar protocol amendment for UCART123? Or is the plan still to maybe not do that? André Choulika: Well, to answer your second part of your question. No, no, no, they were like we're not ruling out probably -- we're -- like I have to start with something, we're starting with 22. We'll -- certainly, we'll try the same with 123. It's very -- it makes sense to make parallel studies because the two -- every CAR and every target behaves differently, so it has to be assessed. So for now, we're starting with 22, we'll do it with 123. It's a systematic study of the lymphodepletion condition. We'll move probably other parameters around that, as I said, potentially like the way to administer the doses, et cetera. We are talking with our partners, also in the way they're doing it, and I think the way to precondition the patient is probably a good guarantee to success. And as we have here, an array of different parameter we can play with, it gives us a lot of opportunities. And the better we make the study now, the better the expansion will be and the cleaner the shot to BLA will be at the end. To answer the first question you have on the patent that was upheld in Europe. It's essentially the use of CRISPR-related technologies, CRISPR-Cas related technologies and the same family in T-cells to edit genes, in general. And that's what we have, like we have a patent in the U.S. that has been granted, covering this technology and a patent in Europe. In Europe, you have a procedure that is called -- that had to be take place before the end of a period in Europe, which is 1-year period, you file an opposition. And opposition was filed against this patent, and the patent was upheld, unchanged. And we're happy about this outcome because it shows that we have been very early inventors on the use of CRISPRs in T-cells. Very early stage, we started using CRISPR in T-cells because we're gene-editing geeks. So whatever new technology comes, we're usually trying to use it, tweak it, et cetera. And at the time we used it, that was early 2013, it was -- there was no real idea of using CRISPR in T-cells and we've been very early stage in doing this. And we got the same patent in the U.S. and we continue like developing the technology and filing patents with it. And you will see that in the future, Cellectis will strengthen its position in there. And nevertheless, we compare all the technologies. So far, we're using TALEN for precision, efficiency and safety reasons. But CRISPR is a very attractive technology, for example, for research, we believe, and -- but has some limitations compared to TALEN, for example, concerning homologous recombination, time of efficiency, et cetera. And the things that we're working on still today.

Our next question is from Amanda Murphy with BTIG.

I actually had a question on CS1, and I may be reading into this a little bit too much. So forgive me if I am. But I thought last quarter, you said that you would move pretty quickly past cohort 1 if you weren't seeing efficacy. And I think there's -- I don't know the number if there's 28 days between patient one and 2, but since you're still in cohort 1, am I reading too much into that there based on the comments that if you didn't see efficacy, you would move quickly into cohort 2? André Choulika: Francisco do you -- would you like to answer this question, please?

Yes, sure. Yes, we are just finishing the first cohort, and then we'll move to the second one, but we need to wait 28 days to make sure there are no dose-limiting toxicities. So that's -- the cohort is going on as planned.

Okay. The other question I had was around the ALLO-501 milestone payments. I know that they get a high level increase as a part of the renewal or of the agreement. Is it still $350 million in terms of preclinical? And I guess, you might have said this, but how much have you received thus far? I'm curious how we should expect that to trend [indiscernible] Allogene has talked about having a data readout coming up here in the middle of the year? André Choulika: Well, we have not disclosed exactly the amount we received so far, but it's not quite easy to make the calculation because it was disclosed as most of our filing, but we have not given the full number. It goes back to 2014, it was a big milestone that we're paying in 2015, end of 2015, and a series of milestones that were paid during 2017, 2018 and in 2019. We strongly believe that this product is set to go for BLA in series of different indications. As we said, B-ALL for pediatric and adult, which was conducted through the PALL and CALM studies by Servier, with the DLBCL study that is conducted with -- by Allogene on ALLO-501 and ALLO-501A, and we think that this product can also go in other type of indications also. So there is an array of options that are covered by UCART19. It's the first-in-class, and I believe, also the best-in-class product in this category as an allogeneic CAR-targeting CD19 and the partners we have are pushing this on very -- in a very strong way. So we are very confident that this product will hit probably the market at the end. And that it would provide us with a series of milestones paid to the company. It's definitely worth everyone's willing on the partnership side that we have. So there is $410 million of milestones that are supposed to be paid on UCART19, like on the CD19 target, in general.

And also, we'll receive a $27 million as upfront in March 2020.

Yes. And also the lower -- I'm sorry, the slightly higher royalty rate is somewhere around 10%, and I guess, that's part of that. André Choulika: Yes, and plus it's flat now, which is good.

Our next question is from Jim Birchenough with Wells Fargo.

Congratulations on all the progress. Couple of questions. So first, just wondering if Francisco can comment on implications of AbbVie's failure with their CS1 ADC, whether you think that has any implications for the target itself? Do you think that was drug specific? And how do you think that affects your approach with UCARTCS1? And then the second question is just on manufacturing. And with the planned expansion to the GMP facility, will you be supplying to partners? Is that covered in the existing agreement? Or is there an opportunity to amend to provide more supply to partners with the expanded capabilities?

This is Francisco. So regarding the ADC for CS1, obviously, it is a different approach. The CAR-T is much more -- supposed to be much more potentive in a way once it finds the target, but obviously it's something we're watching very carefully, and we'll -- it's hard to tell. It's hard to compare the ADC approach to the UCART. Obviously, we'd like to see better good responses with ADC type of therapy in terms of validating the target. But we still believe CS1 is an important target in multiple myeloma, especially after patients have received and progressed through BCMA therapy, is considered one of the best targets regardless of this clinical trial you just mentioned. André Choulika: Well, I say something, Francisco on the same question from Jim. CS1 works with the monoclonal antibody, but have difficulty with the bi-spec. And let me tell you our analysis for this. For example, UCARTCS1, we have to add it out from the T-cell, CS1, unless you would induce fratricide killing because CS1 is expressed on the surface of T-cells. So the problem is that the bi-spec, if it's a CD3 on one side and CS1 on the other, how would you bind it to T-cell and how the bi-spec would know exactly what's the T-cell versus what's the cancer cell. And so it makes a problem and it wipes out a lot of the CD8s, which is not great. You don't obtain this problem when you're using monoclonal antibody because the monoclonal antibody just induced the killing by ADCC. In the case of a bi-spec, it's problematic. That's why we're very surprised that AbbVie has pushed a bi-spec there just because like the bi-spec don't know what to bind on the T-cell, would it be CS1 or CD3.

That's very helpful, André. And then just on the manufacturing question, and whether you can leverage the expanded capabilities with your partners? André Choulika: Well, it's not our plan currently. The plan is more to leverage our manufacturing capacity on our side and Allogene. They have like 125,000 square foot building in California, where they're building a manufacturing plant also. And we intend to build -- there are so many things to be built out from this plan because here we're talking about our 3 CAR-Ts. We're looking for a proof-of-concept. But standing in line, we have like a full pipeline of things that are in there. So if, at the end, we have like a positive outcome in one or several of these trials, then it would be a kicker to a series of different type of product that we would like to develop, and I'm wondering if our capacity would be enough only to build our pipeline. So no, for now, we're only going to use this in-house. So no Servier.

Our next question is from Salveen Richter with Goldman Sachs.

This is Andrea on for Salveen. I'm just wondering if you could provide a little bit more color on your announced partnership with Iovance and, particularly, some color on the structure of the agreement and the specific milestones that you anticipate receiving? André Choulika: Simon, you want -- do you want to take this one?

Yes, absolutely. So we're very excited about the Iovance partnership. We are going to share much more detail about this as we're getting closer to an IND. It's been a partnership that has been going back actually over a year already. We've been working as part of a research agreement that then transformed into a full partnership. And when the time is right, we will share more about that as well as economics.

So our next question comes from Christopher Marai with Nomura.

I was wondering if you could further elaborate on your ongoing dose-finding studies for each of the 3 products? It seems reasonable to me that each one of these products is going to be active. And some of that activity might be tied to some of its toxicity. So I'd be curious to understand quantitatively, how you are going to determine that safety activity balance for each of these compounds? And then maybe how you look at potential for re-dosing? And I'll jump back in the queue. André Choulika: I'm going to take a bit of the question, then I'll pass it to you, Francisco. So it's a very -- Chris, by the way, it's a tough question because there is a lot of question mark on like the doses. And even though when you talk with a lot of specialists, such as -- like Steve Grupp [ph] or others, it's -- most of the time, as -- it's a product that is expandable and size up to the size of the tumor burden, people wonder if there is a real dose. You have a floor, but you might not have a DLT at a time, like it's difficult to know. So either you have an embedded toxicity with the product that you inject, that is linked to the product itself. For example, 123 hitting the myeloid progenitor, CS1 hitting the T-cells, then NK cells versus lymphopenia for 19 or 22, et cetera. That is like the embedded toxicity that is linked to it, or something else, I don't know, like an on-target, but of tumor side effect that you would get. But the more you increase the dose, you might not increase the effect of it because you just see that the patient with enough cells to induce a response and you just have to know exactly what's enough cells to induce the full response. That's why we're doing with dose escalation, then you might increase the dose to a certain point, but always obtain exactly the same result at the end. And the discussion we're having with our partners, the data we've seen so far, gold in the sense of having a floor or threshold that you hit where it's becoming effective. And under this, it's not working really. And there is a differential between also autologous and allogeneic CAR-T than the way they're used, actually, due to the freshness of the cell that you're using. And yes, we're going to use actually the dose escalation and the way you were doing the older preconditioning to try to determine the best toxicity versus efficiency in terms of tumor response and ID potentially behind this, is coming also to the way, for example, this amendment that we're trying to file now, is to try to expand maybe the window and like tune up with the window we're obtaining in order to develop a strategy with the right dose to do potentially a different type of regimen given to the patient with potentially 123, I don't know X doses, given or maybe only 1 dose. But at the end, the way to dose the patient is something that will be determinant of the success of the drug at the end. And each target will be a different story. It's not going to be easy to set up the dose directly. Maybe at the end, we'll come up and say, okay, I don't know, it's like 80 million cells per kg for an adult and you come with a flat dose period per adult and flat dose period for like a pediatric, but so far we don't know that. Maybe, Francisco you would like to add on this?

Well, yes, thank you. André, it's a very important question, obviously, is that the main endpoint of these Phase I trials. And these, all of them being first-in-human trials, safety is our main concern. So we have to make sure they are all safe. So we are starting at a very low dose, and we may -- we are not -- may not see optimal responses until we go higher on the dose. But it is very important to ensure it's safe. Some of the toxicities will be related to the product itself, like CRS, when it works as expected. So that should not be an issue. But finding a dose that is safe, effective and where we see CAR expansion, for example, maybe what we're -- is what we are looking for, as supposed to a classic MTD, maximum tolerated dose that which we may or may not see, but it will be or it may be different from -- for each product and each target. So we are taking a very cautious approach and trying to move as quickly as possible within the dose escalation, optimize the lymphodepletion and then find the dose, but still too early to tell.

Okay. So it sounds to me that you're really looking to understand performance of the T-cells in these dose escalation studies. So how many patients do you think you need to get comfortable with their performance in each of these indications just to sort of declare your Phase II dose? And then how should we think about Phase II trials? Are they going to be completely separate? Or will you run expansion trials from the -- based on these phase Is and then I would assume those expansions or these Phase IIs are also designed to be registration worthy? André Choulika: So the number of patients, it may range from 2 to 4 patients per cohort, it's a patient design approach called mTPI. So we -- then it depends on where we see the optimal dose in the -- for example, in the UCART123, we are looking at potentially four dose levels. In the other products, the UCART22 and CS1, we are looking at the three dose levels. And so the number of patients, again, may range from 2 to 4. Usually in Phase I trials, we usually enroll three patients per dose cohort and then expand to another 3. This is a little bit of a modified statistical design. And once we find the right dose, then we'll move into expansion. And at the same time, start planning the registration trials because we want to get more safety data on the expansion to make sure what we are doing in the pivotal studies is safe, and that is a part of the Phase I development. But we will start very quickly thinking about registration trials as well as we do the expansion trials.

Okay. And then just in terms of thinking about discussing your registration trial plans with the Street, what kind of timing might we anticipate for those types of updates? It's my last question. André Choulika: So the registration approach has not been finalized yet. We need to get more data from the ongoing study, which, as you know, is still relatively early. But based on what we've seen from recent FDA approvals, there are basically 2 approaches, either go with a single-arm study if we -- for a particular indication, and 1 of these 3 products, we see very active responses in selected populations, for example, 123-positive refractory AML or CD22-positive ALL and so on. If we see exceptional responses, we could get approval based on a single-arm study. Otherwise, we'll have to proceed with a controlled study, and those will be discussions with the agency, both the FDA and the EMA, once we have additional data, but both options are on the table at the moment.

Our next question comes from the line of Wangzhi Li with Ladenburg.

Maybe two quick questions. One is for alemtuzumab. Could you provide us with the color, are you going to test different doses of alemtuzumab or just a fixed dose, maybe based on partners' experience? André Choulika: Francisco, maybe this one is for you also.

Yes. Our first approach, we will just use a single dose. This is not a study to look at alemtuzumab. It's just a slight modification of our current lymphodepletion regimen. So if we need to do additional studies in the future, that's a separate approach. But right now, it's going to be just a single dose. But it is an important question because the optimal dose of alemtuzumab is not well defined, or there is a possibility we might use in the future alemtuzumab without chemotherapy. All these questions are on the table, but for the time being, we can just make -- we want to make very slight modifications to our ongoing protocols.

Got it. But at this moment, do you expect maybe similar doses across different indications? Or you think each indication needs different optimal alemtuzumab dose, if is needed?

Yes, we don't know. It may be different for each indication, and we are talking about B-ALL and AML because we are not -- we will not use alemtuzumab in the CS1 product. But for B-ALL and AML, the dose and the schedule of alemtuzumab may be slightly different.

Okay, got it. And then just for the enrollment speed, should we expect -- because you have the 28-day waiting period, should we expect about every patient two months' type of enrollment speed?

Yes. Well, we have to wait 28 patients after the first patient, and then in... André Choulika: 28 days.

28 days, yes. 28 days after the first patient on each cohort, and then we can recruit the second and third patients a little bit faster, but -- and then we have to evaluate all the data, the safety, exploratory data CAR expansion process, et cetera. So yes, it takes about three months per cohort usually, if there are no delays.

I see. Got it. And I think maybe I missed, I think you mentioned maybe a potential data in the second half this year? André Choulika: We hope we'll have data at the end of the year to present, yes.

Okay, great. Last question, maybe a clarification of the dose? Is that a CAR positive T-cells or total T-cells? André Choulika: The classification of the dose is always cells per kilogram. Is that your question, Wangzhi?

It's a CAR-positive cell numbers? Or is it total T-cell numbers? For example, if it's 1 million per kilogram, is 1 million CAR or total T-cells? André Choulika: It's CAR positive, it's CAR cells, yes.

Our next question comes from Biren Amin with Jefferies.

Just on the Servier agreement. Can you talk about the 5 targets that were returned to you and disclosed what your plans are for those targets? André Choulika: We haven't disclosed it yet, Biren, I would love to do this. We will definitely develop some of them, yes. We're super excited because there are great targets in the portfolio, and it's like this is the strong intention to put these targets into the clinic, not this year, but in the coming years, definitely, yes. They are both liquid and solid tumors' targets, extremely relevant for some of them, all of them, by the way, but some of them, we think that have some interesting proof-of-concept that would push us to give like a very favorable environment to put the targets into the clinic. So we're extremely excited by the portfolio. But I will not say we would like to work a bit on them, and then we'll bring you -- we'll update you later on what kind of target we'd like to push. It's a very competitive space. And everyone is eyeing on each other on these targets.

And what was the rationale for Servier for returning those targets? André Choulika: Well, we've signed this agreement back in February 2014, so more than 6 years now. With a lot of things that were like in hand, and we consider that it's probably time to trying to push 1 of the 2 targets after 6 years into clinic, and it was part of the negotiation.

Got it. And then maybe if I could ask a couple of questions on UCART22. André, can you just talk about are these pediatric patients, are they adult patients with ALL, just the baseline characteristics? And also are you enrolling in terms of restricting for baseline blast count in these patients? So if you could just give us some more color on this trial design, that would be helpful. André Choulika: It's an adult trial, it's not a pediatric trial. But maybe on the way we recruit a patient, Francisco, as you're in the front line in there, I think it would be probably better if you add.

Yes. The study we are accruing at the moment is based on adults. We are -- we'll plan to do pediatric studies next year. And the protocol requires very high levels of CD22 expression, at least 90% expression on the blast. So we are looking for a very specific population, which is very common to -- for B-cells to be CD22 positive, even if they low -- even if they lose CD19, for example, patients who are treated with autologous CD19 CAR therapy may lose CD19 at the time of progression, but they -- those blasts retain CD22 expression. So that's the population we are looking for in adults on this Phase I trial at the moment.

Okay. And then, I guess, these are patients that likely would have failed CD19 CAR coming into the study?

We are not excluding them. They are allowed. It's something that probably will happen in the future. We are allowing any -- it's not one of the eligibility, but it would be an interesting population to the study. But yes -- so we are accepting them, but it's not one of our eligibility criteria. We're very open. André Choulika: This is what's great with -- this is like -- Biren, this is what's great with UCART22, actually. It's -- it brings the real unmet medical need. All the -- like -- as Francisco said, this is not the criteria in our trial, but every CD19 negative, relapse or for any CAR-T or bispecific study or treatment, whatever, has no alternative, like it's an unmet medical need. So you need to have a potential to come and bring a new product, 22 can address these patients at our CD 19 negative relapsers or the one that failed CD19 treatment. And that's what makes us very excited, and it's like a potential niche that gives a very exciting potential for this product itself.

And then maybe a question on the AML. Can you just talk about -- I think you said the starting dose is 250,000 cells per kilogram. How does that compare to the prior, I'll just call it, version 1.0 that you had in the clinic? And can you just compare and contrast the dose between that versus 2.0? And I guess, what's your sense in terms of lymphodepletion regimen because I think on the last trial, you had capped the dose of cyclophosphamide to 4 grams over 4 days. Is that still being required for this new AML study?

Yes, the... André Choulika: I'm going to answer the -- okay, go ahead, like Francisco. Sorry about that.

No, no, no. It's -- what... André Choulika: Sorry, I kept -- I cut you. So let me answer at the beginning of the question, I'll let you answer on the cyclophosphamide. So the doses that we used initially, we started at 6.25x10^5 cells per kg for the BPDCN and AML trial back in 2017, where we had a clear expansion, engraftment, and we had -- unfortunately, this BPDCN patient that died where we had to hold. We had to deescalate to 6.25x10^4 cells per kg with this initial -- the former product and escalated back to 2.5x10^5, where we are currently. So the amend -- like the new IND, the start dose is exactly the dose where we stopped with the former product, and we're starting with 2.5x10^5. The next dose is where we'll be resuming back to the initial start dose that we had in 2017. Once we'll finish this first safety dose cohort at 2.5x10^5. So you would expect probably 6.25x10^5 on the second cohort, then it's -- we have 2 other dose levels, which are, I think, 3x10^6 and 5x10 -- 5.5x10^6 on the two additional levels. So for now, I think that we're just continuing on what has been done with the former product at the lower doses, and then we'll resume exactly at 6.25x10^5 soon.

Yes. In terms of the lymphodepletion, it's basically the same as the previous version.

[Operator Instructions]. Our next question comes from the line of Raju Prasad with William Blair.

Just a clarification on the comparability study. Because it's after the second cohort, would you be able to use kind of that second cohort dose or is it a third cohort dose to start alemtuzumab? Or would you start kind of from the beginning and dose escalate with the alemtuzumab to do a comparability from the lowest dose kind of through? And then on the higher doses and kind of across the studies, can you just kind of give some thoughts on split dosing and whether or not you'll have to do split dosing kind of the higher dose cohorts?

So in terms of the -- I'm sorry. André Choulika: No, no. Go ahead.

The alemtuzumab, basically for each dose level, for example, if we start at dose level 2, we have to make sure that dose level 2, by itself, is safe before we have alemtuzumab, and then we can either use the same dose, like, let's say, if the dose level 2 safe, we move to dose level 2 plus alemtuzumab or to dose level 3? And just to clarify, we'll do this at the same time, so we're not wasting any time. And in terms of the super -- I mean, the higher doses, we are not planning to do split dosing at the moment, something we could do easily, but our thought at the moment is that if we can get a high dose initially with good lymphodepletion, that will be a big -- a big block to the tumor is probably better than otherwise. But that's what we're thinking right now, yes. And dose may be an important aspect in some of these targets, maybe different from target to target.

Yes. And then on the potential for re-dosing, can you just kind of describe the protocol for that from a lymphodepletion perspective? And if we see data by year-end, is there a potential to see a patient that's going to be dosed? André Choulika: We can possibly re-dose after discussion with the FDA and the investigators and ourselves, but on the 123, on the UCART123, but we are still -- at this moment, we are still very early in development. We may see something by the end of the year, but it's not a routine treatment for any of these protocols at the moment, but it's a possibility. But I do not anticipate a lot of re-dose in this year.

Our next question is from the line of Hartaj Singh with Oppenheimer & Company.

Just have a couple of questions. One is, André, you -- I assume, you got some learnings of the UCART123 manufacturing experience, the new R&D. I assume you've sort of been through some more batches with 22, and with CS1. Is that progressing as you expect, I guess, what I'm trying to ask in a roundabout manner is that are you seeing any manufacturing changes that you'd have to submit to the FDA with either 22 or CS1 that would be analogous to what happened to 123? And I just got a couple of follow-ups. André Choulika: It's a great question, Hartaj. It's a great question because we've hit -- we've been manufacturing CAR-Ts for 6 years, almost something like 6 years since 2014, GMP, I'm talking, not in the lab. And we've seen a lot of different cases around and build tremendous amount of know-how. In 2017 and in 2018, when we were starting hitting certain problems that we had and we had to fix, we've invested a tremendous amount of energy on the process development and the analytical development, and this has led in the second half of 2018 to a process that is extremely robust and the ability to produce CAR-Ts in a very consistent way from batch to batch, donor to donor and from CAR to CAR. And we consider today that we have a very stable process. And we're just working on the BLA version now. It's translating this clinical supply process to a commercial supply process, is something that we have in focus now, and that's where we're building our manufacturing plant. I think this process will be ready in time. The manufacturing plant in Paris is going to go live this year. And the one in Raleigh will go live next year. I think we will be ready for the BLA versions, versions with an S. And that is all the effort of the company that has been made in the past years and the pride we take out from the manufacturing excellence that we have reached today.

That's great, André. Thanks for that color and the robustness now of your manufacturing strategy. Another question I had is just is, can you just remind us, I know you had mentioned about 28 days between patients, I believe, for 123, is that the same for 22 and CS1? If you just remind us what is the time you've got to wait between patients and between cohorts for the other 2? André Choulika: So for 123, it's 28 days between patient one and patient two, 15 days between patient two and patient three. And then you have the end of cohort meeting at 28 days, after -- like the end of 28 days, you have to wait a bit up to time at the end of cohort meeting, and then you can start the second dose level. For CS1, and 22, it's always the same between patient one and patient two for CS1. Once you have the 28 days between patient one and patient two, and you can enroll two patients immediately once you finish this 28 days. So it's a bit faster than 123, don't have these 15 days, and 22 is the same.

Great. And then last question is just a financial question. If you can just give us some insight, I know that you had a big step-up in R&D expense in the fourth quarter of 2019 over the third quarter, which generally tends to happen, but it seems to be a little bit larger this year than previous years, how to think about R&D going forward into 2020? And then just OpEx, in general, any thoughts on OpEx burn for 2020 versus what happened in 2019? André Choulika: Okay. Well, thanks, Hartaj. I'm going to leave, maybe, Eric comment on this. One thing I would like to comment about this is if you notice that there is an increase in expenses, in R&D expenses, which we take pride out of, we've been also becoming more lean on the SG&A with a great effort to try to focus most of the value of the company on building assets to the company. So maybe Eric, you would like to comment?

Yes. And to complement what André just said is more than 80% of the total operating expenses of Cellectis are new for R&D expenses. So we have a pretty lean in our structure. We don't provide any guidance. As you could see, we had $80 million of R&D expenses. If you backed out the noncash stock-based compensation expense, it's $70 million. We expect that, that will increase up to between $100 million and $110 million in 2020, and we'll remain to lean on the SG&A expenses.

Our next question comes from Yigal Nochomovitz with Citi.

This is Samantha, on for Yigal. I'm wondering if you could just kind of help us frame the expectations for the data readout potentially at the end of the year? Should we expect -- what was the criteria that you'll be using to sort of decide if you have enough data to go forward with reporting? And is there a potential to see data from all three, CS1, 22 and 123, by the end of the year? André Choulika: Francisco?

Yes. We expect to be able to present data on all of them. And hopefully, it's really hard to tell how many cohorts we'll have, but we want to be transparent and show what we have by the end of the year. That's our current thinking.

Our next question is from the line of Bertrand Delsuc with Biotellytics.

I have one, actually, on the use of alemtuzumab that you want to introduce through an amendment in UCART22. I was a bit surprised that yesterday when I read that you had to file an amendment for that simply because -- based on the fact that on -- with UCART19 when only cyclo/fluda was used, there was no expansion seen based on data from 3 patients, or perhaps 4, if I remember, I can't remember. So I was wondering about the rationale of using only cyclo/fluda for UCART22 and if it was based on recommendation from the FDA, who wanted you to, as you say, to prove that the UCART22 was safe at the doses you wanted to process with before moving forward. So that is the first question. André Choulika: So to answer your question, actually, it's absolutely not related to any questions from the FDA to do this. It's like role of the company to try to file a comparative study, and it's a study, it's not to try to compare cohorts between with or without later in the year. When you compare, for example, the data that has been provided by companies such as Precision Biosciences, for example, they showed that there was expansion and even they had some complete remission with no preconditioning with alemtuzumab, with their CD19 CAR-T, which hopes that you don't need specifically alemtuzumab to have expansion and a tumor response in there. It's also when you look at the CD12 -- like UCART 123 first patient that we had, we had engraftment and expansion in the 2 first patients where we stopped the study back in 2017 without the use of alemtuzumab at this time. Now we are working on, this year on trying to find the right way to dose the patient, and to do this, we need to have a very comprehensive way to analyze the way to preconditioning -- to precondition the patient and using all the tools that are at our disposal currently, which is cyclo/fluda, alemtuzumab and maybe potentially other things that could be used, we would probably use them. And the way to admit the drug in a single dose, 2 doses that was the question, for the split dosing, it's something that could also be put on the table, and we have a full year to try to investigate all this up to the time we go for the confirmation of the right way to precondition the patient and to dose the patient during the expansion phase to move into the pivotal phase. So this is absolutely not a way to -- which we like to correct what is done currently, but trying to do a comprehensive study of the way to administer these type of drugs.

Okay. Got it. And just for confirmation, the new process -- manufacturing process you introduced for 123, is it also used for 22 or CS1 as of today? Or as you say, it will be used at a later date for the registration of the trial? André Choulika: Francisco, do you want to answer this question?

So the manufacturing will be the same.

This is Simon, just to jump in, the manufacturing we've done continuously evolves. So for 123, specifically, there was a new manufacturing campaign done after our third-party manufacturing organization wanted to remanufacture these batches. But just to give you a level overview of where we are with manufacturing, we have currently thousands of doses done for each of the products we have in the clinic. So we have no shortage of supply for these clinical trials leading up to registration. And our strategy with manufacturing is currently to bring everything in-house, which we are doing with Paris as a bridge and with Raleigh, North Carolina, as the full commercial scale manufacturing organization. And the completion of these in-house pieces go hand-in-hand with us going through the clinical trials, progressing towards potential commercialization of our product. So we would probably be ready with our full in-house commercial facilities, at least 1 to 2 years before the anticipated commercialization of our first product. So I think we're in really good shape. We know the products we currently have manufactured are of top quality, and they are satisfying all our highest quality standards we have internally.

Okay. Great. Just a last one. On 22, you say that the first dose is 100,000. And the -- the second dose is 1 million. So there is a factor 10. Usually, it's more -- it follows more Fibonacci sequence? Or let's say, could you have just a comment on the jump from -- we see from DL1 to DL2. And also, while it's more for the future, but as you have a flexible trial design for 22, last year there was also some publications about to sensitize CD22-positive ALL with bryostatin. Perhaps, it's not really mandatory as of today since you target population with high expression of CD22. But do you also foresee some other arms or other cohorts using other modalities within the frame of this trial? André Choulika: Well, for the doses, actually, we -- it's exactly the same dose escalation strategy as conducted for UCART19. It's same -- it's not different as it's like targeting exactly the same type of indications. That's why we're in the same type of trends. Most of the time, we can calculate over Fibonacci or sometimes also other parameters that enter in there. That we won't disclose here, but it's exactly the same way it has been calculated for 19 and 22 in there, and that's the way it has been done. And like maybe who wants to take the second part of the question, right away?

So some interesting point that we could combine the UCART22 with other agents to try to improve the engraftment and persistence and so on, but it's not something we are planning to do this year because we need to show that the product is safe and effective by itself, but it's something that we could explore in the future.

And just to add, maybe for the benefit of everyone still on the call. I just wanted to reiterate our decision to add alemtuzumab is a very thoughtful one for 22. And it will not be, in any way, extending the time lines for any of the products because it basically goes hand-in-hand with the dose escalation that we're currently conducting. It is for us to have a better overview of what is the best lymphodepletion regimen. We have a lot of experience with both approaches with alemtuzumab, without alemtuzumab. We already know how alemtuzumab performs at different doses, and we will dial in perfectly the right lymphodepletion regimen to make it an optimal outcome for each patient in each different indication that we're pursuing. But all in all, we're very excited that our programs are moving forward through the dose escalation. We're seeing very encouraging signs so far. And we're very excited to share our data by the end of this year. André Choulika: Exactly. And plus like we -- it's too early stage to speak about what has to be improved as we think that the product that we have designed are already well designed.

Thank you. Ladies and gentlemen, this concludes our question-and-answer session. So I'd like to pass the floor back over to Mr. Harnest for any additional concluding comments.

Yes. Again, thank you, everyone, and we're excited to have you guys all on board. And we're, again, happy to share any more updates. You can always call us directly. And we're looking forward to seeing you at different investment conferences and scientific conferences this year.

Thank you. Ladies and gentlemen, this does conclude today's teleconference. Again, we thank you for your participation, and you may disconnect your lines at this time.