Cellectis SA (CLLS) 2018 Q4 法說會逐字稿

Greetings, and welcome to the Cellectis Fourth Quarter and Full Year 2018 Earnings Results Conference Call. (Operator Instructions) As a reminder, this conference is being recorded.

It is now my pleasure to introduce your host, Simon Harnest, Vice President of Strategy and Finance for Cellectis. Please go ahead, sir.

Thanks, Kevin, and thank you, everyone. Welcome to Cellectis Fiscal Year and Fourth Quarter 2018 Financial Results Conference Call. Joining me on the call today with prepared remarks are André Choulika, our Chairman and Chief Executive Officer; Bill Monteith, our Senior Vice President of U.S. Manufacturing; and Eric Dutang, our Chief Financial Officer.

Yesterday evening, Cellectis issued a press release reporting our financial results for the fourth quarter and year ended December 31, 2018. This press release is available on our website at www.cellectis.com.

Just 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 20F on file with the SEC.

Now I would like to turn the call over to André. Please go ahead.

Thank you, Simon, and good morning, everyone. 2018 has been an interesting year for the CAR T-cell space from our perspective. More than ever, it highlights the strength of Cellectis' platform and business model based on TALEN gene-edited allogeneic CAR T-cell therapies. Autologous CAR-Ts have shown encouraging data in B-cell leukemia and multiple myeloma patients. Yet, the limit of the business models based on patient-derived cells are becoming clearer than ever. It will be interesting to follow the autologous sales ramp-up this year and see whether there is enough space for a new autologous CAR-T treatment on the market.

This past year also strongly demonstrated criticality of manufacturing robustness linked to quality, availability and competitive custom goods. This is becoming one of the major challenges of cell-based therapies.

On the other hand, the allogeneic CAR T-cell space is attracting another growing craft. Five years ago, the concept of allogeneic CAR T-cells was deemed to a scientific dream. Today, as all the predicted challenges materialize for autologous CAR T-cell approaches, the race for allogeneic CAR-T therapies become the path to success for all. Cellectis has invented and is pioneering this approach, and we have built an unmatched leadership in this field. Our universal off-the-shelf CAR product candidate, or UCARTs, are far advanced in clinical development.

Cellectis' leadership position has not changed since 2015 and shows on the one hand our ability to develop, manufacture and invent allogeneic CAR T-cell programs into the clinic. On the other hand, it shows the steep barrier of entry in this field. In short, we are seeing the return of our early investment. We have filed 1 IND per year, 3 years in a row. This is a metric that companies of our size are rarely -- have early achieved.

Following the first IND for UCART19 in ALL patient in 2016, we filed an IND for UCART123 in AML patient in 2017 and an IND for UCART22 in BLL patient in 2018. This demonstrates the performance of our technology, robustness of our product and the outstanding quality of our regulatory and clinical team.

Today, we are proud to say that we have achieved our goal with the successful manufacturing of series of TALEN-based gene-edited allogeneic CAR-T product candidates, UCART19, UCART123, UCART22, and each with the tiers of GMP batches with a parameter of high quality standard.

As we have already explained, in the cell therapy and gene-editing space, manufacturing is the key success factor. Over the course of the past 5 years, our cell manufacturing team has experienced a steep learning curve. The [proprietary] know-how of our team, based on hands-on experience, is what truly set us apart of the allogeneic CAR T-cell therapy space. Based on the proof-of-concept that we have already established in early clinical allogeneic CAR T-cell trials, we are ready to move on to the next step in term of manufacturing and that means investing in our self-owned production capacity.

Here, we're building on two pillars: on the one side, we already started building a starting material facility of our -- at our headquarters in France; on the other side, we recently announced a lease agreement for a 92,000 square feet commercial manufacturing facility in Raleigh, North Carolina. These 2 state-of-the-art plant will start operating in 2020 and in 2021, respectively.

Our U.S. facility will be the first plant in the world to manufacture commercial product that combine gene editing and cell therapy. Bill Monteith, our Senior Vice President of U.S. Manufacturing will share more about this later. But first, I would like to provide you with some update on our pipeline and our strategy.

Today, we are focusing on a series of hematological malignancies including acute myeloid leukemia, multiple myeloma, B-cell acute lymphoblastic leukemia and non-Hodgkin's lymphoma. Our strategy is to try to differentiate the tumor associated antigen targets in patient populations with serious unmet medical needs, such as CD123 or CLL-1 in ML, CD22 in ALL, CS1 or CD33 in multiple myeloma, which are based on clear scientific rationale of targeted expression, clinical proof of concept and enabled by our TALEN-based gene-editing proficiency.

Starting in with UCART123, our first wholly owned product candidate, we are currently in Phase I portion of the study, which is designed to assess the safety and tolerability of increasing doses level in AML patients. The study is conducted by Professor Gail Roboz at Weill Cornell Medical Center in New York, one of the world-leading physician centers in the space.

The patient we treat are relapse refractory AML patient with poor prognosis who have failed all other prior therapies. The goal for this Phase I dose escalation study is to find the optimal therapeutic dose for UCART123 to be used in Phase Ib portion of the study. We are currently at a dose level of 2.5x10 to the fifth cells per kilogram. We anticipate an interim data update by the end of 2019.

Moving on to UCART22. This is the third allogeneic CAR-T program that Cellectis has manufactured. UCART22 is positioned as a treatment approach for B-cell, ALL and non-Hodgkin lymphoma patient targeting CD22. With the scene at ASH 2018, there is a large percentage of ALL patient that relapsed following a CD19-directed CAR treatment. Most of these patients have lost the access for the CD19 antigen in their B-malignant cells. Often, this happens by the CAR transduction into the B malignant cell during the manufacturing.

The CD19 CAR expressed on the surface of the B malignant cell blocks the access of the CD19 target to the CAR T-cell. These patients have currently no other treatment option. With UCART22, we would like to offer another treatment option for these patients, utilizing the CD22 directed allogeneic CAR-T therapy with no risk of transducing B malignant cell. The IND for UCART22 was approved by the FDA in 2018 and subsequently, manufactured a series of GMP cell batches and received an IRB approval to initiate clinical trial. The research for UCART22 will be led by Dr. Nitin Jain, Assistant Professor; and Professor Hagop Kantarjian, Chairman of the Department of Leukemia at MD Anderson in Houston, Texas.

We anticipate dosing our first patient with UCART22 in second quarter 2019. This Phase I portion of the study is designed to set the safety and probability at increasing dose levels similar to the Phase I study of UCART 19. We will start dosing patients with 10 to the fifth cell [per kg] of UCART22, dose level 2 and dose level 3 are respectively 10 to the sixth and 5 10 to the sixth cells per kilogram.

During the course of 2019, Cellectis expect to have three proprietary programs in Phase I clinical development with UCARTCS1 as the first allogeneic CAR-T program targeting multiple myeloma entering the clinic in parallel to our UCART123 in AML and UCART22 in BLL.

This is a very exciting time for us at Cellectis as the hard work for our R&D, process development and manufacturing team has paid off and transformed Cellectis into a clinical stage and proved biopharmaceutical company.

To sum up on the clinical update. We will continue to pioneer allogeneic CAR T-cell treatment being the first company with multiple allogeneic CAR T-cell product candidate in clinical development. And as previously announced, we will anticipate an interim data update by year-end.

With that, I would like to hand the call to Bill. Bill?

Thank you, André, and good morning, everyone. I thought I would start by giving a quick overview of my background. I jointed Cellectis from Hitachi Chemical Advanced Therapeutic Solutions, which formerly was known as PCT. Prior to that, I was the Executive Vice President of Technical Operations at Dendreon. My responsibility here at Cellectis is going to be to lead the manufacturing and operations team in the U.S. Over my career, I have developed deep expertise in facility build-out in the field of cellular gene therapy.

Over the recent past, Cellectis has consistently invested in CAR-T process development and manufacturing in order to ensure the development of a robust and reproducible manufacturing process, along with the assurance of quality, high yields and low COGS. Cellectis realized early on the high sophistication that's required to manufacture such a class of products.

There's several dimensions you have to consider when manufacturing allogeneic CAR T-cells. The first is how best to create a drug substance, sourced from different healthy donors, then growing highly potent cells that undergo a series of state-of-the-art genome engineering steps, all the while meeting strict predefined GMP product specifications. It's very important to note that in the aseptic allogeneic CAR T-cells field, the FDA imposes the highest product performance and quality standards. Cellectis has consistently invested to meet those high standards.

As André mentioned earlier, as Cellectis continues to grow and develop into a comprehensive clinical-stage biopharmaceutical company, we're working to ensure the future supply of our various allogeneic CAR T-cell therapies. This is why we're now ready to move to the next step in terms of manufacturing. And that means strategically investing to build our own production capabilities.

To that end, we are literally working on 2 fronts: we are currently building our starting materials production facility in Paris, France; and then, on the other front, we're recently entered into a lease agreement for a state-of-the-art clinical and commercial manufacturing facility in Raleigh, North Carolina. Building state-of-the-art manufacturing capabilities is at the core of our strategy to deliver readily available cell therapy, faster, more reliably and at greater scale.

The first facility to go live will be our starting materials production facility that's based in Paris. We call it SMART. SMART stands for Starting Material Realization for CAR-T products. We target the start of operations around 2020. This will be a 14,000 square-foot facility and it will be designed to ensure our supply of critical raw and starting materials for our clinical studies. Using our TALEN proprietary technology, SMART is going to manufacture the viral vectors, which are essential to the clinical manufacture of our CAR-T products.

By internalizing this supply, we're really accomplishing 2 things: first, we'll significantly reduce our manufacturing cycle time; and then, secondly, we'll improve our flexibility during clinical development of our product candidates. It will also enable us to rapidly implement additional engineered designer cells. This is a capability we view as strategic for cell therapies.

The facility will also have the potential to supply a commercial starting material for our CAR T-cell products following potential FDA approval. The lease for SMART with design -- was signed in December 2018 and construction is starting as we speak, with go-live planned for 2020.

In parallel, in mid-2018, we kicked off our U.S. project, which we call IMPACT, and that stands for Innovative Manufacturing Plant for Allogeneic Cellular Therapies. We're very proud to announce that we just signed a lease for our facility. This will be the first facility in the world that will manufacture commercial UCART products that combine gene editing and cell therapy.

The facility will be around 82,000 square feet and will be located in Raleigh, North Carolina. This was chosen following a really detailed benchmarking of various sites in North Carolina and New Jersey.

IMPACT will be launched with a focus on the production of the clinical and commercial pipeline for Cellectis. And then later in the second phase, we'll develop additional GMP capacity for the production of critical raw materials that will complement the capacity of SMART. The construction of IMPACT will start shortly and our plans to produce the first clinical products within this facility in 2021.

The capabilities of these 2 integrated facilities will enhance Cellectis' overall mission which is to ensure availability of off-the-shelf cost-effective, quality cancer therapies for patients.

The CMO we are currently working with for the manufacturing of our clinical products will continue to be a strategic business partner, which will complement IMPACT and SMART to assure a robust supply chain for the manufacturing of Cellectis' allogeneic UCART therapies.

With that, I'd like to turn the call over to Eric Dutang, our CFO, for a discussion of our financial results. Eric, please go ahead.

Thank you, Bill. The press release we issued yesterday provides an overview of fourth quarter and year-end 2018 financial results. We provide additional detail on our annual report, which has been filed with the SEC. I will provide a brief summary of the consolidated financials of Cellectis and Calyxt, of which Cellectis is a 70% shareholder. Note that you can get the breakdown of the consolidated financials prepared under IFRS between Cellectis and Calyxt in the appendices of our fourth quarter 2018 financial results press release.

Following our follow-on offerings of 2018, we are on a strong financial position and are well-capitalized to further advance our clinical development programs and support the manufacturing build-out Bill mentioned earlier on the call.

As of December 31, 2018, Cellectis and Calyxt together had $452 million in consolidated cash, cash equivalents, current financial assets, of which $358 million are attributable to Cellectis, compared to $297 million as of December 31, 2017, of which $240 million were attributable to Cellectis. The net increase in cash of $155 million, primarily reflects $227 million in net cash proceeds provided by 2 separate follow-on offerings completed by Cellectis and Calyxt in 2018.

Net cash flows used by operating activities in 2018 were $68 million, of which $48 million was attributable to Cellectis. We believe that the consolidated cash, cash equivalents and current financial assets as of December 31, 2018, will be sufficient to fund our operations through 2021.

Our consolidated revenues and other income were $21 million for the year ended December 31, 2018, compared to $34 million for the year ended December 31, 2017. 98% of the consolidated revenues and other income was attributed to Cellectis in 2018. This decrease between 2018 and 2017 was mainly explained by a decrease in recognition of upfront payments already received and R&D cost reimbursements in relation to the therapeutic collaborations.

Our consolidated R&D expenses were $77 million in 2018 compared to $79 million in 2017. 89% of consolidated R&D expenses were attributed to Cellectis in 2018. The $2 million decrease between 2018 and 2017 was primarily attributed to the reduction of noncash stock-based compensation expenses by $6 million and social charges on stock option grants by $1 million. This decrease was partially offset by higher employee expenses by $4 million, notably due to more R&D headcount, and higher purchases and external and other expenses by $1 million.

Our consolidated SG&A expenses were $47 million in 2018 compared to $45 million in 2017. 55% of consolidated SG&A expenses was attributed to Cellectis in 2018. The $2 million increase between 2018 and 2017 was primarily driven by the increase in wages and purchases at Calyxt of $8 million in relation to the ramp-up of their commercialization capabilities and to their expenses associated with being a public company. This increase was partially offset by lower noncash stock-based compensation expenses of $7 million.

The consolidated net loss attributable to shareholders of Cellectis was $79 million or $1.93 per share in 2018 compared to $99 million or $2.78 per share in 2017. This $20 million decrease in net loss between 2018 and 2017 was primarily driven by a significant increase in net financial gains of $28 million, which was partially offset by an increase in operating losses of $12 million, of which $11 million was attributed to Calyxt.

The consolidated adjusted net loss attributable to shareholders of Cellectis, which excludes the noncash stock-based compensation expenses, was $44 million or $1.08 per share in 2018 compared to $50 million or $1.41 per share in 2017.

We foresee focusing our cash spending on Cellectis for 2019 in the following areas: supporting our rich product candidates pipeline, including manufacturing and clinical trial expenses of UCART123, UCART22 and UCARTCS1; building state-of-the-art manufacturing capabilities; and strengthening our manufacturing and clinical departments, including hiring talented personnel.

Calyxt plans to focus their cash spending for 2019 in the following areas: launching their high oleic soybean products, including their Calyno high-oleic soybean oil and soybean meal; supporting their rich innovative product pipeline; and strengthening their commercial and general and administration support.

I now turn the presentation back over to André for closing remarks.

Well thank you very much, Eric, and I would like to finish with a very short comment. In 2018, we did our homework. We successfully manufactured 2 proprietary allogeneic CAR-T product in 2 different [hematological] indications. We received IND clearance from the FDA for UCART22 in ALL patient.

2019 is the year for clinical execution for Cellectis. There will be 3 wholly-owned allogeneic CAR-T products expected to be in Phase I clinical development, with interim patient data by year-end. Our track record truly distinguish Cellectis as a leading allogeneic CAR T-cell company.

Our daily commitment is to bring new hopes to patients in critical medical need, with the commitment to a cure. And we look forward to sharing more about our exciting work in this area at medical meetings through 2019 and 2020. These will be transformational years for us and hence, for allogeneic CAR T-cell -- the CAR T-cells field.

With that, I want to thank you very much for your attention, and I would like to open the call to questions. Joining me for the Q&A will be Eric Dutang, Bill Monteith and Simon Harnest.

Operator, please go ahead.

(Operator Instructions) Our first question today is coming from Gena Wang from Barclays.

So I just had 2 questions, and the first one is regarding manufacturing for the U.S. and the France site. What will be the capacity in terms of the number of patients assuming 1 in 6 cells per kg dose? And the second question is regarding UCART123 Phase I data later this year. Should we expect to see the data at the ASH? And how many patients should we expect to see?

Bill, do you want to take the first part of the question?

Sure André. So we're in the process right now of doing dose escalation, so the final dose hasn't fully been decided. However, the capacity that we are building at both the SMART facility in Paris and at IMPACT is such that we would, based on our current clinical manufacturing at our CMO, have significant capacity to be able to do the initial clinical and commercial product launches. We also have additional space and we're using a modular approach to the construction so that we can very quickly and flexibly add additional capacity should the buying necessitate it.

Maybe just to add to that, Gena. This is Simon. Currently, we've already got hundreds of doses per one manufacturing run we do currently. And our goal is to replicate that. And as you know, you can do a batch of cells with a few hundred doses within 2 to 3 weeks. And so I think as more automization comes in, I think we're trying to become more efficient than that even. So I think there's ample room to grow.

And maybe, André, for the second part of the question?

Well, yes, actually, the capacity today is approximately per months, we can produce around 1,000 -- I can treat like around 1,000 patients. And our goal is to reach probably triple this production for the manufacturing plant by 2022, when it will go live, in '21 and '22, be able to produce for the 3 type of indications, we'll file the BLA for.

Now the first clinical data that we're going to produce for the 3 trials, which is the AML trial, but also the BLL trial and also the multiple myeloma trial that is supposed to start this year, we'll produce clinical data by year-end, essentially, at ASH and we'll probably organize an event for this.

And Gena, just to add to that as well, the fact that we can't give exact numbers of patients, by the end of the year, that really just depends on how quick we go through dose escalation. And I think you've heard this all from many companies in our space, this is the critical part for an allogeneic CAR-T company. This is very, very different to the autologous CAR-T companies.

In allogeneic CAR-T setting, you have to spend a little bit more time in dose escalation and going through this as this is mandated by the FDA. And we are basically, pursuing the same approach of dose escalation in all these 3 programs, for AML, ALL and multiple myeloma.

And so, as you know, we have between 3 to 5 patients per dose cohort, and we have usually between 3 and 4 dose cohorts per program. And as we're progressing through those, I think getting into the second half this year, we will have much more visibility on how many patients we will be able to present by the end of the year.

Next question is coming from Salveen Richter from Goldman Sachs.

So I just wanted to follow-up on the earlier questions around data that could come by year-end. Can you just help us understand based on the cohorts you just mentioned, when you would get to therapeutic dose levels and when we should be able to expect a first look there? And what you'll be looking for really from this initial dose cohort when you do present this data set over to us?

And then the second question just an update here on your plans for the Calyxt business. Is this a business you're going to grow within the firm long term or is it one you're looking to divest at some point?

I'm not sure I understood the second part of the question. Can you say it again please?

At Calyxt, I wonder...

When we expect to divest in our equity stake in Calyxt? That's it. Okay. Got you. Okay, so first of all, well, with the therapeutic dose, so -- are difficult to assess before you start a trial. Normally what is expected in most of the CAR-T cells base, especially, when you go with fitted cells, it's around like 1 million cells per kg.

What we see with UCART123, the first dose we started, which is 6.25x10 to the fifth, it's quite close to 10 to the sixth cells per kg. Currently, we're at 2.5, so we expect to enter the 6.25x10 to the fifth once we will finish this dose escalation, with 2.5x 10 to the fifth, we're trying to be as cautious as possible. We don't want to go again into a clinical hold, so we're trying to be as cautious as possible.

For UCART22, the first dose is 100,000 cells per kg and then the second dose is 1 million cells per kg. For UCARTCS1, we'll update you on this, but potentially, we should start at 1 million cells per kg for the clinical trial, depends of the -- how this would go. But the fact is that we try to get as fast as possible in the range of 1 million cells per kg because we believe that this is the right dose for the patient, based on what we've seen in the past for all of the CAR-T and based on our trial for UCART19, all the trials that are conducted by Servier and Allogene but also for UCART123 in the 6.25x10 to the fifth cells per kg.

Now our stake in Calyxt, yes, of course, we're very excited of the progress of Calyxt this year and I think going to be a big year under the leadership of Jim Blome, the CEO of the company. As I said from the beginning, Cellectis is not meant to remain a long-term shareholder for Cellectis and we're definitely in Calyxt. And we definitely see Calyxt as a potential monetizeable asset.

We are not in strong need for cash immediately. But Cellectis would definitely consider exiting this position when needed, while we still think that Calyxt has a very, very powerful run over the next 12 to 18 to 24 months as in term of sales ramp up and also acreage ramp up and we think that the company is definitely set for success. So we'd like to see a bit of the success transform into equity and then we'll consider the position.

Our next question is coming from Christopher Marai from Nomura Instinet.

Just a follow-up on an earlier one. Could you further clarify on your manufacturing capacity, what's the split between U.S. and EU capacity? And would you expect that EU manufactured product could be used in the U.S. and vice versa?

And then secondarily, with respect to the CS1 UCART product, could I ask what the rate-limiting step is there, sort of relative to the other programs?

And then, given some of the I suppose mild tox profile we've seen for other multiple myeloma CAR-Ts, is your expectation that the tox profile of this product might be a little bit more mild as well? And therefore, you could move through dose escalation more rapidly than, for instance, the CD19 or even CD22 product?

Sure. So let me address your first question on manufacturing and provide clarity. So the European facility in Paris is going to be making the starting materials, the viral vector used in the UCART manufacturing. So they will not be making any product, final product, drug product at this time.

The U.S. IMPACT facility will make commercial and clinical UCART products and will primarily service the U.S. but would have the capability to be inspected and authorized by the EMEA for European distribution. And all of that would be in conjunction with our current CMO.

Okay. And then on the CS1 product, I guess, the rate limiting step as well as the potential to dose escalate more rapidly due to perhaps less profound tox?

(multiple speakers) Yes, sorry, André.

Go ahead. No, no CS1, okay, we will make announcement when CS1 will get, for example, batch release and clearance. But for CS1, we expect to have maybe it's like a faster enrollment as potentially, we started potentially with what UCART123 with 32 days between one patient another, that was shortened to 28 days, which is like it's being under process. And with CS1, we would like to have essentially a full 28 days between one patient another.

You have to keep in mind that CS1 has the potential ability to infiltrate the patient because the cargo that's present on T-cells and also on NK cells, so it should also go after the preconditioning of the patient against T-cell and NK cells, so we have to at least assess the safety of the product for 28 days. And then, we should shorten this -- the time between one patient and another to half of this which is 14 days and potentially, even go for 2 patients at the same time.

So we definitely are focused on trying to secure the product at first and once we see that there is like a clean safety for the product, then we'd like to increase this by speeding up, by doubling the time each time we go from one patient to another, then we'll double the time after this and we can double it by increasing, like enrolling 2 patients together.

Got it. And then, Simon, if you wanted to add something please do. I guess, I was just curious thinking about the update in terms of numbers of patients that you would think would be useful to provide for this program. Is there a rough range that you think might be most meaningful?

Yes, Chris. I think the answer is very simple. It all depends on how the dose escalation goes. And we've seen in the first experience with UCART19 that you have already very good responses at very low doses. And I think these protocols again, are critically judged by the FDA in terms of the expansion and the potential of GvHD of allogeneic CAR T-cells and this already is something that we think the box we have checked long ago.

And so the therapeutic dose level is the next question, and I think that will depend very much from disease to disease. For CS1, it's really a very new approach for us although, alemtuzumab is a very well-proven antibody, it's one of the only antibodies that actually works on its own in multiple myeloma. And we've picked this target for this specific reason. And here also, what André made the point, is that the fact that CS1 is also expressed on T-cells is an interesting factor here for UCARTCS1 because, again, on a lymphodepletion perspective, this CAR-T could have a self lymphodepleting effect.

And I know a lot of the talk in the CAR-T space currently circles around what is the right lymphodepletion, what is the right preconditioning? And as our partner, Allogene, has mentioned this, this is something that we all very critically look at. And we have very critically selected the cyclophosphamide from the [urban] pretreatment as well as the use of alemtuzumab or no alemtuzumab, depending on the target.

And we have a lot of experience, clinical experience in this field already as being the first allogeneic CAR-T company in the space. And we've made a very cautious approach but also flexible, depending on the programs we're looking at.

So I think as the enrollment progresses, you will see that throughout this year, we will have 3 clinical programs ongoing. And that also means that on the almost weekly basis, we will know more about different patients in different programs. And I think currently, we want to keep it on a level playing field and say we will give you more updates as we see more patients enrolled in terms of what is the therapeutic dose level.

Our next question is coming from Jim Birchenough from Wells Fargo.

A few questions. I guess, just first on the manufacturing side. I think the rough estimate of the cost of goods for the auto products is around $150,000. So I guess, the first question is where do you see those costs going with your ALLO approach. And do you expect to meaningfully improve on that? I'll start with that question. And then, I have a few follow-ups.

I can answer this question, because first of all, we would like to not to communicate too much the cost of goods we currently have. The only data that we can give that can be benchmarked with the ALLO prologue comes back to the batches that we've done in 2016 for UCART123, the AML product. And we've calculated very precisely the cost of goods, which were $3,700 the vial per patient. So like the dose was like the per patient at 6.25x10 to the fifth was at $3,700 per dose. But we would not like to communicate too much on the progress that we've made since 2016 and we've made some progress. And we'd like to keep it internal.

Terrific, that's helpful. And then, just on UCART123, just from what you've seen so far, can -- are you looking at persistence of cells? And what can you tell us about what you're learning about persistence of cells and is there any thought that retreatment might be at some point incorporated into the clinical trial protocol?

Well, there is a debate over -- which is a true debate over persistence versus deepness of the action of the T-cells in patients. Do you want your T-cells to persist for a long period of time versus how powerful and how potent your cells that you inject are. And there is a series of papers that came out, there's a series of studies that are now very consistent with the fact that the fitness of the cell that you inject, just like the J Park study, et cetera, that gives absolutely no correlation between duration and the longest term of complete remission for the patient, whereas UCART123, for example, you would like a bit -- it's around like 20 day plus/minus the week where you would like to see like a good action for the T-cells.

You don't want these T-cells to stay for 3 months or a year in the patient because if you knock out permanently all the cells expressing CD123, you will totally eradicate the myeloid progenitors. And myeloid progenitors are responsible for a series of cell effectors, including T-cells and macrophages, but also, red blood cells, so the patient will be in an anemia, like in permanent anemia and will be have to be transfused for a year, which is not consistent.

So it's better to have the real very strong (inaudible) position, where you inject your T-cell, you need to have cells that has a very high fitness state, that means like a very high number of [kills] in a row and also, very strong expansion and mortality and so -- and also, the ability to exhibit, et cetera, to clean up the patient in short period of time and then, let the patient recover.

Our strategy is potentially to go for a series of like doses, but potentially, to alternate between one target and another. So make an injection for UCART123 and next year, that will be the CAR that we'll find the 19 next year which is CLL1. So you can alternate between CLL1 and CD123 or you can alternate between 19 and 22, you can alternate between CS1 and BCME or C38.

So that's the idea behind it, things we're trying to tackle the target, it's for all, like the cells from several targets and these concepts will come once we will exit these tedious phases of dose escalation and that will be definitely series of protocol that we would like to find in expansion trials. But there is no correlation between duration and persistence.

Okay, that's an important point. And I guess, just finally, on UCART22. As you think about enrolling that study, do you expect to enroll patients who are experienced on CD19 directed therapy, whether that's [lymphocyte] or one of the CD19 CAR-Ts? Or would you expect them to be between naive of those kind of therapies?

We would prefer them to be naive of any kind of treatment be as like could patients have not undergone any type of treatment. The problem is that a lot of patients are treating with CD19 autologous CAR T-cell therapies. The problem with autologous CAR T-cell therapies is that the relapse is embedded in the therapy itself. I explained it during the talk.

But the fact is that it has been -- I think that was like a very nice paper by [Mark Weller] from U Penn, showing that the 19 viral vector transduces B malignant cells in the patients, so the patient's B malignant cell. So the CAR is expressed on the surface by the CD19 and shed the CD19 off the CAR-T that expresses CD19. So you cannot access the CD19 target because the B-cell, malignant B-cell expresses the CAR and you cannot prevent that happening. So there is an embedded relapse rate in autologous CAR T-cells.

And definitely UCART22 can be a savior for these types of patients according to performance, of course, potentially, but the fact is that we will treat and we will give the chance to patients that are relapsing after CD19 targeted treatment because very often it is due to the CAR transduction of their malignant cell.

Our next question today is coming from Biren Amin from Jefferies.

So maybe just on UCART22, André, we've seen your partner talk about a little bit about lymphodepletion on UCART19 program. How do you think about lymphodepletion and impact on cell persistence and cell expansion? I think this is one of the critical aspects to see cell activity?

Well for UCART22, we do have the CD20 -- the CD52 actually was integrated, so it's built in. It could be compatible with the preconditioning with alemtuzumab. This is potentially to do combo trials with UCART19, which could be interesting with our partners. And to make an injection of 19 and injection of 22.

However, on our side, we would like to start UCART22 only with the cycles through there, without doing the preconditioning with alemtuzumab and see how it works. And if finally, we see that the engraftment is difficult, then we'll -- that's also preconditioned with the alemtuzumab [vocal] antibody, but what we've seen with UCART123, we've definitely seen engraftment expansion without preconditioning with alemtuzumab.

However, every CAR and every target is a different manner. And every malignancy also, so UCART123 as I said, hits the myeloid progenitors, which could potentially prevent the patient to reconstitute the T-cells during the action of the CAR, which is not the case for 22 or 19, strictly different things.

So potentially, I understand that the preconditioning was on alemtuzumab could be potentially important. Don't forget that it's something that -- it's an idea that came out from Cellectis' first trial with UCART19, so something was definitely right, the preconditioning was alemtuzumab.

And for CS1, or, for example, CD38, which are both of them expressed on T and K cells and a lot of other immune cells, we think that the CAR shouldn't do the lymphodepletion by themselves. And we have the action of alemtuzumab by themselves. So wouldn't require a preconditioning with alemtuzumab. But anyways, we always have the possibility to re-include or not alemtuzumab but we have to investigate first without to see how it works.

Got it. And then, on UCART123, what can we expect entering into dose expansion phase? I know FDA's required, I think, 6 week interval between patients. Is that something that you can reduce at some time point once you came back with safety data and when do you think that time frame would be when you would -- when you construct to enroll patients a little bit faster?

I expect this to come during like second half this year. And so 2020 will be the expansion trials for Cellectis. This year will more like the dose escalation phase and expansion trials should happen next year. We have very much worked with the FDA on our protocols. And we think that we can definitely accelerate probably second quart -- second half this year.

Got it. And then, just a question on CS1. Are you planning on enrolling patients that have received prior BCMA CAR-T or is this something that you're going to exclude from your Phase I study with CS1?

I'm not sure I actually got the question. Actually, what was the question exactly Biren? Sorry, excuse me.

So on the CS1 multiple myeloma program, for the Phase I trial, are you planning on allowing patients that have previously received a BCMA CAR-T therapy? Or are you -- will you exclude these patients from the study?

No, they will not be excluded for starting in the protocol currently. But generally, they can definitely be enrolled the same that we would not like to have is potentially, the targeted cells to be encumbered. But BCMA is a different target and that's fine.

And also, to our understanding, there is like relapses in this space between the target trial and also as discussed it before, there's always a rate of relapse due to the transduction of the malignant cells are here, the multiple myeloma cells by the CAR that shed the BCMA target which is like -- so there is an embedment of the relapse rate within the targets CAR-T treatment and this is also an option that we could offer to the patient during this trial. And there is no reason not to include them. So no, the answer is no, we don't see any of the competitor in this field.

Up on our next questioner is coming from Wangzhi Li from Ladenburg Thalmann.

Maybe a few questions. So starting with UCART123. I know you are now in dose escalation for safety. But maybe can you provide any color in terms of the efficacy kind of bar you're looking for? Should we be looking more for a response rate or durability of response? Do you have any color in terms of the CAR-T cells kill the leukemic stem cells? And currently, the other patients, what's the remission? Are they ingredients for transplantation or kind of watch for durability of response or then re-dose or, as you said, alternate different products?

Well, we would like to -- personally, there's lots of therapy that is outside, like, bone marrow transplantation, bridge to transplant. Even though, for example, for ML, being able to bring the patient to bone marrow transplant is already for a lot of physicians in this space can talked about, a huge success because most of the time, for relapsers, they can't even bring them to a stage where they can transplant them. And for today, the standard of care in this space, like the real success of long durable remission is a transplant, so if you have a prologue that can grant them to transplant, it's a huge success.

However, we believe that our product can potentially be better than this and even avoid a bone marrow transplant for a patient and bone marrow transplant is most of the time compatibility is the problem and also it's like an expensive procedure and also, it's a very risky procedure for the patient. We would like to -- we would rather develop a strategy of free dosing and repeat dosing with potentially different type of target, as I described, so making combos with, for example, 123, with CLL1, et cetera.

And there's potentially other targets and even making combos with our partners. There's like a series of targets that can be combined with our partners, we have to -- the CAR-T, the UCART that are developed by Servier or Allogene, everything developed by Cellectis initially and we're very excited potentially to try to launch these type of things.

But, for now, we have to acknowledge the fact that bone marrow transplant is the standard or like the highest standard for durable remission for these patients and if the physician decides to transplant a patient, then let be it.

When you look at UCART19, for example, there are patients that have been presented at ASH that have not been transplanted post UCART19 dosing and they're still in complete remission, like 6 months after which shows that there is no need for bone marrow transplant after UCART treatment, and we're very optimistic about the fact of bringing this therapy to the patient without the need to do bone marrow transplant afterwards.

Got it. Very helpful. And just also, you mention -- in the press release, you mentioned that you expected 2 to 4 patients per dose cohort. Could you elaborate on the rationale especially why you expected 2 patient per dose cohort?

Well, it goes from 2 to 5 patients. If, after 2 patients, you see no signs of activity of the CAR, then you can jump to the next dose and like -- and because you don't have that much of activity. And there is no need to continue to treat patients with the type of therapy that also requires an Endoxan preconditioning that it's not like nothing.

On the other side, if the CAR gives some signs of tumor response, then you move for more patients and try to test, this is like kind of safety levels around this and before you can move up to the next dose level.

So that's the rationale behind this. So either 2 patients and no response, and then move up to the next dose level, which was the case when we came back to 6.25x10 to the fourth, so at 50,000 cells per kg. And then we're moving up and then we jump to 2.5x10 to the fifth, and then here, we try to take more time before we move also to 6.25x10 to the sixth which was, we think, more an effective dose. And this will come up to this, I hope, very soon and during year 2019.

Got it. And currently, [you should be in the] first dose, right?

Yes. For 123, we're like 2.5x10 to the fifth. For 22, we're starting at 10 to the fifth, which is the starting dose for 19. And for CS1, we'll update you on this.

Got it. So now switch gears to UCARTCS1. As you mentioned earlier, the dose is quite higher than the other products. And so just wondering if you have any safety data for the autologous CS1 CAR-T, I know there's 1 trial in at the City of Hope, but I'm not aware of any data report. Maybe you can update us or further color on what the rationale for the higher dose of UCARTCS1 than the other products?

Well, we think that we would like to start -- of course, we -- there is no risk, for example, like for UCART123 of a potential myeloid ablation like UCART123 on one side. And that was essentially the risk assessment for this. Plus there was also some concerns with the other bispec [trial], so we've been very cautious with UCART123 plus we have the BASF in the space. However, there is not that much difference between 6.25x10 to the fifth and 1 million cells per kg.

However, with 22, we have to go like around like the same type of dose levels as UCART19. So we decided to move directly to a dose that we think should be an effective -- like maybe sub-effective dose that is at [20 million cells per kg], and also, we think that a lot of these cells will be absorbed by the lymphodepletion action. So they will go against NK and T-cells. So there will be a potential higher risk of absorbing the cell by non-tumor targets such as NK and T-cells and leave less of the action on multiple myeloma cells.

So it should -- and has potentially the expansion but maybe we'll need the higher dose to go essentially against the multiple myeloma cells. So that's the rationale that has been developing behind it. But it's very difficult to develop the rationale to what starting dose per CAR because each time it's a different dose.

Got it. Last question is just for the manufacturing side. So the facility is expected to go live in 2021, the IMPACT facility. So do you have any color in terms of the capacity if you manufacture 3 products in this year, next year at the MolMed or CELLforCURE?

Well, CELLforCURE, you have to consider that now they had -- that they have been...

Acquired by...

Acquired, it's -- I'm not sure the closing has already been done because they would have announced it. But maybe they've announced that in '18. But we should consider that we should finish everything with time with them. And so we're continuing to produce constantly CELLforCURE now, but we've moved to MolMed and we're producing at MolMed. This company is expecting to potentially find other channels to backup us on MolMed. But we're extremely happy with the way the manufacturing is happening at MolMed.

We're disappointed that CELLforCURE has been acquired because after all these years of working with them, I think with last year's very high standards of production with CELLforCURE and all the learning curve, which was a very steep learning curve has gone through CELLforCURE and now, why it's since a year we've been producing in a very, very powerful way with them. And we're -- understand the decision Novartis to acquire them. But we'll need probably backup on MolMed.

Now I think that MolMed has all the capacity to produce all the CAR-T that we need. We've produced everything we need for 123, for CS1 and for 22 for the next dose escalation and through the phase expansion -- the expansion phases. So we're not concerned by the need of having more products in the immediate time. This is not a problem for us currently and by '21, we'll have our own capacity for production. So we think that so far, we're on a safe side for this.

Okay. Actually, maybe, if I can add one more question on the financial aspects. So this year, you expect to start multiple trials, maybe, this is a question for Eric, is in terms of the R&D cost, how much increase should we think about in terms of R&D or SG&A?

So we don't disclose any guidance on R&D cost increase. But with 3 clinical trials in 2019 and also the manufacturing capabilities and so forth, so, yes, we expect that the R&D expense will increase in 2019. But R&D expenses at Cellectis without noncash stock-based compensation expense in 2019 was $51 million, just for your information.

So it's -- but in terms of the kind of magnitude, it's -- how significant should we think about the increase, I guess?

So in 2018, we had 1 clinical trials with 2 clinical trials and also the development of UCARTCS1, UCART22 and UCART123, from $51 million in 2018, that could increase up to around $70 million.

Our next question is coming from Hartaj Singh from Oppenheimer & Co.

Just had a few quick ones. Just going back to CS1, André. The target is similar to alemtuzumab, which is approved in multiple myeloma, SLAMf7. I guess, my question is as you're preparing the R&D and talking to the regulators, is any of the development for alemtuzumab and the approval for alemtuzumab over the last 4 years and the data that's come from that treatment of multiple myeloma patients, has that helped you in developing your clinical and regulatory plans for the IND and going forward? And I've just got a couple of quick follow-ups after that.

Well, no, we have never considered alemtuzumab in multiple myeloma treatment for the simple reason that we don't think, for time being, it will be needed. I think [some] lympho quickening seen of the patients with the cyclo/fluda. So by CAR that have a very -- that we think it's a very potent CAR that we're going to use is going to have approximately the same effect as alemtuzumab. Potentially, we might have the patient that could be put in temporary (inaudible). We don't know this. This is potentially a risk, but it has to be assessed. But we think that CS1 will have an affect that is very similar to potential preconditioning that we'll prove that's helpful.

Got it. Just on UCART22, I know that you are also looking at potential for other indications, NHL. I think you had mentioned previously that once you get to a dose that you're comfortable with going forward with that product in ALL, you will potentially file an IND for NHL is that still a probability or possibility next 12 to 24 months?

(technical difficulty)

Ladies and gentlemen, please continue to hold. We are experiencing technical difficulties. (Operator Instructions)

Actually, I was -- I'm back. Hartaj, are you hear -- still hear?

Yes. Yes, André, still here.

Okay. I was turned off. Sorry about that actually. So that was the first part of the question. So no preconditioning as alemtuzumab is potentially an effect that should be lymphodepleting with CS1 by itself. We really like the CAR. I think it's probably the most prudent CAR for the multiple myeloma. First of all, when you look at BCMA, BCMA, there is no proof of concept with the monoclonal antibody, while you have alemtuzumab for CS1, which is definitely very strong proof-of-concept for tumor response. The only thing, you cannot do either bispecs or you cannot do CARs because of the CS1 present on T-cells. That's why we're going to after this. And also, it's an advantage, because it gives you lymphodepletion and it's happening the lymphodepletion and probably engraftment will be enhanced by the CAR itself.

Yes. No, no, that make sense, André. I mean, it's always nice to go after a target that's been validated by a product already in the market, like alemtuzumab and it's a SLAMf7 target, so that's nice. In UCART22, just had a follow-up in that I think you had mentioned previously that once you get to a kind of a dose in ALL that you're comfortable going to expansion with, that you could possibly file further indications, NHL being one. Is that something that still you could see happening in the next 12 to 24 months with UCART22?

Well for UCART22, the deal is very simple for now. For the next 12 months, it's essentially, or I hope less than this, but we're going to start by a dose that is like 100,000 cells per kg for BLL. And once we'll expedite, the first cohort move to millions and then final cells per kg, so that's 3 cohorts. And once we get the right dose, which we think should be within this range between 100,000 cells to 5 million cells, then we will probably expand this to -- this into series of various different trials, going from BLL to non-Hodgkin's lymphoma et cetera, so tiers of different type of indications. And we think also, this CAR can have definitely a very nice run going after most of the refractory relapsation that comes from the autologous CAR-T space and includes any kind of CAR that could be dosed in there. As I said, there is always an embedded relapse rate that people can get if they have their malignant cells transform. So for now, it's like essentially dose escalation from like 100 cells per kg to 5 million and up to -- after this, to NHL, BLL, DHL, DCL, et cetera. There is no restriction to them.

Got it. Got you. I mean could we expect that by the year end, as you should present, when you present data on UCART22, the initial patients or the patients through dose escalation, you could give some insight as to which are the next indications you're potentially filing an IND for?

Yes. Certainly, certainly.

Great. And then the last question I had is just what everybody dreams of with CAR-T has been solid tumors, people have been talking about it for a while, different companies seem to have gotten a little bit quieter about it. Just any thoughts there, André? I know that you had mentioned last year that was something that you were working on and that, at least in 1 or 2 years, it could seem a real feasibility to bring a product into the clinic. Any thoughts there?

You mean outside CAR-T, outside oncology?

No, just ultimately, with CAR-T in general, the allogeneic approach. I mean, if you're thinking about it outside of that also, that would be interesting. But any thoughts on the CAR-T side for solid tumors with the allogeneic approach or even a non-ALLO approach?

Yes, so Cellectis is working on CAR for solid tumors. And okay, so the current stage is you know exactly the CAR that we're pushing in the clinic next year, it will be CLL1, this year it's CS1, next year, it will be CLL1. The one after CLL1, we think that will have probably as decent portfolio for a small company like us in the field of hematological malignancies, and we'll kick -- so the CAR after CLL1 will be a solid tumor CAR. We're working on this. We have very nice technologies that we've been developing at the armored CAR, the one for example, that, for example, when you replace PD-1 by, for example, IL-12. So PD-1 is over expressed when the T-cell engages the tumor. Here, instead of expressing PD-1, which is a checkpoint inhibitor, where you will have no checkpoint inhibition, but you will have a release of IL-12, and we've shown in series of mice models and in preclinical settings that this type of armored CAR can have a very high potent activity for solid tumors, and we'd like to link this to a very neat target and in a very specific kind of solid tumor. But up to now, we're essentially focusing our intention on hematological malignancies. After CLL1, it will be solid tumor. By then, also as -- so I misunderstood your questions. Sorry about that. But by then, we might also file an IND outside CAR-T, outside oncology, because the thing is that Cellectis is a gene-editing company at the basis and we think that we have the potential to develop series of type of indication, good gene-editing outside the CAR-T, outside oncology space and this is something to be excited.

The next question is coming from [Sumeet Roy] from [George Trading].

I just wanted to get your thoughts on how the treatment regimen now in NHL could modify or change in the coming years? Like for example, at the last ASH, we have seen very encouraging and very quite durable data with cocktail approach of CAR T-cells, CD19 with CD22 combination, or sequential treatment prior progression, or even a multi-tumors [associated] antigen cocktail CAR-T approach or TIL approach. So from the last ASH, it looked like the space is moving towards more -- towards, at least, a dual if not a cocktail approach than a single-target approach. So how do you think -- is there a validity you see in the data? Or how do you think that the space will pan out in the next 2 to 3 years?

Well thank you, Sumeet for this question. It's a very good question, because we're deep believers in combo therapies. We definitely think that CAR-T is a very powerful tool, against series of cancers. But if there is already a combo therapy because of the need of the precondition of the patient, and you can have series of different type of combinations which comes from combos of CARs, combos with other type of T-cell approaches that can come from T-cells or TILs, but also combos with, for example, checkpoint inhibitors or other type of chemotherapy, the thing that is absolutely great in the fact of combining gene-editing with the T-cell is that you can make the T-cell compatible with these type of other therapies that come from small molecules. For example, you can knock out the toxicity in tiny pathway in the T-cell, which so takes a big role (inaudible) is the enzymes that processes all the PNA such as [indiscernible], Clofarabine or Cytarabine, Fludarabine, et cetera. And so combining the ability to inject a drug, plus the CAR together and have the CAR resist the drug while the drug has an p impact] effect and then the cell expand in the space, it's something that we're -- Cellectis will definitely investigate it. For year 2019, it will be the dose escalation year for these 3 trials. And then for year 2020, where we'll have all the expansion phases, then our intention is to file series of expansion trials or registration trial also that will go with these type of combo therapies. We will start probably by combo therapies with our, like, other CAR-T but also combo therapies with small molecules or (inaudible) antibodies that can be comparable with our CAR-Ts. We think it's the way to go.

Our next question is coming from Peter Lawson from SunTrust Robinson Humphrey.

Just on CS1, I know you've kind of talked about the presence of it on T-cells and NK cells. How do you think that kind of changes the manufacturing or preconditioning for you or potentially dosing? And then, how do you think that could impact safety and toxicity?

Well, it's a good question. For manufacturing, it's like the manufacturing of CS1 is not a standard manufacturing of our CAR-T (inaudible), because the fact is that you cannot manufacture CS1 CAR-T without ripping off CS1 from the T-cell prior to introducing the CAR. So it's called the reverse process where first, after activating the PBMC, you have to knock out CS1 from the T-cell, which we do with TALEN and once the T-cell are totally ripped off the CS1, then you can introduce the CAR and then do other type of (inaudible) inside the T-cell. And this type of process has been initially an intellectual challenge but that has been very successfully met by the company. We're extremely excited of the product we've been producing during 2018 and are excited also by the potential performance of this product. We just need to have the product released and start the trial for this. Now we are -- we will monitor the CS1 in the clinic for the potential toxicity and putting potentially the patient in [pleasure]. What's great with this target that it's very, very highly expressed on multiple myeloma cells but it's also expressed on T-cells, predominantly in CD8 cells, CD8 positive cells, less on CD (inaudible) but still expressed on CD (inaudible). It's expressed on NK cells, expressed on (inaudible) expressed on [tons] of other cells. And this message on this is that you will knock out these cells with the fresh T-cell that comes out that can go against T-cells to have the kind of alemtuzumab-like activity to maintain the patients in lympho in a low activity, like in creating space let's say for our T-cells to go after multiple myeloma cells. However, as it's an allogeneic CAR-T, we don't expect our CAR to stay forever in the patient. So don't make a memory in this and there. And once the patient will recover, and we can definitely have a suicide switch and can stop our case of activity and let the patient recover and rebuild blood afterwards. But that's obviously there is a monitoring of safety assessment over the CS1 CAR-T. But we think it's potentially one of the dose exciting approaches against multiple myeloma with CAR-Ts so far.

Got you. And then do we get to see -- or does that trial start in the second half, CS1? Do we get the see data this year or do you think that's more of a 2020 event?

No, no, we expect to show by the year-end data with CS1. Probably, not a huge, like not hundreds of patient but like few patients will be, let's just say, presented by year-end.

We reached the end of our question-and-answer session. I'd like to turn the floor back over to André for any further or closing comments.

Well, first of all, thank you very much for all these exciting questions. We're very excited for year 2019 and 2020. We are entering a very exciting phase for Cellectis because it will be the year for [concurrent] execution for Cellectis for all our wholly-owned allogeneic CAR-T, and we'll invite you by year-end for this very big meeting where we'll produce the data. Thank you very much.

Thank you. That does conclude today's teleconference. You may disconnect your lines at this time, and have a wonderful day. We thank you for your participation today.