Adaptimmune Therapeutics PLC (ADAP) CEO James Noble on Q3 2018 Results - Earnings Call Transcript

|
About: Adaptimmune Therapeutics plc (ADAP)
by: SA Transcripts

Adaptimmune Therapeutics PLC (NASDAQ:ADAP) Q3 2018 Earnings Conference Call November 6, 2018 8:00 AM ET

Executives

Juli Miller - Director of IR

James Noble - CEO & Director

Rafael Amado - Chief Medical Officer

Adrian Rawcliffe - Chief Financial Officer

Analysts

Jonathan Chang - Leerink Partners

Marc Frahm - Cowen and Company

Robyn Karnauskas - Citi

Peter Lawson - SunTrust Robinson Humphrey

Michael Schmidt - Gugenheim

Ren Benjamin - Raymond James

Jim Birchenough - Wells Fargo Securities

Ying Huang - Bank of America Merrill Lynch

Operator

Good day, ladies and gentlemen, and welcome to the Third Quarter 2018 Adaptimmune Earnings Conference Call. [Operator Instructions] As a reminder, this conference call is being recorded.

I now like to introduce your host for today’s conference, Ms. Juli Miller. Ma'am, you may begin.

Juli Miller

Good morning, and welcome to Adaptimmune's conference call to discuss our third quarter 2018 financial results and other business updates. We issued a press release earlier this morning and I would ask you to please review the full text of our forward-looking statements. As a reminder, we anticipate making projections during this call and actual results could differ materially due to a number of factors.

James Noble, our Chief Executive Officer; Rafael Amado, our President of R&D, and Adrian Rawcliffe, our Chief Financial Officer are with me for the prepared portion of this call, and other members of our management team will be available for Q&A.

With that, I'll turn the call over to James Noble. James?

James Noble

Thank you, Juli. Good morning, everyone, and thank you for joining us. After my brief remarks I will turn the call over to Rafael for a clinical update and then to Adrian for a financial summary. Then I will give a few closing remarks and open the call up for Q&A.

We have completed the third dose escalation cohort of our MAGE-A4 basket study and our MAGE-A10 triple tumor study. The safety review committee agreed yesterday that the more intense preconditioning regimen and higher cell doses are tolerable and agree that the studies can now move into the expansion phase.

This allows us to use doses of up to 10 billion cells without a stagger between patients. Our MAGE-A10 and MAGE-A4 studies are first-in-man dose escalation studies add a critical aspect to these first three cohorts with the discharge any safety concerns such as off-target toxicity, or dose-related safety concerns.

We have only dosed three patients in the third cohort of the MAGE-A4 basket study and four patients in the third cohort of the MAGE-A10 triple tumor study at target doses of 5 billion cells.

And we need to accumulate an adequate amount of clinical assessment in a broader range of solid tumor indications and a larger number of patients to draw any meaningful conclusions. Therefore we plan to report data on these patients at the same time as some expansion phase data by no later than our first quarter of financial results in May 2019.

I also want to update our progress with our AFP study. It is worth noting that AFP is not a cancer/testis antigen like MAGE-A10 and MAGE-A4, and it is highly expressed it a subset of hepatocellular carcinomas. AFP itself is actually used as a diagnostic biomarker for this malignancy.

To-date, we are very pleased with our progress dosing patients in the first cohort of our AFP study, and we anticipate that we will escalate Cohort 2 early next year.

With that I will turn the call over to Rafael for a clinical update. Rafael?

Rafael Amado

Thank you, James. I will briefly summarize the current status of our clinical programs. As James said, we have been progressing through our studies with MAGE-A10 and MAGE-A4 and AFP. We have completed the transition of our NY-ESO program to GSK, which has enabled more focus on our wholly owned assets.

We have treated nine patients in total with MAGE-A4, three patients in each of the three dose cohorts, seven of whom were women with ovarian cancer. In MAGE-A10, we have treated eight patients in cohort 1, three patients in cohort 2 and four patients in Cohort 3. Eight of these patients have lung cancer and were treated in Cohorts 1 and 2 of the MAGE-A4 lung cancer study.

Assessments for the patients in the third cohort are in progress. The safety review committee met yesterday and reviewed data from seven patients dosed in the third cohort of our MAGE-A4 basket study, and MAGE-A10 triple tumor study and they agreed that these studies can now move into the expansion phase.

At ESMO, we presented data on the first two cohorts for MAGE-A4 and MAGE-A10. As expected, the 100 million cell dose was sub-therapeutic with no antitumor effect on core expansion and persistence. Although a dose of 1 billion cells could potentially be associated with antitumor activity, we did not observe any responses per RECIST criteria in this MAGE-A4 and MAGE-A10 cohort.

This could be due to a number of variables such as type of tumor, treatment history, conditioning regimen, degree of antigen expression, small number of patients and importantly cell dose.

It is worth noting that we only saw responses above 1 billion cells with NY-ESO and that the median dose among responders in Cohort 1 of synovial sarcoma was 5.1 billion cells with a range of 2.2 to 8.3 billion. We have data indicating that the intensity of the preconditioning regimen, as well as cell dose are critical for SPEAR T-cell expansion both infusion and for responses.

Specifically we have observed that certain gamma chain cytokines, such as IL-7 and IL-15, which correlate with T-cell expansion are significantly higher in the peripheral blood of patients who received more fludarabine than those who have less intense preconditioning.

And we will present this NY-ESO data at SITC later this week. In addition, the NY-ESO sarcoma trial shows a relationship between lymphodepletion dose intensity, T-Cell expansion, response rate and response duration. Based on this data, we added an extra day of fludarabine to the third dose cohort and the expansion phases of this study.

Focusing specifically on data from the second cohort in which patients received a target dose of 1 billion cells, we only have data from three lung cancer patients from MAGE-A10 and three patients with ovarian cancer from MAGE-A4. We saw expansion of our transduced cells at peak levels of approximately 20,000 vector copies per microgram of DNA in MAGE-A10 patients

In MAGE-A4 patients, the expansion was slightly higher with one patient reaching levels of around 100,000 copies in whom some evidence of transient anti-tumor activity was observed. Following our recent presentations at ESMO from cohorts one and two of these studies, there have been comments about adverse events with speculation that there may be a correlation between neurotoxicity, cytokine release syndrome, and cell dose.

I want to clarify that to date the most frequent adverse events and treatment related adverse events irrespective of dose were consistent with those typically experienced by patients with advanced malignancies undergoing ferotoxic chemotherapy or other cancer immunotherapy, including low frequency of cytokine release syndrome and low-grade encephalopathy.

All events resolved after treatment and we have not seen any concerning correlation between cell dose level and CRS, encephalopathy or other side effects even in the third dose cohort in which patients received target doses of up to 6 billion cells with more intense preconditioning.

The third cohorts of these studies were designed to include a minimum of three patients with a predetermined lag between dosing. We have dosed three patients in the third cohort of the MAGE-A4 basket study and four in the third cohort of the MAGE-A10 triple tumor study, and we did not see any dose limiting toxicities.

As we move into the expansion phase for these two studies, we no longer have our predetermined lag, and we will be able to gain information more quickly from those patients with a higher preconditioning than the regimen we use for the first two cohorts.

Further, to be able to make meaningful clinical evaluations of potential – or the potential of this therapy, it is important to have evaluate a greater diversity of cancer types than we had in the first two cohorts. And they are eligible patients across a broader range of indications for the expansion phases of these studies, including indications in which T-Cell therapy has been shown to be effective such as melanoma and sarcoma.

As we accumulate clinical data across a broader range of cancer types, we will report them for the third cohorts of these studies, as well as for the patients in the expansion phases. We remain optimistic that with better expansion and trafficking of SPEAR T-cells to the tumor with higher cell doses and more intense preconditioning we may see patient benefit. Now I would like to turn the call over to Adrian.

Adrian Rawcliffe

Thanks Rafael. This past quarter we completed the transition of the NY-ESO program to GSK. In financial terms, we have received payments of approximately $150 million from GSK since 2014, including the $26 million received in Q3 as a result of the transition. We may also receive subsequent development and sales milestones and royalties based on successful development of the NY-ESO program.

Lastly on NY-ESO, there will be an update of the MRCLS data and the translational data in synovial sarcoma, which will both be presented at SITC. The abstracts were published today and details of these data are included in our press release and in our 10-Q, which will be available later today.

Also in the quarter we announced the closing of the registered direct offering of our ADSs with net proceeds of approximately $100 million, which extends our cash runway to late 2020.

And with that I will turn the call back over to James.

James Noble

Thanks Ed. In considering the data presented to date, I want to emphasize that we have only dosed three patients in each of the second cohorts of the MAGE-A4 and MAGE-A10 studies, which is not a sufficient number to draw conclusions particularly given the limited number of tumor indications represented in these early cohorts. It is also important as we look back at 2018 that we acknowledge other progress we have made. First we have seen no evidence of our target toxicity to date. We are now in the expansion phase of the MAGE-A4 study and MAGE-A10 triple tumor study and can dose without a predetermined stagger.

We anticipate moving into the expansion phase of the MAGE-A10 lung study in early 2019, and dose escalating to the second cohort of the AFP study in early 2019. We have added more centers for our MAGE-A10, MAGE-A4 and AFP studies. We have evolving translational data that enables better understanding of our products and how to improve them some of which will be presented in one of the two posters at the upcoming SITC meeting at [SITC].

We are routinely manufacturing cell doses of 5 billion or more transduced cells at our Navy Yard facility, and we are progressing with our Catapult facility in the UK for vector manufacture. We are prepared for the next steps of the company and funded to see our ongoing studies to the next phases of development. We are actively investigating our therapies to determine the best next steps and we remain optimistic.

With that I would like to open the call up for questions. Operator?

Question-and-Answer Session

Operator

[Operator Instructions] Our first question comes from the line of Jonathan Chang of Leerink Partners. Your line is now open.

Jonathan Chang

Good morning. Thanks for taking my questions. First question, you have guided to providing a clinical update by no later than your 1Q earnings call in May, how should investors be thinking about potential updates between now and May, and how are you thinking about your overall strategy in terms of data disclosure?

James Noble

Thanks. That is obviously a key question. Essentially we have been looking forward at the runway, if you like, of the patients and the number of patients who will be dosed and the time of – the data will come from. Essentially if you take a single patient, you have to remember that if they are dosed today we wouldn't take the first scan until four or six weeks depending on the trial and then to get a confirmed responses is a further four weeks after that.

So even if you dose some today, it takes 2.5 months to get a response. So what we have been emphasizing in the press release is that we think that it is the most sensible thing to get a reasonable number of patients and bear in mind we were only just getting out of the staggered dose 3 cohort across indications. I think it is unfortunate and it is just the way these things happen that we ended up with purely ovarian patients in the second cohort, and actually the first and second cohorts of MAGE-A4 and I think it is important for us to test it.

So where we have material data before then we will be telling investors, and of course, we will be attending JP Morgan, but I'm trying to guide to the fact that when I look at the runway we will have a decent number of patients with a decent interval after they have been dosed by that and the latest and that is what I'm trying to guide to. Obviously if we have something very material before them, we will be discussing it earlier and we will be at JP Morgan.

Jonathan Chang

Got it. Thanks. Second question now that the safety review committee has recommended moving into the expansion phases of the MAGE-A10 triple tumor and MAGE-A4 basket studies, can you talk about the overall lessons learnt in terms of the safety of your platform, and you mentioned this earlier, but I love to get any additional color, have you seen any correlation between safety signals and the higher dose levels?

James Noble

Yes. I think overall there has been 2 to 3 areas where the toxicities are focused on. One is the chemotherapy side effects, mostly immunosuppression, which in the first two cohorts was quite mild and then actually added a day of fludarabine. And so far haven’t resulted in worsening neutropenia or febrile neutropenia or infection. But patients do get immunosuppression.

The other is immune mediated responses, and so we have seen, as we said at ESMO, low grade encephalopathy, which has now picked up more readily because there are very thorough assessments at most tertiary care centers do looking for clues of mental status changes, disorientation et cetera, and patients are querying like three times a day.

So it is easier for this diagnosis to emerge. But the highest grade has been grade 2 and resolved with tocilizumab, and the third area is disease related. So some of the patients unfortunately because of the stagger they have advanced cancer, and they may have symptoms related to their disease, or eventually disease progression.

I think the most important feature for us is that we haven't seen a dose related sort of increase in toxicity either one of these three that I mentioned, and clearly no evidence of cross reactivity, which is obviously the toxicity that one worries about in this field.

And so we are pretty happy actually moving into the expansion phase knowing that now that there is a shorter lap, patients won’t have to wait that long and patients would be fitter with other more diverse set of tumors, and knowing that the tolerability, lessening tail cross-reactivity, which would have been very worrisome for this program.

Rafael Amado

Yes, I think in terms of platform I think we have described before how – that we have a very comprehensive preclinical testing program, and I would say that what MAGE-A10 and MAGE-A4 have shown really is that what we saw preclinically. We have seen clinical data.

That is obviously very encouraging, but every product is its own product. So you have to make sure you do it as diligently for every new product as you do in the past. But I think one thing we have also learnt is that while the 100 million cell dose is obviously a nice starting point because of the very limited expansion, we don't really think you get a definitive answer as to whether the toxicity is at that level.

So I think the lowest level probably doesn't show very much and that is why we as a company are so pleased that we have got through the billion and the 5 billion cell dose because at that level you certainly would pick up cross reactivity.

So I feel encouraged by both the preclinical platform, and the fact that the doses even where we have seen reason for expansion we are not seeing cross reactivity, which is definitely the most concerning toxicity you could hope not to see and we haven't seen it.

Jonathan Chang

Got it. And just one last question if I may, can you talk about reasons for confidence in the ability to prove upon the durability of the responses in stable diseases that you see so far? Thanks.

James Noble

I think, you know, the lessons that we have are from NY-ESO, and you know, the durability has been variable. We are having learnings from tissue collection and serum markers such as persistence, cytokines, the type of cell that persists what's in the tumor ability to traffic into the tumor.

We have a nice summary of all those data as we mentioned given in that prepared remark sits in. We think, obviously that persistence is important; that traffic into the tumor bed is important; that the degree of conditioning is important; and that there's a minimum threshold of cell that's required for activity to be seen.

And so, I think if in the right patient with the right conditions, the responses can be quite durable and we have had patients in the synovial sarcoma programs that have enjoyed very longer missions and then patients that have had remissions and then minimum increase in a single lesion which they have irradiated or resected and the patient has remained with functionally controlled disease without any therapy.

So, I think this treatment has the potential to control disease long-term but obviously with these wholly owned programs we need more experience and I think we're now well poised to start testing them.

Jonathan Chang

Got it. Thanks for taking the questions.

Rafael Amado

Right.

Operator

Thank you. Our next question comes from line of Marc Frahm of Cowen and Company. Your line is now open.

Marc Frahm

Yes. Thanks for taking my questions. It's for James and Rafael James, given in your prepared remarks and you mentioned kind of difference of AFP relative to the other targets you've tested not being cancer testing engine.

If you just think back now you more that you've learned about MAGE-A4 and MAGE-A10, are there any kind material differences when you look at the Mesa target versus NY-ESO as a target?

James Noble

So, between MAGE-A4 and MAGE-A10 on the one side and what you saw in the other

Rafael Amado

AFP.

James Noble

And AFP as other. So, AFP is obviously not a cancer tested engine and the reason that we were I think rightly cautious about the trial is that it is expressed on some healthy liver progenitor cells. So, so far so good on that.

So, I don't think we see differences between MAGE-A4, MAGE-A10, and NY-ESO in terms of sort of characterization obviously and in populations of patients and different percentages of patients that actually I think generally speaking NY-ESO is more heterogeneously expressed antigen than the other two MAGE -- that happens the MAGE-A4 and MAGE-A10 and vitro killing is slightly superior.

But I think that I don't think they are fundamentally different. I mean none of the three has any known purpose in terms of any function. So, they're just flags that just target all different T-cell receptors.

Rafael Amado

Yes. I would just add that alpha-fetoprotein is when it's expressed it's expressed at really high level. So, the cell is not only expresses but secretes this protein and the peptide is very well characterized.

We know it's real and it's been described before the same peptide that just TCR mimic or from Eureka is targeting the secretion of the protein allows us to diagnose patients and select patients which is very useful in a clinical program.

And it has the potential to actually be picked up by antigen presenting cells and amplified immune response. But more importantly, it's expressed in most of the tumor cells and as you know can suppress this antigen sometimes has heterogeneous expression.

The only thing that's interesting about this is that is in epithelial carcinoma and the liver is one of the first-class organs were these cells go. And so, the homing and the trafficking which is important for the success of these therapies is almost guaranteed in this disease.

So, obviously we were cautious because of the toxicity but we're pleased that we have not seen at least in the low doses any normal liver injury and so far AFP appears to be presented and seems to be immunogenic. So, we're pretty hopeful about this program as we move forward.

Marc Frahm

Okay, thanks. And then, you mentioned you're having pre manufactured cells for some patients and I think even as our presentation disclosed the number of patients who are awaiting therapy. Now, that you've cleared the mandated safety windows, how fast you think you can actually manage from a clinical trial infrastructure position and from a corporate risk perspective to help?

How many patients do you think you can dose kind of on a monthly basis across the MAGE trials?

James Noble

The manufacturing capacity is around two patients a week in the Navy Yard facility and another two patients a week in the HCATS, Hitachi, which is the old PCT in Annandale, New Jersey.

So, that's a sort of overall capacity but that doesn't mean you got that that many patients. So, that's the sort of maximum capacity we run at the moment. What happens is though that many patients we've made the cells for who are actually not ready to receive the cells.

So, these are patients who are either going through some other form of treatment from their doctor or they're not progressing. So, obviously we have to go through that list, is quite it is a backlog of those if you like to address.

And then, as I say the most sort of the most one could hope for is to go to from each of the two sites but it is very important to sort of think about the dynamics of all of these programs.

What actually happens in practice is if you see some encouraging data, let's say a responsive in a particular indication you'll suddenly find you get enormous numbers of patients suddenly put in for screening and that indication and you can see that in the MRCLS space.

MRCLS cohorts and the previous synovial sarcoma cohorts. What then happens, it goes a bit slowly until you see something and then you get it – you'll get a rush of patients.

So, it'll be indication that would be indication specific I think that and certainly was in sarcoma before and multiple myeloma we did the trial, the same thing happened. We started getting responses. We started getting quicker treatment.

So, and that’s why I say I think we should have a decent number of patients to report on by the three, four, five, month away.

Marc Frahm

Okay. And then I guess you to mention the seven out nine ovarian cancer patients in MAGE-A4 and the lung cancer patients and MAGE-A10 have you now dosed anybody from these tumor types like synovial and MRCLS where we kind know T-cells can work?

James Noble

We dosed one synovial sarcoma patient in cohort 3. We have not dosed and MRCLS patient.

Marc Frahm

Okay. Thank you, very much.

Operator

Thank you. Our next question comes from line of Robyn Karnauskas of Citi. Your line is now open.

Robyn Karnauskas

Hi, thanks for taking my question. So, I just want to ask quickly about the AFP; about AFP. So, a couple questions. So, what are your thoughts about the competitive landscape? What are your thoughts about repeat dosing whether you think that's necessary?

And then expand that into like thoughts about what we're starting to see about good data coming after out of repeat dosing for solid tumors and how you think that will inform your decision when you develop plans going forward?

James Noble

So, I'll start off and Rafael will give the thing. We have repeat dosed a number of patients in the sarcoma studies in the past. It was up to the clinician involved actually and interestingly one of the partial responders in the MRCLS study on repeat dosing became a complete response.

Sorry, synovial sarcoma cohort four was so became a complete response after being a partial response which then relapsed in on second dose. We used to take the view that there was only over any point if the T-cells if the transduced T-cells had disappeared for some reason and the tumor is still there and still antigen positive.

So, that's why we didn't dose very many people in the first cohort of synovial sarcoma. I think we re-dosed two or three patients. It proved that the cells could be kept because they were actually kept 18 months I think at least you can keep them and they're still perfectly viable when you put them back into the patient 18 months later.

But we didn't see much effect. But then in the fourth cohort, we did see this 1PR relapse go to a complete response. So, I think that the data are going to merge.

On AFP, we think the competitive landscape we think is immensely encouraging with the Eureka data to be honest because they are targeting exactly the same HLA despite being a far eastern study. It's actually HLA II. So, it is the same peptide and the same HLA that program and they did get responses. Now I will let Rafael complete it. Rafael?

Rafael Amado

Yes. I mean it is a field that's changing pretty fast after a lot of many years without effective therapies of course you know Serafin even Regorafenib are available. The checkpoint inhibitors are approved a single agent for persistence is in the U.S. and they are from lung trials as you know that are going to redefine I think the treatment of these disease.

So, the fact that it is an immunogenic tumor. I think what a revelation which is welcome by the field. I think as James mentioned the fact that this the peptide appears to be presented to us by the encouraging data from Eureka.

The question that we have is if we have activity in hepatocellular carcinoma, what would be the place in therapy of this product because these patients can decline pretty fast. And I think that is something we'll have to think through and design the trials in the best possible way.

But there's definitely a lot of room for improvement in this disease and I agree with the repeat dosing and I mean fortunately for these patients we had a question or two whether or not we would be able to manufacture.

The manufacturing for patients with liver cancer has been extremely successful. So, we obviously cannot give most of the cells we make because we are still dose escalating. But it is a perfect setting in which to test repeat dosing as we gain more clinical data.

Robyn Karnauskas

So, when would you start doing repeat dosing given the data that you know from the competitive landscape? How might you start incorporating that for AFP and what about your thoughts on safety profile? Do you think you can have a similar safety profile?

Rafael Amado

I think I mean the repeat dosing probably would come into play I think in two ways. One is we are modifying the protocol to treat patients that have received sub-therapeutic doses during dose escalation. So, the patients that are still able to be treated will get higher doses as we clear the subsequent cohorts.

And then, during the third cohort and the expansion phase, I think we will want to see what the activity is with single agent at the top doses which can be up to 10 billion. And then, if we saw responses depending on the durability and depending on what we see on the tumor especially in trafficking, the antigen is still there and there's no evidence that antigen presentation machinery has any deficits; then we would test repeat dosing.

So, I think we have a little way to go before we make that change in general. But as I said in the dose escalation, we will change the protocol to treat patients at low doses now that we know that is safer at least in the initial cohort.

In terms of how we think about safety are main concern, what cross reactivity would AFP expressed in normal liver, we biopsy all patients the normal liver of all patients or the non-tumor liver rather to rule out AFP expression. Of course that doesn't guarantee that there won't be closer activity but as I said so far we haven't seen it.

And what we expect is the normal sort of immune reactions that we see with this adoptive T-cell therapy in other diseases. We started with lower dose conditioning. We are now increasing the conditioning and that the chemotherapy is also well tolerated and the patients in this trial have to have good liver function.

So, overall we're constantly optimistic that this will be tested fully and hopefully we'll see some benefit as we go into higher doses.

Robyn Karnauskas

Great. Thank you, so much.

Operator

Thank you. Our next question comes a line of Peter Lawson of SunTrust Robinson Humphrey. Your line is now open.

Peter Lawson

Hi, thanks for taking my question. Just as we think about trying to drive deeper responses in A4 and A10, what do you think takes that, is it preconditioning or is it trafficking to the tumor bed or is it just hitting the right tumor type?

Rafael Amado

I think I mean trafficking is absolutely required and we are we have shown trafficking in sarcoma and we have some in patients upon relapse we have seen that the tumors are no longer inflamed and we are working on how to overcome that but obviously the cells need to get to the tumor.

There are obviously other factors as I mentioned before the antigen has to be present. The antigen presenting machinery has to be functional. There's a lot of resistance mechanisms that have emerged from the checkpoint field some of which may overlap with cell therapies such as loss of heterozygosity for the HLA in question or mutations in gamma interferon signaling and TAP proteins et cetera, beta 2L.

Those are fortunately rare and we haven't actually seen that in the relapse patients in sarcoma but conceivably that would be an issue. And the patients that we're treating have for the most part already seen checkpoint inhibitors if they are of a disease where or checkpoint inhibitors are approved.

And so, cell dose and conditioning are variables. So, we can control and then as we learn more about what mediates trafficking and persistence and we could modify some parameters to optimize those factors. But we're starting in the immunotherapy studies with maximizing doses which for TCR seems to be critical.

And maximizing lead for depletion which allows for our T-cells to expand and proliferate and traffic and cure. So, those are some of the factors, there are a lot of other factors that we're exploring. These are some of the ones that are readily modifiable in the clinic that we're testing.

Peter Lawson

Sure, thank you. And then, James would you press release -- the first response you see in A4 or A10 or would you wait for really a larger cohort, kind of discuss that data?

James Noble

Well, I think we've talked about how difficult it is. I'd much rather do a larger pool of data but we are obliged obviously to put out material data. And also I think I mentioned before in these calls the difficulty we generally have is that both clinicians and patients have there's nothing to stop them getting on social media which they have done in the past.

So, obviously if they do that then we have to. I just think the reasons I'm so keen on more indications in more patients, if you actually think about a 50% response rate well that means if you dose 10 patients, five respond, it doesn't mean that the first five responded. It means five of the 10 respond for the first three dose.

I mean if you take cohort three of the synovial sarcoma study with NY-ESO, what actually the one without fludarabine, what happen is the first four didn't respond to the first -- to the fifth one got an 80% response. Now had the fifth patient come in first we would have thought crikey, without fludarabine you get an 80% response rate an 80% reduction in the tumor.

And then, we had the reality of seeing one-by-one no more responses where fludarabine in other words I think these terribly small patient numbers can deceive. That's what I'm keen of. But you know responses are material and if we have to report material things or if they leak in particularly, obviously we'll have to talk to people.

It's just that nothing has worked within a solid tumor in every patient so far as I'm aware I'm looking at Rafael. I don't think anybody is at 100% response rate in anything. And therefore, the fact that you do or don't get responses in the first two or three patients doesn't really tell you you have to dose more patients, that's really what I'm trying to say.

But it's something we watch daily that way.

Peter Lawson

Got you, okay. And then, just the final question. What's the longest duration you've had a patient on the higher dose for the MAGE-A10, A4?

James Noble

It must be a couple of months. But simply, I just work it backwards because there was a stagger of two weeks between them and then at a 30-day period before the safety review committee. There must be 30-days plus three times two of those. There must be a couple of months but and the dates that ahead.

Peter Lawson

Perfect. Thank you, so much.

Operator

Thank you. Our next question comes from the line of Michael Schmidt of Gugenheim. Your line is now open.

Q – Michael Schmidt

Hey guys, thanks for taking my questions. I had a couple. The first one is on the efficacy side. How confident are you that the 5 billion cell dose is a high enough dose which is potentially resist responsive and that you don't have to go higher at this point.

And then secondly, you did speak about looking at a greater diversity of cancer types in these expansion cohorts and I was just wondering this is sort of following up on what you said earlier.

That I guess, what are factors or other potential factors besides target expression that could determine whether a patient is could be responsive and into what insights do you have whether there are some of those factors are more noticeable in certain types of cancers with others.

But for example; when thinking about the synovial sarcoma data which obviously look very compelling compared to data that's sort of emerged from other tumor types. Thank you.

James Noble

I'll let Rafael answer. In general, that's why we pointed out to people at seven of the first nine MAGE-4A patients to be in ovarian cancer. This is not like running a cohort of nine synovial sarcoma patients or MRCLS patients or indeed multiple myeloma where we saw responses in the context of autologous stem cell transplant.

I mean, previous data in and just to give you a comparison, you may recall we actually did have an ovarian arm in NY-ESO study and that wasn't one of the ones which GSK has decided to take forward and you could see why because the response rate there wasn't very impressive.

So, I think it may be disease specific and I think that's why we're very keen on in trying more of these different diseases. Now some diseases we know have responded whether in our trials or other people's trials to targeted therapy.

So, that's why for example there'd be no melanoma patients so far and we know within ways in the hands of Dr. Rosenberg at the NCI that he had a 55% response rate in the melanoma trial admittedly with different preconditioning and different is a retrovirus et cetera but nevertheless it could be responsive.

We didn't have any of those patients. We had no synovial sarcoma patients as I said in the first two cohorts. There are just I think we need to give the cells a proper chance is what we're really saying by giving them across different tumors and that's what we're saying is a better test of whether they are actually going to become functional.

I'll let Rafael on the other half of it.

Rafael Amado

I think that the 5 billion comes from obviously the sarcoma clinical trials where we know that that was the median dose for responders. So, we know that can be a therapeutic cell dose. Of course testing trials where people have given many more cells and there is a group that doses up to a 100 billion cells and gives IL2 and so on.

So, it's not like the dose response curve for these technology has been very well defined but we know 5 billion can be active and whereas we also know that doses below a billion we've never seen responses. So, we are testing higher doses. So, in the expansion we can go up to 10.

And we believe again that that if what we saw with NY-ESO is recapitulated with MAGE-A4 and MAGE-A10, we will see activity. In terms of the tumor type, I think in the field people believe that as long as the antigen is relevant you can see responses in any tumor.

And the surgery branch has shown that by cloning the right TCR in a patient with cholangiocarcinoma there's been really very long responses. The same has been seen in melanoma and very recently in breast cancer. So, I don’t think this is necessarily too much specific, provided that you have active cells that the antigen is relevant and it is properly presented in the context of MHC and that the self contraffic to the side of the tumor.

And of course there are indications where patients effected before may have deficit penalty in presentation that cross react with cell therapy. And those having described, I would checkpoint inhibitors. And so, patients have received checkpoint inhibitors, particularly they respond it for a long time and then they re-lap their selves.

They consist of may bear some of those alteration. But in general, we think that we can exclude a particular tumor type at least present and that's why we included so many tumors in a study. And in the expansion phase it's so much larger diversity of indication than during the dose escalation phase.

Michael Schmidt

Great, thank you.

Operator

Thank you. And our next question from the line of Ren Benjamin of Raymond James. Your line is now open.

Ren Benjamin

Hey guys, thanks for taking the questions. I guess maybe just starting off with the manufacturing of the cells. Can you talk towards a little bit about the sort of the mean dose for production running in cohort 3?

I know that you have this range of 1.2 to 10 billion cells, and what the target dose is and where I'm getting at is when you start these expansion studies, don’t you want to be focused on only those patients that are getting let's say above 5 billion cells.

I think somewhere in the press release you mentioned how your add-up at a capacity of where over 50% of your runs can get over I think it is 5 billion. So, can you help us maybe just triangulate how confident you are that these next expansion studies will be at doses that you want to be evaluating.

James Noble

Yes, that's a great question. I mean, I can tell you that in the patients that was treated in the third cohorts of all received doses in the upper bound of the range. And so, we have a number of patients that received 5 billion and the upper bound was actually fixed and some patients have called sick.

We've had a patient where we have fewer cells and we did another run because the patient could wait and we were in the stagger phase and we combined to be able to give something closer to five. So, this runs that we are manufacturing largely, most of them pretty successful and when we can we dose to the upper bound of what's allowed.

Ren Benjamin

Got it, okay. That made excellent sense. And just for setting expectations regarding the data update in the first quarter of 2019. When I'm thinking about the number of patients worth of data we should possibly see, obviously I come with seven for cohort 3.

And then is it fair to say that roughly another seven or so from the expansion cohorts seems reasonable or should expectations be somewhere around 10 patients worth of data at these high doses?

James Noble

Well, it'll be obviously more than seven. It's very difficult to be precise. I mean, patients come into these studies and drop out of the studies, I mean, until they're actually there. I mean, in all kinds of studies that I've been in, people can drop out of the day before, saying not obliged to do any of these things.

It will be a number greater than seven. I actually I just don’t know where will be myself. So, it'll be misleading to give you a number but I hope it will be a decent number above seven.

Ren Benjamin

Yes, no problem. And there is one final question from me and it goes back to this idea of persistence and I think an answer Rafael to which you provided before regarding all these different aspect between trafficking and the lymphodepletion regimen, everything adding up to how you can make these cells more persistent.

But we've talked about I guess a couple of times during this conference call the use of checkpoints in general in oncology and then I think in particular in combination with cell therapies. And it seems that that would be the ideal sort of place to go.

I know that we've talked about this or you guys have talked about this in the past. Can you just remind us kind of where we are with those kinds of programs and what next generation programs could look like to improve upon persistence?

Rafael Amado

Yes. I mean, this clearly we're looking at the immunotherapy activity right now but thinking about how to make the therapy better and one is with new generations of factors which they're actually through safety assessments and we have plans to bring him into the clinic and once we've characterized the benefit risk of this products, that's immunotherapy.

But we're very keen on starting to explore them and will announce exactly when they go at the right time that it is really top of mind for us in R&D. and the second is combination trials. We started one as you know within where you saw in multiple myeloma and that was packed on in flight if you will to GSK.

In multiple myeloma and there are not being an indication where checkpoint inhibitors will be registered. So, we now have an opportunity with the diversity of indications that we have to focus on indications with checkpoint inhibitors that's standard care and sort of build upon that with cell therapy.

There's a lot of enthusiasm for that concept among investigators and among their potential partners that we talked to. And I think we'll say more about that once the discussions program.

Ren Benjamin

Great. Thanks for taking the questions.

Operator

Thank you. Our next question comes from the line of Jim Birchenough of Wells Fargo Securities. Your line is now open.

Unidentified Analyst

Good morning, it's Nichol for Jim. It is very early here on the West Coast. I apologized if I missed this but so this level 2 you'll have to leapfrog I believe A10 triple due to safety in lung.

So, is there reciprocity for those level 3 in lung or do you need to complete both triple and lung at dose level 3?

Rafael Amado

Yes, great question. We actually didn’t discuss that with the safety review committee. I think potentially one could have decided to sort of graduate if you will the non-small cell lung cancer study into the expansion. But we had patients that are ready to be treated and we couldn’t have done it without and then in the protocol and we're just we think that we had enough time to go to there.

And then the IRB and so on before these patients were ready to go. And so, because the cell doses aren’t that different; the maximum is six in the third cohort and turning the expansion. That's really the only difference in addition to the staggered.

We've decided to just proceed with the third cohort. But in theory it's something that we could have done. It's just in practicality would have made that much difference. So, we're going to complete cohort 3 in lung cancer and then continue with expansion.

Unidentified Analyst

Okay, great, thank you. And then, just to clarify on the A10 triple you dosed four patients. So, one patient obviously did not get for with the whole deal three period. Is that correct?

Rafael Amado

That's right. And one patient hadn’t. The protocol allow a minimum of three for evaluation of DLT. And so, but the patient had gone through that the acute phase of hospitalization where you know most of the side effects tend to be seen.

And what's discussed during the SRC review and as I said before as we said before the SRC endorse proceeding with expansion.

Unidentified Analyst

Thank you.

James Noble

I mean the reality is, Rafael just mentioned the acute phase. I think that's very important. In general, if you don’t see something in the first week or two, you're not going to see issues coming out in the later stage. So, that's not a pattern in the cells.

So, I mean, sometimes you see things but not very often whereas you obviously do expect to see things like CRS come along in a few days or within the first week or two.

Unidentified Analyst

Okay, thank you. And then, that's a follow-up from me. So, can you, I know it's early but can you comment on peak expansion whether you're routinely hitting your 100K threshold?

Rafael Amado

So, so far in the third cohort the expansion level are around that number. So, we've seen actually quite a faithful dose response if you will in terms of peak expansion. And so, we're pretty pleased with those numbers. Of course we don’t have long-term persistence and we will like to look at the area under the curve to see how long cells persist at the high level.

But the data that we have so far is encouraging that it sort of recapitulates the expansion that we were seeing in third cohort.

Unidentified Analyst

Great, it's very encouraging. We look forward to next update. Thank you, very much.

Rafael Amado

Welcome.

Operator

Thank you. Our next question comes from the line of Ying Huang of Bank of America Merrill Lynch. Your line is now open.

Unidentified Analyst

Hey guys, this is Alec on for Ying. Thanks for taking our questions. Just a couple from me. And the A4 and A10 studies, it was nice to see the interim data from the first two cohorts that as now. Now that you have a few patients there from cohort 3 from both studies.

Could you maybe give us a sense other than safety how these patients are doing, and whether you have any emerging favoritism between the two programs based on that to-date?

James Noble

We can't update on that because that would just be as we said we're going to try and update when we got reasonable, I suppose. I think the end the preference between MAGE-A4 and MAGE-A10 which I think we mentioned times, many times for that MAGE-A4 is expressed in many more cancer and at a higher percentage of those cancers.

So, in into the commercial terms it's a much more important. But we can't give any updates on the third cohort today about the say went through the safety to-date.

Unidentified Analyst

Okay, thanks. And along the same lines, how has its real-world expression of the antigen levels sort of compare to your expectation?

James Noble

So, that is a really excellent question. And I think in we learn as we go along, I think with as you know with some major -- some NY-ESO programs we got much higher expression. We expect probably that since all came, and I think the multiple myeloma both higher than the database figures.

The normal database people start with is the TCGA and that's what we sort of start with. There were some very difficult low percentages there which did not correspond to the database. And that was for example in ovarian cancer and then ways that was actually much lower than reported and so is adenocarcinoma with the lung.

So, there is some which you get surprised and some which you don’t. I think MAGE-A10 is relatively low expressed compared to MAGE-A4. I think MAGE-A4 bears that the TCGA database in most respects. Obviously we have to go individually by disease but most of them where we've had actual tumor sample if it reflects that database much more closely, I'd say that and why you say they did?

Rafael Amado

Yes. I mean, now we just have we've validated expression with tumor banks at in the Anderson and with we collaborate with that group for our alliance and we saw staying 100s of samples of multiple tumor types that are included in our study. And then contrast that with what we are seeing this with we screen many patients already in the trial.

And so, we now are pretty precise with our predictions, which is excellent because it allows us to size the study correctly and get the right number of centers et cetera. In MAGE-A4, finding MAGE-A4 patients, have not really been a problem actually. And those there it's expressed in the last malignancies that the right levels that there is a healthy run way.

Unidentified Analyst

Great. And last one from me more of a modelling based question. So, now that you have the proceed from the $100 million raise, how do you sort of see capital allocation over the next six to 12 months? Thanks.

James Noble

So, obviously the advantage of transferring NY-ESO to GSK is that essentially all of that capital's allocated to the holding on platform. And the guidance we've given is that that plus that we see from GSK plus the amounts that were previously in the bank, and the cash balance of the end of Q3 is $238 million.

That balance we'll see as through to the end to late 2020 with running the business that we have. So, which is now almost entirely on holding on platform.

Unidentified Analyst

Great, thank you.

Operator

Thank you. And I'm showing no further questions at this time. I will now like to turn the call back over to Mr. James Noble for closing remarks.

James Noble

So, thanks everybody for joining us this morning. I feel that we are at a pivotal stage in the company. It's a very exciting time to have got through the safety cohorts and now have most of the things lined up. We have centers lined up, have patients who've enrolled in the trials. We got through the dose expansion.

We've got through the dose escalation and win the expansion. And we should be able to provide data over the coming months both in MAGE-A10 and MAGE-A4. With that we are making good progress and we hope to get through to the second dose level early in 2019.

So, I see very exciting times ahead of us and I look forward to giving you further updates in due course.

Thank you, very much.

Operator

Ladies and gentlemen, thank you for participating in today's conference. This concludes today's program. You may all disconnect.

Everyone have a great day!