bluebird bio Inc (BLUE) 2017 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the bluebird Quarterly Update and ASH 2017 Abstract Call. (Operator Instructions)

I'd now like to introduce your host for today's conference, Ms. Liz Pingpank. Ma'am, you may begin.

Thank you, operator, and good morning, everyone. Thanks for joining us to discuss our quarterly results as well as the small, but we believe important amount, of new data contained in some of our ASH abstracts. We issued a press release outlining the data this morning, and you can find the release on the Investor Relations section of our website. The agenda for today's call is as follows: Nick will make some opening remarks, and then Mohammed will walk through the clinical abstracts. Philip will share some preclinical updates, and Nick will then conclude and we will open up the call for Q&A.

Before we begin, let me review our safe harbor statement. Today's discussion contains statements that are forward-looking under the Private Securities Litigation Reform Act of 1995. Such statements are based on current expectations and assumptions that are subject to risks and uncertainties and involve a number of risk factors that could cause actual results to differ materially from projected results.

In particular, statements whether the changes to the HGB-206 clinical trial protocol, including plerixafor mobilization, will improve outcomes in patients with severe sickle cell disease and risks that the current or planned clinical trials of LentiGlobin in thalassemia will be sufficient to support regulatory submission or marketing approval in the U.S. and EU. Additional information concerning these risk factors is contained in our filings with SEC, which are available on the Investor Relations section of our website, www.bluebirdbio.com.

While we may elect to update forward-looking statements in the future, we specifically disclaim any obligation to do so even if our expectations change except as required by law. You should not rely on these forward-looking statements as representing our expectations as of any date subsequent to today.

I'll now turn it over to Nick.

Thank you, Liz, and thank you, everyone, for joining us on the call this morning. Our objective today is to highlight our quarterly progress and share the emerging data included in some of our ASH abstracts.

In the past, we've generally not done a quarterly call or a webcast for abstract releases. But in this case, we felt it was important to provide the context for the new data. In particular, we will focus on some limited data on our 206 study in sickle cell disease as well as updates on our 204 thalassemia study. Our goal is to cover 2 simple questions. What is the relevance of the new data? And what are the expectations now going into ASH?

Before we dive into the data, we always start off by talking about why we're here. And simply put, it's about hope. It's about doing all that we can to bring hope to patients and the families in desperate need. True Blue captures that ideal for our patients, and BLUE MOJO captures everything we do inside the company to make that dream come true.

So how do we succeed? What you will hopefully will notice about team BLUE is that we believe success lies in our ability to continuously innovate and learn about both our gene therapy platform as well as the underlying disease biology. We spent much of 2017 talking about key learnings in both these areas.

On the technology front, we've been seeking to improve on our product as well as manufacturing to optimize all aspects of our programs. Specifically, we shared our new manufacturing process for LentiGlobin, the PI3-kinase manufacturing edition of BCMA. And in this call, we'll share our research into the plerixafor -- into using plerixafor for sickle cell disease mobilization.

On the biology front, we've been focused on going deep in our understanding of the pathophysiology of disease and using that to inform innovations in our approach.

We further our understanding of patients genotypes in thalassemia, implement a protocol changes in sickle cell disease to addressing engraftment challenges, and more recently, share a deep cell phenotyping in both thal and sickle cell. The point is simple, by innovating and integrating our learnings in both technology and biology, we believe we'll continually deliver the absolute best for patients.

Today, we announced our quarterly financials and recent operational progress. As we progress through the year, we're achieving many of the goals we set at the beginning of the year. Notably, this quarter, we disclosed significant progress in our BCMA program related to the treatment of the first patient of the expansion cohort of Phase I CRB-401 trial of bb2121 as well as the treatment of the first patient in the Phase I CRB-402 trial of our second-generation candidate bb21217.

We also disclosed our interim data in early October from our Phase II/III trial in ALD, showing that 15 out of 17 of the patients had met the primary endpoint.

Just last week, we also received the Regenerative Medicine Advanced Therapy Designation, also known as RMAT, for our sickle cell program for the FDA -- from the FDA, enabling early informal and formal engagement on our development plans.

Lastly, we continue to be well funded with $1.1 billion of cash on the balance sheet, taking our runway into 2020.

Our focus moving forward is clear and an exciting evolution of the company. We are preparing the company for future product filings and commercial launches, and building towards our 2-2-4 in 2022 vision, 2-plus commercial projects, 2-plus nearing commercial and 4-plus pipeline products. It's very exciting times.

Now let's shift gears and focus on the ASH data. I'm happy to say that this is going to be the biggest bluebird -- this is going to be bluebird's biggest ASH ever and with 11 presentations across our clinical and preclinical programs. Most of the abstracts are placeholders with no data. However, several of the abstracts made public today contain some new and important data.

On the thal front, there is an update to the HGB-204 study, also known as Northstar. On the sickle cell disease front, there is an update on the HGB-206 study, including data on the safety of using plerixafor as a mobilizing agent in patients with sickle cell disease. But for clarity, today, we will not be providing updates on either of the BCMA program or the HGB-207 studies, also known as Northstar-2, but new data on both will be provided at ASH in a few weeks.

We also have several preclinical contracts -- abstracts, one of which will be in touch -- we will touch on briefly today related to our relationship with Boston Children's Hospital and their preclinical program in sickle cell disease.

Before I turn the call over to Mohammed for the data review, let me give you the quick highlights. On sickle cell disease, the question is whether any of our improvements and protocol changes are going to make a difference for sickle cell disease patients? The good news is that the early clinical data is a promising step forward, including our ability to shift towards an apheresis-based product using plerixafor mobilization. On thal, our clinical and regulatory path remains on track based on the updated 204 data.

With that free preview, I will hand it over to Mohammed to review the data. Mohammed?

Thank you very much, Nick. Let me start with a quick overview of our 4 studies of LentiGlobin in thalassemia. HGB-204, also known as Northstar, and HGB-205, our Phase I/II studies conducted in patients of all genotypes using the original manufacturing process. These 2 studies will form the primary basis of our filing in the European Union.

The ASH abstracts provide an update on the 204 study that I will present today.

HGB-207, known as Northstar-2, is our Phase III study in patients with non-beta-0/beta-0 genotypes while are transfusion dependent. The 207 study is conducted using the improved manufacturing process, which substantially increases both the drug product vector copy number and the proportion of stem cells transduced by the vector. The 207 study is ongoing. And as Nick mentioned, the ASH abstract does not include any new data, so we will not update this study today.

HGB-212, known as Northstar-3, is our Phase III study in patients with thalassemia, who have the beta-0 genotype. These patients make no endogenous beta-globin. 212 is also conducted using the improved manufacturing process. We are on track to start 212 this year and will announce when the first patient has been treated with LentiGlobin.

Now that you are oriented, I will walk through an overview of the Northstar Study. As mentioned today, abstracts include an update to the Northstar Study. 18 patients have been treated in the study, 10 patients with non-beta-0/beta-0 genotypes and 8 patients with the beta-0/beta-0 genotypes. And they have a median follow-up of 26 months.

Of the 10 patients with the non-beta-0/beta-0 genotypes, 8 have been free from transfusions for a median of 27.1 months. At the last study visit, these 8 patients had a total hemoglobin ranging from 9.3 to 13.7 grams per deciliter. With T87Q-globin ranging from 3.6 to 9.6 grams per deciliter.

The other 2 patients with non-beta-0/beta-0 genotypes have seen their annual transfusion volumes reduced by 30% in 1 patient and 94% in the other. Notably, these 2 patients received drug product at the low end of the VCN range in the study, 0.3 and 0.4, respectively.

There is a strong correlation between DP VCN and T87Q-globin production in the thalassemia studies. So the outcomes in these 2 patients with low drug product VCNs is not a surprise, but rather underscores the importance of introducing the improved manufacturing process to optimize transduction efficiency.

In the patients with the beta-0/beta-0 genotype, 2 of the 8 have not received a transfusion in more than a year. At the last study visit, for these 2 patients, total hemoglobin levels were 9.0 and 10.2 grams per deciliter, with T87Q-globin levels of 8.2 and 6.8 grams per deciliter.

These patients have vector copy numbers in the higher range of what we saw with the original manufacturing process. With subsequently robust peripheral VCNs of 0.9 and 0.6, respectively.

The remaining 6 subjects with beta-0 genotypes have seen their annual transfusion volume decreased by a median of 63%. Across these patients, we have not seen any new safety signals. And the safety profile to date has been typical for busulfan myeloablative conditioning.

These results show clinically meaningful benefit in the great majority of the patients. With 8 of the non-beta-0/beta-0 and 2 of the beta-0/beta-0 patients, now free of transfusions for at least a year and transfusion reductions in the remainder.

Importantly, these data reinforce the potential clinical benefit from the improved manufacturing process implemented in our Northstar-2 and Northstar-3 pivotal studies. Using this process, we expect to achieve more consistently robust T87Q-globin production to mitigate against patient-to-patient variability and enable an even higher proportion of patients to become free of transfusions with near normal or normal total hemoglobin levels.

Looking forward, the Northstar-2 study will be key to our demonstrating the clinical impact of the manufacturing improvements for patients with TDT. At EHA this year, we presented early data from our Northstar-2 study, which showed that with the improved manufacturing process, we are able to consistently produce drug product with a higher VCN and percentage of cells transduced.

The improvement in drug product VCN translated into higher levels of T87Q-globin in the clinic. And at EHA, we showed that Patient 1 was producing 9.5 grams per deciliter T87Q at 6 months of follow-up, enough to be free of transfusion based on T87Q alone.

At ASH, we will provide updated data on the 3 patients we presented at EHA as well as new data on approximately 5 additional treated patients. With the growing body of clinical data on LentiGlobin in thalassemia, we are making preparations for the filing and launch of LentiGlobin for patients with thalassemia, first in the European Union and then in the United States.

As we described during our Gene Therapy Day event in 2016, we have general agreement from the regulatory authorities in the United States and European Union on the path forward. In the EU, we are working through the framework of the prime designation and adaptive pathways to file for a conditional approval in patients with non-beta-0/beta-0 genotypes on the basis of the NorthStar and the 205 studies and including available interim data from 207.

In the United States, we have breakthrough therapy designation and plan to file for a full approval in patients with non-beta-0/beta-0 genotypes based on data from the Northstar-2 study. We plan to provide more clarity on timelines in 2018.

Let's now switch gears to discuss sickle cell disease. We were fortunate that the first patient we treated, Patient 1204 in Paris, who is the subject of a case study published in the New England Journal of Medicine earlier this year, had a successful treatment response and provided important lessons for optimizing outcomes in sickle cell disease by improving how we manage patients through stem cell collection and transplant as well as driving improvements in our drug product manufacturing.

In the first cohort of patients treated in our HGB-206 study, we saw a large drop from drug product vector copy number to peripheral vector copy number, shown on this slide. And the low peripheral VCN was associated with low levels of T87Q-globin. Consequently, achieving improvements in both drug product VCN and peripheral VCN has been a major focus for us over the past year. We have implemented several changes to our -- to the study protocols intended to address this drop in VCN, which we see largely as a problem of engraftment.

The changes include: pre-harvest blood transfusions intended to reduce inflammation and hypoxia of the bone marrow to improve both the quality of harvested stem cells and improve engraftment of the drug product; more stringent myeloablative conditioning to effectively eradicate competition from residual uncorrected cells; changes to the manufacturing process to increase cell dose; implementation of the improved transduction process as we did in our thalassemia programs to increase drug product vector copy number and percentage of cells transduced. Finally, evaluating the feasibility of using plerixafor mobilization and apheresis rather than bone marrow harvest to improve the number and quality of stem cells collected from patients.

I will start out with the description of the different stages of the 206 protocol to clarify how the changes have been implemented sequentially. Group A, the original version of the HGB-206 protocol using the original manufacturing process, where none of the variables outlined had been optimized.

In contrast, in HGB-205, the single-center study in France, all patients had received preharvest transfusions and stringent myeloablative conditioning. And the differences in outcomes between 205 and Group A helps to inform the protocol changes we implemented in 206.

Group B. We have optimized across almost all of the variables, but stem cell collection is still via bone marrow harvest. Today, we will have a first look at data showing the impact of these changes in 2 patients.

Group C. This group consists of patients whose stem cells have been collected by plerixafor mobilization and apheresis. And today, we will share the safety data and initial drug product characteristics using this approach.

To set the stage for the 206 results, included in today's abstract, next I will show you data from the 205 study that we presented at EHA, which highlights the positive impact of red blood cell transfusions and stringent myeloablation on both in vivo VCN and T87Q production.

The red lines represent the 206 patients in Group A, and the blue dotted lines represent the 205 patients. In the figure on the left, the Y-axis shows VCN on a large scale because of the wide range of values across the patients, and the Y axis in the figure on the right shows the amount of T87Q-globin.

Patient 1204 represents an optimal outcome, but early data from the subsequent 2 patients treated in Paris indicate they are also on a positive trajectory. These patients were optimized across some of the variables we have discussed, including preharvest transfusions and stringent myeloablative conditioning, but their stem cells were collected by bone marrow harvest, and they did not receive drug product manufacturing with the transduction improvements now incorporated into 206.

It is encouraging to see that the baseline transfusions and stringent myeloablation in 205 improved in vivo VCN and T87Q relative to 206 Group A, and we hope to see an even greater favorable impact with the additional protocol changes now in place for the 206 study.

The 206 ASH abstract includes new early data from 2 patients treated in Group B. The first treated patient, 1313, was treated with a mix of 2 drug product lots. One manufactured using the original process and one with the improved process, providing DP VCNs of 1.4 and 3.3, respectively.

The second patient 1312 was treated with 2 drug product lots manufactured using the new process. And these lots had vector copy numbers of 5.0 and 2.9. In the figures, you can see the dramatic improvement in DP VCN and the percentage of transduced cells achieved with optimizing manufacturing, shown in green, compared to the original manufacturing process, shown in red.

The early data in the Group B patient is shown in green on the next figures. And it shows how the manufacturing and protocol changes have achieved our goal of substantially improving in vivo VCN and T87Q-globin production. The first patient treated in Group B, 1313, has a peripheral VCN of 0.5 at the last visit, month 3, which is higher than all of the patients in Group A and also above the patients in HGB-205 with the exception of Patient level 4. At 3 months post-infusion, Patient 1313 is producing 1.5 grams per deciliter of T87Q, which is higher than the results for any prior 206 patient at that time point.

The second patient, 1312, was treated with 2 DP lots manufactured using the refined manufacturing process. At 1 month, shortly after engraftment, the patient has a peripheral VCN of 2.6. This in vivo VCN exceeds all other patients in the 206 study treated to date. At the time of the abstract submission, the initial T87Q value was not yet available for this patient.

These early results provide evidence that the modifications to the study protocol have improved drug product engraftment and in vivo VCN, and we anticipate these improvements will translate into increased production of anti-sickling T87Q-globin. We will provide further follow-up data at ASH on the in vivo VCN and levels of T87Q production in these 2 patients.

The final modification to the study protocol is the implementation of plerixafor mobilization and apheresis to collect stem cells rather than bone marrow harvest. Mobilization and apheresis is expected to yield both improved numbers and improved quality of stem cells but had not previously been used for patients with sickle cell disease because of historical data that GCSF, the glycoprotein typically used for mobilization, can trigger severe crisis by virtue of inducing significant leukocytosis and activation of leukocytes and endothelial cells.

We and the others have been investigating whether apheresis with single-agent plerixafor might be used safely since it induces a less intensive leukocytosis and does not activate cells like GCSF does. Importantly, apheresis would substantially improve the patient experience as it represents a less invasive procedure than bone marrow harvest.

This abstract presents data on 3 patients with sickle cell disease, who have undergone plerixafor mobilization and apheresis. We are encouraged that the mobilization in apheresis procedures have had an acceptable safety profile with no dose-limiting toxicities observed. In these 3 patients, we have collected cell doses of 15.3, 5.6 and 9.0 million CD34 cells per kilogram. This is approximately double the average cell collection of 5 million CD34 cells per kilogram achieved with bone marrow harvest. And no subsequent apheresis or mobilization was required to achieve these doses.

These early results suggest we can safely collect substantially higher cell doses using plerixafor mobilization and apheresis. In addition to patient convenience and increased cell dose, our analysis of cells collected using plerixafor mobilization and apheresis suggest that these cells may have a significantly larger proportion of primitive stem cells compared to those collected by bone marrow harvest.

These true primitive stem cells have high expression of CD34 and may lead to more robust engraftment and repopulation of the bone marrow following transplantation. Analysis of the cells obtained through bone marrow harvest shows that on average 41% of them are called CD34 DIM cells.

These cells are low expressers of CD34, and therefore, probably do not represent the true stem cells that are required for robust engraftment and repopulation of the bone marrow with transduced cells that will ultimately differentiate into the erythroid lineage and express T87Q-globin.

By comparison, on average, only 8% of the cells collected through plerixafor mobilization and apheresis are DIM cells, suggesting that a far larger proportion of these cells are true stem cells. This analysis is preliminary, and we expect to present additional data at ASH to further characterize these cell populations.

Based on the favorable data we have seen on safety and improved cell dose and the early indication that plerixafor yield cells with a more favorable phenotype, we are implementing the next stage in the 206 protocol to proceed with transplant using LentiGlobin drug product derived from plerixafor-mobilized cells.

Additionally, due to the promising initial experience with plerixafor mobilization, we and our study investigators decided to include only 2 patients in Group B and place all subsequent patients in Group C. Further detail on the drug products manufactured for these Groups C patients will be provided at ASH. However, the post transplant in vivo clinical data for these patients will not be available until 2018.

To recap, the updated data in thalassemia from the 204 study indicates that a high proportion of patients with thalassemia of non-beta-0 genotypes become free of transfusions and that the benefit is durable. Even 2 patients with the beta-0 genotype achieved transfusion independence, and the mean transfusion reduction was substantial with the original manufacturing process.

This trial results reinforce the importance of optimizing transduction efficiency to achieve robust production of T87Q-globin and underscore the importance of the refined manufacturing process that has been implemented in the pivotal trials. At ASH, we will provide an update on HGB-207 in which patients have been treated with drug product with substantially improved VCN.

In sickle cell disease, the early data from our Group B patients suggest that the changes to the study protocol and manufacturing process have improved engraftment, even in cells collected by bone marrow harvest. At ASH, we will provide additional follow-up data to show how the improved in vivo VCN translates to T87Q production.

We were encouraged to see that the initial clinical cohort of patients treated in 206 have had a reduction in clinical events compared to baseline. But this observation requires a larger sample size and longer follow-up to confirm. Our initial clinical experience with plerixafor mobilization and apheresis suggests it is well tolerated. It improves both the number and quality of stem cells that are collected.

We will present additional data characterizing these cells at ASH, and the impact the different cell populations may have on clinical outcome. To date, the safety profile of LentiGlobin is comparable to autologous transplantation with busulfan conditioning.

Now Philip will discuss the preclinical abstracts.

Thanks, Mohammed. Beyond our clinical programs, we have several preclinical posters at ASH, highlighting work done by our academic collaborators as well as our team of scientific and development bluebirds in Cambridge and Seattle. These abstract shed some light into our preclinical work in immunotherapy and gene editing, and we look forward to sharing further detail at ASH and, hopefully, in the future as some of these programs progress.

One preclinical abstract worth highlighting comes from a study conducted by Dr. David Williams and his team at Boston Children's Hospital, representing a novel and very promising approach to address sickle cell disease by upregulating expression of fetal hemoglobin, and importantly, simultaneously downregulating the pathological hemoglobin S.

This approach uses our lentiviral vector core technology to deliver an erythroid lineage-specific, BCL11A-targeted microRNA-adapted short hairpin RNA, which I appreciate is quite a mouthful. Put more simply, this vector results in a repression of BCL11A, exclusively in erythroid cells. Reduced BCL11A levels are associated with hereditary persistence of fetal hemoglobin, or HPFH, which is known to ameliorate the signs and symptoms of sickle cell disease.

The preclinical data generated by the bluebird team and Boston Children's Hospital showed that transduction of HSCs from sickle cell disease patients with this novel lentiviral vector, followed by the differentiation of these modified HSCs into red blood cells, results in dramatically increased expression of the anti-sickling hemoglobin F, ranging from 70% to 81% of all beta-like goblin chains. Moreover, this increase in HbF is concomitant with suppression of disease-causing hemoglobin S to levels of just 19% to 30%.

These results compared favorably to preclinical data on HbF upregulation presented to date with gene editing technologies, which have shown approximately 40% HbF under similar experimental conditions. Our lentiviral approach may represent a powerful novel therapeutic approach to sickle cell disease, which stands upon all of the advances we have made with respect to sickle cell biology, manufacturing advances and patient management, which Mohammed walked you through earlier.

We are pleased to be collaborating with Dr. Williams and his team, and look forward to providing details of this exciting preclinical data at ASH. We anticipate this program will advance into the clinic in 2018, and we plan to share more information as the program progresses.

And with that summary of the preclinical abstracts at ASH, I'll hand the call back over to Nick.

Thank you, Philip. As I mentioned in my opening remarks, this has been an incredibly busy and productive quarter. And ASH will certainly continue that trend with 11 accepted abstracts and clinical updates across all of our ongoing -- nearly all of our ongoing clinical trials.

Just to make sure we are clear on what we will disclose at ASH, here is an overview: additional follow-up on 204 as well as 205, which will form the basis -- the primary basis of our EU filing from LentiGlobin in thal; an update on Northstar-2 or 207 study that will include a few more patients that have been treated with the improved manufacturing process and longer follow-up on the patients we showed at EHA; further updates on the 206 sickle cell disease study will include longer follow-up on the 2 patients we just shared in Group B. We will have some product, some drug product data for several additional subjects in Groups C, who've had their cells collected using plerixafor mobilization; we will also have an update on our BCMA program that will include additional follow-up on the 21 patients presented at ASCO.

Lastly, we hope you'll join us on Sunday, December 10, at ASH, where we will host a webcast Investor and Analyst Event to provide more detail and context on our clinical and organizational progress.

That completes the formal part of our call, and we look forward to take some questions. Operator?

(Operator Instructions) Our first question comes from Salveen Richter from Goldman Sachs.

So just with regard to the sickle cell program, could you just remind us what the age or tell us what the age of these 2 patients are? And given you're at 1.5 grams at 3 months, what's the threshold level you're trying to achieve here for this patient? And remind us what Patient 1204, the first patient treated, ever showed at this time point?

Okay. So thank you, Salveen. I appreciate the question. So first maybe, Mohammed, you can address the age, and then we'll step through the other questions.

Yes, so our age range in the study is 18 and older. Our 2 subjects fall within that range, older than the traditional patients treated with sickle cell disease. Then in regards to the HbAT87Q production, so the results that are in the abstract represent the highest T87Q production at that time point of any patient thus far treated, including 1204. So not substantially greater than 1204, but greater than 1204. Sorry, I missed the last one.

What's the third question, Salveen?

Just when we put the 1.5 gram in context, how should we think about what the threshold T87Q level you're trying to achieve long-term here for this to be meaningful?

Yes. No, I think that's an excellent question. So I think part of the answer to that question is we have to see how the data play out and how our clinical results evolve over time. But the short answer is that in subject 1204, we had approximately 4 to 5 grams of T87Q production, and that resulted essentially a conversion of that patient to a biochemical sickle trait. I think that will be a very admirable goal to go for. I think based on historical data, with hemoglobin F production, we know that individuals that have approximately 30% of their total globin being hemoglobin F have significant attenuation of their symptoms. So I would say that shooting for a similar goal for our HbAT87Q, which behaves similarly to hemoglobin F would be something we are trying for.

And Salveen, it's Nick. It's a little too early obviously, right? This has a very nice trajectory as you can see in the data. But we'll provide a little bit more context as we go. But I don't think we all need to get up to the 1204 level, but our point was to really -- and the primary emphasis here is have we addressed some of the engraftment challenges, so we can actually get more visibility into connecting the dots with sort of drug product copy number with in vivo copy number and how that translates into HbAT87. It really wasn't easy for us unless we can address the engraftment challenge, and this is certainly the beginnings of that.

Great. And just one follow-up if I can, and this might be too early as well. But do you have any update on what endpoint the FDA may be focused on here with sickle cell?

Yes, you're right. You answered your own question on that one, Salveen. It is a little too early. We certainly are looking at this data, and as the data continues to evolve, with a keen eye on what's the appropriate sort of level of information, if you will. But we are -- we've got to first get more critical mass and consistency, and then we'll engage with the agency. But right now, the one thing we do know is that the regulatory agency has been quite engaged with us, providing, most recently, the RMAT designation. So those would be a pretty significant engagement. They certainly are aware and very concerned around the plight of sickle cell disease in this country and globally. So we feel, at least if we show up with good data, we'll be able to engage in a very active and constructive dialogue about what the appropriate threshold is for a registration study.

And our next question comes from Josh Schimmer from Evercore ISI.

Can you talk about factors that will cause a difference between the vector copy number in the cells infused versus what's measured in the peripheral blood? I mean, clearly, the percent engraftment will be an important determination, but what are the other factors that might cause those numbers to be different?

Josh, you're -- we're having a little trouble, at least I am, hearing you. Can you just restate the question there? Josh, I'm sorry.

Sorry. Just -- I'm trying to understand what factors cause a difference in the vector copy number between the cells that are infused and what's actually measured in the circulating blood, aside from the percent engraftment, what other forces will cause a difference between those numbers?

Yes, thanks, Josh. This is Philip. Well, I think there's a number of factors that I think you're aware of that impact the relationship between the drug product and the peripheral VCN as measured. Some of those include the absolute cell dose. It was provided including, of course, as you point out, the percentage of transduced cells in that dose. And a second feature that we're learning much more about now are the quality of those cells, whether they have the phenotype required to be a true long-term primitive stem cell. And so I think those are critical and across program. And as you've seen in sickle cell disease, clearly, the disease status makes a big difference. And our recent changes to include preharvest transfusions, we believe, make a big difference to the quality of the stem cells recovered as well as the ability of those cells to engraft. So like many things, it's complex or, I meant, multi-variable. But I think that what Mohammed walked you through is a fairly comprehensive set of changes that we're trying to address each one of those criteria.

Josh, at the highest level -- sorry, just to finish off. At the highest level, this is the part of the reason in sickle cell, we've been so actively trying to move from a bone marrow harvest apheresis to sort of a almost top line objective to really address any number of factors there. And we're certainly excited to see some of this cell phenotyping data because that allows us to dig into it and understand it a little bit more specifically and maybe also start to see why the apheresis product in thalassemia is more consistent in that regard. So that's an exciting step forward.

And when you report peripheral blood VCN, what cells are you actually measuring and capturing in that assessment?

So these are PBMCs basically, so it's just the peripheral blood mononuclear cells.

And does a successfully transected peripheral blood mononuclear cell have a survival advantage over an unmodified cell or is it only the red cell lineage that has that?

Yes. So we don't -- given that the transgene will not be active in any, I think, other than erythroid cell. We don't believe there will be any difference in the marking in the peripheral blood versus the bone marrow. So that -- we think that should be a very good marker of bone marrow engraftment.

And our next question comes from Eric Schmidt from Cowen and Company.

Congrats on the steady progress. Maybe just to touch on a question that Salveen asked indirectly, do you think that age in sickle cell disease is a factor in driving better outcomes?

This is Nick. Thanks, I appreciate it. I think you're asking us to speculate here. But maybe I'll cautiously look over at Mohammed here to see if he has some thoughts. But again, time will tell as we start to treat younger patients.

Yes. As Nick said, time will tell. I think we just don't have that data for gene therapy yet. What we can say is for allogeneic stem cell transplant, in general, in particular for patients with sickle cell disease, there is a huge difference between younger and older patients. And certainly, younger patients do much better with allogeneic stem cell transplant. But that is a different kettle of fish. So we will see how this evolves as we hopefully go into younger patients.

Okay. And then just for clarification in the Group C sickle cell patients that was treated with plerixafor. It sounds like we are going to get several patients with in vitro drug product VCN. Is that correct at ASH but none of those patients have yet to be transplanted?

Yes, at the time of -- actually, you're right. We'll have some drug product information, but there will not be in vivo data this year.

Have you treated any of the patients yet, Nick?

I can't really speculate on that. I don't think -- but for sure, we'll give a full update on where the study stands at the time of ASH.

All right. And I don't want to lose focus on bb2121 either. You talked about having additional...

Neither do we.

I know you're going to have 6 months or so of additional follow-up. Are you going to include any additionally treated patients in that cohort? And what should we be looking for in the longer-term follow-up?

Yes. Thank you, Eric. Yes, so we're look -- we're given the provided in the -- we're providing the update on the 21 patients that were in the sort of escalation cohort. And as you know, we have initiated the expansion cohort, but none of those patients will be far enough along to provide any kind of detailed update. So it will be a pretty significant update in the context of 6 more months across these patients to be able to understand everything from durability to response to safety. So on that sense, the good comprehensive next look at the BCMA program, and both Celgene and us are very excited about that.

And our next question comes from Cory Kasimov from JPMorgan.

One question to follow-up on what Eric just asked about, Group C, the sickle cell trial. That abstract was as of July. Can you say whether or not you've enrolled more patients than the 8 since that abstract?

Again, let's not -- Cory, if you don't mind, let's not speculate. We'll provide an update at that time at ASH on the fullness of that. One of the things that is and we know it's across all our studies, right, given all these variables and the prior transfusions, et cetera, and the scheduling and so forth, the emphasis for us is really trying to understand this data and trying to be able to with very few numbers understand where we want to go next, right. And so enrollment is, obviously as you've seen the abstract, is proceeding. And so we'll provide a specific update on that. But the overall emphasis is try to understand this data like we're outlining here to say, we've got an address -- if we're starting to address the engraftment challenge and do we see consistency? If so, then what's the appropriate and most aggressive path that we can from a regulatory point of view?

Okay, understood. And then for the 2 patients that are being updated in the abstract Patients 1312 and 1313, were they receiving regular blood transfusions prior to the required 2 months pre-infusion of LentiGlobin?

No. They were not.

They were not. Okay. And then lastly, just a quick one on the BCMA program. For the 21217 study, the next gen product, is this initial trial -- should we expect it to look much like bb2121 in terms of size and scope of that study?

Yes, it's very much modeled on the initial study with 2121.

And our next question comes from Jim Birchenough from Wells Fargo.

Just wondering for Patient 1313, is it possible to express their T87Q level as a percent of total hemoglobin, just so we can benchmark against the earlier results from Study 205?

Yes, so certainly we will be doing that in the ASH presentation.

Yes, so it's not in there at this point in time. So we can't really speculate on that, Jim, at this point. But you can see, we shared it here, the trajectory of it. But we'll do the same as we've done in the past across the patients. You know the stacked bar gram that will allow you to understand as a percent.

And then just on Study 205, the second and third sickle cell patients seem to be on a relatively better trajectory than Study 206 patients. Will we get an update on those patients specifically? And anything you should -- you can say about what our expectation should be for those 2 additional patients from Paris?

I think the only thing we'd say right now, Jim, is the expectation is yes, we'll provide an update on 204, 205 and 207 and all our studies. And in this case, the sickle cell patients, certainly, we'll update it as they are part of the 205 study. So I can't really speculate on that. But I think what we've been trying to highlight there is, those are obviously patients who had the benefit of prior transfusions and seem to be consistently doing better than those who did not. So again, we're trying to tease out these variables. And we'll provide an update on that in the full context and also the additional follow-up on the Group B patients.

And I think it's also worth remembering that the 205 subjects were not treated with our refined manufacturing process, right. So we think that process is superior and will give different results.

And then just a final question on the cell therapy side of the business. It seems like you've got some interesting preclinical data on gene editing to track locus and controllable cell therapies. When should we see some of those moving into clinical development? When might we see bluebird move beyond BCMA for a cell therapy product?

Philip, would you like to handle that one?

So obviously, we're beginning to show some of the preclinical data that we're very excited about. As you pointed out, both on the use of editing in T-cells as well as controllable CAR T-cells using small molecule drugs. Those are obviously exciting technologies, and as we approach the client with those, we'll be sure to update.

And the next question comes from Matthew Harrison from Morgan Stanley.

I have maybe 2 or 3. So I guess, first question, can you comment on how much additional follow-up you can expect to those 2 sickle cell patients? And then also, I think in these 2 patients, both had 2 doses of drug product. I think that's different than what we've seen with some of the other patients. Can you just comment on the reasons there? And then I have one follow-up.

Yes. So Mohammed, you want to address those 2?

Yes, so I think the follow-up will be incremental. I think you can imagine probably the next major time point for those 2 patients will be presented at ASH. In terms of the 2 drug...

It has to be -- let's be more specific. One of the patients will be close to 9 months, one will be close to 6 months. Just to give you a specific sense, Matt.

And then in terms of the 2 drugs, so actually the vast majority of our sickle patients have actually acquired 2 or more drug products derived from bone marrow. So in general, we have required multiple bone marrow harvest to achieve adequate cell doses for our patients. So this is fairly typical. In the -- I should mention, one of the impetus to moving to apheresis.

Got it. Perfect. That's helpful. And then just one question from the apheresis process, I saw on the abstract that one of the patients, I think, had a VOC event after the apheresis. Can you just comment sort of the clinical characteristics around that event? And if you think there is any relation to the apheresis process or not?

Yes. So I mean, I think, sticking to largely what's in the abstract, this particular individual did have VOC associated with their bone marrow harvest. So it's hard to speculate exactly what the mechanism of this was. But as we also highlighted in the harvest, they were numerous VOCs that occurred in other patients post bone marrow harvest as well, too. So it was a fairly, unfortunately, a frequent occurrence after obtaining cells.

Nothing new on the apheresis side.

Yes.

And our next question comes from Dane Leone from BTIG.

Two from me. The first one being, for the patients enrolled in the plerixafor study, were -- was there any on constrain on having a VOC in the prior 12 months? And what is -- can you be more specific in terms of what you were looking for from a safety perspective that gave you comfort to move it into the full use format for the study?

Yes, that was a great question. So I think, the first question, no. I mean these patients in the plerixafor arm were, in general, typical for our non-plerixafor arm. They had VOCs. Some of them had lots of them prior to getting mobilized. The only major difference is we did require this 2-month lead-in period of transfusions, which can be helpful at suppressing VOCs. In fact, that's one of the major reasons that was implemented. It was to suppress VOCs during this 2-month period. So that was the only nuance in that respect. And then the second question -- sorry.

Safety.

Oh, yes. Well, we're looking for safety. So all safety events, we were monitoring these patients for any potential complications. Now most relevant, obviously, is the GCSF experience in which patients develop very significant increases in their white blood cell count and then, life-threatening crisis after receiving GCSF. And so were obviously very cognizant of what these patients' white cell counts were doing and monitoring them for very severe sequelae. And I can tell that these VOCs that these patients -- that this 1 patient did have that you commented on did not follow the criteria or follow the same characteristics as the GCSF VOCs.

And so in your perspective, as it's developing with now the clinical data and the formation of what you think will be an ideal protocol for sickle cell. Can you update us on your thoughts around what would be a clinical outcome measure that regulators or what you think clinicians would gravitate towards with these patients?

Yes, this is Nick. This is -- again, we really are trying to be cautious about not speculating. Once we have the fullness of the data here at ASH, we can probably have a little more extensive conversation on that. But again, part of this is aggregating just enough information then we can get a sense about how the different variables are behaving and the consistency of the outcomes sort of what ties to what. It will allow us then to sort of get more specific around which of the various -- I wouldn't call it standard, but what are the various options of endpoints in pivotal study for sickle cell disease to engage in that debate first and foremost with the agency. So too early to speculate on that. But this is certainly a step in the right direction. Like I said earlier, we feel the agency will -- has shown signs in sickle cell to be very open to engage on some creative options to really do the best we can for sickle patients.

And our next question comes from Michael Schmidt from Leerink Partners.

I had just a follow-up on the 2 sickle cell disease patients. The -- remind me what the HbAT87Q production was for Patient 1204 at 3 months. I think you said it was lower than what you saw with this 1313 Patient?

Yes, marginally lower. So I don't have it on top of my head, but it's approximately...

It's [42.5].

Yes, it was probably about 20% less, I think, versus Patient 1313.

All right. And then is it the absolute hemoglobin -- anti-sickling hemoglobin production that's important or the relationship to the overall Hb levels?

Well, the answer is probably a bit of both, right. I mean, I think, obviously sickle cell anemia is at some level an anemia, an increasing level of globin is important to solve that problem. Then there's the problem of the sickling issues, and that is going to be corrected by having a substantial fraction of that globin being anti-sickling in nature. So I think that we're looking to achieve both. And as Mohammed previously said, we're targeting around 30% or hopefully more anti-sickling globin. But hopefully, that will come with a correction of the total hemoglobin levels to a non-anemic set point.

Sounds good. And then question regarding your Regenerative Medicine Advanced Therapy Designation, and I'm just curious what the impact of that is and maybe how that might differ from something like a breakthrough therapy designation?

Would you like to address that, Mohammed?

Yes. I think this is -- certainly, speaks to bluebird's very proactive strategy for engaging with regulators to get as much as frequent feedback as we can. And I think it -- this is an opportunity for us really to engage more frequently with the FDA to get regularly feedback on what they think of our trial design and our potential endpoints.

Yes, and one of the main emphasis here is to try to, obviously, early dialogue. But also what's the appropriate development path? How do you think about what should be registration study sort of, at what point or what time should? It's pretty specific. The good news is, we've really have had a very good relationship with the agency that does spectacular in engaging with us across, whether it's manufacturing or clinical. So this is certainly an important step. But I don't think it fundamentally changes how we interact. The key strategy that we have is, we like to have data before really engaging with the agency in a substantive way. So we get out of speculation as opposed to here's what we think is going on and here's what we think is the right path forward based on the data we're seeing. So obviously, this is a nice step in that direction. We're happy. You'll hear more at ASH. And then as soon as we can, we'd comment more specifically on how we're thinking about U.S. and European development plans for sickle.

And our next question comes from David Nierengarten from Wedbush Securities.

I can ask for lots more speculation here, but I'm actually curious about the CD34 cells in that CD34 DIM phenotype. First off, is it known in prior studies with plerixafor that it generates more primitive stem -- CD34 cells? And then, similarly on that atypical CD34 patient from Northstar-2 that you look at those stem cells and where they, again, a different phenotype with, particularly, low number of -- or high numbers, sorry, of the DIM cells?

Thanks, David. This is Philip. So starting at the top, we've, on the research level, been very interested in understanding the phenotype of CD34 cells and how that correlates with features such as engraftment, and ultimately in vivo VCN. As you pointed out, I think the data that we have on comparison of bone marrow-derived stem cells and sickle cell disease versus apheresis is actually new. I don't think there's -- other people have done that sort of study. So I think that's -- there's nothing to really sort of reference to, although plerixafor is, of course, used usually in combination with GCSF and stem cell mobilizations. And one of the reasons that's done is to improve both the number and quality of the stem cells in those standard mobilizations. With respect to sort of the Northstar-2 subject, these are exactly the same types of analysis that we're doing now across the board. So we're looking for the level of CD34 expression, which we think is a good characterization of the stem that's present in those cell populations as well as whether there are any definitive lineage markers expressed in those cells. And you remember that we mentioned that type of analysis, specifically with respect to that Northstar-2 subject. So this is all part of the same, and I think it's important to understand the drug product as well as we can to be able to predict the outcomes in patients across both thal and sickle.

And David, it's Nick. The thing we've always said and it's glaringly obvious, right, is the major difference between thal and sickle is where we get the cells from, right. And so this is why we've spent so much time trying to understand this. And this is an unusual situation to be able to compare across both. So in that sense, I think it is very new and also very interesting. It really starts to speak to, eventually, being able to figure out how do you think about dose? And how is that different in bone marrow versus apheresis? And so having both sides to compare against is, is it's going to be, we think, very meaningful in starting to unravel or unveil sort of a more fundamental understanding of what's going on. So more to come on that, certainly, as we continue to get more patient data to be able to look across these variables.

And our next question comes from Matthew Luchini from BMO Capital.

So two from me. First, on the plerixafor mobilizations side, looking at the 3 patients, I noticed that one is -- you've got a little bit of a range, one patient slightly lower at 5.6 times 10 of 6 cells. I was just curious if you could provide a little bit more color on this patient, if there's anything that was different about him or her from a baseline point of view or is it experience. And then secondly, on Patient 1312, obviously, it's early here, but we've got peripheral VCN at about 2.6 copies for genome at 1 month. I just wanted to hear from you a little bit about how you're thinking of the significance of that time point. And maybe you can remind us where Patient 1204 or maybe even Patient 1313 were at that same time point.

So maybe, Philip, you can address the first one, the sort of the variability if you will?

Yes. With respect to plerixafor mobilization, obviously, we anticipate some variability from patient to patient. What I can say about that particular subject, which is in the abstract this will be, the absolute mobilization in the peripheral blood was lower, and that obviously explains the harvest dose -- the harvested dose was lower. So it was simply that the drug didn't mobilize as effectively directly in the patient. With the -- perhaps, Mohammed, wants to comment on the 1313 subject?

1312 at 1 month. So it's obviously an impressive in vivo copy number at that stage. And so I think that is certainly considerably higher than others.

I mean, it's safe to say that's very promising because one of the major issues that we've had in 206 was sub optimal engraftment of our drug product transduced cells. And so this patient definitely appears to have robust engraftment of the transduced cells. And then -- so as we've alluded in the abstracts and we've stated before, I mean, comparatively speaking, this is a much better preservation of vector copy number than we've observed in the other 206 studies and comparable to what we saw in 1204.

And even 1204 at this time point, Matt, it wasn't even -- I think, it just trying to engraft at that point. So in that sense, it really -- it's not even really a comparison. That's what it's a little tricky to look at the way we're displaying this just because it's at the last measurement, so it's not -- you're comparing different time points. But the point here is the same, where you sort of ultimately get to them. The fact is that these patients at 2.6 at 1 month is a little bit sort of almost off -- it's off the charts.

And our next question comes from Ying Huang from Bank of America.

I have one for sickle cell. So you guys have done a good job in terms of improving the VCN number in drug products, but what are you trying to improve in terms of seeing a higher peripheral VCN number in the patients for sickle cell? And then at ASH, would you have already finalized the pivotal product called for bb2121? If that's the case, would that include also patients with early lines of myeloma instead of only heavily pretreated patients in Phase I?

So thank you. I can't really address the second question there. We will provide an update at ASH on the study I outlined. As far as the development plans and where we go with early lines of treatment, that's a -- that will be for another time. So we're certainly actively discussing and debating what the data merits in our dialogues with Celgene, who is obviously extremely involved and driving big pieces of that program. So on the first question there, I think maybe, Mohammed, you can hit it, but it really hits all the key points that we've been addressing today.

Yes. No, I mean, I think it's -- to summarize sort of what was discussed, I think in terms of addressing the loss of vector copy number going from the drug product to periphery, I think certainly one thing we can do is increase where we start from, so that we have a better peripheral vector copy number. We're trying to increase the cell dose, so hence, the plerixafor mobilization is we hope points to a huge step. We're trying to increase or improve the purity and the stem-like nature of the drug product, and hence, again, the plerixafor mobilization will be helpful. We are trying to both reduce inflammation in the sickle bone marrow as well as to prime it optimally for reception of the transduced stem cells. And hence, we've implemented the 2 months prior transfusions and made the myeloablation more rigorous.

So overall, it's just a -- it's a combination of all these factors. Exactly what the contribution of each one is there but effectively do the frame shift and then sort of try to reduce the slope. And that's what I think we've seen a nice step towards here.

And our next question comes from Vincent Chen from Bernstein.

A couple of questions on the plerixafor mobilization. First, the VOC seen in the plerixafor patient, is your sense that this is related to plerixafor mobilization? And is there a world where you might consider moving to a lower dose even if that can potentially require multiple rounds of apheresis or a plerixafor mobilization? Also were patients transfused prior to the plerixafor? And what level of hemoglobin S did they have prior to receiving plerixafor treatment? And then I have a follow-up.

Yes, so answering the first -- the second question first. So I think our goal -- and I think this is based on extensive discussions with people who do this is to get the hemoglobin S down below 30%. And so this patient was and all of our patients are transfused to get the total hemoglobin S below 30% prior to beginning the process to minimize the probability of vaso-occlusive crisis. As to the speculation of what causes patients to have a vaso-occlusive crisis, the biology is too complicated necessarily to come up with a definite answer here, so it'll be speculative. All we can say, this patient did also have vaso-occlusive crisis after their bone marrow harvest.

I think that's an important point here. It's just that we didn't view this nor did the data safety monitoring board view this as of any significant concern in that regard, right. Because it was not something that was unique to the apheresis per se. It's also something -- remember we're doing both of these treatments on these initial patients because the apheresis on was really a research purposes and safety purposes. So that's -- from our perspective was not something that got in the way of us moving forward.

I see. That's great. And then with respect to cell quality, how do the cells generated from apheresis compared to the bone marrow harvest ones, in terms of some of the other commonly applied assays, for example, their ability to reconstitute a bone marrow in humanized mouse models or differentiation assays?

Yes, this is Philip. So there's a fair amount of history on plerixafor-mobilized stem cells. And the short answer is on those publications. They behave very similarly to other standard methods of mobilization. So GCSF and, of course, GCSF plus plerixafor. So there's really -- we don't believe this process is substantially different. So I think that's probably the most important feature. So from an apheresis perspective, I think that this specific way of mobilizing, I think, has great promise. And we've shown that in the sickle population, that's really the -- sort of the new piece.

And our next question comes from Biren Amin from Jefferies.

Did Patients 1312 and 1313 have a dilution of the alpha globin gene similar to Patient 1204?

Not to our knowledge. This has not been documented.

And so how do you think these sickle cell patients with [pulmonary] alpha thalassemia would compare to patients that don't exhibit this trait?

Yes, it's an interesting question. I mean, I think the short answer is, we'd have to actually have a reasonable number of patients with this deletion to really make a definitive statement. Yes, I think that's about all I can say.

We can't really speculate on that, Biren. We need more data to understand the nuances of that.

Okay. And then just on the licensed Boston Children's program, what's the thought on how that would be strategically positioned compared to LentiGlobin?

Yes, I think our focus here has been around do everything that we can to leverage the platform that we have and the technology that we have in places as Philip outlined. So we are to some degree sort of agnostic about how we do it. So we want to really do the best we can for sickle cell patients. So when we saw this opportunity here with actually one of our already collaborators on the ALD program and broadly speaking, we thought it was an exciting approach. And so wanted to take a look at it. And as Philip just outlined, that some of the data here is actually pretty exciting. So at the end of the day, there is -- to have more sort of options for sickle cell is our objective here. And ultimately, how these play out, we'll have to see. It will be completely data-driven. For right now, we're certainly full speed ahead on both.

And our last question comes from Gena Wang from Barclays.

So I know it's still evolving, but just wondering if you have any sense like what would be the peripheral VCN you are looking for that will be a good indication for clinical benefit?

Yes, I mean, I think -- this is Philip, so obviously we're -- that final number is, too, going to be data-driven based on what we actually see. But we've certainly had a goal of trying to get into that 0.5 VCN and above range. And so far, that's been -- at least been the goal. But obviously, the final target, if you will, is going to be driven by the T87Q and the relationship with that VCN number.

Yes, Gena, this is Nick. It's also -- may vary a little bit from thalassemia, the sickle cell, and obviously, it may vary in between. The hope is that we can consistently get above the threshold that we'll deliver an important outcome for all the patients. But it's too early to speculative what exactly that number is. But the good news is, our ability to deliver consistently higher and higher and higher is something that we think is a very good thing.

Great. And I have one question regarding the gene editing in sickle cell. So how do you see yourself compared to the competitors in this space? I know there are 2 independent programs that will go to clinic later this year, if you can give some thoughts.

Yes, so just to be clear, the program that we talked about that regulates BCL11A is the lentiviral program, so that's using our core lentiviral technology. Although it hits the same target as many of the gene editing companies we're going after, which is BCL11A. I think that the editing approach is -- are interesting in the sense that they try and achieve a similar outcome. They try and introduce mutations, which drive an HbA [FH] phenotype. The challenge that the editing approaches have is creating a sufficient number of the desirable mutations that drive that phenotypic outcome. So every lentiviral vector that enters and introduced into the cell is if you will fully penetrant to completely changes the biology of that cell, every mutation you make in a gene has to be the right mutation before it actually drives the phenotype you're interested in. And I think that's one of the reasons we're seeing the 70%, 80% change in F versus the 30%, 40% that has been reported by the gene editing community. So we think that program is very competitive. We're also obviously very interested in gene editing ourselves. And we have MegaTAL approach, which we think has some advantages, both in terms of specificity and in terms of how easy it is to deliver, the single polypeptide to cells. And we're working feverishly on that, too. But for today's abstract, it's to see the LentiGlobin BCMA program that we're highlighting.

And I'm showing no further questions in the queue at this time. I'd like to turn the call back over to the speakers for any final remarks.

I just want to thank, everyone, for taking the time. And obviously, if you have any follow-up questions, we'd be happy to take them at any point, just reach out to myself or to Liz, and we will respond. Thank you very much.

Ladies and gentlemen, this does conclude your program for today, and you may all disconnect. Everyone, have a great day.