Precision BioSciences, Inc. (DTIL) Earnings Call Transcript & Summary

Good day, and thank you for standing by, and welcome to the Allogeneic CAR T Update Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. [Operator Instructions] I would now like to hand the conference over to Alex Kelly, Chief Financial Officer. Please go ahead, sir.

Thank you very much, and good morning, everyone, and welcome to our webcast to discuss the progress that we've made with our allogeneic CAR T programs. Before we begin, I'd like to remind you that the statements that we make today on the webcast that do not relate to matters of historical facts or future expectations and our forward-looking statements. Our forward-looking statements are within the meaning of the Private Securities Litigation Reform Act of 1995. Such forward-looking statements involve known and unknown risks, uncertainties and other important factors, including, without limitation, the risk factors and Precision's first quarter 10-Q. All forward-looking statements speak only as of this date, except as required by applicable law, Precision has no obligation to update or revise any forward-looking statements made today whether as a result of new information, future events or change in circumstances. With that out of the way, I'd like to introduce our call participants. First, we have Michael Amoroso, our President and CEO. Next, we have Dr. Alan List, our Chief Medical Officer, who will review our clinical stage CAR T programs, including the latest interim data for PBCAR0191. And we also have Derek Jantz, our Chief Scientific Officer and Co-Founder, on the call to participate in the Q&A. With that, let me introduce Michael Amoroso.

Thank you, Alex. Good morning to our investment community. Thank you for being with us. We're excited to be with you today. As ASCO draws to a close and a few weeks before EHA and summer kicks off. As Alex stated, my name is Michael Amoroso. I'm the CEO of Precision BioSciences. And today, I'm joined by my esteemed colleagues Dr. List and myself who will give you the presentation this morning. And Derek and Alex will be available to the live question and answer. We plan on spending about half the time presenting and half the time for questions. So we'll make sure we have plenty of time to answer anything that's on your mind. Our teams have been fast at work since the momentum at ASH in 2021, advancing our allogeneic CAR T programs, and we're excited to share the progress with you today. With that, I will open up and give some frame of reference of what the team has been doing, and then I'll turn it over to Alan to give us an update on the data. First, a reminder of our identity, Precision BioSciences. We are a gene-editing company first and foremost. Our wholly owned ARCUS proprietary platform created by Derek and our CTO, Jeff Smith, is what we believe to be the preeminent gene editing platform in the world. ARCUS is unique as it originates from the homing endonuclease, an enzyme called I-CreI, that has existed in nature for generations and exist in an inactive form until our world-renowned protein engineers redirect the enzymology for specific gene editing targeting purposes. The ARCUS platform maintains distinct advantages over other gene editing technology, namely safety, logs, lower off-target editing. Gene edition as opposed to not only just gene deletion, and last, ARCUS is small. So it's advantageous for delivery to many tissues beyond just the liver, for example, muscle and CNS. Delivery to these tissues can be carefully chosen within ARCUS, whether it be through a lipid nanoparticle or an AAV vector. ARCUS is the foundation that enables our gene editing ex vivo, the focus of today's talk for our allogeneic CAR T creation, 1 dose, 1 single gene edit which we believe gives us significant advantage in product attributes as well as ARCUS is the backbone of the other side of our business for rare and other diseases, in vivo gene editing for possible onetime gene corrective cures. Today, we'll focus only on our ex vivo host-derived, off-the-shelf allogeneic CAR T lead programs. What you can see here on the next slide is the focus and goals of our multiple allogeneic CAR T programs that Dr. List's teams are currently executing through the clinic at various developmental stages. First, what we call our first-in-class approach is PBCAR0191, which is designed to meet an urgent need for a patient population where we've seen unique and promising results, first unveiled at ASH 2021, the CAR T relapse population. These are heavily pretreated patients, median lines, 5 prior therapies. So double the treatments where autologous CAR Ts are used as standards today as well as where most allogeneic products are being focused and pursuing. These patients truly have the poorest outcomes of all aggressive lymphoma patients. These patients have urgent need and after generating a very unique single at ASH where PBCAR0191 worked quite well in a 6-patient subset of the CAR T relapse patients, we fully dedicated PBCAR0191 and our teams in this pursuit of imminent need. Of note, PBCAR0191 knocks in a CAR at the track locus and out the TCR using our ARCUS proprietary editing technology. Today's data presentation will focus primarily on this first-in-class program. Now in parallel, on your right hand, we're also studying our earlier lines of DLBCL, second and third line, and relapsed/refractory multiple myeloma patients across both our CD19B stealth cell and our 269A BCMA-targeted allo CAR T programs, respectively. Today, in both of these settings, autologous CAR T is the standard of care and the goal of both of these programs are to displace an autologous CAR T with an off-the-shelf, more accessible option with hopes of also improving upon the approved autologous product therapeutic indexes. Today, we will give you operational updates for the next steps in these programs, and our data presentation will focus primarily on PBCAR0191 and CAR T relapse patients. Today, Dr. List will review 12 patients, the 11 evaluable who received PBCAR0191 in only the CAR T relapse setting, again, a heavily pretreated group with a median of 5 prior lines. We'll show 12 patients across 2 cohorts. Starting with 6 patients at ASH, known as the ASH cohort we'll refer to, with dose level 3 and enhanced lymphodepletion, higher intensity chemotherapy and now validated and supported by 6 additional patients treated this year with Dr. List's team in 2022, which we'll call the new cohort, who received a higher dose of PBCAR0191 from our optimized and next stage manufacturing process with lesser lymphodepletion moving down on the density intensity of chemotherapy. At a high level, the results to be reviewed were as follows: 100% overall response rate. 73% of evaluable patients had complete response. Duration of response greater than 6 months of 50% driven by our ASH cohort who have longer outcome data now as we update 6 months later, show a 50% durable response greater than 6 months. Dr. List will also show you improved safety at the new cohort dose level 4B with less, what we call it, modified lymphodepletion that we used in the past during the ASH cohort. Safety results were as follows: greater hematologic recovery versus the ASH cohort, reduced grade 3 infection, no CRS and limited Grade 3 ICANS. And you'll also hear from Dr. List why we think we can still do better and lessen lymphodepletion a bit further to standard levels to optimize the therapeutic index for this very fragile patient group. In addition, Dr. List will talk to you about PBCAR0191 optimized manufacturing process at dose level 4B, the new cohort, while also lessening the lymphodepletion, which we think make our allo CAR T programs quite unique if you scale the landscape. He will show you a proposed reason for our robust efficacy results with durability that we have reached our pursuit of CAR T peak expansion based on flow cytometry, looking quite similar to the peak CARs that were reached in autologous CAR T studies in the patients who had longest-term responses and potentially cure. Last, Dr. List will focus on what's next, starting with our plans for the next cohort dosing as well as plans for an interaction with the FDA later this year to seek guidance on a path forward. Before Alan shares the data, I'd like to share a few details around our Precision Bio's allo CAR T platform approach. Additional information on our optimized CAR T cells and why this matters. And finally, context on the patient population that we're proud to be studying, the growing population of CAR T relapse patients with urgent need. You will see before you now is Precision surgical and prospective pursuit of best therapeutic index, the 5 levers for success as we call it. First, our unique CAR T construct. Then our manufacturing expertise, resulting in favorable cell product attributes for characteristics, a deep understanding of what components we do and do not want in our CAR T products. Next, our predictable dose. Of course, what matters most about dose is product attributes and characterization. The lymphodepletion regimen or lever, as with autologous CAR T, it is clear that lymphodepletion is an important component of CAR T treatment to enable and propel peak expansion. And last, most importantly, the heterogeneity of the patient selection, you must match with all of the prior levers. As the most important component of our success, after all, our vision is to get to safe and effective cures with onetime treatments. Now as I've told many of you over the past few months, I'm very proud of the expertise we have in-house and the groundbreaking science we do at Precision each day, and the teams have been very busy since ASH. The goal coming out of ASH was how do we continue to pursue our very unique and robust efficacy signal in CAR T relapse patients fast as they have urgent need, but even more safe and equally effective. You will see our novel PBCAR0191 construct has a single-gene edit. We believe this is vitally important for efficacy and safety and a unique Precision owned proprietary co-stimulatory domain known as N6. Next, the arrows will connotate the choices we made. We've continued to implement and optimize our manufacturing process to create a yield of only the types of T cells we desire most. We then pushed the PBCAR dose to level 4B and took the important step of lessening lymphodepletion to clean up infection rates as this is in a highly lymphodepleted population with 5-plus lines of therapy. Safety is also paramount. Last, patient selection, as Dr. List will show you the result of the right patient, right permutation of these levers in a few moments. In summary, since ASH, we optimized product attributes. We went up in our PBCAR0191 dose and we went down on lymphodepletion in order to reach the optimal therapeutic index pursuit for these very sick patients. Next, a bit more about optimizing CAR T cells and why we're trying -- what we're trying to accomplish. You will see the key driver or attribute of optimized manufacturing processes we have been employing. Our goal is simple, achieve highest peak CAR T levels with all products in the peripheral blood early. Data published in late 2020 from Blood Advances confirms the only key variable of CAR T pharmacokinetics that prove consistent in long-term outcome patients who received autologous CAR Ts was achieving a peak CAR level within the first 2 weeks that can suppress circulating tumor DNA very early on in treatment. In fact, you'll also see from this slide that CARs at a later time point, persistence as we often refer to it, while, of course, is associated with a higher peak and area under the curve, did not independently have an impact on long-term positive outcomes. It is CAR T peak that Dr. List and Dr. Jantz' scientists have been focused on for the past 2 years. Next, we see in order to achieve the desired early CAR T peak to achieve potential cures, we must know what types of cells we want for efficacy and what types of cells we want to limit for toxicities. Less differentiated versus more differentiated cells. Now one may ask if this knowledge is available to all, why is there varying degrees of success across the allogeneic product portfolios and companies. The answer, we believe, is twofold. It comes down to people and expertise as well as the right toolkit for those people to employ. We have built quite a cell therapy expertise with the expert group at Precision Bio, and our talent is enhanced through their toolbox, ARCUS. ARCUS is the only gene editing tool that makes or enables allo CAR T with one single-gene edit. Something we believe is foundational to our increasingly optimized single doses of PBCAR0191, and hopefully, all of our other CAR products that are following behind 0191 in clinical stages. Last, before I hand it over to Alan, I'll take a few moments to discuss the CAR T relapse urgent need one final time. Since ASH, many of you have asked me questions. Is this a small niche of patients? Why all the focus on the relapsed CAR T population? The shortest answer I can provide is patients need us urgently and the time has never been more precarious. The CAR T relapsed population is expected to grow fourfold to fivefold over the next few years on the back of autologous CAR T becoming the second line standard of care over auto transplant in diffuse large B-cell lymphoma. On the slide, you see the relapsed population today in the maroon color. Based on first indications for auto CAR T in later line, third-line plus. But as we project 2025 in the oncoming years, this population will grow immensely and so too will their need. These patients need us now. Let us take a look at their current day outcomes once auto CAR T fails them. What you see is no true FDA approved standard of care in this population. A conglomerate of salvage therapies being used and finally and most importantly, very underwhelming clinical outcomes, progression-free survival of 1 to 2 months with an overall survival of 4 to 5.9 months. In fact, 1/4 of these patients only go to best supportive care. Finally, as I hand the baton to Alan, I'll show you our clinical teams always practice evidence-based medicine and disciplined drug development, and we will only pursue regulatory paths if we really feel we can move the disease treatment paradigm forward for patients. What you see here is a minimal acceptable as we see it at Precision BioSciences' target product profile in the pink color. Now please do not misunderstand, we always aim for cure even in this very sick heavily pretreated population. But we also want to be laser focused on the product profile we must pursue or achieve and beat to bring solutions to this CAR T relapse population. You will see in gray to the right, not the focus of today's presentation, but the DLBCL third-line plus profile of autologous CAR T that many of you often ask me about. This profile would be more relevant to compare our stealth cell pursuit. As we look at displacing auto CAR T, in fact, third and eventually second line. But for today's focus, the pin target product profile in the more heavily pretreated population of CAR T relapse with 1 to 2 months of progression-free survival and 4 to 5.9 months of overall survival is our key focus. Dr. List, please take us through the data.

Okay. Thank you, Michael. Good morning, everybody. In our last update at ASH, we shared encouraging results from -- with PBCAR0191 in 6 patients who relapsed following treatment with a CD19 directed autologous CAR T. Those patients received dose level 3 or 3 million cells per kilogram with our enhancement lymphodepletion regimen that consists of a fourth dose of fludarabine and 3 doses of cyclophosphamide at 1,000 milligrams per meter squared. All subjects responded with 3 of 5 patients evaluable at the time experiencing remission inversions compared to their autologous CAR T response. In other words, the duration of response to 0191 was longer than the original auto CAR T response. This encouraging signal in the autologous CAR T relapsed population was associated with deeper host T cell [ maters ] as well as delayed T cell recovery, which together mitigated rejection of 0191, consistent with prior reports that patients who relapsed following response to a CD19-directed autologous CAR T typically have poor product cell fitness due to inherently impaired immune integrity. As a result, these auto CAR T relapse subjects experienced earlier and higher peak cell expansion with more prolonged persistence compared to the autologous CAR T naive subjects that were treated with the same cell dose and lymphodepletion. Although the results were encouraging for this high-risk population, there is a higher frequency of prolonged neutropenia and infection with the enhanced lymphodepletion regimen. Since then shown here in blue, our goal was to improve the safety of the lymphodepletion regimen by lowering the cyclophosphamide dose to 750 milligrams per meter squared to reduce their duration of myelosuppression and the associated infection risk in this relatively fragile patient population. And secondly, to improve efficacy by reaching potentially curative levels of peak CAR T expansion, we increased cell dose to dose level 4B or 500 million cells as a flat dose given the absence of Grade 3 or greater CRS at dose level 3. With these changes, our intention was to treat an additional 6 patients to assess the impact of these changes while monitoring durability of response in the ASH treated cohort. The demographic features of the dose level 4B subjects are shown on the far right-hand column and the dose level 3 subjects on the left. Overall, these are heavily treated patients with a median of 5 lines of prior therapy. The distribution of aggressive lymphoma histologies are comparable between the 2 groups, with the exception of one subject who had B-cell ALL in the ASH data set that relapsed following 2 allogeneic stem cell transplants and the autologous CAR T. The notable differences between the patient groups are the higher median age of the dose level 4B subjects of 72 years compared to 50 years, while the frequency of prior stem cell transplant was higher in the ASH cohort. A high-level summary of responses in the 12 total CAR T relapse subjects is shown here. Among the 11 patients currently available for response at day 28 or later, each experienced a response to PBCAR0191, including a complete response in 8 or 73% of subjects with response duration currently exceeding 6 months in 3 or 50% of the initial 6 subjects in the ASH update. Six patients or 55% remain in an ongoing response. Overall progression-free survival currently exceeds the median for standard of care that Michael showed you in 7 or 70% of the 10 evaluable patients. The swimmer plot provides further detail regarding the depth and duration of response for each patient, as shown here according to cell dose. As we shared in the previous slide, each of the 11 evaluable subjects responded with 8 complete remissions and 6 subjects with ongoing response. Patient A at the top of the slide, for example, is the ALL subject who had relapsed following 2 allogeneic stem cell transplants as well as the autologous CAR T and remains in an MRD-negative complete remission that is ongoing in 18 months. Overall, 4 of 6 evaluable subjects have experienced remission inversions, meaning the duration of response to PBCAR0191 surpassed that for the prior autologous CAR T. Knowing the published progression-free survival in this high risk population is only 1 to 2 months as reflected by the dotted line. These preliminary results are highly encouraging. Okay. Focusing now on the swimmer plots for the 6 dose level 4B patients treated with lower intensity lymphodepletion, you can see that each of the available subjects achieved a complete response. Four of the subjects had relapsed following response to a CD19-directed autologous CAR T and 2 relapsed following treatment with an allogeneic CD19 CAR T. As you can see, each of the 5 evaluable subjects achieved complete response despite reduced intensity of the lymphodepletion regimen likely a result of the higher cell dose and optimized PBCAR0191 cell attributes. I'd like to point out that subject L who has the longest follow-up experienced a partial response lasting only 28 days with the initial CD19 CAR T treatment and is now approaching 5 months in complete response. Subject J, who died suddenly on day 23 had a CT scan on day 21 that showed complete resolution of the prior disease. PBCAR0191 cell expansion is shown on this slide, as measured by flow cytometry. As you can see in the blue curve, mean peak cell expansion with dose level 4B and the lower dose lymphodepletion was threefold higher than those subjects reported at ASH, accompanied by a 3.5-fold increase in the area under the curve with abundant CAR T cells still circulating through days 21 through day 28. The clinical relevance of such robust expansion is illustrated here by comparing PBCAR0191 peak expansion in the autologous CAR T relapse patients by cell dose to that reported in the ZUMA-1 (axi-cel) trial according to response and relapse status. On the far right-hand side of the figure, you can see that the peak expansion for dose level 4B subjects receiving the lower intensity lymphodepletion either matched or generally exceeded the median peak expansion in the durable responders in ZUMA-1, illustrated by the dotted blue line. Whereas peak expansion for dose level 3 subjects clustered around the median of the ZUMA-1 responders who relapsed shown by the dotted red line. These data indicate that the peak cell expansion achieved with dose level 4B and reduced lymphodepletion offers the prospect for long-term durable response following auto CAR T relapse. We are aware of no other allogeneic CAR T that has demonstrated this level of peak cell expansion. Okay. Let's transition now to the safety data. Select adverse events of special interest according to PBCAR0191 dose level and grade are shown in this table. Despite the robust cell expansion, there was no cases of Grade 3 or greater CRS in either dosing cohort -- CAR cohort while there was one subject experiencing Grade 3 ICANS at each dose level that resolved to Grade 1 within 48 hours. There was no evidence of graft-versus-host disease consistent with the known insertion of our CAR into the T cell receptor alpha gene locus. The frequency of Grade 3 or greater infection was much lower in the dose level 4B cohort, who received the reduced dose of cyclophosphamide. Occurring in only 1 subject or 17% compared to 4 or 67% in the enhancement of depletion cohort. Despite the reduced infection rate, there were 2 deaths on study at dose level 4B that were each associated with late occurring encephalopathy that is suspected related to fludarabine-associated neurotoxicity. This patient population is at particular risk by virtue of prior fludarabine-based lymphodepletion for the prior CAR T. And both patients had neurologic symptoms that emerged at day 21 or later that did not resolve with corticosteroids. Both patients had features placing them at higher risk for fludarabine neurotoxicity as reported in the stem cell transplant literature, including age over 70 years prior neuro inflammatory insult related to ICANS of Grade 2 or 3, prior intrathecal chemotherapy treatment and hypertension. One of the 2 deaths occurring in the enhanced lymphodepletion cohort also had delayed encephalopathic changes, suspected related to fludarabine neurotoxicity. Because of these adverse events, we will return to a 3-day fludarabine lymphodepletion regimen going forward to maximize safety. Consistent with lower frequency of Grade 3 infection that I showed you earlier, hematologic recovery with a modified lymphodepletion occurred earlier than was seen with the enhanced lymphodepletion with a median neutrophil recovery to a safe level above 500 per microliter occurring between days 10 to 14, as shown here in the blue curve in the figure. To summarize what we have shared, the efficacy of PBCAR0191 and CD19 CAR T relapse subjects exceeds results reported with the standard of care. The overall and complete response rate as well as preliminary durability validate the original signal reported ASH in CAR T relapse subjects with 100% of subjects responding and 73% achieving a complete response, including all 4 evaluable subjects treated at dose level 4B. Greater than 6 months durability was evident in 50% of the ASH cohort that received the lower cell dose ranging from 7-plus to 18-plus months with an additional 4 ongoing responses that have not as yet reached 6 months follow-up. Optimization of PBCAR0191 functional attributes and higher cell dose achieved a desirable and competitive therapeutic index for CAR T relapse patients despite reducing the intensity of lymphodepletion. Peak CAR T expansion in these subjects reach levels achieved in the auto CAR T patients experiencing durable responses and probable cure. More rapid hematologic recovery was associated with reduced frequency of Grade 3 or greater infection. However, the 2 deaths at dose level 4B was suspected fludarabine-associated neurotoxicity indicate that the fourth dose of fludarabine in this CAR T relapse population imposes excess nonhematologic risk. Given the robust expansion achieved with dose level 4B that generally exceeds the median observed and durable ZUMA-1 responders, we will transition to standard lymphodepletion to minimize toxicity in this relatively frail in heavily treated patient population. Moving forward, in the second half of this year, we are expanding the number of participating institutions to accelerate enrollment to the study with plans to request an end of Phase I meeting with the FDA before year's end and discuss possible registration strategy. Let's transition now to our best-in-class program, 19B and our BCMA program. PBCAR19B or stealth cell is our candidate for best-in-class allogeneic CAR T to compete with second- or third-line CD19-directed autologous CAR T. For those not familiar with stealth cell, there are 2 additional genetic edits to the CD19 CAR that were incorporated to evade host rejection by T and NK cells. These include an shRNA to beta-2 microglobulin that reduces MHC Class I antigen expression to prevent rejection by T cells as well as insertion of an HLA-E transgene to prevent rejection by NK cells. This is accomplished by incorporating all of the cargo into the CAR vector that is inserted into the track locus thereby accomplishing this with only 1 gene edit. As we shared at ASH, 3 patients were dosed at the first dose level, 270 million cells with standard lymphodepletion this past year. And due to the performance of 19A as a potential first-in-class product, 19B was strategically paused to implement manufacturing process improvements, suitable for best-in-class therapy. We have now completed production of lots with improved product attributes and yield that are expected to be released by midyear, permitting us to commence dosing in dose level 2 in the third quarter of 2022. And finally, PBCAR269A is our BCMA targeted allogeneic CAR T that we are currently investigating in combination with a gamma-secretase inhibitor, nirogacestat in relapsed/refractory multiple myeloma. We completed dose level 2 or 2x10 to the 6 cells per kilogram with nirogacestat that showed clear biologic effect that was evidenced by peak CAR T expansion that was boosted to levels equivalent to that achieved with dose level 4 monotherapy or 960 million cells as a flat dose. There were no dose-limiting toxicities. While the efficacy at this low cell dose remained below that of recently FDA-approved products, for that reason, we have proceeded to dose escalate to dose level 3, 480 million cells in combination with the gamma-secretase inhibitor, and we will provide a program update on this in all programs at ASH later this year. That concludes my presentation. I'll turn it back over to Michael.

Thank you, Alan. What to expect from us going forward in H2? As Alan talked to you about, a laser focus on continuing to pursue PBCAR0191 forward for patients in the CAR T relapse setting. We have continued to validate the signal from ASH while lowering the lymphodepletion. We know there is still in this very fragile patient population, toxicity is very, very serious and we want to make sure that we're giving the best potential therapeutic index. We're incredibly excited about the efficacy. We like the improvements in safety. We still want to see if we can do that even better for patients. We'll move to standard lymphodepletion. And we will look to speak with the agency on what a path could be for this patient population in the second half of this year. With 19B stealth in our pursuit of displacing an autologous in earlier lines of DLBCL where most allogeneics are looking today. Like Alan said, we've optimized our next batch of cells and we will go into dose level 2 in the second half of this year, beginning in Q3. And with 269 in combination with the GSI, we moved to dose level 3, and I believe we'll commence dosing this week or today. So those are the next steps that you can expect from us in the second half of this year as well as towards the end of the year program updates. With that, I'm going to turn it over to Alex to open us up for Q&A.

Okay. Thank you, operator. We're now ready for questions and answers. So if you please queue it up, please.

[Operator Instructions] Our first question comes from Ben Burnett with Stifel.

I appreciate you holding this call. I had a question. You showed some interesting peak expansion data with dose level 4B relative to dose level 3. I guess as you move forward with a 3-day fludarabine lymphodepletion, what's your expectation for what that means in terms of peak expansion? And I guess, do you anticipate that this will be compensated by the higher dose?

Yes. Thanks, Ben. And yes, we're very excited about that data. We've looked at other lymphodepletion regimens, including one with just 3 days of fludarabine with the enhanced lymphodepletion Cytoxan dose and which made no change at all. So we fully expect that these levels that we're seeing right now all relate to the improved product attributes alone.

Yes. So Ben, it's Michael. Great question. I think we feel really good that we are reaching the peak expansion like Alan said, and seeing all of the new dose patients in CRs, deep in responses. I think we expect to keep those responses and still be able to go down to standard lymphodepletion. It's a higher dose, but it's also the new optimized and the next round of manufacturing for our CAR T cells. We continue to get better and better at predicting the fixed ratio of exactly what we want in our doses.

Our next question comes from Soumit Roy with Jones Research.

Congratulations on the robust data. Could you give us a little bit more detail on the 2 Grade 5 events we saw in the new dose cohort? Are they like much frailer patients? Did they have a higher disease burden? Anything that stands out would be really appreciated.

Yes. Thanks, Soumit. These 2 patients were older. One was 78; the other one was 72, which was a known risk factor. If you're looking at the literature publishing with stem cell transplant, where patients that were routinely using fludarabine in the conditioning regimen, this is one of the -- probably the most predictive features for this. And of course, prior fludarabine exposure is key, too. But any prior CNS insult within proximity to the fludarabine conditioning can then increase the risk that someone is going to have some kind of fludarabine-related neurotoxicity. It predictably occurs late generally on day 21 or later, and that's what happened in these 2 cases as well and does not respond to corticosteroids like ICANS generally will. So this raised enough suspicion for us, particularly when looking back at enhanced lymphodepletion and seeing that we had another case very similar. And we feel going to the 3-day regimen will give us a lot -- much more improved safety profile.

Soumit, it's Michael. Thanks for the kind words. I think one of the key things I would say here is, obviously, we take patient safety very, very seriously. I think as Alan said, moving to the dose level 4B up, we very significantly changed the infection rate that was plaguing us a bit with the hematologic recovery at ASH. I think the one thing that's really important to note is, look, we have a really high bar here, single-dose looking for cures and we're looking for zero Grade 5 events. That's a goal in our target profile than anything we do. These heavily pretreated population, unlike the auto CAR T space today, third line plus, these are patients that are excluded for most trials. For this reason, they're very, very sick. They're a median of 5 prior lines. They're really patients on the razor's edge. And if you look at almost all the allogeneic programs, most exclude this population for that reason. So it's really a tinkering of those levers we showed, Soumit, to try to get it perfectly right the therapeutic index for this population.

Our next question comes from Gena Wang with Barclays.

This is Tom for Gena. We have 2 questions. In the current setting of late line, as you show here post all the CAR T patients, how many patients should we think about would be eligible for PBCAR0191 given the relatively short, medium survival response? And what is your experience in screening and enrolling this population on trial? And secondly, going to the potential larger opportunity of the post second line auto CAR T patient as you outlined, how should we think about the efficacy bar as the survival and response are likely to improve with current options, like in earlier line settings?

Yes, Tom, it's Michael. Thank you for the question. It's a great question, and it's the slide. I know we move fast through and the slides will be posted. But if the slide we try to show you guys, if you think about just the U.S., I'll get throw some numbers out and we probably put a conservative number up. But if you look at the original approval of Yescarta, the third line plus setting, you're looking at somewhere between 5,000-plus commercially available patients. What do I mean by commercially available? There's probably another 2,000 in that setting in clinical trials. From the best epi numbers we've seen somewhere between 5,500 to 8,000 patients. And if you think about relapse we're only curing or giving longer-term outcomes. I'm comfortable saying cure since auto CAR T has been in the setting for over 5 years now. We're curing 3.5 out of every 10. So that first slide kind of showed you 65% of the 5,000 are in that relapse setting today, almost 2,000 per year. But immediately on the back of the ASH data 21 ZUMA-7, we know that auto CAR T, as you just stated, will displace auto transplant as a second-line standard of care. That immediately opens up a more than doubling of the second-line population. It's almost 15,000 patients in the United States alone, you would double that with Europe and the EU 5, another 1/3 for Japan and about 1/10 of Canada, if you just look at the G8 major markets. So you go from somewhere like 2,000 a year CAR T relapsers to almost 8,000 overnight as the standard of care and therapy continues to take place for auto CAR T in the second line. Right now, you asked about the outcomes, another great question. I think it's really important, and I hope it was clear in the presentation. While this is 12 patients, 11 evaluable, it started to become more of a robust package. You look at the progression-free survival of 1 to 2 months, that's the standard of care in the United States today in this population. There is really salvage that it is -- it's not a dearth of therapy. There's a lot of different therapies. They just don't offer great outcomes in this population. You're looking at best supportive care in 1 out of 4 of these patients. So you have survival somewhere from 4 to 6 months. Alan showed you a 50% greater duration of response at 6 months, that's the median survival best case for this population today. So look, we want to continue to generate more patients of evidence and efficacy, continue to get even better on the safety and we did improve from ASH. We want zero Grade 5 events, but we feel we're really well ahead of the clinical hurdle right now that could potentially move this path forward for human beings in patients who have no options. That will be the genesis and kind of the backbone of the dialogue we look to have at the FDA later in the year. As the last point you made is, what does that hurdle, Michael, in the CAR T relapse patient? Let's just say, in 2025, when you have a 70% market share of second-line setting of auto CAR T and I think you'll start to see those PFS and those overall subsider outcomes, I sure hope so, increase and get better. But I'm sure hoping they're getting better on the back of an allogeneic CAR T approval like PBCAR0191.

Thanks, Tom. Appreciate the question.

Our next question comes from Andrea Tan with Goldman Sachs.

Michael, maybe one question for you. Just curious what data set exactly you expect to have before you engage with the FDA to discuss next steps. And then I have one follow-up.

Yes, Andrea, great question. Good to hear your voice. Thanks for joining us this morning. I think we know we want to treat the next 3 to 6 with SLD. Like Alan talked about, we feel really good about the next optimization of our CAR doses. Of our attributes of products at the CAR dose, we think we've got some room to keep our peak expansion where we need it to be. We know peak expansion ties to the long-term outcomes, Andrea, we want. So I think you can imagine we'll start conversations with the FDA probably in Q3 as we're generating another 3, 4, 5 patients, but we'll stop that interaction. You first have to request the meeting. So there's things we'll have to do there. We want to collect some of these patients at SLD.

Got it. And then if I can ask a question on the BCMA program. Just wondering if there's any details you can share on what that combo response looks like for the dose level, too. Just curious how that compares to the 29% ORR you saw with monotherapy.

Well, one of the things I pointed out, I think we shared this at ASH is that the peak expansion was equivalent to what we saw with dose level 4. So it was pretty impressive.

I think, Andrea, overall the...

Yes.

Go ahead, Alan.

I was just going to say the hurdle is very high when you look at the data from Legend and BMS' approved agent and the bispecific antibodies that are out there now. It's a very high hurdle. Our focus right now is to see what happens at the higher cell dose with the combination. This is a very low dose. And certainly comparing this to the BMS, it's about half the dose. So our hope is we're going to maximize response and biologic activity with the GSI the next cell dose.

Yes, Andrea. And the last thing I would say there, it's one of the reasons not to be coy, one of the reasons we didn't calculate all the response numbers. We're not where we need to be for that hurdle yet. As Alan said, it's a lower dose. I don't know that you can linear figure out, hey, you're this much of the way there. I think we've got to see what the next dose. We like the expansion with the GSI. There's no doubt about it. And we've got to see what the next dose level looks like. But as we promised you, we'll be fiscally disciplined and will be evidence-based. And we're not going to move into an area that multiple myeloma is unlike the late-stage DLBCL setting. It has the opposite of a dearth of options, right? It's pretty quite saturated. And that legend TPP is, as you would have discussed prior, is a pretty high hurdle. So we will be relentless at making sure we do not move forward in a cohort, in a setting if we're not seeing the results we want. So let's calculate the response rate or the deaths here later in the year and we look at the next dose level.

Our next question comes from Maury Raycroft with Jefferies.

This is [ Farzin ] on for Maury. You have noted optimized manufacturing is gating dosing for 19B. Can you talk more about what was optimized? And what gives you confidence the optimization will actually translate to improvement?

That's good to hear your voice. Thanks for joining us, it's Michael. So look, I mean, this is the secret ingredient of the soup, right? So I know all the companies are not necessarily saying exactly what they're doing. I think you saw the slides we put up. We know you want to have less differentiated so versus more, and I'll let the good doctor chime in, in a minute. I think we feel really good that we know what our targets are for efficacy, what types of T cells we want in distribution, what we don't want to have for toxicities. That's clear. That's in the publications I put up on the slide here today. The question is, do you have the expertise in-house to do it? And I think we've proven that we do. And do you have the toolbox to do it? We think a single-gene edit. All the conversations that have come up have been about translocations with gene editing, and we only talk about safety. That's super important, no doubt about it. But what's also important, and we think equally important is the single-gene edit puts less stress on your cells when you're manufacturing, allowing for those key ingredients to stay viable. Alan, is there anything else you'd want to share about at a general level? Obviously, we won't talk about exact distributions we do or types of cells we want to have versus not haves.

Yes, absolutely. And I mean, there's -- I think Michael spoke to this very well. But when you look at the data that he shared with you that [ Jason Weston ] talked about from ZUMA-7, there are 2 specific attributes that are very important to efficacy. And one is a juvenile phenotype, that needs the naive T cells. It also means the central memory cells. And in fact, the higher the amount of differentiated cells in your product, the higher the toxicity. That's what they found from ZUMA-7 and that's what we're focused on as well. Having said that, a single-gene edit is so critical to get to this kind of profile. For every additional gene edit that occurs in a T cell, it increases the differentiation capacity. In other words, once you have more than one single-gene edit, you're going to have more differentiated T cells, which only adds toxicity and reduces benefit. So I think those are the things that we're striving for, and we're pleased with the profile that we're seeing.

Last point I'll make is very important, our manufacturing platform is a platform. It's the same platform and the same platform, different constructs with a similar platform as far as the characterization attributes across all CARs. So it is very, very viable to believe the implementations we make on one program matter for the other. That's why we -- as we're moving fast to 19A, we started to make some other process implementations into the 19B program. That will be important that we're at next stages of our optimization for all of our programs. Derek, anything on your end? We do have Derek. Derek's is on the room with us. He's back in derm, keeping the fort down. Derek, is there's anything you'd like to add?

No, I think you hit everything.

Our next question comes from Patrick Trucchio with H.C. Wainwright.

Congrats on all the progress. Just a follow-up on Slide 26. So just regarding the potential approval for PBCAR19A. I'm wondering if you can discuss more what the ideal registrational program can look like in terms of the number of patients enrolled or other elements of the study design and specifically, are there any limitations to enrollments such as CD19 expression or other prior lines of treatment?

Yes. So Patrick, it's Michael. Thanks for the kind words. How do I know you were going to ask me the regulatory question? So look, I think we're very hopeful here. We're measured. There's work to be done. I can tell you the way we think about it. We want to be clear with all our investors, credibility is really important to us. We have not spoken to the FDA about the path forward yet. We plan on doing that in the second half of this year, discussed with Andrea, as we collect a little bit more data here. But when I think about this population, it really rewinds me to 2015 and '16 when we were getting ZUMA-1 approved. And I think about the third-line population using the historical control, the Scholar data. We knew that median survival in that population was -- what was the backbone of why 6-month CR that you folks always ask me about even got created that CR at 6 months clearly showed it was better than what was out there in the standard of care. So there was no FDA-approved comparator. So we think about a possible historical control similar to ZUMA-1. Alan has showed you the data. PFS at 1 to 2 months. There's a reason he plotted how far out in front of that hurdle we are right now. But look, we're measured. We want to continue to clean up the safety one more round. This patient is incredibly frail. Unfortunately, this patient group has an absolute 100% Grade 5 rate if we don't intervene with therapy. This patient has got no options. So we want to speak to the FDA about that. But if you think about the ZUMA-1 path, that's kind of how we think about things. We've shared with you the aggregate of the meta analysis we've done of the literature out there on what patient outcomes look like. So I think you get back into that a little bit, but I don't want to put any words in the FDA's mouth yet. We'll work closely with them. And I think the urgent speed and need for this population. Hopefully, we can align with the FDA on what that is, what the appropriate robust package. The last comment I'll make is coming out of ASCO, we really wanted to make sure we digested data in the clinical development area coming out of ASCO, maybe not necessarily CAR T, right, other monoclonal antibodies out there. And there are some other groups looking at this space. And I think we feel really good of where our data stands today with, again, the hopes to continue to get that Grade 5 to 0. But from an efficacy standpoint, we feel pretty good about that versus anything we've seen that's used today or even that's in the clinical development landscape.

Patrick, you asked about CD19 expression. It's been somewhat controversial in the auto CAR T literature. But at least in our 12 patients, we had expression ranging from low CD19 expression to high and we saw no difference. And that includes our last 6 patients where it's 100% complete response rate.

Yes. So that's a great point, Alan. So we don't believe we'll have to have a subset of patients we allow into this, Patrick. We're looking to try to help all these patients in need. We did look at -- right now, I believe we have at least 2 lines in our inclusion criteria. The ASH cohort went up to 7, 8, 9 lines past your auto CAR T. This last cohort, we've got a median of 5-plus prior lines and we capped it at no more than 2 lines after your auto CAR T. We're really trying to hone in on what is that clinical measuring stick we should be comparing to and in like types of line of therapy.

Our next question comes from Raju Prasad with William Blair.

I wanted to get a sense for maybe Dr. List on the blend rep plus GSI data we saw at ASCO. Is there a reason why we should think that using the GSI with the CAR T might have a different profile than with blend rep? Or -- I mean, how are you kind of thinking about that data set in the context of your study?

Hey Raju, Alan. I think Raju is asking a question of any data we've seen with the GSIs in general outside of even the pairing with CAR Ts and, Raju, correct me if I'm wrong. Is there anything we've seen in that usage that makes us think the reason we're combining with the GSI, obviously, for increased epitopes and things like that. Anything you've seen from us with non-CAR T that would guide you one way or the other with our [ combinatorial ] approach?

Raju, is that kind of what you're asking?

Yes.

Yes. I mean the data even with bispecifics in the combination obviously looks good. And we know from our own data, at least our preclinical data, that with low expression in myeloma cells, low BCMA expression that our CAR does not behave as well. So our goal really just like with any of the therapeutic, the targets BCMA, our goal is to increase the density. And the GSI appears to do that from the data that we have from changes that we see in soluble BCMA as well as an increase in surface expression with it, which I think is what generated in the peak expansion that we got at dose level 2. So we're very excited to look at what we're going to see with the combination of dose level 3.

Yes, if I can just have one follow-up. You mentioned the CD19 antigen rates kind of vary, which I think is pretty interesting. Were any of these patients that were treated considered a CD19 like negative relapse? Or I assume none of them were in that category here about?

Yes. Thanks for asking, Raju. They're all CD19 positive. They have to have some CD19 expression. So all these patients do get biopsy before they're eligible.

And Raju, you might remember from the data we've seen since the autologous evolution, really escape antigen -- CD19 escape antigen where they take another pathway where they would be CD19 negative is really a small percentage of the population, somewhere 9% to 12% is some of the anecdotal kind of evidence we talked about when I was in my past life. So really most of these folks are progressing with CD19 positive disease.

Okay. Thank you, Raju.

Our next question comes from Justin Zelin with BTIG.

So we've heard some discussion at ASCO around cell expansion PK and its relation to durability of response. Can you speak to some of your translational data and what you consider to be meaningful durability to shape up for 191?

I can say that meaningful durability for us is going to be a cure. But we know we're not going to achieve that in all patients. Autologous CAR T doesn't do that. And we're going to see relapse in about 60% to 65% of patients with autologous CAR T. But we do think we're encouraged by what we saw at least from the ASH data, and this is a dose level 3. So we had 50% of those -- 3 of those 6 patients that were beyond 6 months. And one still out 18-plus months, another one ongoing at 7-plus months. So we do think with this kind of peak level that we're achieving now, above the median from ZUMA-1, we're going to have very durable responses, I think, in the majority of patients that get CR.

Yes. Justin, it's Michael. Good to hear your voice. Just to add to that, when Alan showed you the second swimmer plot of the dose level 4B, you'll notice the depth of those responses, they were all CRs. So on top of the 3 of 6, which are along those runners from the ASH cohort, the new cohort has another 4 patients that haven't reached 6 months yet. So we do believe and that's why I shared the data today that the greatest predictor of long-term outcomes and/or cures is reaching that peak CAR T level in the first early phase, the first, let's call it, 2 weeks in order to get that circulating tumor DNA down fast. Everyone asks about persistence. There really isn't any data out there that persistence is, of course, a function of area under the curve, the higher peak, the more persistence you'll have. But there is no data out there that says that is mutually independent variable that drives outcomes. It truly is getting to that peak. So we're really hopeful that those dark blue stay dark blues. And when we started this mission at ASH, I was hoping we could get a possible drug for these patients even if it wasn't cures. Dr. List said to me, if we get to these peak levels, we've got a chance to cure people, and that's what we're staying focused on. And that's how we think we really differentiate ourselves versus bispecifics or anything else coming into this space, which, by the way, we welcome patients need options. They don't only need PBCAR0191. I hope we have 5 approvals in this space. So I hope that makes sense.

And Justin, one other thing I want to add to that, just to give you some other validation of this. There are several publications out there with autologous CAR T. The patients that are cured that have long-term responses, clearly, they're circulating tumor DNA by day 7. It's not day 28 of day 42. It's day 7. So peak -- and peak is the best predictor of clearance of tumor DNA. And the only way you get to high peaks is good functional attributes, juvenile correction. That's why we're so focused on the process of manufacturing.

Our final question comes from Kelsey Goodwin with Guggenheim Securities.

I know you said you haven't met with the FDA yet, but I'm just wondering if maybe you could provide a bit of color on how you think about a path to market in post CAR T. I guess do you anticipate you could enroll a line agnostic study and kind of be tied to autologous label expansion into earlier lines? Or do you anticipate kind of needing a trial per line of therapy?

Yes. Kelsey, thanks for the question. Yes, I want to tease that out a little bit because I want to make sure we have clarity, yes. No, this is absolutely not in the autologous CAR T setting, and this is not where most of the allogeneics are pursuing, which is really third and second line. This is -- that is all of the program. This is always going to be tied to the patients, the 65% of patients who relapse after CAR T therapy. Even the second line study, ZUMA-7 is starting to show an EFS that will have a similar group of patients, maybe about 60% who relapse. So this is later line you are looking at. If you think about what all the world's CAR T use has occurred in diffuse large B cell to date, it is third, fourth and fifth line and majority in the third and fourth line. That means you're going to be looking at fourth, fifth, sixth line for this population and that's where the clinical hurdle that we shared with you today, progression-free survival 1 to 2 months overall survival, less than 6, a real tremendous lack of therapies out there to give robust outcomes. It's that fourth to sixth line. It's the clinical hurdles we shared with you today. Those are the exact same literature from the consortium from all the published data that we'll put in front of me today, and we'll talk with them and see what their thoughts are. The other thing I'll share, which was the earlier question. There's no comparator in this setting. There's nothing approved in this space. So hopefully, we can work with the FDA to do something Scholar like with a literature analysis as a historical control. But this is the program for CAR T relapsers. Of other CAR T programs, stealth and some of the other things we're looking at, that will compete for the earlier lines of DLBCL, third and eventually second line and hopefully try to rival Yescarta and Breyanzi be as good, if not better, and displace autologous CAR T. That would be a randomized trial in my mind and a very different pathway to regulatory.

And I'm currently showing no further questions at this time. I'd like to hand the conference back over to the CEO, Mr. Michael Amoroso, for any closing remarks.

Well, operator, thank you for your assistance today. I want to thank my team who has worked diligently here before ASH, but even since ASH, to have this 6-month review ready for the community as we talked about. I want to thank the patients and their families who have been brave enough to embark upon this journey with us. And finally, I thank you to our investment community. Thanks for your patience. We promised you folks we will continue to show up rain or shine, and we're pretty excited about today's data and building upon it. So thank you for your thoughtful questions, and we look forward to catching up soon.

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