Welcome to everyone tuning in. My name is Ashley Vanderbeck, and I am an associate here at H.C. Wainwright. With me is Gavin MacBeath, PhD, the CEO at TScan Therapeutics, who also serves as a member on the board of directors. Welcome, Gavin. We are very happy to have you here with us.
Thank you. Thank you for having me.
I think we can start off with a very simple and basic question. For the investors that are less familiar with TScan, it would be really helpful if you could give us a brief overview of the company and your pipeline assets.
Sure. Thank you. TScan is a clinical stage, fully integrated TCR T-cell therapy company. We were actually founded back in 2018, and over the years have built a pipeline of various TCR-engineered T-cell therapies for cancer. Our lead program is in heme malignancies, where we're treating patients that are undergoing bone marrow transplants, really targeting any residual disease in the patient post-transplant to prevent relapse. That's the most advanced program. We've been running a phase I trial, which we call the ALLOHA study, that's been going for about three years now. We recently reached agreement with the FDA on a pivotal trial design and are actually launching our phase III study this month. In addition, we also have a program in solid tumors.
There, we're now taking the strategy of doing in vivo engineering, so targeting various cancer-specific antigens with TCRTs, PRAME, MAGE-A4, but doing that with lentiviral particles that get infused directly into the patients and engineer their T cells right in their body. For that program, we're in sort of mid to late stage preclinical development, and now have timelines to get us into the clinic with that program mid-next year.
Great. That's very helpful. Hopefully, we'll be able to touch base on all of those aspects of the pipeline. I think a good place to begin our real discussion is to start with your lead asset, TSC-101. One question we get is, what do you think is still most underappreciated about the HA-2 targeting TCRT approach to preventing relapse post allogeneic hematopoietic cell transplant?
This program is based on a long-term clinical observation that patients undergoing transplant, if they develop T cells against minor histocompatibility antigens that are present in the patient's blood cells, but not the donor's blood cells, those patients perform better. They don't relapse at the same rate. We are really taking advantage of a natural mechanism that occurs in transplant, but really augmenting it with an engineered T cell therapy product. One, just the science behind it makes sense. It's consistent with clinical observations. I'd say some other underappreciated aspects of the program is one, unlike a lot of CAR T programs, once this becomes a commercial product, it doesn't have the same commercial challenges that CAR T has. For example, we don't need to go looking for where these patients are. They've already been referred to a transplant center.
If a patient gets diagnosed with AML or MDS in a community hospital, if they're eligible for a transplant, they're getting referred to a transplant center, and there's only about 120 transplant centers in the U.S. that perform allo transplants. They're already a concentrated market in those transplant centers. Those transplant centers are, by definition, cell therapy centers, so you're not having to teach them how to do cell therapy. In addition, our product is being extremely well tolerated, so we're not going to be subject to any of the REMS programs, things of that nature that are a real challenge in the CAR T commercial world. I think from many perspectives, this is just a sort of made for an easy commercial rollout program once the product gets approved.
Yeah, that's really interesting and really interesting to hear about the history of the product coming from a natural observation that clinicians have had over the years. Thinking a little bit more about what the outcomes look like in these patients, how thinking about the predictive nature of early MRD conversion and complete donor chimerism for long-term relapse-free survival. How do we really interpret the cases where that early complete donor chimerism is achieved but not really sustained beyond one year?
Yeah. As you're well aware, but maybe not everyone else is, we have been using donor chimerism as an early marker for if our product is successful, because again, our product is intended to eliminate any patient-derived blood cells while sparing the donor-derived blood cells. Complete donor chimerism just means that there are no detectable patient cells anymore. By definition, that's a sign that the product is working. Also because relapse comes from the patient's cells, if you've eliminated the patient cells, that should reduce the risk of relapse. We have now a growing body of data that shows that achieving complete donor chimerism early on is highly predictive of long-term benefit.
We have been using a very sensitive assay to detect chimerism, the standard clinical assay has a sensitivity of 2%-5%, whereas this AlloHeme assay that we've been using, developed by a company called CareDx, it has a sensitivity, a detection limit of 0.2%. We're very sensitively detecting any patient cells. What we found in our phase I study, and we reported this at Tandem this year in February is that if you achieve complete donor chimerism at two months post-transplant, you have a very low probability of ever relapsing, even long term. That's data from our own clinical trial based on about 40 patients' worth of data. In addition, CareDx has been running a study called the ACROBAT study, where they've been, again, using their assays to assess patients post-transplant, and they're seeing the same result.
That's why we really feel that achieving that early donor chimerism is a very meaningful measure for us as to how well the product is working. Now, there are a few instances in which a patient does achieve early chimerism, and then subsequently, you see the recipient chimerism rising again. That's rare, though. We only saw it twice in our cohort A, and we've seen it once in cohort C, that's basically three times we've observed that out of now about 35 patients. It's a fairly rare event that occurs. The probability of relapse is not zero, right? By the way, those patients don't necessarily relapse, they just have increasing chimerism. Overall, the results have been that is predictive of long-term benefit.
Yeah, that makes a lot of sense. Thinking holistically about that data set and as you described so well what MRD negativity means and how that is correlated to how the patients fare post-transplant. If you look at that data set across the complete donor chimerism, the RFS persistence, and particularly this subgroup data, what of all of those factors really gives you the most confidence that the activity is real and durable rather than a function of small numbers or maybe a favorable patient mix? Which part of that data set do you think the investor community might still be overlooking?
Yeah. We kind of have now two distinct data sets, right? We have cohort A, which is the patients that we've been treating for the last three years. Many of those patients have been on study for a long time. Obviously, ultimately, what matters is relapse-free survival, right? A patient not relapsing, patient not dying. Cohort A is now a very mature data set. What we've shown, and we did have a control arm in the studies from the very beginning, the control arm is also very mature now. What we recently reported is that both of the patients that have been on our study for over three years are still on study, disease-free, with complete donor chimerism. That compares to none of the three patients on the control arm that would have been out three years.
100% versus 0% at three years. At two years, 71% of our patients are still on study and disease-free versus 43% on the control arm. Cohort A gives us that long-term benefit that is ultimately going to be important to getting this product approved. The primary endpoint in our pivotal trial, which I'm sure we'll get into, is relapse-free survival. That's ultimately what matters. Now we have cohort C, which is patients treated with our new commercial-ready manufacturing process. Those data are simply not as mature because we only started enrolling that cohort at the end of last year. For that cohort, we really are depending on this early chimerism read because that's what we have right now, but we've also shown that that is predictive of long-term benefit.
As that data set matures over time, and we'll continue to report on that data set every six months at major medical conferences, we'll start to see that early chimerism translate out into relapse-free survival. For now, because those data aren't as mature, we're really looking to that chimerism data as the key readout.
Yeah, that makes a lot of sense with the time lag. Before we really dive into what else you're seeing with cohort C and what it might mean, maybe it would be nice to provide the audience with understanding how cohort A and cohort C are different beyond the time points of when the different cohorts started. Would you be able to walk us through the key differences between the 17-day process that was used for cohort A and the new 12-day commercial-ready process used for cohort C, particularly the move from the magnetic bead selection to drug selection, and how that translates into both manufacturing economics and the higher proliferation and activation markers that you reported?
Yeah, sure. When we started this study a little over three years ago, obviously we started with a manufacturing process that was phase appropriate, a process built for phase I. As the program matured, we worked on ways to make that process more efficient and bring down the cost. Basically improve on that process and make it a commercially feasible manufacturing process. Really the big difference, as you noted, between the initial process and our commercial-ready process, is that in our initial process, we used a magnetic bead-based purification step on Day 9 of the process to purify the engineered T cells away from non-engineered T cells. Unfortunately, that step was actually quite inefficient. Although it purified the cells, we would also lose over 90% of the engineered cells at that step.
That then required us to spend another eight days regrowing those purified cells in order to get up to the dose that we wanted to administer to the patients. We really wanted to eliminate that inefficient purification step. We introduced a drug selection marker into the T cells that makes the cells resistant to methotrexate. Now we don't use magnetic beads at all. Instead, once the cells have been engineered, we simply add methotrexate to the culture media, and that enables only the engineered cells to grow and proliferate, and the non-engineered cells to die. That is much more efficient. We don't lose any engineered cells, which means we can stop the process five days earlier, 12 days versus 17 days. It also has dramatically lowered cost of goods. It's simplified the process. Overall, much more commercially friendly process.
Also, we think, a better product because, ultimately, these cells are younger. They haven't been amplified for 17 days. They've been only for 12 days. That has translated in our data that we've seen in patients now to a product that has higher proliferative potential as measured by Ki-67. It's got more T cell activation as measured by CD28, and more cytotoxic potential as measured by granzyme B. All of the translational markers are pointing to a product that is more active in the patients.
That's really helpful. From my own days in the lab, I know how many cells you lose with magnetic bead purifications in any immunological model. It'll be interesting to see also how those higher proliferation and activation markers translate to persistence of T cells in the patient for cohort C. Thinking more about outcomes in that cohort, we know that cohort C carried a higher proportion of MRD positive patients pre-transplant. I think it was 75% versus 53% in the control. They haven't shown any relapses to date, which is exciting. With this in mind, which patient subsets do you think are most likely to benefit? Is it those MRD positive pre-transplant patients, and/or high-risk genotypes like TP53 or MYC, or some broader post-transplant population?
Yeah. I'd say, obviously, these genetic risk factors like MYC rearranged or TP53, they're certainly a very challenging disease to treat. It's almost like a death sentence to have one of those mutations. Clearly, we've been very excited about the fact that these high-risk genotypes are benefiting from our product, certainly, in comparison to the control arm. That said, ultimately, every patient that's undergoing a transplant is going to benefit from our product. We're not just focused on these high-risk groups. Sometimes it's difficult to know what the risk is prior to transplant. For example, with MRD positivity, as you noted, we did have more challenging patients in cohort C relative to our control arm because more patients were MRD positive prior to transplant. That's not something that can be assessed a priori.
Typically, a patient gets that bone marrow biopsy right before they have the transplant, and the results of that biopsy take several weeks to come out. By that point, the patient's already undergone transplant, and you've already given them our product. You can't actually run a trial in which you're only selecting MRD positive patients prior to transplant. Just logistically isn't feasible. What we are doing in our pivotal study is we are ensuring that we are not just enrolling high-risk patients. We're specifically gearing the study to enroll both intermediate and high-risk patients, because ultimately, we feel that this product is going to be of benefit to anyone undergoing transplant, because you can't always tell what the risk is for the patient prior to going to transplant.
If you determine that too late, the product is out of the window in which you want to be treating the patient.
Another thing that you had briefly mentioned in passing was that the pivotal design is going to use a biological assignment rather than randomization. Again, yes, this avoids ethical and feasibility problems, but it does raise the question of unmeasured confounding factors between the HLA-A*02:01 positive and negative patients. Beyond the stratification and Cox modeling that you presented, I think it was slide 34 on your last presentation, what gives you and the FDA confidence that those arms are truly comparable?
We've had multiple meetings with the FDA about this question. One, in terms of does being either HLA-A*02:01 positive or HLA-A*02:01 negative influence outcome? There we have done a large study with CIBMTR, which is the repository of every patient undergoing bone marrow transplant in the U.S. All centers are required to enter data from their patients into this database. This database has over 700,000 patients' worth of data in it. Very robust database. We're able to show, in collaboration with CIBMTR, that there is no difference in outcome between patients that are A2 positive versus A2 negative. We presented those data to the FDA. The FDA agreed with us that there was no significant difference. That is why the FDA agreed to this genetically randomized controlled trial that we're running in our pivotal study.
There's no real question of the A2 positive versus A2 negative population. With respect to trying to maintain balance between the arms, we are stratifying the study based on, one, obviously disease, AML and MDS, but also on various high-risk genotypes like P53 or MECOM rearranged disease. We will be stratifying this and controlling enrollment in each arm of the study to ensure that there's balance between the arms on these various stratification factors. In addition to that, we are also going to be able to adjust for any remaining imbalance between the arms using Cox proportional hazards model. That's all in our statistical analysis plan that we submitted to the FDA, and it's our intention to take that approach to ensure balance moving forward.
Diving a bit deeper, you mentioned all of the ways in which the pivotal design clinical study will be designed. Could you give us a broader overview of how should the investors be thinking about this population, specifically around the details you mentioned, the exclusion of ALL patients, which were in ALLOHA phase I, the donor types allowed, which was a bit different, the enrollment stratification that you just mentioned, and also the clinical eligibility around maintenance medications like menin inhibitors, FLT3 inhibitors, and hypomethylating agents that might make these patients even more profoundly cytopenic than they would be otherwise?
I would say overall, the pivotal trial is really designed to make this as accessible and as general a therapy as possible. One, the exclusion of ALL, this is really to focus on AML and MDS, largely because there's no good treatment options for those patients. Obviously, ALL patients are eligible for CAR T, but also to make the study a little more homogenous. Beyond that, really the goal of this study is to provide a therapy that would be used in the real world, and integrate seamlessly into how transplant physicians are actually treating their patients. One, the introduction of mismatched unrelated donors, that is to make the therapy more accessible. It's easier to find a donor that's A2 negative if you're looking not just for haploidentical donors, but also for mismatched unrelated donors.
In fact, both of those types of donors are more and more frequently used in the transplant world. It's easier and easier to find donors due to the availability of haplo and mismatched unrelated donors. In terms of maintenance therapies that we're allowing, again, you have to conform with what's actually being used in the real world. If physicians are going to use FLT3 inhibitors or IDH1 or IDH2 inhibitors in the real world, it makes sense to allow those on both the treatment and control arm in the study. If physicians feel that's going to be a benefit to the patient, why would you deny the patient use of those therapies?
Really the only thing that we're just making sure that we're controlling a little bit on the treatment arm of the study is although we are allowing menin inhibitors and hypomethylating agents, we're just asking that those be delayed relative to the infusion of TSC-101, simply because menin or HMAs could actually be harmful for our engineered T cells. They're being allowed on the control arm, and they are being allowed on the treatment arm with the appropriate delay following infusion.
Yeah, that makes a lot of sense. Talking more about the discussions you've had with the FDA and the alignment that you've reached with them, what was the most important thing in reaching agreement on the phase III design? One of the questions that kept coming up for us is how are you going to balance optimizing those enrollment kinetics and execution against preserving balance by disease biology, and genotype with this enrollment stratification?
We've just been really working closely with the investigators on the study to ensure that they obviously understand the importance of enrolling the control arm as well as the treatment arm, the importance of enrolling the same types of patients on the control arm as the treatment arm. With about 30 clinical sites already enrolled in this study, I think we have very broad representation and a very strong desire to run this study appropriately with the appropriate balance between treatment and control arm.
As I said, on top of that, we are implementing a strategy to stratify for disease type and risk factors. We do that by, at 50% enrollment, assessing all those strata, then we have the ability to turn off and turn on enrollment into these strata to make sure that we're getting balance between the treatment and control arm. That's all part of our study plan.
That all makes a lot of sense and is pretty exciting. Thinking about the data set to come, what do you view are the key elements of a clinically and regulatorily compelling data set beyond your primary endpoint RFS, and particularly thinking about chimerism, safety, and maybe consistency across subgroups?
Obviously, relapse-free survival is the primary endpoint. That's obviously what's most important to be positive at the end of the study. We're powering the study at 85% power to achieve a hazard ratio of 0.52 or less for relapse-free survival. That is ultimately going to support full approval of the product. Yes, as you say, on top of that, we are collecting various secondary endpoints. Our key secondary endpoints, event-free survival, as well as overall survival. That's very important to be trending in the right direction, obviously. As an exploratory endpoint, we are collecting chimerism data using this sensitive assay, which we hope will support the use of this as a surrogate endpoint down the road for future studies.
It's obviously not a surrogate endpoint for this study, but we want to continue to collect those data and build the case that donor chimerism is a meaningful surrogate endpoint for relapse-free survival or overall survival. We also have a variety of sensitivity analyses planned in the study to look at subgroups, including disease subgroups, risk factor subgroups, MRD positive versus negative subgroups, et cetera. Overall, building a compelling data set that could then be used to support commercial launch of the product.
We're looking forward to the data set and everything that is to come. I think we can end, just one more question. TScan has cash of $128 million, which you've guided towards funding operations until second half of 2027, but the phase III top-line data is targeted for mid-2028. How are you thinking about bridging that gap, and what data flow between now and then do you see as the most value-creating?
I think there's a number of mechanisms by which this trial can be funded. Capital markets is one, but business development opportunities are others, and I think we've been getting quite a bit of interest in this program now from a variety of other companies. In terms of the data flow, there's multiple points at which we'll be presenting both updated data on cohort C, which I think is going to be most meaningful for people, since this is reflective of the manufacturing process that will be used in the pivotal study. We plan on updating publicly those data at major hematology conferences. I believe there's one in December and one in June. We'll do that on a regular basis. There'll be data flow every six months on this program.
In addition, we are expanding this program to other HLA types with TCRs that target antigens derived from CD45. That is a phase I study that we're going to launch in Q4 of this year. Data flow from that program will also continue to provide a catalyst for funding the company and moving forward.
Great. That is very helpful. I think we can stop here. Thank you again for your time and giving us an overview of TScan. We look forward to hearing from you again.
Great. Well, thank you for your time as well.
Thank you.