Are you ready to kick it off?
Yeah, whenever.
Whatever you guys want to do.
Okay.
All right. Welcome back to the Citizens Life Science Conference. My name` is Silvan Türckan , and I'm a senior analyst covering precision medicines. Now it's my pleasure to introduce Senti Biosciences. Presenting will be Tim Lu, CEO and co-founder, and Jay Cross, CFO. Thank you so much for coming and joining us today.
Thanks a lot, Silvan. Appreciate the opportunity to be here. I'm here to tell you about Senti Bio. We have some forward-looking statements here. We're trading on the NASDAQ. Senti Bio is a company that's really focused on transforming the treatment of oncology with our logic gating technology. What I'm going to tell you about is our logic gates allow us to go after cancer indications where there is no clean single target. We can essentially recognize multiple antigens on the cancer cells and kill. We can also recognize antigens on healthy cells and protect those cells from being affected. We think that's a really differentiated mechanism compared to other ADCs, T -cell engagers, or CAR T -cells out there in the market. We've implemented this into our lead program, SENTI-202.
It's an off-the-shelf CAR-NK program going after two AML targets, CD33 as well as FLT3. It protects healthy hematopoietic stem cells by recognizing a target called EMCN or endomucin. This is an off-the-shelf program, meaning that we have the cells cryopreserved. We're shipping them on demand to patients. Patients are getting quick treatment. We recently announced at AACR some really exciting positive data, as I'll share with you. We've treated now seven patients for which we have reportable data on. Four out of seven were able to achieve MRD negative complete responses with longest durability, eight months and counting. This gene circuit platform does offer us the opportunity to expand into solid tumors downstream and also works in both CAR-T and CAR-NK cell platforms. Senti was founded originally out of technology from a variety of universities, including at MIT, where I used to be on faculty.
Basically, we've developed now four classes of what we call genetic circuits or technologies that we can put into any cellular gene therapy product. Today, I'm going to primarily focus on this logic gating concept, and I'll walk you through that in a little bit. We also have the ability to multi-arm ourselves, so engineering a cell therapy that not only contains a CAR, but has one or more payloads, including cytokines built into it to enhance its efficacy. We have something called the regulator dial, which basically allows us to control using an FDA-approved small molecule. It allows us to control what a gene or cell therapy is doing in the body, as well as a library of synthetic promoters that constitute what we call smart sensors that can turn on at the right place at the right time in the body.
What does a logic gate allow us to do? If you think about current CAR-T cell therapies, ADCs, T -cell engagers, you're oftentimes faced with finding that clean target that's only expressed on the cancer cells and not expressed on the healthy cells. If you find a target that doesn't have that clean profile, what happens is you basically will achieve not only killing of the cancer cells, but killing of healthy tissues, and that limits your ultimate therapeutic window. Fundamentally, that comes from the fact that there aren't just that many clean targets that are out there. I think we're limited by the biology. That's just the reality of cancer. The way we've chosen to overcome this problem is instead of recognizing just a single target, why not engineer a product that can recognize multiple targets?
We have what we call an activating CAR that recognizes multiple cancer targets. In the case of AML, this is CD33 as well as FLT3. We have what we call the NOT gate, which is implemented through this inhibitory CAR or iCAR that recognizes the healthy cell and delivers a do-not-kill signal into the product. That basically keeps those healthy cells from being affected. We have a pipeline of programs that are using this technology. The most advanced is SENTI-202, which is what I'm going to focus on today. SENTI-202 is focused on AML. AML is a really difficult-to-treat hematological cancer with over 20,000 patients newly diagnosed in the U.S. per year. Current standard of care for patients that are in that relapsed refractory condition, unfortunately, these patients are seeing CR rates with current standard of care maybe in the 15%-25% range.
Median survival for these patients is about five months. There are several challenges with AML. It's a very heterogeneous disease, meaning that there is sort of a wide diversity of AML clones as well as leukemic stem cells in patients that basically make it difficult to eradicate sort of deeply the cancer. To achieve longer, better durable responses, we think we have to hit that heterogeneous nature of the disease. Oftentimes, probably have to go after more than a single target to do so. Secondly, all of the AML targets that we know of on the cell surface are also expressed on healthy cells. Many of them are expressed on healthy bone marrow cells or sometimes in cases on endothelial cells.
If you just go after a single target by itself, what's been seen in clinical studies over and over again is you see a lot of toxicity and limited therapeutic window in treating these patients. I'm going to talk you through SENTI- 202, how it's designed, and the detailed clinical data. Just to give you a highlight of what we presented about a couple of weeks ago, SENTI-202 is enrolling currently in a relapse refractory AML patient population. Out of the seven patients for which we had data to report, five out of seven achieved an ORR response, four out of the seven achieved a composite CR response, including two out of three in our preliminary RP2D cohort.
All four of those CCR patients did achieve MRD negative status, which is a really important part for us to determine, are we really hitting the tumor as deeply as we can? All of our complete response patients are continuing to maintain their responses anywhere from four plus to eight plus months and ongoing. From a safety perspective, SENTI-202 is quite clean. We do have the ability to enroll patients in an outpatient setting. No major DLTs detected, and we did declare our preliminary RP2D based on and part of this data set. The other thing that's really exciting about that data set we put out at AACR is, in addition to that oral presentation of this data on the clinical side, we also showed a lot of detailed cytoph data on the patient samples themselves.
We take bone marrow from these patients every cycle, every 28 days. From that, we can actually look deeply into are we hitting the AML blast, are we hitting the AML leukemic stem cells, and are we preserving the hematopoietic stem cells? Indeed, that's actually what we're seeing. I'll show you some snippets of that today. All right, dive in a little bit further. This is a product that is sourced from healthy adult donors. We get PBMCs from healthy adults. We can extract the NK cells. Then we transduce the NK cells with a single retroviral construct that contains three genes. That activating CAR I mentioned earlier recognizes two different targets that are well known in AML, CD33 as well as FLT3. These targets are chosen to be complementary with each other.
Over 95% of patients in AML have expression of one or both of these targets. CD33 tends to be found more on the AML blast, while FLT3 tends to be found more on the leukemic stem cell population. There has been a lot of efforts in the past actually to go after FLT3 as a tumor antigen target. One of the issues with this is that these targets are also expressed on those healthy stem cells in the blood, in the bone marrow. To avoid this type of toxicity, what we did was identify a target. In this case, it is called EMCN, endomucin. We did this through single-cell RNA-seq to look at endomucin being highly expressed on those healthy cells and not expressed on the cancer cells or minimally expressed on most cancer cells. This serves as that do-not-eat-me signal.
Finally, we have this IL-15 construct that we've designed that allows us to provide stimulation to the NK cells themselves as well as to the surrounding host immune system. I mentioned earlier this is an off-the-shelf process. It basically involves expanding cells in the manufacturing facility, cryopreserving the cells, and we're basically then able to ship out these vials in a frozen format anytime we have a patient study ready to come out to study. Our ongoing phase one trial, SENTI-202 here, is enrolling adult patients, relapse refractory, CD33 or FLT3 positive, heme malignancies. Right now, we're primarily focused on AML, although we do have the ability to expand into high-risk MDS as an example. These patients in general are receiving anywhere from one to three lines of prior AML treatments. We looked at two different dose levels.
We have a protocol for two different dose levels and two different schedules. In terms of dose level, we can treat with either 1 billion cells per dose or 1.5 billion cells per dose. The two different schedules are shown here on the right. There is a lymphodepletion step where we treat with fludarabine. What we can do is give patients either three doses of drug at day 0, 7, 14 or five doses of drug, 0, 3, 7, 10, 14. We then assess at 28 days, take a bone marrow there. Patients are allowed to get multiple cycles, including a consolidation cycle upon a response as well. In terms of patient characteristics in general, the study enrolled a high-risk relapsed refractory AML population with multiple baseline adverse characteristics. We have a little bit more breakdown a little bit later in terms of the exposure that the patients have received.
These patients have received, for example, exposure to previous treatment with fludarabine and Ara-C. That's the lymphodepletion agent. They've relapsed on those treatments as an example. Many of these patients have been exposed to venetoclax in the past. In terms of preliminary safety in these patients, generally, SENTI-202 was well tolerated. Most of the AEs that we observed were heme-related. A majority were deemed unrelated to SENTI-202 and were consistent with what you would expect with lymphodepletion. As I'll show you later, one of the key response criteria in AML, as many of you know, is that you get recovery of the healthy cells. That's certainly part of the mechanism that we've built into this product. We actually do see that quite robustly happening in our patients that do respond. Many of these AEs are essentially transient related to the LD.
In terms of responses across the cohorts, I'll show you a waterfall plot a little bit later. As I mentioned earlier, five out of seven of the patients did achieve an overall ORR response. That includes whether it's a CR, a CRh, or an MLFS type response. Four out of seven of the patients overall received a composite CR response, which is basically summing up the CR and the CRh. Two out of three in our preliminary RP2D core did achieve that composite CR response as well. All the patients were MRD negative. The patients that did respond were MRD negative. We are assessing MRD negativity right now through local site. Some sites use NGS to assess that. Other sites use flow cytometry. Flow cytometry sensitivity is something like 10^-4 in terms of sensitivity.
Diving in a bit more detail here, we are seeing in all of our patients significant reduction of bone marrow blasts in the majority of patients compared to their baseline. We had certain patients who came in with an extraordinarily high blast count. It was something in the range of not over 90%. We were able to see a significant reduction in all the patients across the dose cohorts. All right, in terms of walking through the waterfall plots here, we look at several characteristics in terms of the patients, in terms of are they deemed as sort of primary refractory disease. Many of our patients were. Do they have adverse risk criteria based on ELN 2022? Many of them were as well. Have they been exposed to fludarabine or Ara-C in the past and have failed on that treatment?
One question we sometimes get is, are you seeing these blast reductions simply by the fact that these patients are being treated with the lymphodepletion? We are able to see for several of our patients, especially in that RP2D cohort, that we have patients that are fludarabine refractory. These are patients that have shown responses. In our longest patient that we've followed so far, patient one, we've seen eight plus month durability in terms of that response. The shortest patient so far has been four plus months. We're continuing to follow and monitor these patients.
All right, so one of the things that we did as we collected this data set was not only look at the clinical characteristics, are these patients responding or not, but actually we take bone marrow samples and actually do pretty detailed cytoph analysis on those patient samples to get a sense of, is this product working the way we designed it? One of the key things we looked at in addition to seeing blast count decreases earlier is looking at the leukemic stem cell population. Currently, there is no product on the market that directly targets leukemic stem cells. As you can see here on the left-hand side, many of these leukemic stem cells are in the G0 phase. They're not expected to be susceptible to chemotherapy. This highlights some of the challenges with treating leukemic stem cells with that sort of approach.
If we look on the right-hand side here, looking at leukemic stem cells through this particular set of biomarkers, you can see in the majority of cases, we see a reduction in the leukemic stem cells in the bone marrow as we had anticipated. In the poster that we presented at ACR, which is available on our website, we've actually looked at this through a variety of different ways. There's a couple of different ways to measure LSC, and we see a consistent trend here. OK, I mentioned earlier one of the key things that we did when we designed this product is to try to protect those healthy hematopoietic stem cells or at least minimize killing against them. On the right-hand side, these are some of the patient samples we were able to obtain and look at HSPCs in the bone marrow sample.
In general, we see for several of our patients either a maintenance or an increase in the amount of the HSPCs in that bone marrow. This, we believe, reflects itself actually in the peripheral blood cell counts. One of the key response criteria in AML is not just looking at blast reduction, but also are the healthy cells coming back above a certain threshold? That threshold for CR oftentimes looks at platelets as well as ANC, which is what we are showing here on the left-hand side, illustrated in that blue line there. We see for many of our patients that do respond, certainly if they go above that sort of minimal blue line there, they have achieved a CR. Even those patients that have achieved CR, many of them actually are going way back up to normal again.
That was an exciting observation that we made. Again, we believe it's consistent with the mechanism here of being able to hit the cancer hard. You're trying to suppress the tumor burden, therefore allowing the normal blood system to reconstitute itself, but two, also sparing those HSPCs from over toxicity so that we can achieve this recovery in the healthy cell counts. In terms of PK, we are seeing a PK profile that's consistent in general with allogeneic NK cell therapy and other allocells as well. We are not expecting this product to be like an autologous product that lasts for years and years. The mechanism of action here is to go into the patients, hit the leukemic stem cell, hit the blast population as hard as possible, enable multiple cycles of treatment, and then hopefully that translates into longer-term outcomes for patients that are favorable.
In general, we haven't seen significantly different trends between the different dosing schedules and the dose amounts. We basically see cells that are detectable across about a 14-day time period. As you'd expect, when the immune system starts to reconstitute itself post-lymphodepletion, the cell population goes down. Where are we going with this? We're certainly very excited about the SENTI-202 program itself and the ability to continue to generate responses for patients in AML. We do want to expand the opportunity into other types of cohorts. We have a lot of clinicians interested in this program, thinking about, are we going to go into MDS or pediatric or certainly sort of earlier lines of AML? I think that's all on the table for our SENTI-202 product. Exciting opportunity to do that.
In addition to that, we think with increasing data collected on the logic gating working in humans, we have the potential to expand this into other cell therapies, especially for solid tumors, which is obviously a major indication. As I alluded to earlier, one of the challenges in this space has always been to find a clean target. You see a ton of crowding around the same indications or same targets from the ADCs or T-cell engagers or the CAR-Ts of the world. There is a large number of solid tumors that actually do not have a clean target like that. We have begun work internally to really mine for those additional indications and to develop programs against those.
Just to show you a highlight of what this actually looks like visually, because I think it goes a long way to kind of showcase how the logic gates work in a realistic form. This is under a microscope where we can actually mix healthy cells and cancer cells together. In this case, the cancer cells are in red, healthy cells are in green. Literally what we do is we put our product in with the different components of the gene circuit. We look at, are we killing the healthy cells? Are we sparing the healthy cells? Are we killing the cancer cells or not? You can look at all aspects of that under the same experiment. In this particular case, we've gone after a solid tumor target known as CEA.
CEA is a target that people have been trying to address for a while in colorectal cancer. The problem with CEA in previous clinical studies is that, in part, CEA is expressed on healthy epithelial cells, including in the colon and in the lung. There is this therapeutic window challenge. What we did here is we actually identified a different target called VSIG2, which is present on the healthy cells on green, but not present on the cancer cells. That serves as that do-not-kill sort of signal for the product. We've implemented the same type of logic gating in our CAR-NK. The cool thing about this is you can see no treatment. Both the green cells and the red cells grow at the same time. In the middle here, if you only go after CEA by itself, you kill both populations.
On the right-hand side here, you can see with the NOCKI technology, you're actually able to selectively enrich for those healthy cells while maintaining killing of the cancer cells. We have shown this works in T -cells as well. I think that gives us a lot of opportunity, both from an internal pipeline development and the ability to partner with other companies that have T -cell platforms available. This is going after the same set of targets. You can see the same type of behavior where on the right-hand side, no treatment, both healthy and cancer grow. The middle here, CEA CAR-T cells are very potent at killing both populations. The CEA not VSIG2 constructs allow us to essentially suppress the cancer cells and kill the healthy cells.
We're quite excited about where this can take us in terms of additional programs that the company can bring forward. With that, just stepping back a little bit from a company perspective, we are laser-focused as a company on SENTI-202. We think this product has a tremendous amount of opportunity for the relapsed refractory AML population in the near term and the ability to expand into nearby indications in MDS, in newly diagnosed AML, and pediatrics. In addition to that, this gene circuit technology has the ability to expand into solid tumors. That's part of our pipeline expansion. We closed an investment from several groups, NEA, Celadon Partners, Bayer Leaps, back in December. That is continuing to support the company as we go forward with additional clinical data that we're enrolling in SENTI-202 and hope to bring that to the market.
We've been really excited to have a great group of team members working with us. I'll just call out, in addition to Jay, who's here, our CFO, Kanya Rajangam, our President, Head of R&D and CMO, a lot of experience in developing cell therapy products in the past. Also for us on the right-hand side here, previously at Kite, Gilead, played an integral part in their scale-up of the Kite CAR-T products and somebody we brought in to really think about how do we move this product from the small-scale phase one work that we're doing now and really get this ready for a pivotal commercial scale. With that, I'd like to thank you guys for your time and happy to take any further questions.
Thank you so much for coming and joining us today.
Transforming the treatment of oncology with our logic gating technology. What I'm going to tell you about is our logic gates allow us to go after cancer indications where there is no clean single target. We can essentially recognize multiple antigens on the cancer cells and kill. We can also recognize antigens on healthy cells and protect those cells from being affected. We think that's a really differentiated mechanism compared to other ADCs, T-cell engagers, or CAR-T cells out there in the market. We've implemented this into our lead program, SENTI-202. It's an off-the-shelf CAR-NK program going after two AML targets, CD33 as well as FLT3. It protects healthy hematopoietic stem cells by recognizing a target called EMCN or endomucin. This is an off-the-shelf program, meaning that we have the cells cryopreserved. We're shipping them on demand to patients.
Patients are getting quick treatment. We recently announced at ACR some really exciting positive data, as I'll share with you. We've treated now seven patients for which we have reportable data on. Four out of seven were able to achieve MRD negative complete responses with longest durability, eight months and counting. This gene circuit platform does offer us the opportunity to expand into solid tumors downstream and also works in both CAR-T and CAR-NK cell platforms. Senti was founded originally out of technology from a variety of universities, including at MIT, where I used to be on faculty. Basically, we've developed now four classes of what we call genetic circuits or technologies that we can put into any cellular gene therapy product. Today, I'm going to primarily focus on this logic gating concept. I'll walk you through that in a little bit.
We also have the ability to multi-arm ourselves, so engineering a cell therapy that not only contains a CAR, but has one or more payloads, including cytokines built into it to enhance its efficacy. We have something called the regulator dial, which basically allows us to control using an FDA-approved small molecule. It allows us to control what a gene or cell therapy is doing in the body, as well as a library of synthetic promoters that constitute what we call smart sensors that can turn on at the right place at the right time in the body. What does the logic gate allow us to do? If you think about current CAR-T cell therapies, ADCs, T-cell engagers, you're oftentimes faced with finding that clean target that's only expressed on the cancer cells and not expressed on the healthy cells.
If you find a target that doesn't have that clean profile, what happens is you basically will achieve not only killing of the cancer cells, but killing of healthy tissues. That limits your ultimate therapeutic window. Fundamentally, that comes from the fact that there aren't just that many clean targets that are out there. I think we're limited by the biology. That's just the reality of cancer. The way we've chosen to overcome this problem is instead of recognizing just a single target, why not engineer a product that can recognize multiple targets? We have what we call an activating CAR that recognizes multiple cancer targets. In the case of AML, this is CD33 as well as FLT3.
We have what we call the NOT gate, which is implemented through this inhibitory CAR or iCAR that recognizes the healthy cell and delivers a do-not-kill signal into the product. That basically keeps those healthy cells from being affected. We have a pipeline of programs that are using this technology, the most advanced, SENTI-202, which is what I'm going to focus on today. SENTI-202 is focused on AML. AML is a really difficult-to-treat hematological cancer with over 20,000 patients newly diagnosed in the US per year. Current standard of care for patients that are in that relapsed refractory condition, unfortunately, these patients are seeing CR rates with current standard of care maybe in the 15%-25% range. Median survival for these patients is about five months. There are several challenges with AML.
It's a very heterogeneous disease, meaning that there is sort of a wide diversity of AML clones as well as leukemic stem cells in patients that basically make it difficult to eradicate sort of deeply the cancer. To achieve longer, better durable responses, we think we have to hit that heterogeneous nature of the disease. Oftentimes, probably have to go after more than a single target to do so. Secondly, all of the AML targets that we know of on the cell surface are also expressed on healthy cells. Many of them are expressed on healthy bone marrow cells or sometimes in cases on endothelial cells. If you just go after a single target by itself, what's been seen in clinical studies over and over again is you see a lot of toxicity and limited therapeutic window in treating these patients.
I'm going to talk you through SENTI-202, how it's designed, and the detailed clinical data. Just to give you a highlight of what we presented about a couple of weeks ago, SENTI-202 is enrolling currently in a relapsed refractory AML patient population. Out of the seven patients for which we had data to report, five out of seven achieved an ORR response. Four out of the seven achieved a composite CR response, including two out of three in our preliminary RP2D cohort. All four of those CCR patients did achieve MRD negative status, which is a really important part for us to determine are we really hitting the tumor as deeply as we can. All of our complete response patients are continuing to maintain their responses anywhere from four plus to eight plus months and ongoing. From a safety perspective, SENTI-202 is quite clean.
We do have the ability to enroll patients in an outpatient setting. No major DLTs detected. We did declare our preliminary RP2D based on and part of this data set. The other thing that is really exciting about that data set we put out at ACR is in addition to that oral presentation of this data on the clinical side, we also showed a lot of detailed cytoph data on the patient samples themselves. We take bone marrow from these patients every cycle, every 28 days. From that, we can actually look deeply into are we hitting the AML blast? Are we hitting the AML leukemic stem cells? Are we preserving the hematopoietic stem cells? Indeed, that is actually what we are seeing. I will show you some snippets of that today. All right, dive in a little bit further.
This is a product that is sourced from healthy adult donors. We get PBMCs from healthy adults. We can extract the NK cells. Then we transduce the NK cells with a single retroviral construct that contains three genes. That activating CAR I mentioned earlier recognizes two different targets that are well known in AML, CD33 as well as FLT3. These targets were chosen to be complementary with each other. Over 95% of patients in AML have expression of one or both of these targets. CD33 tends to be found more on the AML blast, while FLT3 tends to be found more on the leukemic stem cell population. There has been a lot of efforts in the past actually to go after FLT3 as a tumor antigen target.
One of the issues with this is that these targets are also expressed on those healthy stem cells in the blood, in the bone marrow. To avoid this type of toxicity, what we did was identify a target. In this case, it's called EMCN, endomucin. We did this through single-cell RNA-seq to look at endomucin being highly expressed on those healthy cells and not expressed on the cancer cells or minimally expressed on most cancer cells. This serves as that do-not-eat-me signal. Finally, we have this IL-15 construct that we've designed that allows us to provide stimulation to the NK cells themselves as well as to the surrounding host immune system. I mentioned earlier this is an off-the-shelf process. It basically involves expanding cells in the manufacturing facility, cryopreserving the cells.
We're basically then able to ship out these vials in a frozen format anytime we have a patient study ready to come out to study. Our ongoing phase one trial, SENTI-202 here, is enrolling adult patients, relapsed refractory CD33 or FLT3 positive heme malignancies. Right now, we're primarily focused on AML, although we do have the ability to expand into high-risk MDS as an example. These patients in general are receiving anywhere from one to three lines of prior AML treatments. We looked at two different dose levels. We have a protocol for two different dose levels and two different schedules. In terms of dose level, we can treat with either 1 billion cells per dose or 1.5 billion cells per dose. The two different schedules are shown here on the right. There's a lymphodepletion step where we treat with fludarabine and aracy.
What we can do is give patients either three doses of drug at day 0, 7, 14 or five doses of drug, 0, 3, 7, 10, 14. We then assess at 28 days, take a bone marrow there. Patients are allowed to get multiple cycles, including a consolidation cycle upon a response as well. In terms of patient characteristics, in general, the study enrolled a high-risk relapsed refractory AML population with multiple baseline adverse characteristics. We have a little bit more breakdown a little bit later in terms of the exposure that the patients have received. These patients have received, for example, exposure to previous treatment with fludarabine and Ara-C. That is the lymphodepletion agent. They have relapsed on those treatments as an example. Many of these patients have been exposed to venetoclax in the past. In terms of preliminary safety in these patients, generally, SENTI-202 was well tolerated.
Most of the AEs that we observed were heme-related. A majority were deemed unrelated to SENTI-202 and were consistent with what you would expect with lymphodepletion. As I'll show you later, one of the key response criteria in AML, as many of you know, is that you get recovery of the healthy cells. That is certainly part of the mechanism that we've built into this product. We actually do see that quite robustly happening in our patients that do respond. Many of these AEs are essentially transient related to the LD. In terms of responses across the cohorts, I'll show you a waterfall plot a little bit later. As I mentioned earlier, five out of seven of the patients did achieve an overall ORR response. That includes whether it's a CR, a CRh, or an MLFS type response.
Four out of seven of the patients overall received a composite CR response, which is basically summing up the CR and the CRh. Two out of three in our preliminary RP2D cohort did achieve that composite CR response as well. All the patients were MRD negative. The patients that did respond were MRD negative. We are assessing MRD negativity right now through local site. Some sites use NGS to assess that. Other sites use flow cytometry. Flow cytometry sensitivity is something like 10^-4 in terms of sensitivity. Diving in a bit more detail here, we are seeing in all of our patients significant reduction of bone marrow blasts in the majority of patients compared to their baseline. We had certain patients who came in with an extraordinarily high blast count. It was something in the range of not over 90%.
We were able to see a significant reduction in all the patients across the dose cohorts. In terms of walking through the waterfall plots here, we look at several characteristics in terms of the patients, in terms of are they deemed as sort of primary refractory disease. Many of our patients were. Do they have adverse risk criteria based on ELN 2022? Many of them were as well. And have they been exposed to fludarabine or Ara-C in the past and have failed on that treatment? One question we sometimes get is, are you seeing these blast reductions simply by the fact that these patients are being treated with the lymphodepletion? We are able to see for several of our patients, especially in that RP2D cohort, that we have patients that are fludarabine Ara-C refractory. These are patients that have shown responses.
In our longest patient that we've followed so far, patient one, we've seen eight-plus month durability in terms of that response. The shortest patient so far has been four-plus months. We're continuing to follow and monitor these patients. One of the things that we did as we collected this data set was not only look at the clinical characteristics, are these patients responding or not, but actually we take bone marrow samples and actually do pretty detailed cytoph analysis on those patient samples to get a sense of, is this product working the way we designed it? One of the key things we looked at in addition to seeing blast count decreases earlier is looking at the leukemic stem cell population. Currently, there is no product on the market that directly targets leukemic stem cells.
As you can see here on the left-hand side, many of these leukemic stem cells are in the G0 phase. They're not expected to be susceptible to chemotherapy. This highlights some of the challenges with treating leukemic stem cells with that sort of approach. If we look on the right-hand side here, looking at leukemic stem cells through this particular set of biomarkers, you can see in the majority of cases, we see a reduction in the leukemic stem cells in the bone marrow as we had anticipated. In the poster that we present at AACR, which is available on our website, we've actually looked at this through a variety of different ways. There are a couple of different ways to measure LSCs. We see a consistent trend here.
OK, I mentioned earlier one of the key things that we did when we designed this product is to try to protect those healthy hematopoietic stem cells or at least minimize killing against them. On the right-hand side, these are some of the patient samples we were able to obtain and look at HSPCs in the bone marrow sample. In general, we see for several of our patients either a maintenance or an increase in the amount of the HSPCs in that bone marrow. This, we believe, reflects itself actually in the peripheral blood cell counts. One of the key response criteria in AML is not just looking at blast reduction, but also are the healthy cells coming back above a certain threshold?
That threshold for CR oftentimes looks at platelets as well as ANC, which is what we're showing here on the left-hand side is illustrated in that blue line there. We see for many of our patients that do respond, certainly if they go above that sort of minimal blue line there, they've achieved a CR. Even those patients that achieve CR, many of them actually are going way back up to normal again. That was an exciting observation that we made. Again, we believe it's consistent with the mechanism here of being able to hit the cancer hard. You're trying to suppress the tumor burden, therefore allowing the normal blood system to reconstitute itself, but two, also sparing those HSPCs from overt toxicity so that we can achieve this recovery in the healthy cell counts.
In terms of PK, we are seeing a PK profile that's consistent in general with allogeneic NK cell therapy and other allocells as well. We are not expecting this product to be like an autologous product that lasts for years and years. The mechanism of action here is to go into the patient, hit the leukemic stem cell, hit the blast population as hard as possible, enable multiple cycles of treatment. Hopefully that translates into longer-term outcomes for patients that are favorable. In general, we haven't seen significantly different trends between the different dosing schedules and the dose amounts. We basically see cells that are detectable across about a 14-day time period. As you'd expect, when the immune system starts to reconstitute itself post-lymphodepletion, the cell population goes down. Where are we going with this?
We're certainly very excited about the SENTI-202 program itself and the ability to continue to generate responses for patients in AML. We do want to expand the opportunity into other types of cohorts. We have a lot of clinicians interested in this program of thinking about, are we going to go into MDS or pediatric or certainly sort of earlier lines of AML? I think that's all on the table for our SENTI-202 product. Exciting opportunity to do that. In addition to that, we think with increasing data collected on the logic gating working in humans, we have the potential to expand this into other cell therapies, especially for solid tumors, which is obviously a major indication. As I alluded to earlier, one of the challenges in this space has always been to find a clean target.
You see a ton of crowding around the same indications or same targets from the ADCs or T-cell engagers or the CAR-T's of the world. There's a large number of solid tumors that actually don't have a clean target like that. We've begun work internally to really mine for those additional indications and to develop programs against those. Just to show you a highlight of what this actually looks like visually, because I think it goes a long way to kind of showcase how the logic gates work in a realistic form. This is under a microscope where we can actually mix healthy cells and cancer cells together. In this case, the cancer cells are in red, healthy cells are in green. Literally what we do is we put our product in with the different components of the gene circuit.
We look at, are we killing the healthy cells? Are we sparing the healthy cells? Are we killing the cancer cells or not? You can look at all aspects of that under the same experiment. In this particular case, we've gone after a solid tumor target known as CEA. CEA is a target that people have been trying to address for a while in colorectal cancer. The problem with CEA in previous clinical studies is that, in part, CEA is expressed on healthy epithelial cells, including in the colon and in the lung. There is this therapeutic window challenge. What we did here is we actually identified a different target called VSIG2, which is present on the healthy cells on green, but not present on the cancer cells. That serves as that do-not-kill sort of signal for the product.
We have implemented the same type of logic gating in our CAR-NK. The cool thing about this is you can see no treatment, both the green cells and the red cells grow at the same time. In the middle here, if you only go after CEA by itself, you kill both populations. On the right-hand side here, you can see with the NOT gate technology, you are actually able to selectively enrich for those healthy cells while maintaining killing of the cancer cells. We have shown this works in T-cells as well. I think that gives us a lot of opportunity, both from an internal pipeline development and the ability to partner with other companies that have T-cell platforms available. This is going after the same set of targets.
You can see the same type of behavior where on the right-hand side, no treatment, both healthy and cancer grow. The middle here, CEA CAR-T cells are very potent at killing both populations. The CEA not VSIG2 constructs allow us to essentially suppress the cancer cells and kill the healthy cells. We are quite excited about where this can take us in terms of additional programs that the company can bring forward. With that, just to give you just stepping back a little bit from a company perspective, we are laser-focused as a company on SENTI-202. We think this product has a tremendous amount of opportunity for the relapse refractory AML population in the near term and the ability to expand into nearby indications in MDS, in newly diagnosed AML, and pediatrics.
In addition to that, this gene circuit technology has the ability to expand into solid tumors. That is part of our pipeline expansion. We closed an investment from several groups, NEA, Celadon Partners, Bayer Leaps back in December. That is continuing to support the company as we go forward with additional clinical data that we're enrolling in SENTI-202 and hope to bring that to the market. We've been really excited to have a great group of team members working with us. I'll just call out, in addition to Jay, who's here, our CFO, Kanya Rajangam, our President, Head of R&D and CMO, a lot of experience in developing cell therapy products in the past. Also for us on the right-hand side here, previously at Kite, Gilead, played an integral part in their scale-up of the Kite CAR-T products.
Somebody we brought in to really think about how do we move this product from the small-scale phase one work that we're doing now and really get this ready for pivotal and commercial scale. With that, I'd like to thank you guys for your time and happy to take any further questions.