Thank you for waiting, everybody. We'll now start Daiichi Sankyo's WCLC/ESMO 2024 highlights. This is Ken Asakura from Corporate Communications, I will be the moderator today. First, about the language for this event. All presentations will be in English, Q&A will be in both English and Japanese. Simultaneous interpretation is provided, please click the interpretation icon at the bottom of the screen and select either English, Japanese, or original audio. When original audio is selected, you will hear the original sound. English presentation slides will be shown on the Zoom screen as well as live streaming screen. The presentation slides have been uploaded to IR Presentation Material page of Investors section on our corporate website. Please view or download as needed. Our presenter for today is Dr. Mark Rutstein, our Head of Global Oncology Clinical Development. We will take questions after the presentation.
Dr. Ken Takeshita, our Head of Global R&D, will open the meeting and will join the Q&A. Please note that this event will be recorded. With that, let's start the event. Ken-san, please.
Okay, Asakura-san. Thank you very much for that very nice introduction. Today's session is an update on all the data we have presented at the World Conference on Lung Cancer and the ESMO, the European Society for Medical Oncology conferences. We have a lot to cover, I'm not going to spend too much time on these introductory remarks, except just to say that what you'll see is a very steady progress in our DXd ADC portfolio, a lot of productive information and data coming from our industry collaborators, as well as our academic collaborators. Also now finally, some emerging data on our new next generation ADC. We're very excited with all the progress we are making. Let me just turn it over to you, Mark, to start to go over the large amount of slides that we have.
Yes. Ken, thanks so much, it's my great pleasure to be here today. I'm excited to be here. My name is Mark Rutstein. I lead oncology clinical development at Daiichi Sankyo. Following on Ken's comments, the key themes we have across World Conference on Lung Cancer and ESMO include really making progress toward creating new potential standards of care, not only in lung cancer, which was the highlight of World Conference on Lung Cancer, but also in additional tumors as we saw at ESMO, ovarian cancer, endometrial cancer, breast cancer, et cetera. This is an exciting time for us for sure. Indeed, we also look to create new innovation for patients, including the introduction of our second ADC platform, which we presented at the ESMO meeting, we'll touch on that today. Okay. With that, we'll get right in.
The first topic is the final overall survival analysis from TROPION-Lung01. Next, please. People may recall, this was a positive phase III trial in second line non-small cell lung cancer with Dato-DXd monotherapy, statistically significant progression-free survival. Of course, we saw the benefit was really driven by the non-squamous histology, where we saw a clinically meaningful benefit in that subpopulation. The PFS is there on the middle of the slide, 0.63 hazard ratio, and then more than doubling of the overall response rate. Next. I think this is just the schema and simply that this is a second line plus population in non-small cell lung cancer. The design was, of course, Dato-DXd monotherapy versus docetaxel for this phase III trial called TROPION-Lung01. Maybe just a quick reminder that we had dual primary endpoints, progression-free survival by BICR and overall survival. Next.
Baseline characteristics. Because we've shown this before, I'll be quite brief. You can see prior lines of therapy. The majority of patients had it at, well, second or third line, but even greater lines of therapy. Definitely high unmet need pretreated population. The AGA population represented about a minority of patients at 17%. Not a surprise that in terms of prior systemic therapies, patients would have been pretreated mostly with platinum and anti-PD-L1 or targeted therapy if AGA. Three-quarters of the patients approximately were non-squamous. Next, please. Here's the intent to treat overall survival analysis, remembering this is the dual primary endpoint. You can see there's a numerical improvement with Dato-DXd over docetaxel. However, this primary endpoint did not meet statistical significance.
You can see P value 0.530 and hazard ratio 0.94. Next. Looking into the subgroup analyses, that tells the tale of this pivotal trial data set. Down bottom of the slide in this forest plot, you can see that by histology, we saw the difference, overall survival by histology. Hazard ratio 0.84 for nonsquamous patients, 1.32 for squamous patients. Also difference in AGA as well, with AGA present having hazard ratio of 0.66. Next, please. Here are the Kaplan-Meier, looking more closely at efficacy by histology. As I noted, nonsquamous population, 0.84 hazard ratio. The medians are worth pointing out. Median survival, 14.6 versus 2.3, so actually a 2.3 median improvement in overall survival, which we've indicated is clinically meaningful in this population. In the squamous patients in the right, you can see there is no evidence of benefit there. Down the bottom of the slide, very briefly, we looked at subgroups of subgroups. Including whether or not in the nonsquamous subgroup, patients benefited regardless of AGA status, and indeed, there is evidence that they did. Next slide, please.
For safety. Not much new here. This is 11 months additional follow-up, and the safety profile was consistent with the previous analysis. No new late onset toxicities. It remains to be the case that there were fewer Grade 3 and higher treatment-related adverse events or discontinuations or dose reductions in the Dato arm compared to the docetaxel arm. Of course, there was a difference in median treatment duration favoring Dato-DXd. Next.
This is just more closely at the adverse event profile, stomatitis events were common, but primarily Grade 1 and Grade 2. Hematologic events, as we showed previously, were much more common with docetaxel. Importantly, with this 11 months additional follow-up, there were no new adjudicated drug-related ILD events or deaths since the previous analysis for PFS endpoint. Overall, a really consistent profile with what we know. Next. Conclusions here is that TROPION-Lung01 did meet its primary endpoint of PFS, statistically significant improvement. Dual primary endpoint of overall survival did not meet statistical significance, showed a numerical improvement, but consistent benefits was seen in that nonsquamous subgroup across the endpoint structure and now with the overall survival that was shown, no new safety signals. We will talk, this last bullet really talks about that TROP2 Normalized Membrane Ratio, that QCS assay.
We'll move right on into the next presentation, which will follow up on this. Next, please. Okay. At World Lung Cancer, there was not only the presentation of the final overall survival analysis for TROPION-Lung01, but also this biomarker presentation that was actually at the presidential session. This was Normalized Membrane Ratio of TROP2 by Quantitative Continuous Scoring, showing that it could be potentially predictive of outcomes in that same study that we just presented the overall survival update, TROPION-Lung01. Next. As a background, Dato-DXd is an ADC that has a plasma-stable linker. Because of that, Dato-DXd is believed to rely heavily on internalization, then to release its cytotoxic payload. We had looked at conventional IHC scoring, and it was not predictive of response to Dato-DXd in non-small cell lung cancer.
Seeing that conventional IHC was not going to be predictive, there were additional biomarker discovery efforts, starting with TROPION-PanTumor01, the phase I study where Dato-DXd monotherapy was given to patients with non-small cell lung cancer. It was hypothesized that a more precise and quantitative assessment of TROP2 expression on both cell membrane and cytoplasm may predict efficacy of Dato-DXd. The thinking behind the cytoplasmic expression is that this could be representative, a surrogate for internalization, where when you have an ADC with a plasma-stable linker, the cytoplasmic expression therefore may be important and beyond just the cell membrane expression with a conventional IHC. Next. This is the methodology, and this is an assay technology that's been developed by AstraZeneca, our partner. It starts with conventional immunohistochemistry. You deploy a TROP2 assay using conventional immunohistochemistry, and then there's a whole slide imaging.
The slides undergo automated image analysis, QCS or Quantitative Continuous Scoring . What is QCS? QCS is an artificial intelligence-informed computational pathology approach. It is capable, and I'm on the bottom of the slide now, of differentiating tumor from non-tumor. Then very importantly, it would look within a tissue slide at all the tumor cells within that slide and be able to quantitate the analytes, including not only membrane TROP2 expression, but subcellular or cytoplasmic expression. The assay algorithm that was developed on the far right is to calculate this Normalized Membrane Ratio, looking at the ratio of membrane expression over total expression, where total expression includes membrane expression and cytoplasmic expression. You'll remember on the previous slide, I mentioned why cytoplasmic expression may be relevant to an ADC that has a plasma stable linker.
On the upper right, patient biomarker status using this algorithm was determined as follows. Greater than equal to 75% of tumor cells with TROP2 NMR or Normalized Membrane Ratio less than 0.56 is positive, less than 75% of tumor cells with that NMR, 0.56 was negative. What's important here is that as the cytoplasmic expression goes up, the ratio goes down, we're really considering the cytoplasmic expression in this NMR ratio. Next. These are the TROPION-Lung01 progression-free survival results in the original intent-to-treat analysis. As we mentioned a few slides ago, PFS hazard ratio in the ITT population was 0.75, statistically significant. We move to the next slide.
The biomarker evaluable population was a subset of the total ITT TROPION-Lung01 population, these are patients with available tissue samples for QCS determination. The biomarker cut points were optimized, in this case, for PFS in the non-squamous, non-AGA patients from TROPION-Lung01. Biomarker cut points were rigorously evaluated using robust statistical analysis. On the bottom left, because we're going to show two PFS analyses, progression-free survival using these TROP2 assay QCS. The first is the total biomarker evaluable population, that includes non-squamous, non-AGA, non-squamous AGA, and squamous patients. We're going to show the PFS for that subgroup of patients, non-squamous, non-AGA, for which the assay is being optimized.
On the right, you can see the prevalence of NMR positivity or biomarker positivity, 66% for non-squamous, within that, 63% for non-squamous, non-AGA, 76% for non-squamous AGA, squamous at 44%. It's notable to us that the biomarker positive prevalence is higher in non-squamous than squamous patients, perhaps this is one contribution to the observed differential efficacy by histology with Dato-DXd in non-small cell lung cancer. Next slide. It's very important when you are looking to validate a biomarker that the biomarker evaluable population and the baseline characteristics are similar to the ITT or overall population, that is the case. We won't go over any individual numbers here, but just to show you that the baseline characteristics are very similar between the populations for which we're now going to show the biomarker evaluated efficacy analysis and the total population. Next, please.
Here is that first analysis, that overall biomarker evaluable population. You can see that when you compare efficacy by NMR positive patients and NMR negative patients, so biomarker positive by TROP2 QCS versus negative by TROP2 QCS, you can see that Dato-DXd versus docetaxel has an improved PFS hazard ratio. You can see it's 0.57, median PFS 6.9 versus 4.1, overall response rate was significantly higher with Dato-DXd versus docetaxel. On the other hand, in the biomarker negative cohort on the right, you can see the PFS hazard ratio is 1.16. This is evidence in this exploratory analysis, this is a retrospective exploratory analysis, that the TROP2 QCS assay may provide potential to predict efficacy to Dato-DXd in this setting. Next.
This is the second analysis where there was optimization of the assay in the population of interest, which is non-squamous, non-AGA. This is the progression-free survival with overall response rate. Here again, you can see improvement in PFS and higher response rate in patients that are TROP2 QCS -NMR positive. PFS hazard ratio now down to 0.52, median PFS 7.2 versus 4.1, and also an improvement in response rate. Again, evidence, this time in the biomarker optimized population, that this assay has the potential to predict efficacy for Dato-DXd. Next. Here are the safety data. What you can see in the biomarker positive population versus the negative population, overall similar safety. Grade 3 and higher treatment-related adverse event rates were similar.
There were some adverse events of interest where there was a modestly higher rates in the biomarker positive population compared to the negative population. Not sure why this would be, or if it's maybe related to the smaller subgroups. Overall, fairly consistent safety. Next, please. In summary, this TROP2 Normalized Membrane Ratio as measured by QCS, demonstrated several key outcomes and the potential ability to predict outcomes in this exploratory TROPION-Lung01 analysis. Biomarker positivity was more prevalent in non-squamous than squamous patients, and we saw that together. Patients receiving Dato-DXd who were biomarker positive had a higher response rate and longer PFS compared to those who were negative. The overall Grade 3 treatment-related adverse event rates were similar.
Importantly, further investigation of this promising biomarker and assay is ongoing in two pivotal trials with Dato-DXd, AVANZAR in frontline non-small cell lung cancer, with Dato-DXd durvalumab and platinum. TROPION-Lung10, which is the phase III frontline trial, PD-L1 50% and higher with Dato-DXd, plus the immunotherapy IO bispecific agent, TIGIT PD-1 inhibitor rilvegostomig. QCS-NMR has the potential to be the first TROP2 biomarker and the first computational pathology biomarker to predict outcome for clinical response to Dato-DXd in non-small cell lung cancer. Next, please. We're going to stay with Dato-DXd here, but now we're going to show data in an earlier line of therapy. It's important that when we study our DXd ADCs, we consider not only the advanced and metastatic setting, but also even potentially the resectable setting.
NeoCOAST-2 is a study of the efficacy and safety of neoadjuvant durvalumab plus novel anticancer agents, including Dato-DXd plus chemotherapy, and then continued with adjuvant durvalumab, and in some cases, novel agent in resectable non-small cell lung cancer. Next slide. Here we're going to focus on what's in red. Because there are multiple combinations tested here, including with some of AstraZeneca's agents within its pipeline, like oleclumab, which is a CD73 inhibitor, and monalizumab, which is an NKG2A inhibitor. We're going to focus on Dato-DXd in red. This is resectable non-small cell lung cancer. These are EGFR/ALK wild-type patients. You can see Arm 4, they get Dato-DXd plus the PD-L1 inhibitor durvalumab plus single agent platinum. There are 54 patients in that cohort. They undergo surgery, and they get durvalumab alone.
What we're going to focus on here are the primary endpoints. A primary endpoint of pathologic complete response rate. That means clearance of the tumor, no residual tumor found in the surgical specimen. There's also major pathologic response rate, which is less than or equal to 10% residual tumor in the surgical specimen. We're also going to look at some safety, including feasibility of surgery. Next slide, please. Okay. Again, I'm going to focus on the red. Dato-DXd plus durvalumab plus chemotherapy. We can see 54 patients started neoadjuvant therapy. Very importantly, there was a higher rate of patients that could go to surgery. If you give a neoadjuvant treatment before surgery, a key aspect is to understand if patients can tolerate that pre-surgical regimen enough to continue on to surgery.
Indeed, in the Dato-DXd containing Arm 4, only two patients did not make it there to surgery. There was also a high rate of what we call R0 surgical resection, which is where you look at the tumor margins and you see negative tumor at the margins. A majority of patients were able to start the adjuvant therapy. This is a good sign toward tolerability. Next, please.
Okay. Again, focusing on the red on the right. Dato-DXd, durvalumab, and chemotherapy showed a pathologic complete response rate of 34.1% and a major pathologic response rate of 65.9%. This is a relatively modest size phase II data set, it is a preliminary signal, but it is a promising preliminary signal when one considers the historical control pathologic complete response rate of chemotherapy and immunotherapy alone, which is really in the mid-20% range. Next slide, please. Just quickly showing that, again, focusing on the red on the right, that among patients who had that pCR rate of 34.1% in Arm 4, there were pathologic complete responses observed across PD-L1 status, with a gradient, though, with higher pathologic complete response rate and higher PD-L1 expression levels. Next, please.
Here's the safety profile, I already showed you that the majority of patients could go on to surgery after receiving the Dato-DXd containing neoadjuvant regimen. I'll focus to the left where we see in the neoadjuvant phase where Dato-DXd is given, the grade greater than equal to three treatment emergent adverse event rate was 18.5%, and AEs leading to discontinuation was less than 10%. On the right side, when we look at the types of adverse events that we saw in the neoadjuvant phase, you can see that cytopenias were observed, which isn't surprising with chemotherapy on board. Otherwise, some very typical adverse events associated with what you'd expect from that triplet regimen. The majority of events being Grade 1 to 2. Next, please.
Okay, fundamentally what we say here is that in this phase II study in perioperative non-small cell lung cancer, the combination of Dato-DXd durvalumab and chemotherapy showed a promising pathologic complete response rate and major pathologic response. As I indicated, it was a manageable safety profile with the ability to continue to surgery comparable to already approved regimens. This is the first global phase II study showing encouraging efficacy and manageable safety of an antibody-drug conjugate in the neoadjuvant setting for patients with resectable non-small cell lung cancer. Next, please.
Okay, we'll continue with Dato-DXd. Now as we'd explained when we provide our R&D strategy, we look to what we call expand, which is we focus originally on breast cancer and lung cancer, but then look at earlier lines of therapy like we just presented with NeoCOAST-2, also additional tumors beyond breast and lung cancer. The TROPION-PanTumor03 study was presented, this is monotherapy Dato-DXd in ovarian cancer and endometrial cancer. Next slide, please. Okay, in this study, there's a cohort of 35 ovarian cancer patients. They progressed on at least one line of platinum, but no more than two lines of therapy for advanced or metastatic disease. There are 40 patients in the endometrial cohort, and all of these patients, by the way, are unselected for TROP2 expression.
In the endometrial cohort as well, they progressed on at least one prior line of platinum therapy and had no more than two prior lines. They get Dato-DXd monotherapy, the primary endpoint is overall response rate and safety. Next, please. Okay, here are the baseline characteristics for this ovarian and endometrial cohorts. We'll focus on the previous lines of therapy, third from bottom. You can see these are significantly pretreated patients. In the ovarian cohort, nearly 69% of patients had received two or more regimens. In the endometrial cohort, about 28% of patients had received two or more regimens. These patients are all pretreated. You can see the pretreatments there, including 100% of patients receiving platinum, a number of other standard of care regimens.
On the bottom, in the ovarian cohort, you can see that the majority of the patients are platinum resistant, which represents a particularly unmet medical need. Next slide, please. Here are the efficacy in this signal finding study for ovarian cancer. You can see the overall response rate in the total population was nearly 43%. Duration of response, close to six months. You can see in the smaller cohort with platinum sensitive disease, nearly a 67% response rate, but a small sub cohort. The platinum-resistant population with a nearly 35% response rate. If you look to the waterfall plot on the right, you can see what is a promising figure. If we move to the next slide, please. Now, the endometrial cancer cohort. Confirmed overall response rate of 27.5%.
We can see, interestingly, the median duration response for those who responded was 16.4 months. The waterfall plot on the right, also favorable. Again, demonstrating the potential of Dato-DXd in additional tumors, including pretreated ovarian cancer and endometrial cancer. Next slide, please.
We look at the treatment duration, and it was 5.6 months in the ovarian cohort. When you look at safety overall, adverse events were fairly low and significantly less than 10% in each arm. There were no deaths. In the event of treatment limitation, about half the patients were allowable, and there were only three patients with Grade 3 adverse events. Next slide, please. Overall, there were no deaths. We look at the adversity, there were non-adverse events that were across the board expected with this regimen, which already the majority is what we report here. ILD was reasonably low. There was only one ILD in each cohort, Grade 1 to 3. Then there were ocular surgeries. A minority of those were Grade 3. In fact, there were no Grade 3 events in the ovarian cohort.
You can see the most common adverse events below in the slide are typical of what we'd expect for Dato-DXd. Next slide, please. Okay. Overall, Dato-DXd monotherapy demonstrates encouraging efficacy in ovarian cancer and endometrial cancer. We've already covered the numbers, but we would say encouraging early efficacy and the safety profile is manageable and consistent with what we already know about Dato-DXd. Next slide, please. Okay. We'll now transition to trastuzumab deruxtecan. We know that in HER2 or T-DXd is approved in HER2 mutation-positive non-small cell lung cancer. This is a signal finding cohort in HER2 overexpressing non-small cell lung cancer. This is DESTINY-Lung03, part one. Next slide, please. Here's the study design, and you can see it has multiple cohorts.
We're going to focus on part one, which is T-DXd monotherapy, patients getting 5.4 mg/kg every three weeks. Now, patients in this part one had one or two prior lines of therapy, and those who were eligible to receive targeted therapies, had to have received previous targeted therapy. Key endpoint included overall response rate and other efficacy parameters as well as safety. Next slide, please. Here are the baseline characteristics. I'm going to focus on the right, and you can see that about half the patients were IHC 3 + or IHC 2 +, and all patients were required to be IHC 2 + or IHC 3 + to get into this HER2 overexpressing cohort. For PD-L1 status, the majority of patients were less than 1% or 1% - 49%.
Importantly, on the bottom right, there were a significant percentage of patients who had received EGFR TKI, therefore indicating EGFR mutation positivity. Next slide, please. Here is the efficacy. In the 36 patients enrolled in this Arm 1D, part one, T-DXd monotherapy, a confirmed response rate of 44%. You can see, the median duration of response was 11.0 months. A notable and encouraging signal of T-DXd in the HER2 overexpressing population. Next slide, please. Here we show the responses according to PD-L1 status and HER2 expression status. You can see on this waterfall plot that there were responses across the spectrum of PD-L1 expression and across the two subgroups of IHC 3 + and IHC 2 + IHC expression.
What I'm not showing, but will be mentioned on the conclusion slide, is that there was a higher overall response rate in patients who were actually EGFR mutation positive. I think we'll see that on the conclusion slide, but I wanted to verbalize that here now. We looked at multiple subgroups within this data set. Next slide, please. For safety. Overall, we don't see anything new here related to T-DXd. On the left, drug-related AEs related to discontinuation were less than 10%. About 20% of patients had dose reduction. Importantly, in this lung cancer population, the rate of ILD was about 5%. There were two events, and they were both Grade 2. There was one death, and it was due to neutropenic colitis.
On the right side, you can see the most common adverse events, which are really consistent with what we already know about T-DXd. Next slide, please. In summary, we see an encouraging signal for T-DXd monotherapy in pretreated IHC 3 + and IHC 2 + HER2 overexpressing metastatic non-small cell lung cancer. We did see responses regardless of IHC status and PD-L1, we also did see, and it's at that third bullet down, high response rate in EGFR TKI pretreated patients than those who weren't pretreated. I want to say that these are pretty small subgroups. Overall, if we take the total cohort, we do see an encouraging signal. These data are suggestive that T-DXd can improve outcomes in this setting. No new safety signals identified. This study, DESTINY-Lung03, continues with T-DXd-based regimens, including combination with immunotherapy and chemotherapy and immunotherapy.
Next slide, please. Okay. Now we're going to shift to our HER3-DXd. As Ken alluded to in his comments, we're presenting data here from collaborators. This is a study, a phase II study with HER3-DXd monotherapy in advanced hormone receptor-positive, HER2-negative breast cancer that's run by a collaborating institution, Gustave Roussy in Paris. Next slide, please. Here's the study design. The patients are unresectable advanced hormone receptor positive, HER2 -negative breast cancer patients. They had to have progression on CDK 4/6 and endocrine therapy, and they had to have all received prior chemotherapy. They received HER3-DXd at 5.6 mg/kg every three weeks. Also, I'll mention that they did not receive prior T-DXd. The primary endpoint here is overall response rate, and there was a very robust translational analysis here.
I'm not going to have the time to go over all the translational data because we're covering so much today, we could certainly follow up on that more of interest. I will tell you, if you look at the bottom of the slide, blood and tissue is mandatory at several time points in the trial. When in a study like this, you're able to capture blood and tissue sequentially like that, pre and post-treatment, you can do multiple investigations that were done by our French collaborator, looking at predictors of response and predictors of resistance, and these type of aspects. Next slide, please. Here are the baseline characteristics. On the bottom left, you can see that the majority of patients or the highest percentage subgroup were actually IHC 0 for HER2.
There were some patients that were IHC 1+, then a minority of patients that were IHC 2+ or IHC 3+. However, 30% of patients, it was unknown HER2 status. In the upper right, you can see that HER3 membrane expression by conventional immunohistochemistry was also looked at to be able to correlate HER2 expression with efficacy. You can see a range of HER2 positivity on the membrane. The majority of patients had at least a greater than equal 75% membrane positivity. You can see that the median prior number of lines of therapy was two. Virtually everybody had received prior CDK4/6 inhibitor and endocrine therapy, and all patients had received prior chemotherapy. This is a heavily pretreated population. Next slide, please. Here are the key efficacy data. What you see is a confirmed overall response rate of 53.5%.
Importantly, this is not a very small phase II dataset. You have nearly 100 patients with a sample size of 99. There were two complete responses. Overall, the clinical benefit rate was nearly 63%. The waterfall plot on the left gives you a clear impression of an efficacy signal in this population. With tumor shrinkage in 90 out of 99 patients, there was no significant association between HER2 expression and efficacy. In addition, though it's not shown in what I'm going to present, there was no clear association between HER3 expression and response rate either. Next slide, please. Here is the duration of response in progression-free survival. The median duration of response was nearly nine months. The median PFS was 9.4 months. This is evidence of a very significant anti-tumor effect. Next slide, please. Overall safety.
The safety profile was overall manageable and consistent with what we've seen in our HER3-DXd studies. Patients with treatment-related adverse events Grade 3 or higher was 50%. 11% of patients had discontinuation due to toxicity, there were 20% of patients with a dose reduction. There was one adverse event leading to death, this was not related to HER3-DXd. This was with a pleural effusion. Interestingly, in this population, there were seven ILD events, 7.1%, but they were all Grade 1. Overall manageable safety profile. When we look on the right, we can see the types of events we know for this HER3-DXd ADC, with fatigue and nausea being the most common, also some cytopenia, including decreased neutrophil count as well. Next slide, please.
Just a brief overview of the types of biomarker analyses that were done, but we're not going to have time to show the results of all of them, though there will be a few summary statements on the conclusion slide. Our colleagues at Gustave Roussy, looking at sequential blood samples and tumor biopsies, looked to see whether or not HER3 expression could predict response to HER3-DXd, looked to see whether or not there were genomic alterations or mutations that could predict response, and looked at the characterization of the tumor microenvironment, including the immune aspects of the tumor microenvironment, to try to determine any characteristics that may help predict the response to HER3-DXd. If we move to the next slide, please. This is our conclusion.
HER3-DXd showed a clearly clinically meaningful activity and a manageable safety profile in this population of heavily pretreated hormone receptor-positive, HER2 -negative metastatic breast cancer patients. We reviewed the efficacy results together already. I won't repeat them. The activity was observed across HER3 and HER2 membrane expression levels. Those analytes did not predict response to HER3-DXd. With the limitations of small sample size, and I'll just summarize here, because even though I didn't get into all the details of the biomarker data, the distribution of HER3-DXd in the tumor was seen to potentially play a role in determining better treatment response in addition to up-regulation of genes involved in immune response, such as interferon alpha and gamma. Overall, and importantly, the efficacy and safety profile of HER3-DXd monotherapy in this setting make it an optimal candidate to now study this drug in further larger trials.
Next slide, please. Okay. I think we're getting almost toward the end of my presentation, and I appreciate all your attention. A couple more topics. We're going to talk about I-DXd, our B7-H3-DXd ADC in extensive stage small cell lung cancer. The interim analysis from phase II study IDeate-Lung01. Next slide, please. Phase II study IDeate-Lung01, this update was presented at World Conference on Lung Cancer . This is small cell lung cancer. These are patients with extensive stage small cell lung cancer. They've received at least one line of prior platinum and less than equal to three prior lines of systemic therapy. Overall, this is a second line plus advanced small cell lung cancer population. Patients with brain mets were permitted, and of course, brain mets have a high prevalence in this particular disease state.
Patients were randomized to two doses of I-DXd, our B7-H3-DXd ADC, 8 mg/kg and 12 mg/kg . This is a very important study for our small cell lung cancer program because we were able to use this data to select a dose for our ongoing phase III study that we just started, which is IDeate-Lung02, which is a pivotal global trial of I-DXd monotherapy versus standard of care in second-line small cell lung cancer. You'll see the data that I show you explains why we chose 12 mg/kg for that pivotal trial. Importantly, after this randomized dosing component, we extend enrollment at that chosen dose of 12 mg/kg Q3 weeks, adding another 70 patients in a third-line plus population. We're going to gather additional data in a very advanced population.
Primary endpoint response rate, and of course, safety is included, in addition to these efficacy endpoints. Next slide, please. Okay. On the left, I'll point out that the discontinuation rate due to AE was on the relatively low side. Median treatment duration in that lower dose cohort, 8 mg/kg , was 3.5 months. It was 4.7 months at the higher 12 mg/kg cohort. If we look on the right side, we see the baseline characteristics. 40% of patients had brain mets at baseline, and you can see, again, as stated, this is really a second-line plus population. These are very advanced patients with a high unmet medical need. They've received a number of prior treatments, including three-quarters of them receiving immunotherapy, which is often given in the frontline setting in this disease state. Next slide, please.
Here is the efficacy data. You can see on the left the 8 mg/kg cohort with a 26.1% response rate. On the right, you can see the 12 mg/kg cohort with nearly a 55% response rate. You can see a very favorable waterfall plot. In addition to that, the disease control rate represented in the table below shows you that at 12 mg/kg , the disease control rate exceeded 90%. It was on this basis with a significantly higher, although these were not statistically compared, these cohorts, but just empirically, what we can see is that we can see much higher rates of response at 12 mg/kg . This promoted us to choose that dose for pivotal trials and for expansion in this phase II study. Next slide, please.
Here are what we call spider plots, what they show you is that you get rapid response, and you get durability of response as well. Duration of response, you can see down below, 7.9 for the 8 mg/kg cohort, 12 mg/kg cohort at 4.2. Some might say, "Why is that? You have a lesser DOR in the higher dose." This is likely explained by the limited number of responders at the 8 mg/kg cohort, the smaller subset of responders, in addition to the fact that there were some differences in baseline characteristics between the 8 mg/kg and the 12 mg/kg cohort, with a higher percentage of second-line patients in that lower dose cohort. Next slide, please.
Here's the PFS and OS, remembering that this is a phase II study, these time to event endpoints have limitations in their interpretation. What I would point out to you is that the 12 mg/kg cohort for both PFS and OS were numerically higher, for PFS and OS than the lower dose. In addition to that, at 12 mg/kg , the median survival was nearly a year. That is notable in this very high unmet need advanced small cell lung cancer population. Next slide, please. Okay. The safety summary. You can see median treatment duration was longer at 12 mg/kg . Adverse events leading to discontinuation was also higher at 12 mg/kg , including dose delay and reduction. Overall, we believe there's a manageable safety profile at both doses.
If you look below, there was one case, one related death, which was a pneumocystis pneumonia that was deemed related to study treatment in the 12 mg / kg arm. Otherwise, the rest of the deaths were not treatment-related. If we move to the next slide, please. Here are the most common adverse events at both doses, and what you'll notice is that there are some higher rates of adverse events, both GI and hematologic, at the 12 mg / kg cohort. Relative to what we already know about I-DXd, we have no new safety findings. Overall, we believe the profile is manageable at both doses, including the chosen dose of 12 mg / kg. In terms of ILD, in the lower part of the slide, there were four cases of ILD at 8 mg / kg.
There were five or nearly 12% at the 12 mg / kg cohort. In that 12 mg / kg cohort, one was Grade 1, three were Grade 2, and one Grade 3. There was one death due to ILD in the 8 mg / kg cohort. Next slide, please. In summary, I-DXd continues beyond what we saw in the phase I trial to demonstrate promising efficacy in advanced small cell lung cancer. There was improved efficacy or higher efficacy rates at the 12 mg / kg dose relative to 8 mg/kg. Overall, there were some higher adverse event rates observed at the chosen dose of 12 mg/kg, but manageable for both doses. Though I didn't show it here for the sake of time, I-DXd did show prominent intracranial activity with intracranial overall response rates of 50%-60% across both dose cohorts.
12 mg / kg, as I mentioned, has been selected for the ongoing phase III study in second line small cell lung cancer, IDeate-Lung02. Next slide, please. This is our last topic. As we mentioned, our goal is not just to progress in bringing new standards of care to patients with high unmet medical needs across tumor types, but also to continue to bring new innovation. The DXd platform has been remarkable so far in bringing in HER2, an approved drug, but also some promising efficacy signals with some of our other DXd ADCs, the majority of which have now advanced into late-stage trials. This is evidence that this presentation that we're bringing new innovation, new platform, another ADC platform. This is DS-9606, which is a CLDN 6-directed antibody drug conjugate in patients known to express CLDN 6, and this is a first-in-human study.
Next slide, please. As background, DS-9606 is an ADC composed of a humanized anti-CLDN 6 monoclonal antibody, has a cleavable linker, and a modified PBD payload. That is pyrrolobenzodiazepine, PBD payload. This is different than the topoisomerase I payload that is DXd in our ADC technology that we've presented to date. About CLDN 6, it is a component of tight junctions between cells, and it is nearly absent in adults' normal tissue, but expressed in several tumor types, and we can see them to the right. This includes ovarian cancer, endometrial cancer, gastric, germ cell tumor, including testicular cancer, which we'll talk about in a few slides, and non-small cell lung cancer. The expression of CLDN 6 can be associated with poor prognosis. This is the first report from our ongoing phase I trial of DS-9606 in patients with advanced solid tumors. Next slide, please.
Here is the study design, is a classic dose cohort escalation phase I trial. We're actually still in the dose-finding phase. We're showing you preliminary data. This data will continue to evolve. Key eligibility are patients that were known to have tumors known to express CLDN 6, and they had to have had progression of disease on prior standard of care. This is a typical phase I population with high unmet medical need. If we move to the next slide, please. We have multiple doses every three weeks here and in milligram per kilogram basis of DS-9606. What I'll point to you really is the bottom of the slide and probably the far right column there.
When we look at the majority or the highest percentage of patients had ovarian cancer, second highest germ cell tumor, particularly testicular cancer, also gastric GE junction, lung cancer, some patients pancreatic, breast, and endometrial. Also on the far right column in the totals, the median number of prior therapies was four, this is a very heavily pretreated population. You can see nearly 40% of the patients were ongoing at the time of this analysis. This is a very preliminary analysis. Next slide, please. Here is the safety overview. Importantly, there were no dose-limiting toxicities to date, and there were actually no discontinuations due to treatment-related toxicity. Overall, well-tolerated so far in the dose escalation, and the maximum tolerated dose or dose for randomized trials has not yet been determined.
I would show you that if I focus across all the doses to the far right column, you can see there were only three Grade 3 related adverse events, two related serious adverse events, and only a modest number of patients, or just a few patients, that required interruption or reduction due to treatment-related adverse event, and there were no deaths. Next slide, please. Here's the safety. Again, the safety was tolerable, with a majority of adverse events being Grade 1 to 2, and a clear minority of events, in green, being Grade 3 or higher. There were no Grade 5 events. Most common events were GI in nature, but also there was anemia. There were no Grade 4 or 5 treatment-related adverse events.
I did mention on the previous slide, there were only three Grade 3 related adverse events, and two of those were anemia, and one was rash. Overall, this is an emerging safety profile, but clearly, so far, a manageable profile. Next slide, please. Yes, actually, just to mention actually safety that we also did see some skin tox, which was lower grade as well. Overall, a very well-tolerated drug to date. Here's the efficacy. Preliminary efficacy demonstrates there have been four responses according to RECIST. You can see two of those are germ cell tumor, and one of those is gastric, and one of those is non-small cell lung cancer. Response rate was highest at that 0.15 mg/kg dose, which is one of the higher doses tested. If we would move to the next slide, please.
I want to spend a moment here just on the germ cell tumor. Again, these are patients that are heavily pretreated and refractory. Two out of seven patients that were eligible for tumor assessment with germ cell tumor had a PR or RECIST PR. You can see that there on the left. On the right side, five out of seven patients who were evaluable for efficacy with germ cell tumor had a greater than or equal to 90% reduction in tumor markers. This is an early signal, but it is a very notable signal for anti-tumor activity in a very heavily pretreated, high unmet medical needs setting. Next slide, please. Okay, this is my last slide. Overall, what we can see so far is the safety profile is manageable for this asset.
There is some promising early anti-tumor effect, but we're still in dose selection, so it is early, and particularly in germ cell tumor, where we see that signal so far. Of course, we'll continue with the enrollment and dose escalation to determine the maximum tolerated dose and the dose for randomized trials. That is my last slide. I would like to extend a sincere thank you to listening to the presentation because it was lengthy. I hope you can see the progress across our pipeline.
I hope you can see the breadth and depth of our pipeline, not only in advancing our DXds across tumor types and lines of therapy, but also continuing on the quest to bring new innovation by bringing this second new ADC platform, all with the goal and clear mission to basically provide new treatment options to people who are sick and either don't have options or are running out of options. Thank you for your time.
Thank you, Mark. Now let's go to Q&A. Before we go to Q&A, there were some issues with simultaneous interpretation audio, we will upload a recorded version of this event on our corporate website soon. Please visit our website then. We apologize for the inconvenience. Now let's go to Q&A. When you have questions, please click the Raise Hand icon at the bottom of the screen. I will call your name in order, when your name is called, please unmute yourself and ask your question. You could ask your question in English or in Japanese. If you have multiple questions, please ask one at a time for optimal translation. When you have no further questions, please lower hand and mute yourself again. Please raise hand if you have questions. The first question is from Yamaguchi-san from Citigroup. Please go ahead.
Hello, can you hear me okay?
Yes.
Thank you. I have three questions. I try to be very brief. Thank you for the explanations. It is very helpful. The first question is that, after this TR01 OS data, the financial markets seem to be a bit worried about the probability of success or approval, approvability at the PDUFA date in December. I'd like to make a basic question once again, but it is a dual endpoint data, and the PFS is already statistically significant. OS was a bit weakish, but it's still trending, the number is trending. With these conditions, after this OS data updated, is it fair to say that the probability of approvability is go down or not really? My understanding is that the PFS is good enough to get approval, but is there something new, things happening to make you guys less confident about the probability of the data?
That's the first question. Thank you.
Thank you very much for that question. Currently, we are still in discussions with the FDA on our submission. We will be able to update you once we have some definitive decisions coming from that submission. It's still an ongoing open submission right now.
Mark, do you have any comments?
Yes. No additional comment except to confirm what you stated, which is PFS statistically significant. Overall survival, dual primary endpoint, not significant. What we saw in that subgroup of non-squamous, non-small cell lung cancer patients, this is where we saw the benefit. We're still ongoing discussions with regulatory agencies, so no update to provide yet, but they have been provided these results. We will issue an update as soon as we can.
Right. Still that the dual endpoint, meaning one of which is hit the statistical significance, that the trial itself is a success, right?
Right. The PFS primary endpoint was statistically significant—
Right.
—that means that there's a positive trial. Correct.
Okay. Thank you. The second quick question is that this non-squamous, squamous things, including QCS, it looks like you have some idea of why there's a difference between non-squamous and the squamous, especially the TROP2 membrane or penetration into the cells. Do you have any idea why there's a difference between the non-squamous, squamous? Also how FDA view this sort of a difference, and is there any scientific evidence to see the difference of a non-squamous, squamous at this point?
Thanks for the question. What I showed in the presentation from World Conference on Lung Cancer is that the prevalence of biomarker positivity was higher in the squamous patients-- excuse me, in the non-squamous patients compared—
Right.
—to the squamous patients. Yes. We think this could be one contributor and potential explanation for why we see differential efficacy according to histology. We don't have a firm and clear explanation for this differential efficacy according to histology, but it may not only be related to this different biomarker prevalence. There are also characteristics of squamous versus non-squamous histology that also, at a biological level could potentially explain differences in efficacy to Dato-DXd. At this point, it's still an ongoing investigation, but one good clue is provided with that QCS analysis.
In terms of, I think you commented on the regulatory. Yeah. At this point, we're not able to provide an update on what FDA thinks of the differential activity according to histology. Of course, they can see in the data that we provide that there's very clear evidence of differential activity according to histology. We've seen that not only in TROPION-Lung01 trial, but also in external phase II data sets. We believe this is a significant finding.
Okay. Thank you. The follow-up question is that regarding the new generation DS-9606, it looks pretty premium at this point because you are still in the process of raising doses. I understand that. It looks very safe in overall things. Also there are some commentary, and you pointed out as well, there's a new sort of potential side effect, especially on skin tox, has been emerging. Already only one or two. Do you think this is going to be dose limiting factor in the future, or it really doesn't matter? How long do you see the more response, especially like a cancer, like ovarian cancer, which should be the main target of this indication, the drug. Thank you.
Yeah. In terms of the skin tox, we're seeing some skin tox, and this has been seen with some PBD type payload.
Yeah.
Of course, ours is a modified payload. It's hard to predict whether or not the skin tox will become dose limiting. At this point, we're really still increasing those doses because skin tox is something that's been recognized with this type of payload. It's possible for it to eventually be dose limiting. At this point, it's hard to predict. Then, you commented on, I think, efficacy in other tumors, like ovarian—
Right.
—and gastric. Yeah. What we have to do here is complete the dose escalation and then choose the dose we want to take into further study. What we will do is we will then take expansion cohorts. We'll look at specific tumors of interest based on the signals that we observe in the dose escalation. Of course, as I mentioned, we have seen a response in gastric cancer, as an example. So we will be able to, in those expansion cohorts, look at those key tumors, including ovarian cancer, as you cited.
Yeah. Thank you. Thank you very much.
The next question is from Wakao-san from JP Morgan. Please go ahead.
Hi, can you hear me?
[Non-English content]
It's Seiji from JP Morgan. A question about TL01 upper bounds. Do you think that the upper limit of the 95% confidence interval for the OS hazard ratio exceeding 1.0 could be an obstacle for approval? I'd like to know your thought on this point.
Yeah. I think when we look at the efficacy in the subgroups, which were not statistically tested, so the descriptive analyses, we look at multiple characteristics to determine what we think is clinically relevant. The 95% confidence interval, and I think you're referring to the upper bound going to 1.05, I think in the OS hazard ratio, and this is only one aspect that we look at. Importantly, we look at the medians as well. We look at the hazard ratio. I think here in the case of the medians, we saw a median of 14.6 months overall survival and a delta or difference in survival, a median of 2.3 months. It's really the totality of the data by which the subgroup efficacy results would be considered. We don't just look at that 95% confidence interval alone, but the totality.
Next, about biomarker strategy. I understand that Daiichi Sankyo does not have biomarker analysis technologies like QCS, and I'm concerned that lack of such technologies could pose a challenge for your ADC pipeline, particularly for those who have partnered with AstraZeneca. Could you please clarify whether it is necessary to have in-house technology like QCS?
Maybe I'll start and then I'll hand it to Ken. Yeah. I want to say that biomarker technology is very important to us at Daiichi Sankyo. It's true that AstraZeneca developed this QCS technology, we internally at Daiichi Sankyo are working very deeply and comprehensively on biomarker programs, on establishing predictive biomarkers for all of our assets. We're deploying multiple different technologies. I don't think we have shared the details of all that we're working on, with time, we will. I want you to be assured that we are very capable and very much underway in those efforts, including the remainder of assets beyond Dato-DXd, for which the TROP2 QCS has become prominent. Ken, would you like to add anything additional?
Yes. Thanks. Yes, I just want to add to Mark's comment that QCS digital pathology just represents one of many different types of technologies and modalities to identify predictive biomarkers. Yes, we have a number of technologies that we're looking at to identify predictive biomarkers. In addition AstraZeneca has been working on the digital pathology in QCS for many years now.
More recently many other academic groups as well as industry groups have developed their own methods for digital pathology, so we can certainly tap into them to acquire the necessary digital pathology expertise.
Okay. Lastly, about DS-9606. Could you comment whether the result of this phase I result suggests any difference in aspects such as the linker or payload between DS-9606 and a DXd ADC? What clinical difference are you expecting for DS-9606 or second generation safety or efficacy or something?
Yeah. At this point, I'm not able to comment in too much detail about what are the mechanistic or structural characteristics to this new ADC platform versus DXd. At this point, what we're able to offer is the target, which is CLDN 6. The fact that it is a PBD payload, but it's modified, and we're not really discussing at this point how we modified it. Also, we do say it's a cleavable linker. That's pretty much where we stand now in terms of what we are able to offer. We'll have more information with time. We think this ADC platform, which is a new platform, could be a potential new addition to the treatment landscape.
Okay. Thank you. That's it from my side. Thank you.
Next question is from Hashiguchi-san from Daiwa. Please go ahead.
This is Hashiguchi from Daiwa. Thank you for this opportunity. Concerning NMR -QCS, I have a question. I think we have been researching and studying various biomarkers. Given this data, in order for you to select the optimal patient group, how much this QCS as NMR is promising to select optimal patients? Have you already decided to use this biomarker for the selection of the patients, or you'll be continuously exploring other biomarkers. If you try to roll out this particular biomarker, many of patients with NSCLC need to have an access to this biomarker measurement or testing. What would be the hurdles if you try to do so? How would you try to overcome this difficulty? How you'll be able to propagate this measurement testing in the field? How long will it take?
I can answer that question. There is a considerable amount of work being done by our collaborator, AstraZeneca, on this particular QCS assay. They have already pointed out in their own investor relations conference that we have obtained additional information beyond the TL01 clinical trial in other indications. It's quite possible that this QCS technology is applicable to other diseases besides lung cancer. This is very exciting. Not only this, they have thought about how this might be used in the patient care setting. They have partnered with a diagnostic company, Roche, to be able to provide central pathology services where physicians can submit three slides from a patient sample to get the QCS results.
Now you kindly explained what AstraZeneca has been doing. What about Daiichi Sankyo? For Dato-DXd, has Daiichi Sankyo decided already to use NMR by QCS for Dato-DXd treatment? If so, how long does it take?
AstraZeneca is a collaboration partner. We together are very interested in the QCS for the Dato program. We are working very closely together with AstraZeneca to commercialize the assay as well as at the time of Dato approval. Yes, we're very interested, and here we have a very important collaboration on that.
Thank you very much. That's all.
Next question is from Mamegano-san from BofA. Please go ahead. While we wait for Mamegano-san, before that, we would like to go to Muraoka-san from Morgan Stanley. Please go ahead.
Thank you very much. Muraoka is my name, from Morgan Stanley. Can you hear? Thank you very much. The first question, TL01, regarding this AGA in relation to that. With OS, AGA was related, I have understood that. In terms of PFS, also AGA. Because of AGA, the difference in response, similarly, have you been seeing that? Could you let me know about that, please?
Yes. Thank you for the question. Previously we had presented progression-free survival, and we'd seen this difference in AGA as well. The differential evidence of efficacy or anti-tumor effect within the AGA subgroup compared to the non-AGA has been consistent across some of these endpoints, yes. I have to point out, though, that, as we always do, that the AGA population in TROPION-Lung01 was actually a minority of the patients.
Thank you very much. Next, I-DXd, regarding this. A very good data, I think, you have obtained. In the near future, with this data, are you thinking that application is possible, or will you be waiting for another while in order for the data to be accumulated? Which is the case?
Thank you for that question. We are expanding that phase II study. We're going to add 70 patients in third-line plus small cell lung cancer. In addition to that, we, of course, have announced the start in August of our randomized global pivotal trial in the second-line of small cell lung cancer. We're not commenting on any specific regulatory strategy yet. What I can tell you is that both our partner, Merck, and Daiichi Sankyo realize this is a very high unmet medical need. We will examine different options and we will examine and look for options to bring this drug to patients as soon as possible who are in need and who have a high unmet medical need. As soon as we can update any particular regulatory strategy, we'll do that.
Let me repeat myself. That means that you will accumulate a little more data and then would make the application or the filing towards the authority. Is that the case?
Yeah. We're actually not commenting on the regulatory strategy, in terms of how much more data we would need to accumulate, before we talk to the health authorities about a filing pathway. My comment was just that we are accumulating more data both in phase II and in phase III, and we will seek a registration path that is efficient, and we'll try to get this drug to patients as soon as possible. We're not prepared at this point to discuss how much more data from which study, we will use to seek registration.
Thank you very much. My last question is second generation ADC regarding this very promising, preliminary but promising data I think you have. This second generation ADC, using that newly TROP2 ADC to be redeveloped. That kind of option are you thinking about?
Oh, yes. We do plan to use this platform, this new ADC platform to bring additional agents. DS-9606 will not be the only drug that uses this platform. We're not describing at this point what additional targets that we've chosen or will choose in the additional drugs that we'll bring forward using this platform. We'll do that as soon as we can, as these assets come forward, but I can't comment on specifically whether or not TROP2 will be a chosen target for this particular platform.
Understood. Thank you very much. Thank you. That's all.
Thank you.
Next question is from Barker-san from Jefferies. Please go ahead.
Yes. Steve Barker from Jefferies. Thanks for taking my question. My first question is about the TL01 overall survival data. I was hoping I could get your comment on why do you think the hazard ratio deteriorated from 0.77 at interim to 0.84 in the final analysis?
Yes, I think when we have follow-up of overall survival, there are a number of factors the data can evolve. The hazard ratio is similar, so don't have anything specific to offer as to why the hazard ratio went up a bit. As you say, numerically it did increase. What I can say is what I had stated, which is that when we look at that median delta of 2.3 months and the median survival of 14.6 months in the data arm, we find that very encouraging. Yeah, no specific comment on exactly why it goes 0.77 to 0.84. Just changed with additional follow-up which can happen over time in the conduct of a trial and when we do multiple analyses over time for a time to event endpoint.
Okay. It doesn't make the data less approvable in your view?
Yeah. At this point, we're not commenting on the discussions that we have ongoing with the regulatory agency. No update that I can provide on specifics about the data and how it impacts the ability for approval.
Thanks. Regarding the delta 2.4 months between the docetaxel arm and the treatment arm. The docetaxel arm 12.3 months, it seems high compared to what you saw in the squamous, which was only 9.4 months and high compared to historical controls. Do you have any comment on why that arm was so high?
Yeah. You get a range when you have similar control arm across pivotal trials. You do get a range of outcomes, and this can depend on the specific patient population that gets enrolled into one trial or another. Sometimes it's not readily explainable. What we see with docetaxel in historical trials and sometimes the older trials show different results than the newer trials. Over time, there has been, I would say, a trend toward docetaxel overall survival outcomes improving over time. Typically, we've seen 12 months or less. One of the potential reasons by which overall survival outcomes for docetaxel have improved relative to historical ones is that in the post-IO setting Patients who receive previous IO and then get docetaxel, this may have had an impact, but this is a hypothesis when this topic has been discussed.
Okay, thanks. My last question is regarding DS-9606. Can you share with us any information about the linker and how that differs from the deruxtecan technology?
Yes, I'd like to. Today we can't do that just yet. All we're commenting about the linker at this point is to say it's a cleavable linker, which has implications for ADC structure, but not commenting anything more specific about the linker. As we have done for the DXd technology, we have commented on more structural aspects. At this point, not quite ready to share anything more detailed than the fact that it's a cleavable linker.
Okay, just a follow-up question, again, regarding DS-9606. The PBD payload, highly toxic, effective in very low concentrations. Are there certain targets that are going to be better suited to this platform as opposed to the DXd platform?
Yeah. I think when we look at the spectrum of target indications for this first asset, DS-9606, using this new platform, we're really guided, I'd say, in part by the payload, but really mostly by the target expression of CLDN 6. With ADCs in general and with our ADCs including DXd, and there's not really much of a difference here in terms of when we think about the indication spectrum, we consider multiple factors, but one of the key factors is target expression. When we talk about DS-9606, what we're looking at, what we mentioned, germ cell tumor, lung, gastric, ovarian, and so we're really guided by target expression. There are some cases when we look at the payload and think about whether or not that payload has particular sensitivity in certain tumors versus others. That is a consideration we make.
At this point, with this asset at its preliminary stage, I think what we're focused on is more on the target expression, when we're thinking about what might be potential tumor types.
Understood. Thanks very much.
Next question is from Haruta-san, from UBS. Please go ahead.
I'm Haruta of UBS Securities. Can you hear me?
Yes.
This is my first question. TL01's result, first-line impact. How should we conceive? First-line, combination with immune checkpoint inhibitor, first-line, that would be the central world, without jeopardizing this immune checkpoint, the antitumor effect has to be exerted, therefore, first-line clinically meaningful result. I think the hurdle is high, I don't think it's directly related as a trial. The second-line details are available. How should we think about the impact for the first-line?
Yeah. I think in TROPION-Lung01, what we've seen is clinically meaningful efficacy in the non-squamous population. As a monotherapy, we saw Dato-DXd double in response, increasing duration of response. This is a drug which demonstrates clear antitumor activity in non-small cell lung cancer. When we think about the front-line trials, it is a different situation because we're combining with other drugs. Based on TROPION-Lung01, we think Dato-DXd has potential in non-small cell lung cancer, we're very committed to this. In fact, we look at the program. We have now seven pivotal trials in non-small cell lung cancer and looking to answer multiple questions. Not only adding on different combination partners like, let's say, immunotherapy and TL08, or immunotherapy and chemotherapy in TROPION-Lung07 and the AVANZAR trial being run by AstraZeneca.
In addition to that, even looking at biomarker segment like AGA population, we've already talked about in the AGA population, we saw a nice signal in the TROPION-Lung01. That's the basis by which we have two pivotal trials in EGFR mutation-positive non-small cell lung cancer, TL14 and TL15. What I would say is that it is, as you note, challenging just to extrapolate from monotherapy to combination settings and different lines. I think what we learned from TROPION-Lung01 is that Dato-DXd has potential, we think it could end up being a very important drug in non-small cell lung cancer. We have a very comprehensive program across different lines of therapy, different patient segments, we're even now seeing, as I mentioned earlier with the NeoCOAST, we're seeing preliminary antitumor effect even in the resectable setting.
Overall, we remain confident in this drug and have a very broad, comprehensive non-small cell lung cancer program to provide evidence of that confidence.
Thank you very much. My second question is, I-DXd, that area, second-line accelerated approval. Regarding that, you said you can't say much, let us suppose that you're going to target at the accelerated approval. ES-SCLC progression is fast and limited treatment options at an early stage. Also, from the market, whether it could be accepted by, or like in ES-SCLC compared against chemotherapy, how much benefit there is, PFS and OS, how much incremental benefit could be shown, whether that would be more important. In the case of ES-SCLC, the treatment option is limited. Regarding that point, what is your idea?
Yes. I mean, we do have the ongoing randomized pivotal trial, it's hard to say exactly what kind of efficacy increment we'll see, how much better in PFS and OS we would see compared to the investigator's choice of therapy. I think we're encouraged, though, because we see in IDeate-Lung01, I presented nearly a 55% response rate. This is a response rate that's holding up over time. We saw in phase I, now phase II. When you see a response rate like this and I showed preliminary PFS and OS from phase II, when you see those kind of consistency of endpoints and this type of efficacy signal, that gives us confidence that when we compare in a randomized trial to the standard of care, we wouldn't do the trial unless we thought we potentially could have a positive outcome.
It's hard for me to hypothesize exactly what kind of efficacy increment that we'll see at this point. What I can say is that we're very interested in taking I-DXd into small cell lung cancer, I'll add, it's not only in monotherapy when we compare to the second-line standard of care. Also, we've launched a phase I-B trial to combine I-DXd with chemotherapy and cytotoxic therapy, that's in the frontline setting. In thinking about the ability to eventually go into the frontline setting. I'll also add that in small cell lung cancer, we recently extended our co-development and co-commercialization agreement with U.S. Merck for MK-6070, which is a tri-specific T-cell engaging DLL3 agent. We've announced that we're going to combine that with I-DXd in small cell lung cancer.
We're very optimistic here, but of course, can't think of exactly what kind of outcomes we'll see, and we'll just have to let the data mature.
Understood. Thank you very much.
Next question is from Mamegano-san from BofA. Please go ahead.
Hello, can you hear me?
Yes, we can.
Yes.
Sorry, giving me the multiple opportunities to ask the question. I have two questions. First is regarding that Dato-DXd TL01. This time, the biomarker clearly showed the differentiation of the benefits in different subgroup. The European lung trial and AVANZAR trial, I think they use the same biomarker. Concerning second line filing, is my understanding correct that you cannot use this? I think currently you are preparing for the filing, but this biomarker is not available for that filing. Is that understanding correct?
Yes. To your first comment or question, yeah, it's the same assay, the TROP2 QCS assay, that has been shown in this retrospective exploratory analysis of the World Conference on Lung Cancer of TROPION-Lung01. The same assay is to be used in AVANZAR. The same assay is to be used in the additional TROPION-Lung10 study that I referenced. You ask, could the QCS analysis we show in TROPION-Lung01, could that be used for regulatory submission for TROPION-Lung01? As we stated, we're not commenting on specific proceedings at this point of what discussions have been with the regulatory agency about the QCS data. The agency is aware of the QCS data, of course they have it. It's important to remember that the analysis we presented at World Conference on Lung Cancer was a retrospective exploratory analysis.
As our colleagues from AstraZeneca offered earlier this week, that assay will be validated in the other studies, including AVANZAR.
Thank you very much. My second question is about the HER3-DXd and the press release. I think you announced, showed the positive trial results. MARIPOSA-2 is going to be your competitor. They also shared the positive data of OS at ESMO. Also amivantamab will be also your competitor, I guess. Now I think that you have a positive data PFS, so you'll be probably going for filing. With amivantamab, do you consider any combination with your asset? What is your strategy going forward?
Thanks. We did have a press release today announcing top line and meeting the PFS endpoint. We're not able to comment on the details about that data, or details of our regulatory strategy, but we'll do that in the future for sure. We do recognize, as you cited, actually, it's great for patients, right? There have been positive trials reading out in EGFR mutation positive non-small cell lung cancer recently, like MARIPOSA, FLAURA2, MARIPOSA-2. We do recognize this is a competitive space and we'd be glad if we can add to the treatment armamentarium in this space, because we think there are still unmet medical needs. In terms of whether or not we would combine HER3-DXd with amivantamab specifically, nothing to share on that.
I would say generally though, we are looking at multiple potential combination partners for HER3-DXd and our other DXd ADCs in multiple settings including non-small cell lung cancer. No specific combination yet to share, including for amivantamab.
Understood. Thank you very much.
Next question is from Sogi-san from Sanford Bernstein. Please go ahead.
Thank you very much for a very comprehensive presentation. I have a question regarding QCS. Currently QCS really focusing on identifying patients' TROP2 status, whether it's on the surface of the cell or inside the cell. It also appears to be a kind of a static matter. It doesn't really reflect the dynamic, the movement of TROP2 between the cell surface to the inside the cell. I think that's, please correct me if I'm wrong. The next question is, you are saying that you are exploring different approaches of the biomarkers and QCS is just one of them. My question is, are these other biomarker options that you're exploring, are they also really focusing on the location of the target of ADC, or is it also looking something else?
Yeah. To your first question, you're right. It's not a dynamic assessment of the expression. It's using immunohistochemistry at first, then deploying the artificial intelligence assisted computational pathology. It's sweeping all the tumor cells in that tissue slide to identify the cytoplasmic expression and the membranous expression. That's looking at that point in time, so it's not an ongoing or a dynamic assessment, so to speak. I would point to the results, though. First of all, the results themselves suggest nice potential, albeit an exploratory analysis, for this particular assay with this methodology to potentially predict efficacy to Dato-DXd. I would also say that there are other assays, and I would say the majority of assays would use this type of non-dynamic type of assessment when we characterize analytes or biomarkers that can predict response to efficacy of drugs in general.
Sorry, your second question—
Yes. My second question.
—I forgot. Yeah.
Yeah. My second question is around the other biomarker options that you mentioned.
Oh, yes.
Is it also around just the location of the targets?
Yeah, good question. We have a broad campaign for biomarkers. We are interested in digital pathology, which allows you to look within the tissue slide at different analytes, certainly. Beyond that as well. Not just this type of methodology. Digital pathology is an important biomarker campaign for sure. We recognize that we have to look further than this. We look at additional technologies as well.
Could you actually give us the examples of what kind of things you are looking at?
Yeah, sure. There could be, as an example, we can look at genomic alterations as an example, which is commonly done, right? We could also look at plasma-based biomarkers that circulate in the blood, just as examples. Those are not uncommon. I don't have anything too much more specific to offer. We're looking broadly at methodologies and types of biomarkers, whether it be genetic, plasma, digital assisted, different types. Yeah.
I see. The second question is around second generation ADC. You mentioned that the second generation, the ADC, has a linker, it's a cleavable linker. I understand that Daiichi Sankyo usually takes a really kind of specific approach in terms of developing a new ADC, new modality to solve some of the challenges that exist in the current modality. I'd just like to understand, you are saying that this new version is a cleavable one. What was the kind of challenges you identified behind it to just develop such a new modality?
Yeah, it's a good question, and we really can't speak too much about this quite yet. Really what we're saying is it's a new payload, a modified PBD payload, and we talk about the target, CLDN 6, we talk about a cleavable linker. We're not describing at this point how this is different than the DXd technology and what specifically was thought about that needed to be overcome. I would say in general, we'll have additional new ADC platforms in the future, right? This is the next one we're presenting, the second one, but there'll be more. I think when you step back and think more broadly, we look at new targets, as an example, which would allow us to address new unmet needs and new tumors. We could look at new linker technologies, we could look at different payloads to optimize benefit risk in multiple ways.
We look quite broadly when we consider next generation ADC platforms and what they could bring in terms of advance. In terms of this specific platform and how it compares to the DXd ADC technology, I'm not able to offer details quite yet.
Thank you very much.
The last question is from Tony Renson, Macquarie Capital. Please go ahead. Looks like we lost Tony Renson, so we will now go to closing. Before we close, again, we apologize for the issues with simultaneous interpretation audio, and we're going to upload a recorded version of this event on our corporate website very soon. Please visit our website. Now we will close and conclude Daiichi Sankyo's WCLC and ESMO 2024 highlights. Thank you so much for joining today.
Thank you.