ALX Oncology Holdings Inc. (ALXO) Earnings Call Transcript & Summary

Good morning, and welcome to the ALX Oncology Conference Call. Today's call is being recorded.

Hello, and thank you all for dialing in to today's conference call. I'm Pete Garcia, Chief Financial Officer of ALX Oncology. Today's call will focus on the clinical data of ALX148, a CD47 myeloid checkpoint inhibitor and specifically our Phase Ib study in gastric cancer combining ALX148 with trastuzumab, ramucirumab and paclitaxel that was presented at the recently completed European Society of Medical Oncology World Congress on Gastrointestinal Cancer, also known as ESMO GI. A slide deck for this webcast is available on our website in the Investors Events section. Dr. Jaume Pons, our President and CEO, will provide a brief overview of ALX148; and Dr. Sophia Randolph, our Chief Medical Officer, will then discuss the updated Phase Ib gastric cancer combination study. At the end of our prepared remarks, we will open the line up for questions. Before we begin, as reflected in Slide #2, we would like to remind you that today's presentation will include forward-looking statements based upon our current expectations and beliefs. Such statements represent our judgment as of today and involve substantial risks and uncertainties that may cause actual results to differ materially from the results discussed in the forward-looking statements. Please refer to our filings with the SEC on our website. With that, I'd like to turn the call over to Dr. Jaume Pons. Jaume?

Let's move to Slide #3. ALX Oncology is an immuno-oncology company focused on the CD47 pathway. The CD47 pathway is a checkpoint that bridges the innate and adaptive immune system. Our lead candidate, ALX148, is a CD47 blocker that is designed to be used in combination. In the clinic, the molecule has shown a very good tolerability profile that has enabled high dosing, and this higher dosing has enabled us in the clinic to achieve better efficacy. We already have clinical proof of principle in both hematological and solid tumors. Our initial focus is on MDS, AML and solid tumors. Please, Slide #4. CD47 is highly expressed on tumor cells, but as normal cells. In the left panel, you have CD47 mRNA expression in tumor cells in red and in adjacent normal tissue in black. So if we want to use CD47 as a tumor-associated antigen and directly kill CD47 positive cells, we would also kill normal cells, and the therapeutic window will be narrow. By [ mixing ] CD47, an interesting target to me is its function as a myeloid checkpoint shown in the right panel. CD47 interacts with receptor SIRPalpha that is expressed on macrophages and other myeloid cells. This interaction is a significant interaction that we call a do not eat me signal. Many anticancer agents like anticancer antibodies or chemotherapy provide eat me signals that can target cancer cells for destruction. For example, anticancer antibodies [ bite ] antigens on cancer cells and through the interaction of the [ CF ] antibody with Fc gamma receptor, so macrophages provide an eat me signal, and macrophages would eat cancer cells. In that situation, cancer cells have regulated CD47 to provide a don't eat me signal, preventing the full activity of many anticancer agents. What's important here is that in absence of a positive signal, CD47 blockade is not enough to activate macrophages. In Slide #5, we see that if we try to use antibodies with an active Fc that directly kills CD47 positive cancer cells will also, at the same time, will be targeting normal cells for destruction, for example, droplet cells and platelets. In the clinic, CD47 antibodies with active Fc have been associated with anemia, thrombocytopenia and neutropenia. This toxicity limits dosing, and in the combination setting, they cannot be dosed high enough to totally block CD47, and they cannot maximize the activity of the combination drug that specifically target cancer cells. We saw this problem. We decided to create a molecule that potently blocks CD47, but is not able to provide the positive signal in Slide #6. Despite sparing normal cells, it does not destroy them, but in the combination setting, we can safely dose very high totally block CD47 SIRPalpha interaction and maximize activity of the combination drug that provides the required positive signal in a cancer-specific manner, for example, an anticancer antibody. In Slide #7, you can see that our molecule, ALX148, comprises what we call a high-affinity SIRPalpha. This is the extracellular domain of SIRPalpha, the receptor for CD47 that has been affinity mature to bind CD47 with picomolar affinity. Natural tumor SIRPalphas bind CD47 with an affinity of 500 nanomolar to 1 micromolar, so we have signs of all higher affinity for CD47 than the natural SIRPalpha. We have use that high-affinity SIRPalpha to complete and activate the Fc of an antibody that is not able to interact with Fc gamma receptors and is therefore not able to activate macrophages. However, this Fc does provide the same half life of an antibody, so we have 30 days half life at a steady state at 10 mgs per kg and 15 mgs per kg. The molecular grade of ALX148 is half of an antibody, so we have twice the number of molecules per gram [indiscernible]. So 10 mgs per kg of ALX148 is the equivalent of 20 mgs per kg of in terms of binding sites. Also the smaller size may provide better tumor penetration, enabling activity in solid tumors. The manufacturing of ALX148 is completely antibody standard with very high yields of more than 4 grams per liter at 2,000 liters bioreactor scale, and we already have 3 years of the stability of 4 degrees as a liquid formulation. It is truly a very well-behaved molecule. Now I want to introduce Dr. Sophia Randolph, ALX Oncology Chief Medical Officer, that will describe the clinical data that we disclosed last Saturday at ESMO GI.

Thanks, Jaume. Good morning. Today, I am going to present updated Phase I data summarizing ALX148 in patients with second line or greater advanced HER2-positive gastric and gastroesophageal cancer. As represented on Slide 8, ALX148 is designed to safely maximize trastuzumab's antibody-dependent cellular phagocytosis of cancer cells by targeting CD47 as a myeloid checkpoint inhibitor. On Slide 9, I'm pleased to present the initial Phase I data from ASPEN-01 summarizing ALX148's activity in patients with advanced HER2-positive gastric cancer who have progressed on prior anti-HER2 targeted therapies. This data was presented recently at the virtual ESMO GI 2021 meeting on July 3, 2021. On Slide 10, in the ASPEN-01 gastric portion of the Phase I study, ALX148 was administered in combination with trastuzumab with and without standard ramucirumab plus paclitaxel in patients with second line or greater disease. The primary endpoint of the study is the determination of the maximum tolerated dose of ALX148 in combination with objective response rate, PK and PD as key secondary endpoints. These data cutoff date was May 3, 2021. The baseline characteristics of patients enrolled to the study are seen on Slide 11. Patients were enrolled from the United States and South Korea, with predominantly Asian subjects enrolled into all cohorts. The majority of patients had an ECOG score of 1. 38 patients were enrolled across the 2 cohorts of patients with gastric cancer. All but 1 patient in each group had tumors that progressed upon prior anti-HER2 therapy, and slightly less than half in the ALX148 plus trastuzumab cohort had tumors that had progressed upon 2 or more prior anti-HER2 agents as well as prior checkpoint inhibitor therapies. In the ALX plus tras-ram-pac cohort, 3 patients received ALX148 at 10 mgs per kg Q week, plus the standard tras-ram-pac, and 15 patients enrolled into the second dose escalation cohort receiving ALX148 at 15 mgs per kg plus tras-ram-pac. In the ALX plus trastuzumab cohort, all patients received ALX at 10 mgs per kg Q week in combination with standard trastuzumab. All groups had significant burden of disease as measured by a percentage of visceral metastasis. Slide 12 shows that ALX148 was well tolerated in combination with trastuzumab and ramucirumab plus paclitaxel. The safety of ALX plus tras has been reported elsewhere. The most common treatment-related adverse events were low-grade diarrhea, rash, urticaria and pruritus. On Slide 13, we can see that there were minimal Grade 3 and above ALX148 associated adverse events. There were no dose-limiting toxicities. There were no on-study deaths or ALX148 associated serious adverse events. A maximum tolerated dose was not reached, and the maximum administered dose in combination was 15 mgs per kg Q week. This initial data suggests that ALX148 may be safely administered with trastuzumab, ramucirumab and paclitaxel and is consistent with ALX148's molecular design in the absence of exposure cytopenia relationship that has also previously been reported. On Slide 14, we see the PK profile of ALX148 following combination therapies with trastuzumab as comparable with and without chemotherapy. On Slide 15, we see confirmed objective responses, including 1 confirmed CR in patients receiving ALX plus trastuzumab and ramucirumab/paclitaxel. In addition, PRs have been reported in patients receiving ALX plus trastuzumab. Slide 16 shows the waterfall plot in patients receiving ALX plus tras-ram-pac. All patients with partial responses had confirmed responses. And of course, that patient with the complete response was confirmed after data cutoff. 9 of the responding patients, including the patient with the confirmed response, remain on study. Objective response was seen at both dose levels as indicated by the asterisk. In the spider plot, anticancer activity was observed with the initial disease assessment at 8 weeks, with continued shrinkage of the measurable tumor lesions, and that appears to deepen over time. Here on Slide 17, we summarize initial anticancer activity of ALX148 in combination with trastuzumab and ramucirumab/paclitaxel in the predominantly South Korean population enrolled into the ASPEN gastric cancer population. We've compared that with other second line or greater study populations. The ASPEN objective response rate was 72.2%, which compares favorably to the 52% objective response rate seen in the South Korean study, with TRP or tras-ram-pac from Rha et al at ASCO this year. It also compares well with the 41% objective response rate seen from the -- in HER2 DESTINY-01 study in Japanese and Korean patients, and with the 34% objective response rate seen with the current global regulatory comparator, ramucirumab/paclitaxel from the RAINBOW study, specifically the Asia Region 3 subgroup. Although the sample size of patients treated with ALX148 in combination at this time is small, the initial data is very promising. Consistent with immuno-oncology therapy experience, we may also be seeing a potential clinical benefit evident in the later survival-based end points where the ALX tras-ram-pac estimated overall survival rate at 12 months is 75.8%, with a 10.5-month median follow-up. Updated clinical activity data with the doublet ALX148 plus trastuzumab, this is a chemo-free regimen, is summarized after a median follow-up of 27 months. Objective response rate was 21.1%, and median overall survival was 8.1 months with an estimated 12-month overall survival of 38%. As ALX148 is not designed to have monotherapy activity, and based on the randomized TACT, or T-ACT study, trastuzumab added minimal activity to paclitaxel in a similar population. The enhanced ADCP activity seen here with ALX tras doublet supports the rationale for adding it to the current global standard care ramucirumab/paclitaxel. On Slide 18, you can see in the swimmer plot that 9 of the 13 responders remain ongoing on this study. On Slide 19, near complete CD47 target occupancy is maintained throughout the ALX148 dosing interval in combination with chemotherapy-containing regimens. And on Slide 20, in an exploratory evaluation, we see a numeric increase in the percent of baseline tumor infiltrating immune cells in the on-study clinical responders compared with that of nonresponders. In summary, on Slide 21, ALX148 is a high-affinity CD47 myeloid checkpoint inhibitor with an inactive Fc gamma domain and favorable safety profile, which demonstrates exciting initial objective response in combination with trastuzumab with and without chemotherapy in patients with second line or greater gastric cancer. Preliminary data suggests that ALX148 can be safely combined with no maximum tolerated dose reached and a maximum administered dose of 15 mgs per kg Q week in combination with standard trastuzumab plus ramucirumab and paclitaxel. ALX148 demonstrates a promising 72.2% objective response rate in patients with second line or greater disease in combination with tras-ram-pac, and this compares well with the historical data and supports our plans to evaluate this combination in an upcoming randomized Phase II/III study in patients that have progressed upon prior HER2-targeted therapy, where the Phase II portion will confirm ALX's activity on top of tras, ramucirumab and paclitaxel. To our knowledge, this is the only CD47 targeted agent in the clinic to date that has shown this type of activity in the solid tumor setting. So on Slide 22, we look forward to updating you on the trial design and the beginning of our Phase II trial in gastric cancer in the second half of this year, and this program has received Fast Track Designation by the FDA. We would like to thank the participating patients, their families and research teams, and thank you for your attention. We will now take questions. Operator, could you open up the line for questions?

[Operator Instructions] Our first question comes from the line of Michael Yee with Jefferies.

And congrats on the data. I had 2 questions. One was, I guess, given the higher response rates and potentially higher signals of survival, is the plan to move to a Phase IIb type of [indiscernible] against a ramu-based regimen? Is that the thought? So maybe just talk about the next steps. And then the second question is, I guess, taking the learnings from this in head and neck, can you just describe your thoughts around the MDS study that is ongoing, the progress there and how to think about what you would show there versus other CD47 that have shown data in MDS?

Sure. This is Sophia, so I'll take that. So thank you for your question. For the first question, in terms of what is our next steps for the gastric study, so certainly, we are looking at a phase -- really a Phase II/III type study design, the Phase II portion. What we'd like to be able to definitively show, especially given the Rha data presented recently at ASCO is that ALX is providing added benefit onto that backbone of the tras-ram-pac. So the Phase II portion of the study will look at ATRP versus TRP. And there, we'll be looking for not only improved delta between the 2 arms, but also looking at the magnitude of ATRP versus historical controls for RP. So certainly, those are the goals -- the main goals of the Phase II portion of the study. Now as you know, TRP is not a regulatory comparator at this point in time. So beyond the Phase II, we do intend to do the Phase III where we will be combining ATRP -- excuse me, comparing ATRP versus RP in the Phase III portion of the study. For the second question, as you were alluding to, in our head and neck data, which will be updated at a conference later on this year, we'll also be looking at a fully accrued cohort looking both at overall response rate as well as some of these later time points -- survival-based time points. And then in terms of how that is impacting our MDS study or what our MDS study is doing right now, for the MDS study, where we have our Phase I portion looking at ALX148 in combination with azacitidine, there, we anticipate being able to read out that Phase I portion of the study at a conference towards the end of this year and then seamlessly moving into the Phase II portion of the study, which will be randomized, looking at ALX plus azacitidine versus regulatory comparator of azacitidine alone.

I guess the question was, given what we've seen in totality that you've presented, do you have more confidence in MDS and what others have seen? Do you expect that to be similar, better? How do you think about that given all the data that we've seen now. That was kind of the question.

So yes, I mean, I think that, clearly, we've shown that this -- even with this early data, it's an active drug. And so our hopes that are in the MDS study, where we can also capitalize off of the proof of principle that other CD47 targeted agents have shown in that space, that our drug, ALX148 in combination with aza would also have at least the same or better activity. I think our nonclinical data in MDS supports that. And then certainly, the signals of activity that we're seeing really across these different tumor histologies and different combinations suggest that ALX is truly providing added benefit via its myeloid checkpoint inhibition.

Our next question comes from the line of Colin Bristow with UBS.

And congrats on the data. Just any other details you can give us on the Phase II design, anticipated timing of enrollment, readout. I was also curious, are you going to explore a higher dose? You didn't hit any DLT. I'm just curious if that -- do you have any appetite for that? And then just one final question on the head and neck program. Can you remind us when we should expect to get the next update on this data set and what that will consist of?

Sure. So for the ASPEN, we're calling it ASPEN-06, so for the Phase II/III study of ALX in the gastric cancer population, here, as we just discussed, we will be looking at ATRP versus TRP. We anticipate that we'll be opening in the second half of this year. As a typical, it's -- and is anticipated to be approximately 100 patients. And as a typical Phase II, we would anticipate about 2 years or so to complete enrollment and be able to have some idea of study readout. I think this will be a global study. And given the footprint of gastric cancer, it will certainly be enriched for sites in the Asia-Pac region, but it will be a global study. In terms of the head and neck cancer studies, the Phase Ib cohort that enrolled head and neck cancer in our ASPEN-01 first-in-human study, we anticipate having that -- an update of that towards the end of this year at a conference, hopefully, SITC. But obviously, we haven't submitted the abstract yet.

And potentially higher dose.

Yes. Sorry. So for the gastric cancer study, ASPEN-06, there and -- this is one of the benefits of our drug given its safety profile. We'll be dosing ALX at 30 mgs per kg every other week, and the idea there is we will be using an every other week dosing strategy for trastuzumab. And then ramucirumab and paclitaxel are also being dosed on a standard 28-day regimen. So it puts everything on to a 28-day cycle. So it's more convenient for patients and for the treating physicians. So the ALX148 will be dosed at 30 mgs per kg every other week. That's a dose that we have already shown to be safe in our Phase I study. The exposure levels -- steady state exposure levels, we've also shown to be similar to 15 mgs per kg Q week, and that added Cmax of going from 15 to 30 has been safe.

Our next question comes from the line of Chris Raymond with Piper Sandler.

Just a couple of questions. Just on the inclusion criteria, I just want to just clarify. In terms of prior platinum exposure, just kind of curious on your wording that patients in this cohort had to have progressed on prior trastuzumab and fluoropyrimidine or platinum chemo. I just want to clarify. Did they all see platinum chemo? Or -- and if not, can you sort of clarify the percentage of patients that did -- that were platinum-naive? And then the second question -- yes, go ahead. I'm keeping the second one.

No. That's okay. Go ahead, please.

Okay. Yes. So second question is on -- just on the one patient that had the CR. I know this was confirmed after the data cutoff, but just can you clarify, is the waterfall plot in your deck representative pre or post data cutoff?

So good questions. I'll just do the latter one first. So the waterfall plot is preconfirmation, and the confirmation date -- excuse me, the patient was, as you just said, had the confirmed response after the data cutoff. So their response did deepen. And then the other piece about the prior therapies, so in -- yes, it's just the way it's worded because there are sometimes different first-line regimens that are used, whether patients are in clinical studies or not. But in this study, everybody had progressed upon a regimen that contained both a fluoropyrimidine and a platin. And the eligibility stated they had to have progressed on prior trastuzumab. We did have one patient who only received the 5-FU based and then the platin and did not have the trastuzumab. But everybody else had progressed upon that doublet plus trastuzumab.

Our next question comes from the line of Alethia Young with Cantor.

Congrats on the data. I guess, a little bit more on the CR. I just wanted to get -- like was there anything notable -- notably particular about that patient that drove CR? And then when you look at like Slide 17, where you have the comparables, can you talk about where and if CRs were seen there in comparable populations? And then I guess, as far as for CD47 target occupancy, do you think this is something that in like all tumor types, the occupancy levels need to be about for the same duration? And so one can make kind of a read across to some of the other indications you have going on as well.

Yes. Thanks. So for the patient with the CR, beyond that particular patient did have, well, one notable thing is that they were -- their HER2 scoring was 2-plus as opposed to 3-plus, and I think that is significant. We haven't -- actually in this study, we haven't looked at FISH status and looked at amplification. So for eligibility purposes, HER2 scoring is done locally at the site. However, the HER2 scoring that's presented on the waterfall plot is an exploratory analysis where we've taken archival samples and then central assessment to look for HER2 score. And in that analysis, that complete responder did have a scoring of 2-plus. So that was actually sort of significant for us. Other than that, though, the patient had sort of typical standard first-line therapy. And again, consistent with some of the other responders in this population, their response does appear to deepen over time. In terms of the second question, looking at CR rates and other studies, in all of these studies, there is about -- I'd say, about a 5% or so CR rate, maybe a little bit less. So I think you would expect to see some -- hopefully, some CRs. And I think in the smaller data set, where we're seeing -- the 1 CR, it kind of puts it at about 6%. I think the tras-ram-pac data coming out of ASCO a couple of weeks ago had a 2% CR rate, but our numbers are pretty small. So I think the presence of the CR is significant and in the smaller data set, and then we'll see what happens as we get into the Phase II setting. In terms of receptor occupancy, so this is a very interesting area where what we've shown, I think, over the course of ASPEN-01 is that having full receptor occupancy in the periphery is important. But even more than that, being able to push that dose past full receptor occupancy in the periphery, the hope is that it drives that drug into the tumor microenvironment where in the solid tumor setting drug exposure may be limited. So going back to previous presentations of this data set in our non-Hodgkin lymphoma cohort, certainly, there, we found an exposure response relationship where, even after having full receptor occupancy in the periphery, as we increased our exposure, that led to increased responses in that population. And so for that reason, in these other solid tumor indications, we've moved towards a exposure level that is equivalent to 15 mgs per kg Q week. So here in our gastric population, you'll notice we have increased the dose to 15 mgs per kg Q week, and it's that exposure that we'll be bringing into the Phase II setting at the 30 mgs per kg every other week. Similarly for our MDS study, there, we are dosing up to 60 mgs per kg Q month. But again, that's a building block. It will have the same exposure as 15 mgs per kg Q week. So we're doing that program-wide.

Our next question comes from the line of Swayampakula Ramakanth with H.C. Wainwright.

Most of my questions have been answered, but I just want to have an understanding of the patient profile or the patient baseline characteristics in those studies that you were trying to compare on the Slide 17 of your presentation. Any commentary on how universally similar they were in terms of the baseline characteristics? And I know ASPEN stacks better than the rest of them, but I'm just trying to understand how similar the baseline characteristics as well.

Sure. Well, I think as we just sort of go down the line, as you look at the Rha et al data from ASCO 2021, there, overall, the patients, the demography was fairly similar. I think one key difference is they had a much higher percentage of IHC3+ patients. In their reports, they had 78% of their patients were -- again, all of these are done with archival. In this particular slide on 17, these other study populations have been designed based off of archival blocks, and we'll talk about DESTINY in a second. So for the Rha et al, there, they had, I think, about a 78% reported IHC3+ population as compared to us where, in terms of our known 3+ made up about 22% of our population, and then about 28% of our population was IHC2+. And then -- but there was about 1/3 of the patients who we did not have sufficient archival biopsies to be able to adequately characterize our HER2 score. So in terms of now IHC3+, our population had less. So that actually, I think, is significant given that even with that, we're seeing a 72% objective response rate across the board in our cohort. So that's one thing. One would anticipate that with the higher IHC3+ status and certainly the DESTINY-01 data that was presented at ESMO GI as well where they did characterize the patients with regards to IHC as well as ctDNA exploratory analysis there, they do see that with the higher amplification or with the higher CT score that does seem to correlate with increased response. In terms of DESTINY-01, I think the key difference there is that, in that population, all patients enrolled were third line or greater. They had failed 1 prior trastuzumab, but had failed 2 prior regimens. In our study, we are a second -- true second-line study where all of the patients in this particular cohort had failed 1 prior regimen. I think we had 1 patient who had failed 2 priors. So very much a standard second-line population, but all patients with the exception of 1 has failed prior trastuzumab. In terms of the IHC status there, I'd have to go back and look. I actually don't know off the top of my head the percentage of IH3 versus 2+ in that study. But certainly, they saw the correlation that would increase IHC3+ or amplification. There was greater response. And then in terms of RAINBOW Asia -- or RAINBOW overall, there, this is also a second line or greater population. The study was done -- obviously in time -- was done earlier in time compared to the more recent DESTINY-01 randomized data. So their standards of care, they would have also received trastuzumab and presumably 5FUs and a platin in the first-line setting. So there, I think it's probably most similar to our population. I think, overall, about half the patients in the RAINBOW overall study were Asia, about half were ex-Asia. And then what's presented on Slide 17 is really just the Asia Region 3. Certainly, just to be complete, DESTINY-01, all Asian study with enrolled by subjects coming from Japan and South Korea, and then the Rha data at ASCO 2021 was -- although single arm was also predominantly single institution coming out of South Korea as well.

Great. I was just trying to understand how -- what's the impact of biology on the doublet, the ALX148 plus tras that you could only achieve 4 PRs at this point? Is -- how should we think about that data? Or is that because some of them were tras failures or nonresponders? That is the reason why you could not see any better data than what you have at this point.

I think for the doublet, I would look at it through a slightly different lens. I think for the doublet, the significant thing here is that this is also a chemo-free regimen. So anytime you put chemotherapy in the mix, you're going to have a higher response rate. And I think you can see that in all of the other studies that we've presented, including our chemo combination. However, in the doublet data here, now, we are in a chemo-free regimen. And so here with ALX, as we stated earlier, based on the molecular design, no anticipated monotherapy activity. When you look at the trastuzumab component, to our knowledge, there isn't a second-line tras-only published data set out there. So the closest that we can do is to look at tras contribution to a chemotherapy backbone. So in the randomized study, TACT as you guys are familiar with, trastuzumab patients who had failed prior tras-containing regimens in the first line went on to be randomized to receive paclitaxel alone or trastuzumab plus paclitaxel, so tras after tras. There, the data between the 2 arms was very similar. There was about a 33% objective response rate. Again, chemotherapy is in the mix. However, when you look at the delta between those 2 arms, it was, I think, on the order of 1%. So the trastuzumab did not seem to be adding any additional activity in that second-line setting to the chemotherapy backbone. So in our data set, when we look at the doublet to see an objective response rate of 21%, in the absence of chemotherapy, we think, is pretty significant. When you go on to look at the other survival-based end points, so for example, I think in Slide 18, I have it compared to probably a more recent chemo readout, which would be the control arm for DESTINY-01, where patients received either irinotecan or paclitaxel, here, you can see the median overall survivals of ASPEN-01 of 8.1 months and DESTINY control arm of 8.4 months, and then the sort of intriguing 12-month OS rates a little bit now later in time, you're seeing a 38% OS rate at 12 months from the ASPEN-01 population of the doublet versus 29% from the DESTINY control arm. So again, in a chemo-free setting, we think this is significant, and I think it's really representative of the enhanced ADCP activity of trastuzumab. So we're no longer relying on inhibition of ERBB2 signaling by trastuzumab, but rather exploiting this ADCP activity, and that's the rationale for adding this doublet onto the current global standard of care the, ram-pac.

Our last question comes from the line of Adam Evertts with LifeSci Capital.

Just curious if we can get a better understanding of where the new responses came from versus SITC. How many of the 4 new responses came from the newly enrolled patients versus sort of delayed responses from the patients we saw at SITC with stable disease?

Let's see. So if we go to roughly to Slide 16, the new -- like I say, certainly, the complete response that was a newly enrolled -- one of those 4 newly enrolled patients, the confirmed CR. The other -- the others, out of the remaining 3, I think they all had PRs. I'd have to go back, but I think the stable diseases were -- I'd have to go back and look. I think the stable diseases may have been from the earlier group. But of those 4, they were definitely one CR, and there would have been some of the PRs in that mix as well.

And Sophia, do you know the 2 unconfirmed responses from SITC confirmed?

One of them confirmed -- one of them, the patients actually came off due to -- let me see. One of them, the patient came out for a reason. So this is one of the stable disease other than an adverse event. And one of the PRs that came off was due to patients' decision. The PRs came off due to disease progression. So it's a little bit of a mix. One was disease progression. One was patient's decision.

There are no further questions at this time. I would now like to turn the call back over to Dr. Jaume Pons for closing remarks.

Okay. I just want to thank you for your attention this morning to our call. And with that, we can conclude the conference.

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