Alx Oncology Holdings Inc Q3 FY2025 Earnings Call

Thanks, everyone, and welcome to our Q3 2025 results. I appreciate everybody spending some time with us this morning, and I'm looking forward to this update. On Slide 2 here before we start our presentation of housekeeping here are our forward-looking statements for your review. So on the next slide, Slide 3 here, here is the agenda and our plan for today. We're going to be providing an update on our key accomplishments in the third quarter of 2025. Most notably, we are very excited to share with you the data set that will be presented at SITC this weekend from a preplanned analysis of our ASPEN-06 trial that showed CD47 expression as a key predictive biomarker for increasing durable clinical response with evorpacept in HER2-positive gastric cancer patients. So our goals for today are most importantly to share these detailed results with you as we believe this data set now clearly validates the role of CD47 in HER2-positive cancers. We will then give you a sense of how this data now impacts our development strategy for evorpacept going forward. We will also be providing an update on our novel ALX2004 EGFR-targeted ADC, which is now in the clinic. Today, we are also excited to be joined by Dr. Peter Schmidt from Barts Cancer Institute in the U.K., who is a key opinion leader in breast cancer and investigator in our evorpacept Phase II breast cancer study. He will be presenting his views on evorpacept data and its potential within the current treatment paradigm for HER2-positive metastatic breast cancer. Then our CMO, Barb Klencke, will provide an update on our novel EGFR-targeted ADC, ALX-2004, which is currently dosing patients in our Phase I trial. Now on Slide 4. In the third quarter, we made significant advances in both evorpacept and ALX2004 clinical programs. We again are excited to present the full data at SITC that is demonstrating the potential of CD47 expression as a predictive biomarker and highlight a clear opportunity to now identify patients who are most likely to achieve the greatest benefit from evo. As Barb will present in detail in our clinical section in this analysis, we saw that patients with high CD47 expression derived the greatest benefit across all key efficacy markers, response rates, duration of response, median PFS and overall survival from evorpacept versus those with low expression. The data is very clear as the magnitude of benefit across many of these metrics was double or even triple those observed in the control arm and also clearly compare very favorably with the large benchmark studies in second-line gastric cancer. Most importantly, these results support the potential to pursue targeted oncology approach to additional tumor types with evo. And given the broad overexpression of CD47 in both solid tumors and heme malignancies, it gives us a real opportunity to really focus evo now as a targeted IL therapy. Our Phase II clinical trial in breast cancer, which is designed to pursue a CD47 and HER2 biomarker-driven strategy based on this strong data is on track to dose its first patient this quarter. And as we've discussed, evo has the potential to represent the first and only option for metastatic breast cancer patients who overexpress CD47, which we know can lead to worse outcomes and a poor prognosis for these patients. And with our second pipeline product, our novel EGFR-targeted antibody, ALX-2004, which is a highly differentiated ADC, we presented preclinical data and design of our Phase I trial at the Triple Conference a few weeks ago. So we're excited to announce today that we are currently enrolling patients in the second dose cohort. We're rapidly clearing the first dose cohort in this Phase I trial. Turning to our financials quickly. We reported a total cash balance of $67 million, and that cash is expected to provide us runway into the first quarter of 2027, which positions us to achieve the value-enhancing data milestones for both ALX2004 and evorpacept that we have coming next year. Now turning to Slide 5. It has been very well established that CD47 is widely overexpressed across almost every type of cancer. And it is also clear that CD47 overexpression matters as it is clearly a negative biomarker for patients. So when you look at research in CD47 over the last decade plus, it is a very strong foundation that CD47 is clearly a negative prognostic biomarker. And what you can see here is a meta-analysis of 38 cohorts across 17 publications, which includes over 7,000 patients. And there really is no question that CD47 is clearly associated with shorter survival and worse outcomes. And you can see on the right, the wide range of tumor types where this has been established. Now turning to Slide 6. And as a reminder that during our Q2 call just a few months ago in August, we presented the top line data, which support that CD47 overexpression is a clear predictive biomarker for response with evorpacept in HER2-positive gastric patients. In this analysis, patients with both confirmed HER2 positivity and CD47 high expression had a dramatic response to evorpacept as compared to those in the control group who did not have vivo. And as you can see here on the ITT population, we saw a strong response of a 41% ORR in the evo arm versus 27% ORR in the control arm. And if you look at the data now in patients that clearly have CD47 high expression, there is a magnitude of benefit for those patients where we had an ORR of 65% in the treatment arm versus 26% in control with a nominal p-value of less than 0.05. Now as you can see on Slide 7 and what we're very excited to share with you now and at SITC later today, this strong ORR benefit with evorpacept in combination with TRP and CD47 high patients was also reflected and translated well to DOR, PFS as well as survival as it's clear that patients who overexpress CD47 and retain HER2 expression is driving the effect here. This is very important as this clearly validates EVO's dual mechanism of action. And again, this is a second-line plus gastric population, which has historically been a very tough cancer to treat. So in addition to the ORR benefit, which had a delta of almost 40% versus control, the median duration of response here for those patients is over 2 years, which is more than triple the control. The median PFS was over 18 months in the evorpacept arm versus just 7 months in control with an impressive hazard ratio of 0.39. And then we were also pleased to see these gains further translate to a benefit to overall survival where we saw a median OS of 17 months with evo versus about 10 months in control and also a strong hazard ratio of 0.63. Barb will walk through this data in more detail and the full data set will be shared at SITC here soon. What is clear is that this data shows the potential for evorpacept to drive really substantial benefit for these patients with high CD47 expression. On the next slide, this just shows the focus set of milestones that we're driving to now. In summary, we're laser-focused on these 2 programs. First, driving evorpacept into ASPEN-Breast, which is our study investigating patients post in HER2 and again, focused on CD47 high and understanding the impact of that biomarker. We continue to execute well against the milestones that we've communicated in the past and are anticipating first patient in Q4 of 2025, with interim data expected Q3 2026. ALX2004 also remains on track and continues to proceed very well. We dosed our first patient in August of 2025, and we continue to expect initial safety data first half of 2026. Turning to the next slide, and in summary, before I hand the call over to Barb, we had a strong quarter, both in terms of execution, continued tight discipline around our capital and are excited about the key value catalysts for ALX in 2026. And as you can see here on Slide 9, this is a snapshot of our current clinical pipeline. As we have communicated previously, we are pursuing a focused development strategy for evo in combination with anticancer antibodies, given the consistent proof of concept that we have seen in various clinical studies with different monoclonal antibodies, and this data here today further builds on that. In addition to our HER2-positive breast cancer program with a CD47 biomarker-driven approach, ALX2004, again, our EGFR-targeted ADC continues to progress well, and we are also very excited about our partner program with Sanofi Sarclisa in multiple myeloma, which is now in dose optimization phase. So next, we'll turn this over to Barb, who will take over and provide more details on the evorpacept CD47 biomarker data presentation coming here at SITC.

Thank you, Jason. I will begin by explaining how evorpacept works. CD47 is commonly overexpressed on cancer cells to evade immune detection by sending a signal that instructs immune cells not to consume these cancer cells. Evorpacept, which is a fusion protein, is specifically engineered to block this signal. Its Fc region is deliberately made inactive, making it particularly effective when used with anticancer antibodies like Herceptin. The active Fc domain of the antibody can then promote effective phagocytosis, which would have otherwise been inhibited by the CD47 signal. The approach of evorpacept to block CD47 differs fundamentally from the traditional methods used by other CD47-targeting agents. Although CD47 is overexpressed in cancer cells, it is also found in healthy cells, including red blood cells. Traditional methods that block CD47 by using antibodies that also activate macrophages through an active Fc have led to serious side effects in some patients, resulting in these approaches largely failing. In contrast, evorpacept's strategy of using an inactive Fc spares normal cells, and our safety data from more than 750 evorpacept-treated patients supports the safety of this method. Now, moving to the ASPEN-06 gastric study that Jason mentioned earlier. We included 127 patients with HER2-positive gastric cancer who were in the second or third line of treatment, all of whom had prior HER2-directed therapy. Patients were randomized to receive either evorpacept with trastuzumab, ramucirumab, and paclitaxel, or the TRP alone. The main goal was to assess objective response. Since HER2 expression may decrease after prior therapies, we focused on looking beyond HER2 status from archival tissue. Evorpacept is less effective if HER2 is underexpressed on the cancer cell surface. Therefore, we collected ctDNA at baseline from all patients, and additionally, 48 patients had biopsies either at the beginning of the study or after their previous therapies. In total, 95 patients, or 75% of those enrolled, were verified as retaining HER2 positivity through either ctDNA or fresh biopsy. Of these, 90 patients were assessable for CD47 expression in tumor cells using archival tissue or fresh biopsies. High CD47 expression, determined by IHC3+ staining in at least 10% of tumor cells, was found in 48% of these patients. The expanded results of this preplanned exploratory analysis regarding efficacy based on CD47 expression in patients who maintained HER2 positivity will be the focus of today's discussion. We previously reported a 48.9% response rate in the evorpacept arm versus 25% in the control arm among the 95 patients who retained HER2 positivity. In terms of CD47 expression levels, evorpacept produced a 65% response rate compared to 26% in the control arm for patients with high CD47 expression. Control arm response rates were consistent across CD47 expression levels and were lower than those in the evorpacept arm for both CD47 high and low groups. Now, let’s look at the duration of response in these subgroups. We again observe that CD47 expression serves as a strong predictive biomarker for the benefits of evorpacept. The duration of response for all HER2-positive patients, regardless of CD47 expression, was 15.7 months for those receiving evorpacept and TRP, compared to 9.1 months for responders in the control arm. For the CD47 high group, the duration of the response was three times longer for those in the evorpacept trastuzumab RP arm compared to the control arm, with medians of 25.5 months versus 8.4 months respectively. In the CD47 low group, the evorpacept TRP had a duration of response median of 11.2 months compared to 12 months in the TRP group. Next, we will review progression-free survival in patients with confirmed HER2 positivity and high CD47 expression. The hazard ratio was 0.39, with a median PFS of 18.4 months for the evorpacept trastuzumab RP arm, more than double the 7 months seen in the TRP alone arm, again underscoring the predictive value of CD47 expression for evorpacept benefit. Likewise, in the HER2-positive CD47 high evorpacept arm, the median overall survival was 17 months compared to 9.9 months in the control arm, yielding a hazard ratio of 0.63. All of the data being presented this week at the SITC conference is based on thorough follow-up, with median survival follow-up around 25 months. We have examined various thresholds for CD47 expression based on IHC testing strength. We analyzed medium or high-intensity staining defined as IHC 2+ and 3+ in at least 10% and in at least 25% of tumor cells, as well as high-intensity staining defined as IHC 3+ in samples with 5% or more or 10% or more of cancer cells expressing that staining. The prevalence of high CD47 expression ranges from 40% to nearly 60% among HER2-positive patients, depending on the threshold used. Notably, we see consistent improvements in response rates, PFS, and OS in patients treated with evorpacept across different cut points for CD47 expression. I will also present a comparison of our evorpacept efficacy data for patients with retained HER2 positivity and high CD47 expression against benchmark trials in HER2-positive gastric cancer. While considering the usual caveats of cross-trial comparisons, evorpacept’s data appears to favorably compare with the recent ENHERTU outcomes from the DESTINY gastric04 trial in the second-line setting. In that study, close to 500 patients were required to confirm HER2 status via fresh biopsy following a trastuzumab regimen, and they were randomized to ENHERTU or RP as a control. Although ENHERTU demonstrated effectiveness in the second-line setting, our data for second and third-line evorpacept in patients with high CD47 expression, a recognized negative prognostic biomarker, appear to be substantially better. Now shifting gears from gastric cancer to HER2-positive breast cancer, I will introduce a Phase Ib/II trial conducted by Jazz evaluating the safety and efficacy of evorpacept plus zanidatamab in HER2-positive breast cancer patients who progressed after prior HER2-directed therapy. These patients had extensive previous treatments, averaging six prior lines. In 9 patients confirmed to retain HER2 status, the response rate was 56%. The median duration of response in this group ranged from 5.5 to nearly 26 months, with the median not reached, while the median PFS was 7.4 months. These findings provide a favorable comparison to benchmark data, such as the SOPHIA trial, where the response rate was 22% for the drug MARI2’etuximab in patients primarily in the second and third lines of HER2-positive breast cancer therapy. In these two studies, we have demonstrated the potential of evorpacept to activate the innate immune response, which validates its mechanism of action when combined with targeted antibodies in both HER2-positive breast cancer and gastric cancer. This reinforces our confidence in evorpacept's potential and its future in treating HER2-positive breast cancer, which we will discuss next. The opportunity for evorpacept in breast cancer has a high probability of success, having been supported by two positive outcomes across different HER2-positive contexts. A CD47 HER2-positive biomarker-driven strategy with evorpacept offers a highly targeted approach to address significant unmet medical needs in the evolving breast cancer landscape, particularly for patients who have progressed on ENHERTU. I am pleased to introduce Dr. Peter Schmidt, a leading Professor of Cancer Medicine and Center lead at the Center of Experimental Cancer Medicine at Barts Cancer Institute. He is a well-known global lead investigator for several ongoing Phase III trials in both metastatic and localized breast cancer. Notable trials under his leadership include the pembrolizumab KEYNOTE-522 study and the atezolizumab IMpassion130 study. I will now turn it over to you, Peter.

Thank you, Bob. The treatment options for patients with metastatic HER2-positive breast cancer are currently evolving, and this is leading to significant changes. We've witnessed a very active drug, trastuzumab deruxtican, moving into second and third-line treatments, but there is now data supporting its use in the first-line setting. This is excellent news for patients as it provides a robust new first-line treatment option. However, the challenge we face is that there is no standard of care for patients who have been treated with trastuzumab. The previous treatment sequence has been disrupted. Currently, we have several options to choose from, but none have been specifically approved for patients who have already undergone T-DXd therapy. Our options include tucatinib in combination with trastuzumab and capecitabine, PD-1 inhibitors, chemotherapy along with trastuzumab, and the use of HER2 TKIs, although their usage is decreasing. We are also hopeful for additional HER2-targeted therapies. There remains a substantial unmet need for patients with HER2-positive breast cancer who have progressed after T-DXd. I see a potentially promising role for evorpacept, which has shown activity in patients after trastuzumab deruxtecan, especially when combined with other HER2-targeted agents. Our goal is to introduce new agents that can address the resistance seen with T-DXd, necessitating the discovery of novel agents that ideally employ different mechanisms to target HER2. Evorpacept operates through enhanced ADCP, providing a different mechanism than traditional payload-based ADCs we use today. We also aim to have a drug that demonstrates efficacy following HER2-directed treatments, specifically after ADCs and monoclonal antibodies, with trastuzumab deruxtecan being a focal point. Data from gastric studies, as well as studies in HER2-pretreated breast cancer, indicate that evorpacept has demonstrated efficacy following trastuzumab therapy. We seek treatments that can complement and strengthen the existing standard of care rather than replace it; evorpacept is designed to work synergistically with key therapies, primarily trastuzumab or similar antibodies. We prioritize safety and aim for a profile that is more favorable compared to established ADCs, which can have considerable toxicity. Additionally, we recognize the importance of incorporating immunotherapy agents to enhance the long-term survival benefits observed in HER2-positive breast cancer. Some patients experience remarkable survival durations on HER2-targeted treatments, and we believe that augmenting the immune response through a CD47-targeted drug can potentially extend these survival benefits. Importantly, we have a strong biomarker approach with CD47 as part of our selection strategy, supported by gastric data showing a more significant signal in patients with high CD47 expression. About 1,000 patients have been studied, consistently revealing a CD47 high expression rate of around 50%. This substantial proportion of patients enables us to advance our program without being limited by a small target group. Some preclinical data highlight the biology of CD47 expression in HER2-positive cases compared to HER2-negative cases. Our findings show higher CD47 expression in HER2-positive breast cancer cells and an increased positivity in recurrent disease. This aligns perfectly with our targeted population. Emerging data from cell lines pre-treated with trastuzumab deruxtecan demonstrate a higher percentage of CD47-positive cells among those previously treated, underscoring this as our target demographic. The ongoing ASPEN trial serves as a key focus for us. As a clinician, I am eager to move forward into a Phase III trial swiftly because we understand its effectiveness and the significant need following T-DXd therapy. However, we still need to fully grasp the statistical requirements for proper trial design—specifically regarding response rates and progression-free survival. Thus, the ASPEN trial is structured as a nonrandomized Phase II trial for patients with measurable HER2-positive metastatic breast cancer previously treated with trastuzumab deruxtecan, receiving treatment with evorpacept alongside trastuzumab and physician’s choice chemotherapy. This pragmatic design reflects real-world practices. From this trial, we hope to learn about response rates in patients with ctDNA positive results for IL-2, as well as the duration of progression-free survival and overall survival in both CD47 high and low tumors, which will help validate the biomarker data we obtained from gastric cancer studies and refine our Phase III design. Thank you, everyone. I would now like to hand it back to Barb.

Thank you, Peter. Well, let me wrap up this section with a brief breakdown of the addressable patient numbers in the core markets. As you can see, there are roughly 48,000 breast cancer patients in the second plus line setting who are HER2-positive. Of that, we believe that at least 60% to 80% of these patients will retain HER2 positivity following prior therapy. Of that group, 50% to 70% will have high CD47 expression. As Peter highlighted, there are a number of publications to support that CD47 overexpression in HER2-positive breast cancer patients will be upregulated post ENHERTU treatment. We believe that this represents approximately 20,000 addressable patients who are both HER2-positive and CD47 high. If you boil this down and use conventional estimates on pricing, we get to roughly a $2 billion to $4 billion market opportunity, again, just in patients that are CD47 high and HER2-positive, representing a significant opportunity for evorpacept. On Slide 31, I now want to provide a quick update on ALX2004, our EGFR antibody drug conjugate program. As shown on Slide 32, our company's first ADC, the ALX2004 molecule was a result of rigorous internal drug design process. Our goal is to create a best and potentially first-in-class drug designed to maximize the therapeutic window and to overcome the historic toxicity challenges that others have encountered in targeting EGFR with an ADC. With ALX2004, we have optimized all 3 components to do this, including the payload, the linker antibody to create a truly novel molecule against a very well-validated target. ALX2004 uses matuzumab-derived EGFR antibody selected to minimize skin toxicity and to maximize the therapeutic window. Its binding epitope is distinct from the U.S. FDA-approved EGFR antibodies such as cetuximab and panitumumab. Additionally, ALX-2004 has a proprietary linker payload and TPO 1 inhibitor payload engineered to offer improved linker stability for on-target delivery of payload and enhanced bystander effect. At the recently concluded Triple Meeting in Boston in October, we presented preclinical data highlighting these elements in greater detail. Moving to Slide 33. Here are the preclinical data highlights. Both in vitro and in vivo animal models support impressive dose-dependent activity and a differentiated safety profile. Importantly, nonhuman primate toxicology studies did not demonstrate EGFR-related skin toxicities at clinically relevant doses, and there was no evidence of payload-related ILD in the animals. This overall profile supports our conviction that this molecule could potentially demonstrate efficacy with a manageable safety profile in patients. Slide 34 shows a snapshot of the efficacy data from our in vivo models. ALX2004 showed regression and tumor suppression across a panel of xenograft models, representing a broad spectrum of cancer types and EGFR expression levels. Notably, ALX2004 was effective in models harboring KRAS, BRAS and p53 mutations. ALX2004 shows excellent tumor suppression activity at doses as low as 1 milligram per kilogram given either once or once weekly times 3, leading to complete tumor eradication in several of the models. These results confirm the broad applicability of ALX2004 in targeting EGFR-positive cancers. Slide 35 shows the key findings from our 6-week repeat dose with 6-week recovery period in the GLP nonhuman primate tox study. All findings were minimal to moderate and fully recoverable. Thus, these data support the design of the ALX2004 study and the likely safety margin for clinical use. Slide 36 highlights our clinical development plan. We are targeting EGFR-expressing tumor types, namely lung, colon, head and neck and esophageal squamous cell carcinoma in this dose escalation and dose expansion trial. We dosed our first patient in August, and we have completed our first dose cohort at 1 milligram per kilogram without any DLT. We are currently dosing patients in our second dose cohort at 2 milligrams per kilogram. We are on track to provide initial safety data from the Phase Ia portion of the study in the first half of 2026. Our goal in this Phase Ia, Phase Ib trial is to identify the dose that optimizes safety and activity in tumor types, which we believe have the highest potential for success. These data will then set up the program well to advance into a future registration study. With that, I turn the call back over to Jason.

Thanks, Barb, and thanks again to Dr. Schmidt for sharing his perspectives on the program as a KOL in the field. Again, Q3 was a strong quarter, both in terms of execution and new data. What we're most excited about now is driving a targeted IL breakthrough in a first-in-class drug with evo as well as are very encouraged by AOX2004's fast start in the clinic and building momentum. In sum, our CD47 blocker has been successful where no other has, both in terms of its manageable toxicity profile as well as activity as we've now demonstrated efficacy in a randomized study, and we've identified an actionable and predictive biomarker for response to evo in our gastric cancer study. This further reinforces the benefit we have seen in terms of DOR, PFS and OS. And again, this biomarker is on mechanism. Going forward, we're developing a CD47 biomarker, and therefore, it is really of no surprise to see that CD47 overexpression shows such a strong impact on our data. So what this allows us to use is CD47 to select for patients in both current and future trials with the goal of replicating the results we have seen here with gastric cancer and demonstrating the same significant and transformational benefit for patients in our HER2-positive breast study. Again, there are no approved therapies for patients overexpressing CD47 and no options in late development to address this known path of evasion. So we remain focused on delivering for them. In 2004, there are also no approved EGFR-targeted ADCs. And although clearly a validated target, there remains a substantial unmet need for these patients as well. ALX2004 is off to a very strong start in the clinic, and we believe also has the potential to redefine standard of care across a range of EGFR-expressing cancers. So with that, I'll open up the floor to Q&A. Again, thank you for the time this morning.

And our first question will come from Lee Watsek with Cantor Fitzgerald.

This is Daniel Bronder on for Lee. This is an exciting update, and we're curious to hear your thoughts on how to correlate the CD47 positivity that you showed on Slide 30 with the kind of CD47 expression cutoffs that you showed in the gastric data on Slide 18. What would you say is CD47 high in this context? And how should we think about the patient population that would be matching that in your trial?

Thanks, Daniel. Appreciate the question. So 30, just thinking about what we saw in breast or what we've observed in the literature versus gastric, is that the question?

Yes, basically, yes.

It's a great question, and one we've explored. We're fortunate to have a strong scientific foundation behind CD47. We see promising agreement across the two indications. In gastric cancer, about 50% of the group is CD47 high. Looking at the benchmarks, we have five different publications that focus on CD47 specifically in HER2-positive cancer, and we observe strong agreement there as well. Adding those numbers, around 50% of the patients in those studies are also CD47 high. Interestingly, these publications use different clones and methodologies. This gives us confidence that our findings are applicable not only to breast cancer but also to a wide variety of tumor types.

And if I may, can I ask a follow-up question?

Yes, sure.

How should we think about your companion diagnostic development? Are you doing that yourself in-house? Are you using the same kind of evorpacept construct? Or are you using an independent antibody? Can you shed any light on that?

Yes, sure. I mean I'll take it at a high level and then maybe ask Barb to weigh in on the path to a CDx. We've done the testing with a partner for the gastric study, plan to do the same in breast. And then, of course, as this data builds and I think as we continue to understand the right cutoff and how this translates, we'll pursue further work. But Barb, do you want to add to that?

I would just say that the assay is an IHC. It's a research use assay that was applied to the gastric data. Our ongoing or our soon-to-be enrolling trial in breast cancer, the 80-patient single-arm trial will use the same research-based assay. And then we are working already with partners to think about the operationalization of the process prior to the initiation of a Phase III trial so that we will be ready for a companion diagnostic, but again, via a partner.

And our next question comes from Roger Song with Jefferies.

Very interesting data. Maybe related to the efficacy in the CD47 high population, do you have any data in your breast cancer trials with Jazz and any new data you can maybe give some comments on the CD47 high versus low? And then in terms of historical breast cancer, do we have any evidence for the CD47 high population, the traditional or the standard of care is performing less than the CD47 low population? Have you done any retrospective study as well? Because I know the benchmark is using the SOPHIA or any other HER2 chemo combo, but that's in the broad HER2 positive, not the CD47 cutoff.

Yes, those are both excellent questions. First, on the comparisons in the zani study, this data and its impact on our development plan is relatively new. We're excited about what we're observing, particularly the strong results for CD47 high in gastric, which is clearly driving the study's effects. Naturally, we want to explore where else this may be effective, whether in our studies with Jazz, Sanofi, or other anticancer antibodies. Currently, we see a 56% overall response rate in patients post-ENHERTU who have undergone extensive HER2-directed therapy. This is significantly higher than what is expected with the margetuximab comparator. Recent ESMO data reflects a typically low response rate, with a real-world study showing under 20% in patients' overall response rate post-ENHERTU, making our 56% quite impressive. We're actively working to evaluate the differences between CD47 high and low responses. Regarding your second question on benchmarking, Barb outlined the comparator with ENHERTU in the DESTINY-Gastric04 study. From what we understand, the control arm in the RAINBOW studies is performing on par with various benchmark studies. Those benchmarks likely represent the best outcomes we can expect, considering CD47 high is a negative prognostic factor, suggesting those patients would typically perform worse. Clearing those benchmarks and expecting to compare favorably, particularly with the insight that CD47 high patients could do even worse if selected from studies, certainly strengthens our confidence.

And we'll go next to Sam Slutsky with LifeSci Capital.

Just on the interims next year, both the EGFR ADC and the breast cancer program with evorpacept, curious on how many patients you're hoping to have in each of those data sets? And then just how you view a win as you think about safety on the EGFR side and then just delta efficacy on the evorpacept side?

Yes, both great questions. Thanks, Sam. I'll take 2004, and I'll ask Barb to weigh in on the breast front. I think 2004, as you know, targeting EGFR, one of the most well-validated Trode targets in oncology, there's just no question that EGFR is effective. So I think it's led to a natural question from investors and partners, and that's can you target this target with an ADC when you have a payload involved. And as a reminder, again, I think we're very encouraged by what we see in the primate work. That tends to translate very well. And so far, so good, right, to clear 1 mg per kg quickly, I think, is a strong start at already a relatively high dose and now on to the next cohort, which, again, I think is moving fast is what you want to see. So as we go into the next year, early next year in terms of what we'll share, I think it's it depends, right, which is the reality of a dose escalation study. I think our goal is to answer the safety question as best we can in a Phase I and then put up data that will answer that. And again, the study is marching very well here. And I think we feel real confident that if this continues, of course, we'll be able to share something going into early next year. And then on the breast front, in terms of benchmarks, Barb, do you want to weigh in on that one?

Yes. I think, Sam, thank you. I think you were asking what might our expectations be both for number of patients as well as the bar. The bar I'll start with. There's a lot of data with trastuzumab and chemotherapy, which really is the backbone upon which we add evorpacept in our trial. Chemotherapy, trastuzumab at best will have about a 20% response rate. Interestingly, there was new data coming out of ESMO looking at the post-ENHERTU setting and response rates continue to drop, not unexpectedly. And as we noted, our trial will enroll all patients post ENHERTU, where we do anticipate that CD47 overexpression becomes part of the mechanism of resistance, we attack that directly, and we anticipate having good outcome data in our evorpacept trial. So I think the benchmark is going to be in the range of 15% response rates. Again, 20% might be the upper bound, but with the combination of the 2 things, the poor prognostic effect of CD47 as well as the evolving standard of care and the fact that there really isn't anything that has shown up well post ENHERTU really bodes well for us. What do we expect in our bar? I think doubling that would be nice, 35% to 40%. We certainly in our gastric data that I showed you in the gastric setting did even better, and we anticipate that the opportunity is there to do quite well, but I think we would be very happy with a 35% to 40% response rate in our breast trial.

And this now concludes our question-and-answer session. I would like to turn the floor back over to Jason Lettmann for closing comments.

Great. Thanks, everybody. Really excited to share this data with you and continued good progress across both evo and 2004. So a real positive update today. And again, I appreciate the engagement and support and look forward to future updates. Thanks so much.

Ladies and gentlemen, thank you for your participation. This does conclude today's teleconference. You may disconnect your lines, and have a wonderful day.