Hi everyone, and welcome to Affimed's financial community call to discuss data from the AFM24-102 study in non-small cell lung cancer. As a reminder, today's conference call is being recorded. I would now like to introduce your host for today's call, Alex Fudukidis, Head of Investor Relations at Affimed. Please go ahead.
Thank you, Livia. Good day, everyone, and welcome to our call today to review data from the AFM24-102 study. Before we begin, I'd like to remind you that we also issued a press release which may be found on the news section of our website. On our site, you will also find the presentation which we'll be sharing with you today. Presenting on behalf of Affimed today, we have Dr. Shawn Leland,
Affimed's Chief Executive Officer, and Dr. Andreas Harstrick, our Chief Medical Officer. The rest of our management team will also be available to answer any questions that you may have in the Q&A session. We are also honored to have with us today Dr. Valentina Boni, a distinguished Medical Oncologist and Director of Clinical Research at Next Oncology based in Madrid, Spain. Dr. Boni has been a pivotal figure in early drug development and clinical trials.
She has played a key role in the development of several approved anti-cancer drugs, and her research interests span molecular genetic targets, immunotherapy, antibody-drug conjugates, and precision medicine for cancer therapy. Dr. Boni is a key investigator in our AFM24-102 study. I'd also like to remind you that today's presentations contain projections and forward-looking statements regarding future events. These statements represent our beliefs and assumptions only as of the date of this event. Except as required by law, we assume no obligation to update these forward-looking statements publicly or to update the reasons why actual results could differ materially from those anticipated in the forward-looking statements, even if new information becomes available in the future.
These forward-looking statements are subject to risks and uncertainties, and actual results may differ materially from those expressed or implied in these statements due to various factors, including but not limited to those identified under the section entitled Risk Factors in our filings with the SEC and those identified under the section entitled Forward-Looking Statements in the press release that we issued today and filed with the SEC. With that, I'll turn the call over to Shawn to present the data.
Sounds good. Thanks, Alex. Good morning, good afternoon to those joining the call. Thank you for joining, and we look forward to providing a company update today on AFM24. Today's call will focus primarily on the EGFR wild-type cohort, but we will also provide an update on the EGFR mutant cohort, along with some encouraging data that we have discovered related to the exposure-response relationship that not only have we seen with AFM24 but with our prior NK cell engager programs. With that being said, I'll turn the call over to Dr. Boni to take you through the data. Dr. Boni?
Hello, good morning, everybody. Thank you for the kind invitation. I am really very happy to share with you the results of this study in which I am involved as a Principal Investigator at Next Oncology Madrid, in the University Hospital Quirónsalud Madrid. As said before, I am going to present two parts of the study, particularly the report for non-small cell lung cancer, EGFR wild-type, and for those that are EGFR mutant. Let's start with slide number five that is regarding the population involved and analyzed in this part right now for non-small cell lung cancer, EGFR wild-type. Here we have a résumé which patients have considered for the analysis at the data cut-off that was in November 2024. The patients included in the study in this cohort were 43 patients. However, the analysis was done in 33 patients.
Ten patients were not included in this analysis. Two, because they don't have already done a CT scan, because too early for them. Six, because they have early progression, so they don't receive enough amount of the drug in order to be considered evaluated in terms of efficacy. And two discontinued the drug due to unrelated adverse events. Move to the slide number six. Here we can see demographic characteristics of the patients included in this analysis. The majority of patients were male, a median age of 68, and the majority were white patients with adenocarcinoma as histology. ECOG was one for the majority of them. Of remark, the majority of the population was a pretreated population with a median number of prior lines of therapy of two.
All of them received previous platinum-based chemotherapy and CPI, and the majority of them have also received taxane. The majority of the population included was resistant to previous PD-1 therapy. Also, of note, the patients included more than 80% of patients who were not expressed PD-L1 expression in the tumor. The median number of metastatic sites was two. In slide number seven, we can see the safety profile of the drug. Here it is, of course, the combination of the AFM24 with atezolizumab. No new safety issues were found here. Also, of note, the most important adverse events were infusion-related reactions. That was the most common AFM24-related adverse event. But the number of Grade 3 reactions was small, with just four patients that have infusion-related reactions more than or equal to Grade 1.
It was seen also transaminase elevation, but it was just in two patients more than Grade 3 , and it was really transient, and of note, just one pneumonitis and no ocular toxicity were seen with this combination. Regarding the efficacy, as shown in slide number eight, it was achieved a very nice disease control rate with 76% control rate in this population, and the tumor shrinkage in 48% of patients analyzed here. The objective response rate was 21%, with one complete response and six partial responses, one pending to be confirmed. Of the seven patients that achieved response, five of these patients have not achieved previous response with CPI, that is important, and all patients included here that achieved response here have documented disease progression to previous CPI, so these are patients that have resistance to previous immunotherapy.
In this slide number nine, we can see the duration of the response. The combination of the two drugs induced a durable response in this heavily pretreated population, and the 56% of patients included in this analysis were ongoing and are still receiving treatment. The median progression-free survival is shown in the Kaplan-Meier plot in slide number 10, and here it was seen that the progression-free survival achieved was 5.6 months, with a median follow-up of 7.5 months. In this analysis, it is also reviewed the efficacy of the population in these patients that have received previous taxane therapy and those that never received taxane and that was included in the trial. As you can see here, the patients that are not receiving prior therapy with taxane were 12.
In this population, we achieved a more and higher number of response, with an overall response rate of 25% in comparison to those patients that have already received taxane. The median progression-free survival was also different, with a benefit for those patients that have not received previous taxane. A median progression-free survival of 7.4 months versus 4.4 months for those patients that have previously received taxane. In slide number 12, we see a patient that was included and a sample of a patient that was included in this cohort. This is a male, 66 years old, diagnosed with metastatic lung adenocarcinoma with stage IV lung cancer at the time of the inclusion. The patient has already received treatment with carboplatin and pemetrexed, discontinued due to progression, and achieved as best response to the stable disease.
Then a second-line patient received nivolumab from March 2021 to September 2022, and discontinued due to progressive disease. This patient received also third-line with Taxol and was discontinued before the patient joined the trial. As you can see here, the patient started the treatment with AFM24 in May 2023, and the patient has a response with the treatment, and then the median progression-free survival was approximately of eight months. Regarding the patient with EGFR mutant included in this study, here we analyzed 20. We have 28 patients included, but the patients that were reviewed for this analysis were 17 patients. This was because four were not included due to early progressive disease, two discontinued due to unrelated adverse events, and five of these patients are not included because too early and not evaluable yet because not done the first CT scan.
So regarding the patient's demographic characteristics that are shown in the table in slide number 15, you can see that the median age of this population was 67 years old. The majority of these patients were female. Another number was Asian population. All of the population was adenocarcinoma, and the majority of them were with PS cohort one. The number of metastatic sites, the median number was two, and the median number of previous line was three, of which all of them received TKI. More than 80% of the population received the third-generation TKI, and importantly, a majority of the population, more than 70% of the population, received also a platinum-based therapy. So it means that it is a heavily pre-treated population.
As shown in slide 16, we can see that the tumor shrinkage achieved for this population was in 41% of patients included, with a disease control rate of 71% and an objective response rate of 24%. One complete response confirmed, and three partial responses also confirmed. In slide number 17, we can see the duration of the response. In these, five patients were still ongoing for more than 10 months at the time of data cut-off that was in November 2024. In the next slides, we can see an example of a patient included in the trial, and this is from my side. This is a lady 75 years old, diagnosed in October 2021 of stage IV adenocarcinoma, harboring EGFR exon 19 deletions. She was a lady heavily pretreated, including osimertinib, platinum-based therapy, and also another TKI in the context of the clinical trial.
At the time of the inclusion, she had progression to these three lines, and we can see that she had a very nice response from the very beginning from the second CT scan that we have, and this was a very nice example of a durable response because despite a pseudo-progression seen in the summer, she again achieved response beyond the progression, and actually she is still on treatment with very nice tolerance of the drug and very nice clinical benefit to the treatment. In slide number 19, we showed the median progression-free survival achieved in this population with five to six months of progression-free survival and with signs of a plateau that was a duration, an example of duration of response at 30%, with a median follow-up of more than nine months. Thank you for your attention.
Yeah. Thank you, Valentina, for the great overview and summary of the data from our two non-small cell lung cancer programs, and also from my side, a welcome to everybody on the call, so if we continue, what we will show you is also a very interesting data of a post-hoc exposure-response analysis that we conducted in our non-small cell lung cancer patients. As you can see on slide 21, before we go to this post-hoc analysis, let me briefly summarize what we know and what we have learned from our study so far. I think we see a quite robust and meaningful activity for the combination of atezolizumab and AFM24 in heavily pretreated non-small cell lung cancer patients. The data are very consistent across EGFR wild-type and EGFR mutant cohorts, demonstrating that the mechanism of action of our ICEs is independent of the mutational setup of the tumors.
For EGFR wild-type, we see tumor shrinkage in roughly half of the patients and objective responses in 21%. The actual PFS is 5.6 months with an option to increase over time as roughly one-third of the patients is still ongoing and receiving therapy. I think 5.6 months is already very compelling and competitive considering that standard of care docetaxel will produce progression-free survival values of only four months. In this context, it's important to note that most docetaxel data are generated in patients who are largely taxane naive. And if we look at the taxane naive subgroup in our study, we see a very compelling progression-free survival of 7.4 months. In EGFR mutant patients, we see comparable results with tumor shrinkage in 41% of patients and objective responses in 24%.
Importantly, roughly one-third of the patients remain on treatment for more than nine months, indicating a plateau of the PFS curve and long-term clinical benefit for a meaningful percentage of patients. Also important, we achieve these results with a very well-tolerated and chemotherapy-free regimen. Now, to further understand the mechanism of action and possible ways to improve on these already compelling data, we also investigated the relationship of AFM24 exposure and outcome in this post-hoc analysis. As you can see on slide 22, we collected all patients with non-small cell lung cancer who have received a dose of 480 mg of AFM24 in our various studies. For each patient, we calculated a mean exposure using weekly trough levels, and then we initially divided these populations just by the median into a low and a high exposure group.
As you can see on slide 23, the main baseline characteristics are very comparable between the low and the high exposure group. What is important on line four is that the body mass index was similar, so low exposure is not a function of higher body weight or body size. The number of lines of prior therapy seemed to disfavor the high exposure group as these patients seemed to be more heavily treated with a median line of three of prior therapies. Baseline LDH was higher in the low exposure group compared to the high exposure group. Also importantly, the low exposure was not a function of dose reduction or dose omissions as patients in both groups received roughly 90% of their planned and intended dose intensity. On slide 24, you see that high exposure was not associated with more or more severe side effects.
If you look at drug-related treatment emergent adverse events or infusion-related reactions, they are balanced between the two groups. Serious or severe treatment emergent adverse events are even more frequent in the low exposure group. And as you will see later, this is a function of a risk of very early tumor progression in patients who achieve only low exposure to AFM24, which then is associated with tumor-related complications. On slide 25, you see the outcome according to exposure group. And if you focus your attention on the left part of the slide, these are patients who received AFM24 in combination with atezolizumab. And as you can see in terms of response rate, in blue, patients with high exposure to AFM24 show an objective response rate of 46% compared to only 7.7% in the low exposure group.
Correspondingly, the disease control rate in high exposure is twice as high as in low exposure. To further understand this relationship, we did also a sensitivity analysis looking now at exposure quartiles. What you can see on slide 26 is if you go from left to right from the lowest exposure quartile to the highest exposure quartile, you see almost a linear relationship between exposure and outcome with 0% responses for patients with the lowest exposure, then escalating 10%, 20%. In the highest quartile, we see a very impressive overall response rate of 54%. On slide 27, you see the impact on low and high exposure on progression-free survival. I think these are very interesting, very compelling data.
As you can see, the fate of the patients is really decided within the first eight weeks, and low exposure is associated with a very significantly increased risk of early progression, resulting in a median progression-free survival of less than two months for patients with low exposure compared to 7.4 months for patients with high exposure. This also translates into survival as seen on slide 28, and again, if you focus your attention on the left part here, this is a combination of AFM24 with atezolizumab, and I think you would agree that the blue survival curve for a patient population with pretreated non-small cell lung cancer is extremely impressive, with almost a survival plateau at roughly 80% and a significant number of patients in follow-up already for more than half a year.
This is a type of survival curve that you are very unlikely to see with cytotoxic chemotherapy or cytotoxic ADCs, and it's consistent with our hypothesis that by activating the innate and adaptive immune system, we are inducing, by restarting the immunity cycle, extended tumor surveillance through the immune system. Now, what can we do? What are our next steps, and how can we put patients into the high exposure group? And as you recall, in our phase I study, we have also tested a dose level of 720 mg, so higher than the 480 mg. What you can see on the right side of the slide is that the higher dose of 720 mg is not associated with a higher incidence or more severe side effects. So this dose escalation, we believe, can be safely done.
And what you see in the PK curve here in green is the PK profile of 720 mg. The black horizontal line represents the cutoff of the high and low exposure populations. And as you can see, with 720 mg, even looking at the lower end o f the confidence interval, already by week two to week three, basically all patients show exposure levels that are above the cutoff between high and low exposure. And with continuation of 720 mg in cycle number two, you see that basically all patients are also above the red dotted line, which is the lower cutoff of the highest exposure quartile, which was associated with this 54% overall response rate.
On slide 30, you see the kinetics highlighting that this low and high exposure is not a function of accumulation over time, but this separation really happens very, very early, consistent with this high risk of very early progression that we have seen in patients with low exposure. As you can see in red, that is the low exposure group, and in blue is the high exposure group. These two groups separate already after the second infusion and stay separated for the full course of the first two cycles. Reassuringly, the green, this is the overlaid 720 mg curve, basically is above the high exposure group already in week two and stays above the high exposure PK profile for the whole course of the two cycles.
To summarize our findings and where we are in our non-small cell cancer program, I think we could show compelling and competitive data with meaningful clinical activity both in the EGFR wild-type and in EGFR mutant cohort in very heavily pretreated patients. We believe that the observed response rate around 20% to 25%, as well as the PFS of 5.6 months, is already quite competitive considering that standard of care docetaxel in these patients will be associated with response rates of around 10% and PFS of roughly four months. The data are more impressive even if you consider that the real comparator would be taxane-naive patients where we do see a progression-free survival of 7.4 months.
And with the data we obtained from our dose response analysis and the option to increase dosing of AFM24 in order to push patients into an exposure level that was associated with significantly better outcome, we believe that the combination of AFM24 and atezolizumab has a good likelihood to finally result in overall response rates north of 30% and PFS values of north of seven months in our ongoing and future clinical development programs. And to represent our next steps and also summarize some of the exciting data that we are able to share at ASH, I will give it back to Shawn for his concluding remarks. Shawn, please.
Great. Thanks, Andreas. 2024 has truly been a transformative year for Affimed. AFM24 continues to demonstrate very encouraging data, which is competitive with existing and emerging treatments in the non-small cell lung cancer space.
We're very enthused with what we have seen at the current dose, and we believe there's great potential to show further differentiation with the dose intensification and excluding patients with prior taxane exposure and future development. In addition to AFM24, we have demonstrated our ability to replicate the data we generated with acimtamig previously with MD Anderson with the current phase II study where we combined acimtamig with allogeneic NK cells in patients with double refractory Hodgkin lymphoma, where we've shown potential best in indication data with an 86% overall response rate and a 55% CR rate. Not only does this demonstrate enthusiasm from KOLs given the recent presentation at ASH, but we also have strong validation from the FDA with receiving the recent RMAT designation.
Lastly, AFM28 was presented as an oral presentation at ASH, demonstrating compelling clinical proof of concept with a 40% composite CR rate and a time on treatment that ranges from 2.5-6.5 months in responding patients. This is highly encouraging given that the median overall survival expectation for these patients with relapsed refractory AML is typically three to four months. In closing, I would like to thank the patients and providers for their participation in our clinical trials, and now we can open the call for questions. Operator?
Thank you. La dies and gentlemen, to ask a question, you will need to press star one one on your telephone and wait for your name to be announced. To withdraw your question, simply press star one one again. Please stand by while we compile the Q&A roster, and our first question coming from Maury Raycroft with Jefferies is now open.
Hi. Congrats on the progress, and thanks for taking my questions. Wondering what the next steps are with the 720 mg dose relative to the 480 mg dose. Are you planning to enroll more patients at the 720 mg dose? And if so, how many patients? And yeah, maybe I'll start with that question.
Sounds good. Thanks, Maury. Andreas, do you want to address Maury's question?
Yeah, sure. Yeah, Maury, so as you know, we have our 102 program study still open. And after seeing these compelling data, we have already filed an amendment, so we will be able to open an additional cohort. This cohort will use 720 mg in combination with atezolizumab. Regarding atezolizumab, we stay with the current and approved dose.
The protocol has the option to enroll in a two-stage design up to 34 patients on this intensified doublet, which is intended to confirm that the true response rate is really above 30% and corresponding PFS value. The protocol also has the option, after we have seen initial data both in activity and safety, with the 720 mg and atezolizumab to open two additional cohorts, which would be triplet cohorts. One would add ramucirumab to the combination of AFM24 and atezolizumab, and the other one would add docetaxel. Our goal here is to provide data sets that basically allow us to identify the best of these three options, either intensified doublet or one of the two triplets, to take one into a randomized controlled regulatory-intended trial against docetaxel in the course of 2027.
Got it. That's helpful. And maybe one other question, just the post-hoc analyses that you provided are really helpful and I think make sense. Just wondering how to reconcile these data with the prior PK data that you reported at ESMO 2022, where you showed comparable CD16A receptor occupancy with both the 480 mg and the 720 mg dose. Maybe just talk a little bit more about that.
Yeah, yeah. This is something that we really looked at in some detail. I think, again, CD16A receptor occupancy was a completely new field when we started it. And you're correct. If we look at circulating NK cells, we see very comparable CD16A receptor occupancy between 480 mg and 720 mg.
I think what we did not take into account sufficiently is the fact that the true anti-tumor activity, and we have seen this from our AFM13 data where we had also serial biopsies, the real anti-tumor activity is mainly driven by the tumor infiltrating and tumor residing NK cells. We believe that there is still a difference in terms of tissue penetration, in terms of the gradient between plasma and tissue you need to push a large molecule like AFM24 really into the tumor microenvironment and to fully load the tumor residing and tumor infiltrating NK cells. So I think by just looking at circulating NK cells, we may have been a little bit too optimistic and, again, underestimating tissue penetration as an additional important factor.
Got it. Understood. Okay. Thanks for taking my questions. I'll hop back in the queue.
Thanks, Maury.
Thank you. Now, next question coming from the line of Li Watsek with Cantor Fitzgerald. Your line is now open.
Hey, good morning. Thanks for taking my questions. I guess just follow up on the amendment that you just talked about. I guess how quickly can you generate the data from the 720 mg dose? And will you be looking at taxane naive patients as well? And maybe just a little bit of color in terms of partnership discussions.
Li, thanks for the question. Andreas, maybe you want to address the first question, and then I can address the question around partnership.
Yeah. So in terms of how quick we can start, as I said, we have submitted the amendment. It is in the normal, I would say, review process. We believe that if everything is going according to plan, we could start enrollment into this new cohort in the first quarter of 2025.
It will be an open label study, so we are free to take data snapshots. We believe that the first meaningful data snapshot with the doublet should be possible probably in the first half of 2026. And then depending on what we see also in terms of safety, we are free to also start the two triplets. For the triplets, we would think to see or would expect to see first data probably towards the end of 2026 or very early 2027, more end of 2026.
Yeah. And then, Lee, on the partnering question, I mean, I think as you can manage with the significant data updates that we've had at ASH with the RMAT designation for acimtamig, there has been increased interest on the partnering front.
So I mean, I think we have several meetings up and coming in the months ahead, and obviously look forward to progressing those discussions with potential partners. I mean, I think having clinical proof of concept now across all three of these programs really provides nice validation for the platform. And I think we've been able to answer a lot of the questions that potential partners were looking for us to answer. So we'll continue those dialogues in the months ahead and look forward to the potential to partner and bring these lifesaving medications to patients.
Okay. Thank you.
Thank you. Now, next question coming from the line of Daina Graybosch with Leerink Partners. The line is now open.
Hi. Thanks for the question. Just thinking ahead and following up to combo data in 2026 and 2027, I like the combination with VEGF ramucirumab, but of course, we're all looking at these VEGF bispecifics that look to have even better activity and better therapeutic index. So why ramucirumab versus going and partnering with one of the bispecifics? And also looking that far ahead, we all know that IO does better in front line. Is there an opportunity rather than taking these three years to understand combos in a refractory setting to see what you can do in a combination in front line? Thank you.
Daina, thanks for the question. Andreas, do you want to respond to Daina's question?
Yeah. So let's start with the bispecifics. I think it's an interesting class of drugs. It confirms, I think, what we know that VEGF plus PD-1 can work synergistically. We have seen it across a number of tumor types. So the question remains open whether you have to do it in one molecule or whether you get the same by just combining the PD-1 and the VEGF. So mechanistically, I think it makes sense.
Again, none of the bispecifics has yet achieved a regulatory approval or a clear pathway to registration. So I think for the time evaluating the concept whether you can increase PD-1 mediated anti-tumor activity by VEGF, I think the ramucirumab combination is a valid combination given also the fact that ramucirumab has an approval in refractory non-small cell lung cancer. The other question I would agree is first line is always an option. Currently, I think for proof of concept, you get clearer data in the refractory setting where you also have a very clear line of sight what your comparator will be.
So this is where we currently are focusing on. But in terms of lifecycle management, clearly a move into first line or upfront treatment is something that would be the next step.
Great. Thank you.
Thank you. Thank you. And our next question coming from the line of Bill Jiang with Truist Securities. The line is now open.
Hi. This is Bill on for Kripa. Thank you for taking the question and very interesting post-hoc analysis. We're wondering if you guys have any data or any way to suggest that at 720, you are saturating the circulating NK cells. And are the circulating NK cells, I guess, the limiting reagent for lack of a better phrase in this case? And could you go higher given there might be heterogeneity in patients?
Yeah. Bill, thanks for the question. Andreas, do you want to respond to his question?
Yeah. So I mean, in terms of circulating NK cells, this is what I already said. What we have seen is that we are saturating, if you will, the circulating NK cells already between 320 and 480 mg. Still, the exposure that we achieve with 480 mg seems not to be optimal for all patients. So in our opinion, it's really the tumor residing NK cells that you have to address to get to the full benefit. And these, of course, are very difficult to measure in terms of receptor occupancy. We did some PK modeling, also looking at higher doses. But according to our PK modeling, as we also showed in the curve, our assumption is that with 720 mg weekly, we would produce exposure levels that basically are consistent with the high exposure group, even a little bit higher than the whole high exposure group.
Again, this is a group that has produced a response rate of 45% and resulted in a median progression-free survival of 7.4-7.5 months. To replicate this data, I think, is very meaningful and would basically generate a completely new standard in patients with platinum and PD-1 refractory non-small cell lung cancer. Significantly higher doses have not been tested yet, and according to our PK modeling, would likely add very little to the overall efficacy. Given the exposure data that you just presented, doesn't it make sense to increase the threshold overall by adding NK cells? By adding NK cells? This is a question that we currently are focused more in the hematological malignancies where we know that due to the underlying disease, patients have simply not enough functional NK cells. We have proven this, obviously, in Hodgkin's lymphoma.
We have some PTCL data, very few though, that suggest that this might also work. I think leukemia would be a great area, AML, to test allogeneic NK cells. For the solid tumors, again, our data indicate that there is a good number of functional NK cells probably in the tumor microenvironment. So here we are currently focusing on the synergy between the adaptive and the innate immune system. And I hope it did not go without notice, but the survival curves that we see for high exposure, especially from the doublet combination, for me, is absolutely compelling. So if we can follow this lead and can reproduce such a survival curve in a larger patient population, I think that would be also paradigm-changing.
And I guess just one last one. As you follow up, and I noticed there's some patients that had PD but are still on treatment. Is that considered sort of like a maybe it might be a secondary endpoint, but is there a chance you could utilize that and keep them on treatment to push OS further and use that both PFS and OS as a primary endpoint?
This is something, and then I think Valentina already commented on one of her patients, that we really do see in some patients that patients go into something like a response, stay on treatment, then show some increase in their tumor manifestations that, according to strict readers, would be a progression. What we hear from all investigators, and also Valentina, I think, highlighted this, is patients are in excellent performance status. If they had tumor-associated tumor symptoms, usually the symptoms do not come back even with radiological enlargement.
As you have seen in the patients that Valentina has presented, this patient had two measurements where the size was corresponding to a progression, and then the patient went back into a second response again. We do see this. We allow our investigators, as we learn more and more about it, to keep patients on drug as long as patients derive clinical benefit. I think it's one of the strategies to have a larger extension of survival compared to progression-free survival. Since the progression-free survival data may underestimate the real value of this treatment for overall survival.
Thank you for taking all my questions. Sorry I asked a few. Appreciate it.
Thanks, Bill.
Thank you. Our next question coming from the line of Yanan Zhu with Wells Fargo. Your line is open.
Great. Thanks for taking our questions. So first off, I'm curious about the patients that were excluded from the efficacy analysis due to rapid progression. Could you talk about the rationale and whether that's standard practice for data analysis in second-line lung cancer?
Hey, Yanan. Thanks for the question. Andreas, do you want to respond to Yanan's question?
Yeah. Yeah. So these patients were, especially in the early phases of the trial, where it took a very long time between the starting of the screening procedure and patients actually going on treatment. So these patients were usually for several weeks without any treatment, already presenting in a deteriorated status. They may have gotten one infusion, as Valentina said, or they did not receive sufficient drug to even allow the effect of the drug to be assessed.
If you look at earlier phase II studies, especially non-controlled phase II studies, often these patients are removed from the primary analysis exactly for the fact that these patients do not allow you to assess the true value of an efficacy. If you would consider a phase III randomized trial, such patients would likely be included in the ITT population. The difference is here that in a controlled trial, these patients would basically show up in both arms and largely offset each other, which is not possible in an early phase II trial.
Got it. Got it. That makes a lot of sense. Then on the exposure analysis, I was just curious whether could there be reasons that are disease intrinsic that explains the high versus low exposure, in which case, I guess, increasing dose may not proportionally increase the exposure? Or do you think this is as simple as a numbers game? Increasing the dose could lead to increasing exposure. And if that can.
Yanan. We kindly believe that this is a real effect. But sorry, I didn't—just repeat your question and then I'll give the answer.
Oh, okay. Sorry. Yeah. I was also wondering what if you analyze the data by looking at a total exposure to see if that has—for example, that's more uniform and doesn't show a difference between high and low exposures.
Yeah. So let's start with AFM24. We believe that this is a real relationship. This was also the reason why we did the quartile analysis. Again, if you see this really linear relationship, basically exposure and efficacy goes up over several subgroups. I think that's a very strong indicator that this relationship is real.
What we have not shown here, but what I can tell you, what we have also done, we have also done the same analysis across the whole AFM24 program, so this includes now patients with gastric carcinoma, patients with hepatocellular carcinoma, renal cell cancer, colorectal cancer, and if we do this for the whole AFM24 program, we see exactly the same. We also see that exposure independent of underlying disease is associated with better outcomes, so I think that's a very strong supporting argument that the exposure really matters. We cannot comment on PD-1 exposure. Again, we have no PK data for atezolizumab.
As far as I recall, the early PD-1, PD-L1 studies, which in part use significantly higher doses than the doses that are used now, did not provide any evidence that going over a certain threshold with a PD-1 or PD-L1 inhibitor, that you are able to increase efficacy further. I think some of the earlier PD-1 approvals for pembrolizumab, in fact, were at higher doses, and then the recommended dose was reduced because there was no evidence that going over the threshold dose with PD-1, you will increase efficacy.
Great. Thanks for all the color.
Thank you. And our next question coming from Yi Chen with H.C. Wainwright. Your line is now open.
Hello, everyone. This is Yi Chen. My first question is about the safety management of safety profile of AFM24. So specifically, what's the prophylactic treatment across the centers? How uniform is the treatment done across those centers? I'm asking this question because I was wondering if over-treatment of the premed would negatively impact on efficacy of you. Given that you have still pretty good room on the safety side, would you be willing to sacrifice the safety a little bit in the 720 cohorts in the future?
Hey, Yi. Thanks for your question. Andreas, do you want to respond to Lee's question?
Yeah. So the treatment, the supportive treatment, is relatively uniform across sites. We have mandatory steroid premedication with around 12 mg of dexamethasone and some antihistaminic premedication. This is routinely done for the first probably three to four infusions.
After that, patient pretreatment is individualized, and physicians, if there's no infusion-related reaction or only Grade 1 infusion-related reaction, can taper and do taper, in fact, quite a bit the steroid premedication, and they can also reduce the infusion duration. So if we look at patients who are on treatment for a couple of months, most of them are on an outpatient one-hour easy infusion. And as Valentina said, the patient in her site is now on treatment for over a year with excellent life quality and just getting her short one-hour infusion. The only side effect that in our hands was really associated with not really with the absolute dose, but with the dose increase over time initially was infusion-related reaction.
What we have already shown at our ASCO presentation when we presented the 101, the phase I study, 720 mg for the first two infusions we'd give as a so-called split-day infusion. So the first two infusions will be basically half of the dose on day one, the other half of the dose on day two. And this is repeated in the second week. And then after two weeks, we go also with 720 mg to the one-day infusion. Again, the risk of a more severe Grade 3 IRR is over after two infusions. We have rarely seen because we have seen one patient with a Grade 3 IRR at later infusions. But interestingly, it was a patient who had a four-week treatment gap, a treatment interval, where the levels are basically returned back to zero. And then if you increase, you have again a risk of an IRR.
But other than that, the 720 and 480 mg were very comparable in terms of safety, again, the only difference being that the first two infusions with 720 mg are given over two days.
So will that regimen also be used in the upcoming 720 cohort?
Yeah. Yeah. We will implement exactly the same. So the first and the second infusion will be split days, so over two days. And then if there's no severe IRR beginning from the third infusion, these infusions will be given at one day as we currently do.
Okay. Thanks. I'll continue my question.
Thanks, Yi.
Thank you. And our next question coming from the line of Bradley Canino with Stifel. Your line is now open.
Hi. Thanks for this update. This is Dara Azar on for Brad. Maybe I could ask the efficacy denominator question in the context of contribution of the parts. What kind of efficacy would you expect from an atezo in this setting after applying the same denominator exclusion criteria? As follow-up, have you done any work to remove confounding effects? You already addressed Taxol , but from prior taxane experience aside from the high AFM24 exposure group in the post-hoc analysis work that you've done. Thank you.
Hey, Dara Azar. Thanks for the question. Andreas, do you want to respond to his question?
Yeah. So atezolizumab retreatment, again, depends a little bit on the context. Interestingly, only the response rate is really impacted by the level of resistance. If you take CPI-resistant patients, atezolizumab retreatment results in roughly a 3%-4% response rate. If you go to the very other end of the extreme, so patients who have had a good response to PD-1, the response rate would be around 10%.
Interestingly, no matter PFS with atezolizumab is only around 2.4-2.5 months quite consistently in all of the re-challenge studies. So especially if we look at PFS, we would not expect to see any meaningful contribution from atezolizumab independent of the pretreatment characteristics and also independent of the denominator question. As I stated, absolutely consistent. The second question was—what was the second question? Sorry.
Yeah. I can ask you again.
The answer is the second question.
Asking about prior taxane effect on your exposure work that you've done in post-hoc .
Yeah. We looked both at prior taxane in the main analysis, high and low exposure, as well as in the quartile subgroups, and prior taxane exposure is equally distributed across all of the exposure groups.
Okay. Thank you.
Thanks, Dara Azar.
Thank you. And I'm showing there are no further questions in the Q&A queue at this time. Ladies and gentlemen, thank you for participating in today's conference. This does conclude your program, and you may now disconnect.