Good morning to our stakeholders dialing in from the U.S., and good evening to our stakeholders dialing in from China, and good afternoon to anyone dialing in from Europe. I'd like to thank you all for standing by, and I'd like to welcome you all to the I-Mab Biopharma discussion highlighting the phase II data of lemzoparlimab in combination with azacitidine in patients with higher-risk myelodysplastic syndrome that we recently presented at ASCO. Today's session will be followed by a brief Q&A session. The press release can also be accessed on the investor portion of our website. Joining me on today's call from I-Mab senior management team are Dr. Andrew Zhu, President of I-Mab, and Dr. John Hayslip, I-Mab's Chief Medical Officer.
Please note that today's discussion will contain forward-looking statements relating to the company's future performance and are intended to qualify for the safe harbor from liability as established by the U.S. Private Securities Litigation Reform Act. Such statements are not guarantees of future performance and are subject to certain risks and uncertainties, assumptions, and other factors. Some of these risks are beyond the company's control and could cause actual results to differ materially from those mentioned in today's press release and this discussion. A general discussion of the risk factors that could affect I-Mab's business and financial results is included in certain filings of the company within the Securities and Exchange Commission. Company does not undertake any obligation to update this forward-looking information, except as required by law. With that, I'll now turn the call over to Dr. Andrew Zhu, our President. Please go ahead, Andrew.
Thanks, Tyler. Good morning to those who are joining us from the States, and also good evening to those from Asia. As all of you know, I-Mab is taking the pride in developing innovative assets of best-in-class and first-in-class potential in immuno-oncology field. We have 20 assets currently in development and 10 in the clinical stage. Today, we will focus on our discussion on our highly differentiated CD47 antibody, lemzoparlimab. Shown on the left side, you can appreciate, lemzo is uniquely designed by differentiation. Based on the phage display library, we screened and eventually developed lemzo molecule to minimize the on-target hematological side effects through reduced binding to RBCs while maintaining strong antitumor activity. The underlying mechanism of RBC-sparing property is attributed by the recognition of a unique glycosyl epitope of CD47 that's shielded by glycosylation on RBC.
The unique glycosylation around the binding epitope of lemzoparlimab serves as a natural barrier to prevent lenzo from engaging RBCs. By contrast, the binding site on tumor cells does not have similar glycosylation and is fully exposed, which leads to a strong binding of lenzo to tumor cells. That's why, you know, lenzo can uniquely distinguish tumor cells from RBCs to avoid the severe anemia that's commonly seen with other anti-CD47 monoclonal antibodies while retaining strong antitumor activity. This molecular differentiation has been validated preclinically, shown on the right side. In the preclinical models, we have demonstrated a decreased RBC binding and potent antitumor activity in comparison with other molecules in this class. Over the past few years, we have embarked on a very comprehensive clinical development program across various tumor types and also with different combinations.
Of approximately 200 patients who were treated with lemzo, either as monotherapy or in various combination, we have seen a compelling safety profile up to now. At the ESMO meeting on September 10th, we, for the first time, presented our data with lemzo in combination with azacitidine in higher-risk MDS patients. Today, we are very happy to take this opportunity to share the results in details with you and also, to answer some questions you may have. Without further ado, I now want to turn to Dr. John Hayslip, the CMO at I-Mab, to present the data. John, please.
Thank you, Andrew. Prior to the data we will review today, we tested lemzoparlimab as a single therapy to ensure that it was safe for patients with MDS or AML. We tested doses from 5 to 30 mg per kg, and upon completing this portion of the study, we conducted a safety lead-in with azacitidine combination because azacitidine is a standard therapy for these patients, and preclinical evidence suggests synergy between azacitidine and CD47-directed therapy. Today, we'll be reviewing data from cohort two, the MDS phase II cohort. The baseline characteristics of these patients make it clear that our patients indeed have severe disease at baseline. 79% of the patients had high or very high international prognostic scores for myelodysplastic syndrome. 87% of the patients had excessive blasts, and 55% of the patients were dependent upon transfusions when they entered the study.
74% of the patients had severe, that is grade 3 or 4 anemia, and 51% of the patients had severe thrombocytopenia. As you can see from the graph on the left, most patients remain on study at the time of the analysis, as indicated by the purple triangles on the right side of the plot lines. The median follow-up time is 3.7 months. The median time to first response is 1 month, and the median time to complete remission is 2.7 months. Those CR rates continue to improve even out to 5-6 months. Twenty-seven of the 33 responding patients remain on treatment, and the longest duration of response is about 10 months. As you can see on the graph on the right, most patients who have received combination treatment have improved their bone marrow blast count percentage.
Indeed, only 4 out of 57 patients have had blast count worsening as their best response. The interim efficacy results are encouraging and demonstrate that responses deepen over time on therapy. As noted on the previous slide, the median time to first response is faster, and the overall response rate at months three, four, and six are 80.6%, 86.2%, and 86.7%. The complete remission rates vary more over these time points because the median time to complete remission is 2.7 months, with the observed CR rates of 27.8%, 31%, and 40% at the time points of three, four, and six months. Shown here is how anemia and thrombocytopenia from the disease improve with treatment. Low blood counts is a hallmark feature of myelodysplastic syndromes and a common reason that the diagnosis is made and treatment initiated.
As you can see in the upper panel, at baseline, the average hemoglobin is just over 7 gm of hemoglobin per deciliter. As patients remain on study, the hemoglobins improve such that by month 6 or 7, the hemoglobins are approximately 12 gm per deciliter on average, and this is within the normal range for hemoglobin. To the right, you can see the red blood cell transfusion burden on average for patients over time. At the beginning of therapy within the first month, because many patients begin treatment being anemic, the average red blood cell transfusion is over 5 units per patient per month. As patients remain on therapy over time, this decreases such that by month five or six, on average, patients require less than one unit of red blood cell per month. Platelet counts are shown just below and follow similar trends.
At study initiation, the mean platelet count was approximately 90,000. As patients remain on therapy, the mean platelet counts improve over time, so that by month 6 or 7, the average count is between 150,000 and 200,000, which is within the normal range for platelets. Similarly, as early as patients enter treatment, average patient requires nearly 4 units of platelet transfusions per month. As patients remain on treatment over time, this transfusion burden has decreased to less than, on average, one unit of platelet transfusion per month. Along the same line of logic, approximately a third of the patients who were transfusion-dependent for either red blood or platelets when they entered the study have become completely independent of the need of transfusions while receiving treatment.
Although the specific genes involved in MDS can vary from patient to patient, we conducted analyses to look for MDS-associated changes and potential response on treatment. As you can see in the left panel, most gene mutations in the hematopoietic stem cells in the bone marrow decreased with treatment. On the right, we can see the time course for substantial reduction between months 4 and 6. Remission have also achieved minimal residual disease negativity. Lemzoparlimab can be safely administered in combination with azacitidine for patients with higher risk MDS and does not require priming dosing. Infusion-related reactions, a potential risk with antibody therapies, were reported in 5 patients. All were grade 1 or 2 and resolved with standard supportive care. Treatment emergent adverse events leading to treatment discontinuation occurred in 6 patients, and grade 5 treatment emergent adverse events occurred in 3 patients.
Specifically, those were one case each of pneumonia, acute coronary syndrome, often referred to as heart attack, and metabolic acidosis. To put the observed rates of anemia, thrombocytopenia, and neutropenia in context, here we have listed the rates observed in this study and also rates reported from a separate study. In this external study, patients with higher risk MDS treated in China were treated with the standard doses of azacitidine, the same doses used in this study. As you can see, we observed similar to modestly lower rates of anemia, thrombocytopenia, and neutropenia from what has recently been reported for patients receiving azacitidine monotherapy in the Chinese population. In conclusion, lemzoparlimab is a novel CD47 antibody designed to minimize red blood cell binding by targeting a unique glycosylation specific binding site.
Patients enrolled three months or more before the analysis, the overall response rate is 80.6%. Among those enrolled six months or earlier, the overall response rate is 86.7% and CR rate of 40%, and follow-up for all patients remains ongoing. Lemzoparlimab does not require priming dosing, and no new safety signals are observed in combination with azacitidine. Lemzoparlimab has now been dosed to over 200 patients across the program. A randomized phase III trial for patients with higher risk MDS in China is planned. With that, I thank you for your attention and turn this back over to Tyler.
Thank you, Dr. Hayslip, for going through those results and thank you Dr. Zhu for your introduction. With that, we'll turn it over to the Q&A session. I see a few hands raised. The first question will come from Kelly Shi at Jefferies. Kelly, please go ahead.
Thank you, Tyler, and congrats on the progress. I have two questions. First is, so if the drug action of this combination is rapid, why should we think the clinical outcome will continue to improve over a long time for up to 4-6 months after starting the treatment? The second question is, could you elaborate a little bit on why the patient baseline in terms of anemia and thrombocytopenia from China trial looks worse than the U.S. trial? And how should we assess the safety profile for them, so relative to other CD47 antibodies? Thank you.
Thank you, Kelly. Dr. Hayslip, do you wanna go ahead?
Yeah, glad to do that, Tyler, and thank you for those two questions. Those are both very good. To answer the first question, the median time to the first response is one cycle, and the median time to complete remission is about 2.7 months. When I say median, what I mean to say is that at least half the patients may respond at that time or more quickly, and half of patients may respond at that time or more slowly. Because the median time is one month for the first response, we see a rather modest improvement in the overall response between months three to four to six. It's fairly stable even by month three in terms of overall response rate.
We do see that oftentimes these responses deepen over time, and then can be a complete remission. Whereas the median is about 2.7, we pick up about half of those patients who are gonna achieve a complete remission by month three. We continue to see some additional patients improve their response between months 3-6. So far in the program, most of the patients who do achieve CR, if they are to achieve CR, do that by month six or before. I think that's our understanding of why we continue to see those numbers change and also the kinetics of how we're seeing them change over time on treatment. I think regarding your second question, this'll be a pretty speculative answer, because I'm talking now about across-trial comparisons.
I think it would be fair to say that oftentimes in the United States we can see higher usage of transfusions. In some other parts of the world and in China in particular, sometimes the uses of transfusions is not so low. Sometimes, in other words, patients can be considered to be safe even if their blood counts are running a little lower. Therefore, they may not receive the transfusions as early in the course of the situation. Of course, if a patient receives a transfusion, the numbers, the hemoglobin number, the platelet number, it will read higher on the test.
You can see differences in standards, differences between standard practice that can then lead to some differences in these numerical countings. This is all speculation. I mean, these are across trial comparisons. I think what we can say most assuredly is, you know, the results that we have before us. Yeah. Thank you.
Thank you, Dr. Hayslip. Thank you, Kelly, for your question. Next, we'll go with Louise Chen from Cantor. Louise, please go ahead. Okay. We'll go to the next question in that case. Next question, I see a hand raised from Zhong Ping from Huatai. Zhong Ping, go ahead.
Thank you. With some more details about the data of those, TP53 mutated patients. Thank you.
Dr. Hayslip, if you'd like to go ahead.
Yep, certainly. Glad to share about that. I should preface my comment or the data by noting that some of the patients with TP53 mutation enrolled fairly close to the analysis. In other words, they may not have all yet achieved their best response. Having said that, we've enrolled 7 patients with TP53 mutation to this group, and amongst those 7 patients, 2 patients have achieved a complete remission, and 2 other patients have achieved a marrow complete remission. Marrow CR refers to that state where you've cleared all the bone marrow blasts from the bone marrow space, but the blood counts may not have completely normalized.
In myelodysplastic syndrome, not only must we remove all of the dysplasia in order to have a CR, we also have to have normalization of the blood counts. Two patients with CR and two patients with marrow CR. The duration of response for the two patients of CR is fairly impressive. One of the patients is the patient with 10 months duration of response, and the other has been seven months at the time of the analysis. As I mentioned, some of the others were enrolled fairly close to the analysis, and so we'll be updating that in the future.
Great. Thank you, Dr. John Hayslip. I see, Louise has unmuted herself. Louise from Cantor, do you wanna go ahead?
Yes. Thank you. Can you hear me okay now?
Yep.
Okay. Sorry about that. Congratulations on the data, and thanks for letting me ask a question here. I wanted to ask you how you feel about these results compared to the results of magrolimab plus azacitidine in MDS that was presented at ASCO in 2022. Thank you.
Thank you for that question, Louise. Dr. Hayslip, can I turn that one over to you?
Yeah, certainly. I would say I feel encouraged. I think we should acknowledge, kind of like one of the earlier questions. I'm now making an across trial comparison, which will always lead to an interpretation being imprecise. That said, I see the results as being generally comparable with a few key differences. In both reports, most patients achieve a response, and I think that's really impressive, compared to what we had with azacitidine monotherapy. Approximately it's 80%-90%, and just about less than half of those achieve CRs. I believe that patients in both combinations appear to be doing better than I would have otherwise expected for azacitidine alone. Also, both combinations have reported improving blood counts, and reduced mutational burdens, after treatment.
I think one of the key differences that I see is that lemzoparlimab does not require the priming dosing. Also because we do not expect the relevant hemolytic anemia with lemzoparlimab, which and this is speculation, but we may find that the patients may not require such frequent monitoring and support with lemzoparlimab. Thanks.
Thank you, Dr. Hayslip. Thank you, Louise, again for your question. Next we'll have Joe Catanzaro from Piper Sandler. Joe, would you like to go ahead?
Hey, thanks, guys. Hopefully you could hear me clearly. Thanks for the update here. Maybe first one. There were 47 patients in the evaluable analysis set, but just 36 in the analysis of patients with at least three months of treatment. Can probably figure this out maybe from the Swimmer plot, but can you say for those 11 patients that are excluded, how many are still on study and have the opportunity to reach that 3-month time point with longer follow-up? Then second question, maybe related to one earlier, maybe beyond TP53, just looking at all relevant mutations, whether there was any correlation between a specific gene mutation and clinical response and whether that could be considered for the design of the planned phase III. Thanks.
Thank you for that question, Joe. I'll actually call on Dr. Zhu to take this one. Dr. Zhu, would you go ahead?
Yeah. Joe, thank you for the question. Maybe actually I will start, you know, with the second part of your question. Clearly, measuring molecular, you know, defined minimal residual disease is commonly practiced in oncology. You know, in some tumor types, we actually have very strong data that the negative MRD actually correlates with the clinical outcome. In the case of MDS, I have to emphasize, the current standard practice is still looking at the clinical remission, you know, with clinical features, as well as bone marrow aspirate assessment, looking for the disappearance of the blast. Having said that, looking at potential molecular mutations associated with MDS pathogenesis is also considered to be potentially valuable and definitely there are studies looking at this concept.
In our study, just to address your question directly, we have also attempted to assess the gene mutation frequency of the paired pre and post-treatment bone marrow aspirates. We have, you know, nine patients that meets this criteria. We use the next-gen sequencing to assess the various mutation frequency, and the results definitely demonstrate mutation frequency of several well-characterized MDS prognostic genes. You know, such as TP53 that we just mentioned, along with a few other genes like RUNX1, TET2, and those were also significantly decreased after the treatment with the variant allele frequency less than 5%. We are continuing this effort.
You know, currently, as the data cutoff date, we have actually, you know, analyzed the MDS-related driver gene mutations in bone marrow aspirates from a total of 47 MDS patients with at least one assessment of clinical response. We're in the process of trying to really assess the potential correlation of these mutation changes with clinical efficacy. Hopefully, you know, we'll share the results in the future. Now, with regard to your first question, you know, you have to really appreciate that, you know, our trial is still ongoing, and also we're actively, you know, obviously monitoring the outcome of these patients.
At the data cutoff, you know, we use the best judgment, and also the best strategy, in a very unbiased way to assess, if patients receiving, longer treatment duration, whether that will have any impact on the overall response rate, including, you know, the complete remission rate. So we actually assess patient who started the treatment at various time points, from the analysis, and that will give us an unbiased way, to assess, you know, those patients, you know, at different landmark time points, what's their, response rate. For that reason, our analysis is actually an unbiased approach. And also our analysis really dictates at certain points whether a patient finished the treatment or not finished treatment. We include them in the analysis, you know, in an unbiased way.
John, I don't know whether you have anything else to add.
Yeah. Thank you, Andrew. I would just emphasize, as you were saying, the results which the authors have presented are not those who actually received all three, four, or six months, but it's patients who were enrolled that amount of time or earlier. As you know, this is an unbiased intent-to-treat way to handle the data. Specifically to I think it was Joe who asked the question, but specifically to the question that was asked, I was quickly looking because we want to answer you know, exactly factually accurate. I don't believe I have the exact answer to your question, which is, of those 11 patients, what's the. I don't have it calculated for each one. I can say that I meet with the team regularly.
We're not aware of any change in the overall course as the study has progressed, but I could not give you at this moment the detail on those 11 specific.
Thank you, Dr. Hayslip, and thank you, Dr. Zhu. I see that we're at the 830 mark, so we'll end the Q&A session there. If there are any follow-up questions, please feel free to reach out to your local IR representative. Thank you again for everyone who's dialed in. Thank you again to Dr. Hayslip and Dr. Zhu for the detailed presentation as well as going through the Q&A session. Yeah, hope everyone has a nice day, and please feel free to reach out for any questions. Thank you.
Thanks.