Good morning, everyone, and thank you for dialing in today. Yesterday, after market hours, we published a press release sharing our updated phase 2 clinical data with a data cutoff date of March 29, 2022. In today's discussion, we'll go through uliledlimab's key differentiating factors, the most updated phase 2 clinical data, as well as our global clinical development plan. With that, I'll hand it over to Dr. Zhu.
Thank you, Kyler. Thank you all for joining this call, particularly, you know, with ASCO just approaching us next week. During today's meeting, we would like to share with you the latest encouraging phase 2 clinical data of our differentiated CD73 antibody, uliledlimab. In early 2022, we submitted an abstract of uliledlimab initial phase 2 data to ASCO, with a data cutoff date of December 13, 2021. Since then, we're very pleased that the data has matured significantly. Therefore, after the ASCO embargo, we would like to take this opportunity to provide you with the latest clinical data, with March 29 as a new data cutoff date, and also share with you our future clinical development plan. Uliledlimab is an internally discovered differentiated CD73 neutralizing antibody.
I want to highlight a few key features here. Uliledlimab is not only globally competitive, just behind AstraZeneca's oleclumab, but also has the potential to become the best-in-class CD73 antibody. It is designed to avoid the abnormal PK/PD relationship commonly seen in other CD73 antibodies, also known as the Hook effect. What is Hook effect? Simply put, the Hook effect is characterized by an abnormal phenomenon that the drug molecule paradoxically loses effect at higher doses. Uliledlimab's differentiation comes from a unique binding epitope at the C-terminus, while oleclumab binds to an epitope at the N-terminus. Uliledlimab works through an intra-dimer monovalent binding mode, while oleclumab inhibits CD73 through an intra-dimer bivalent binding mode.
For uliledlimab, this actually leads to complete inhibition of CD73 with no Hook effect, and also for oleclumab, a partial inhibition of CD73 with Hook effect at higher doses. As a result, in the clinic, uliledlimab is used at 20 mg per kg every 3 weeks, while oleclumab was given at high dose of 3,000 milligrams every 2 weeks for the first 2 cycles, followed by every 4 weeks. We believe this differentiation actually gives uliledlimab a better therapeutic window and more flexibility when combined with other anti-tumor drugs. In addition, uliledlimab also has the non-competitive inhibitory effect that is not blunted by high levels of CD73 enzyme substrate abnormally accumulated in the tumor microenvironment.
The AMP concentration can be as high as 500 micromolar in these settings, which could be expected for small molecule competitive blockers. For this reason, we believe uliledlimab has potential advantage over small molecules as well. Next slide. We have carried out extensive research on the underlying mechanism as to how uliledlimab works. As we reported during the AACR 2021 meeting, this slide demonstrate the detailed mechanism of action and molecular differentiation of uliledlimab. The crystal structure shown on the left side demonstrate that unlike oleclumab, which binds to the N-terminus of CD73 molecule, uliledlimab interacts with a unique epitope located at the C-terminus of the CD73 dimer.
Because of this unique epitope, uliledlimab works through an intra-dimer binding mode, and that is one antibody molecule of uliledlimab binds to one CD73 molecule to prevent the conformational change of CD73 from inactive to the active form, resulting in the complete inhibition of CD73 enzymatic activity. This working mode does not cause the Hook effect. In contrast, oleclumab binds to the N-terminus of CD73 and inhibits CD73 through an intra-dimer binding mode. That is one antibody needs to bind to two CD73 molecules to prevent the conformational change of CD73 from inactive to the active form. This intradimer binding is associated with the Hook effect. How does this molecular differentiation translate into a functional advantage?
As shown in this slide, uliledlimab can achieve complete inhibition of soluble CD73 enzymatic function without causing the hook effect shown on the left side with a blue line. While oleclumab has the paradoxical loss of enzymatic activity at higher concentration, shown on the right, and thus resulting in partial inhibition of CD73 function, shown in the right line. In terms of the antitumor activity, consistent with the in vitro results, when tested on various tumor models, uliledlimab is found to have an enhanced antitumor activity when combined with a PD-1 antibody on the left panel or a PD-L1 antibody on the right panel. The right line is the combo and has the best antitumor effect as opposed to monotherapies.
The experimental evidence does suggest that uliledlimab in a combo treatment with either PD-1 or PD-L1 therapy works synergistically and has more potential inhibition of tumor growth than monotherapy, providing the rationale of our current clinical development. We're now in the process of testing uliledlimab in other combinations, including small molecule antitumor agents, to potentially expand the clinical application of uliledlimab.
With that, I think you know uliledlimab's differentiated antibody design, as well as our latest clinical data, support the key advantage of uliledlimab that allows for its best-in-class potential for the following reasons. First, it's really due to uliledlimab's differentiation. It binds to a novel epitope on the C-terminus through a unique intradimer monovalent binding model, leading complete CD73 inhibition without a hook effect. Secondly, uliledlimab is in a very strong competitive position, both globally and in China.
As you know, globally, uliledlimab is right behind oleclumab in clinical development effort, and it is very close. While in China, uliledlimab is currently the most advanced CD73 antibody currently in development. Thirdly, we have observed very encouraging clinical data. Uliledlimab has demonstrated a favorable safety profile, and we have the clinical experience with more than 100 patient data with the safety profile. The safety profile really allows uliledlimab to be easily combined, with additional agents, you know, for better tolerability. The RP2D was determined at 20 mg/kg every three weeks, and, you know, it has the expected PK/PD relationship and is validated, in the clinical setting without a hook effect.
In both our U.S. phase I and China phase II trials, we have observed encouraging efficacy signals as well as a potential correlation between higher CD73 expression at baseline and clinical activity of uliledlimab, indicating tumor CD73 expression could serve a potential predictive biomarker. In particular, in the phase II trial, we observed encouraging efficacy signals in advanced, mostly stage IV non-small cell lung cancer, PD-L1 low or negative population, and previously untreated non-small cell lung cancer patients. Our CMO, Dr. John Hayslip, will share with you more data and his insight with you in a few moments during today's presentation. Lastly, in terms of the future clinical development efforts, I think we're actually focusing on the following three strategies. First, we will continue our phase II trial in both U.S. and in China.
In particular, we're going to expand our clinical experience in specific non-small cell lung cancer treatment-naive cohort with encouraging efficacy signal to further evaluate the efficacy and also to continue to assess the biomarker and correlation between CD73 expression and clinical response. Based on our current data and the expansion of cohort 3, we are advancing our uliledlimab clinical development towards a phase III clinical registration trial in selected non-small cell lung cancer patients in 2023 in China, if approved by the NMPA. Meanwhile, we plan to initiate a new phase 2 clinical trial in the States in other selected cancer types and beyond the combination with a checkpoint inhibitor therapy at some point in the second half of 2022.
Also, with the phase three clinical trial in plan, we're actively pursuing the development of a companion diagnostic kit of detecting CD73 expression. Hopefully, this will be in place as we are ready to launch our phase three trial. With that, I'm now going to invite our CMO, Dr. John Hayslip, to go through uliledlimab's clinical data as well as our current development plan. John?
Thank you. Hello, everyone. Here we would like to highlight our completed and ongoing uliledlimab clinical trials. In a completed phase 1 study of uliledlimab in combination with atezolizumab, treatment for patients with resistant or recurrent solid tumors. Among 13 evaluable patients, the response rate was 23%, and the disease control rate was 46%. The results, though preliminary, are very encouraging, and I will review more on this in a moment. Our phase 2 trial in China in combination with toripalimab, a PD-1 antibody, has 5 dose expansion cohorts at the presumed efficacious dose, with dose expansion in non-small cell lung cancer and selected tumor types. The primary focus of this trial is on different groups of NSCLC patients, and we will share with you our latest clinical data in a moment.
Our phase 2 trial in the US focused on ovarian cancer, as well as biomarker-enriched solid tumor arms, including NSCLC, head and neck squamous cell cancer, gastric cancer, and triple-negative breast cancer in combination with atezolizumab, a PD-L1 antibody, and saw the first patient dosed in December 2021. This trial is ongoing. We believe that uliledlimab has the potential to become the best-in-class CD73 antibody in the global market and first-in-class CD73 antibody in the China market. This slide summarizes our US phase 1 data, which was presented at ASCO last year.
The US phase 1 dose escalation study of uliledlimab in combination with atezo showed that treatment is safe and well-tolerated with no dose-limiting toxicities observed. All treatment-related adverse events were either grade 1 or grade 2. All patients who participated in the study had advanced cancers and had exhausted other cancer therapies.
Uliledlimab demonstrated a linear PK profile at doses of 5 mg/kg and greater, indicating target saturation. 20 mg/kg every three weeks uliledlimab in combination with atezo reached a C_trough concentration of 50 μg/ml. A population PK analysis indicates no difference between patients in China and the U.S., with an effective half-life of approximately 13 days. In pharmacodynamic analysis, we observed a decrease of free soluble CD73 in serum and an increase of CD73 receptor occupancy in a dose-dependent manner with no hook effect.
Complete CD73 inhibition and receptor occupancy saturation were achieved at 20 mg/kg given every three weeks. Given the challenges that other CD73 targeting therapies have faced, we view having overcome these key obstacles in CD73 development as key validation of our clinical differentiation. Our recommended phase 2 dose was determined at 20 mg/kg every three weeks.
Among the 13 efficacy-evaluable patients dosed at 10 mg/kg or greater, three patients had complete or partial response for an objective response rate of 23%, and three patients had stable disease for a disease control rate of 46%. The clinical activity was observed in both checkpoint inhibitor treatment naïve and refractory patients, as well as cold tumors, for example, ovarian cancer. One patient with ovarian cancer achieved complete response and as of March this year has remained on study for over 21 months.
Two patients with NSCLC achieved partial response, including one patient who had previously been failed by nivolumab. The three responders were identified to exhibit higher co-expression of CD73 and PD-L1 as compared to non-responders, indicating a correlation between higher CD73 expression and clinical activity of uliledlimab and a potential role of CD73 as a potential predictive biomarker warranting further investigation.
We highlight the ongoing phase 2 clinical trial in China of uliledlimab in combination with toripalimab, a PD-1 antibody developed by Junshi in China, the first approved PD-1 on the market in China developed by a Chinese company. Uliledlimab was administered with an entry dose of 2 milligrams per kilogram and moved up both in monotherapy escalation to 30 milligrams per kilogram every 3 weeks and in combination with tori to 30 milligrams per kilogram every 3 weeks. In dose expansion at the recommended phase 2 dose, that is 20 mg/kg of Uli every 3 weeks, there are 5 cohorts of patients. Cohort 1 patients with NSCLC where standard therapies had failed to control their cancer and the previous regimens did contain a checkpoint inhibitor.
Cohort 2 patients with NSCLC where standard therapies failed to control their cancer and the previous regimens did contain a checkpoint inhibitor. Cohort 3, patients with NSCLC who were unsuitable or had declined other standard therapies. Cohort 4 were other solid tumors other than NSCLC, and cohort 5, platinum-resistant ovarian cancer, tubal cancer, and female primary peritoneal carcinoma. In these 5 cohorts, the NSCLC patients have the most mature data. For today's discussion, with compelling efficacy signals seen in cohort 3, I'll focus my remarks on this cohort. In the phase 2 trial in China, 115 patients have been dosed as of the cutoff date, March of this year. 12 patients received uliledlimab monotherapy, and 103 patients received uliledlimab in combination with toripalimab.
Following a thorough safety review, uliledlimab appears safe and well-tolerated up to 30 mg/kg as a monotherapy and in combination with PD-1 and PD-L1 antibodies. We are pleased to note that a maximal tolerated dose was not reached. No dose-limiting toxicities have been observed, and we were able to safely escalate even beyond the dose our earlier studies indicated would be the optimally effective dose. Most treatment-related adverse events were grade one or two, and no new safety issues have been identified at this time. As I previously mentioned, our phase two trial in China primarily focuses on different groups of NSCLC patients.
The trial was designed to evaluate efficacy signals in patients with NSCLC who required additional treatment after standard chemotherapy alone or in combination with checkpoint inhibitors, cohorts 1 and 2, and cohort 3, who had advanced NSCLC and were either ineligible or declined standard chemotherapy treatment. With the data cutoff of March 29th this year, encouraging efficacy signals are observed in cohort 3, advanced non-small cell lung cancer patients who are either unsuitable for or decline standard chemotherapy treatments. In this cohort, most of the patients have stage 4 non-small cell lung cancer. About 80% of the patients in this cohort have low PD-L1 expression in their baseline tumor samples. That is a tumor proportion score of 1%-49% or negative with TPS less than 1%. These patients are generally considered less responsive to PD-1 therapy.
For example, in the KEYNOTE-042 study, patients with low TPS, that was 1%-49%, and patients with negative TPS appeared even less likely to benefit. Of 19 efficacy-evaluable patients in our study, the response rate to date is 26%, and the disease control rate, with 5 patients achieving PR and 9 patients achieving stable disease, is 74%. This data readout, although preliminary, shows early signals of promising efficacy for uliledlimab in combination with a PD-1 inhibitor in patients with PD-L1-low or negative-expressing tumors, which gives hope to these lung cancer patients who often do not benefit substantially from currently available immunotherapies. These data may represent a step forward for advanced cancer patients who previously had limited treatment options.
Cohorts 1 and 2 patients who had received previous therapies displayed a lower clinical response rate to date, and these patients continue to be treated, and efficacy and safety continue to be evaluated as the study is ongoing. In this cohort 3, we've also observed a compelling correlation between CD73 expression in the tumor samples and clinical responses. Expression of CD73 and PD-L1 were measured in baseline tumor specimens by immunohistochemistry. Overall, high expression, defined as more than 35% expression, was measured in 8 tumor specimens from patients who had either partial response or stable disease. 80% of the patients who had a partial response exhibited high expression of CD73. In comparison, 44% of stable disease patients had high CD73 expression. All of the 5 patients with progressive disease as best response had CD73 low expressing tumors.
As we saw in the earlier phase 1 experience, this points to a compelling case for a correlation between CD73 expression and clinical response. This table summarizes information from across two different studies, one with uliledlimab and another with oleclumab, a separate CD73 targeting antibody undergoing studies and other trials, to highlight a few of the reasons we are optimistic about uliledlimab at this stage in development. In terms of dose, the RP2D, the recommended phase 2 dose of uliledlimab is 20 mg per kg every 3 weeks, a relatively low volume treatment with a simple administration pattern. Whereas in oleclumab trials, they've tested 3,000 mg for the first 2 cycles every 2 weeks, followed by 3,000 mg every 4 weeks. With regard to the pharmacodynamic pharmacokinetic relationship, uliledlimab data showed the expected features based on the design, as Dr.
Zhu reviewed, and achieved full receptor occupancy saturation at doses of 5 mg and greater without the hook effect, which has caused challenges for other drugs in development. We have ongoing efforts to compare tumor biopsy samples from before and after treatment to further understand how this impacts patients. For oleclumab, as you recall, they reported incomplete CD73 inhibition in tumors at the highest doses, 40 mg/kg. With regard to the early phase 1 efficacy signals, for also cohort of mixed tumor types, uliledlimab with atezo demonstrated a 23% ORR and a 46% disease control rate, which appears numerically higher than reported for oleclumab combination at a comparable stage of development.
With regard to the phase two studies, in the recent COAST trial in stage three non-small cell lung cancer patients receiving oleclumab plus durvalumab, reported a 30% response rate in the combination arm versus 17% for durva alone. This was after a median follow-up of 11.5 months. These patients had all received platinum-based chemotherapy and radiation treatment, showing no disease progression prior to entry. In the study of uliledlimab with toripalimab, in cohort 3 of the patients with advanced treatment-naive non-small cell lung cancer, of the 19 patients, the median follow-up is just 3.3 months, in suggesting that with longer follow-up, these numbers may continue to improve, and already have demonstrated a response rate of 26% and a disease control rate of 74%.
This again is in the unselected patients, not accounting for the biomarker strategy that I mentioned earlier. Although not directly comparable, this is a higher than the benchmark KEYNOTE-042 study, which showed an ORR of 16.9% in a similar cohort of patients. Lastly, I always consider carefully the baseline tumor features. The presented information from the COAST trial that I've seen reported, the PD-L1 expression profile was available for 30 out of the 60 patients, and of those, 23.3% of the patients reported a TC of less than 1%. In uliledlimab combination, 88.2% of the patients in this study had low PD-L1 expression. To sum up, both atezolizumab and uliledlimab have demonstrated encouraging efficacy signals in NSCLC patients.
Based upon the highly encouraging interim findings in our study of uliledlimab with toripalimab, we are expanding this patient cohort and will further explore the potential of CD seventy-three as a predictive biomarker. Here, we'd like to highlight our current clinical development plan for uliledlimab. We plan to focus on and expand cohort three in NSCLC patients to further evaluate the clinical efficacy and continue to evaluate the correlation of CD seventy-three expression and patient responses as a predictive biomarker.
In parallel, we are developing a standardized companion diagnostic kit to be employed in future phase 3 clinical trials. Additionally, we are working closely with external lung cancer experts as we plan to initiate a pivotal phase 3 registration trial in China of uliledlimab in combination with a PD-1 antibody in selected NSCLC patients. We are working diligently to determine the protocol to commence this study in 2023.
Meanwhile, we plan to initiate a new phase 2 clinical trial in the U.S. in the next 12 months in other cancer types as we continue to explore the therapeutic role of uliledlimab in combination with PD-1 and PD-L1 therapies. In closing, I'd like to thank the many patients and families who have joined us in these studies, the teams of people working across the CD73 biomarker project, and the tireless efforts of the healthcare professionals who conduct these studies, so that through our collective efforts, we might bring innovative treatments to patients who need them. Thank you.
Thank you, Dr. John Hayslip, and thank you, Dr. Zhu, as well, for both of your insights today. Next, we'll move over to the Q&A session. If there are any questions from the audience, please use the raise your hand function on Zoom, and we'll unmute you for your questions. I see a first question coming from Kelly Xu at Jefferies. Kelly, could you please go ahead?
Thank you. Thank you for taking my questions, and congrats on the data. The first question, do you also see a similar correlation between CD73 expression level and the clinical response in other tumor cohorts? Second question is, for cohort three, there are 19 ineligible patients for PD-1 treatment. What percentage of the patients has TPS smaller than 1%? What, how many patients actually have a PD-L1 expression level in the 1%-49% range? Thanks.
John, could you elaborate on that question?
Across the various stages of development, we have seen signs of this correlation with CD73 expression in clinical response. With the more focused design of the current cohort three, perhaps it's easier to see. We have seen, for example, even in phase one, initial phase one study, we've seen correlation with CD73 expression and response. Regarding the second question, it is a mix of patients who truly have a negative or less than 1% and the patients who have between 1% and 49%.
Thank you, Dr. Hayslip, and thank you, Kelly. Our next question comes from Louise Chen at Cantor.
Hi, thank you for taking my question here. The 26% OR looks very compelling, and you mentioned that you will start pivotals in 2023. Could you share with us some more thoughts here? Do you plan to focus only on first-line non-small cell lung cancer? Thank you.
John, please.
Thank you very much. You know, we're very excited. As you mentioned, this is a compelling response rate. The fact that we've identified in cohort 3, eight subjects with the higher expression of CD73 and amongst them, you know, each of those patients has achieved either a partial response or stable disease. None of those patients have had progressive disease as their best response, though early. We find that a very encouraging signal. We're moving forward at our best speed in collaboration with Junshi to identify a predictive biomarker companion diagnostic kit. That work is ongoing. In parallel, we are in consultation with leaders and this and external experts in the non-small cell lung cancer field. We'll be coming forward to describe the specifics of those phase 3 trials in later calls.
Thank you, Dr. Hayslip, and thank you, Louise, for your question. Next, we have Joseph Catanzaro from Piper Sandler, if you wanted to go ahead.
Yeah. Thanks for taking my question, and thanks for the data here. I know you mentioned that the overall cohort was enriched for PD-L1 low patients, but would you be able to say how the five responding patients broke out in terms of their PD-L1 expression levels, whether they were high or low or a mix or mostly PD-L1 low? You know, I know some other companies are developing small molecule inhibitors against CD73, which also claim to not have a hook effect. You mentioned high levels of AMP maybe impacting those strategies, but maybe you could just discuss what you see as some of the potential advantages of using an antibody like uliledlimab versus small molecule. Thanks.
Joe, this is Jingwu. Maybe I can take this question and then, John, if you have anything you can add, later. Uli-levimab has advantage over CD73 small molecule inhibitors because it works through a non-competitive inhibition to completely block CD73 enzymatic activity and shuts down the entire adenosine pathway when the effective dose is reached. This non-competitive mechanism of inhibition is particularly effective in the presence of abundant levels of AMP, which is a substrate of CD73 accumulated in tumor microenvironment. On the contrary, CD73 small molecule inhibitors works through a competitive blocking mechanism and must compete with a high amount of AMP accumulated in tumor microenvironment in order to inhibit the CD73 enzymatic activity.
Now, since the endogenous AMP level in tumor microenvironment is so high, as Andrew mentioned, about 500 micromolar level as reported, it makes inhibition by small molecule inhibitors less effective and often incomplete. We have done research in that direction and have validated the conclusion as just discussed here. John, do you have anything to add to that question?
Yeah. Thank you. Yeah, I can just add that the patients who we've observed today with the most dramatic tumor shrinkage have largely been the patients who were CD73 elevated and not high for PD-L1. For example, one case I recall very vividly the most dramatic drop in the waterfall plot, the patient furthest to the left on the slide I showed, that patient had PD-L1 expression less than 1%. As I look at, I think you asked about the five patients who had the partial response and only one of those patients had 50% or more PD-L1 finding. These are very early data, but as I mentioned earlier, the responses seem to correlate the strongest with the CD73 expression.
Thank you, Dr. Hayslip. Thank you, Dr. Zhu, as well. Next question we have is from Andres Maldonado from H.C. Wainwright. Andres, if you wanna go ahead.
Yes, thanks for taking my questions, and congrats on progress. My first question, do you think as the data stands now, it's enough to attract the global partnership you guys have been pursuing, or does the data need to mature further for that? Any guidance on the timeline to such a deal and how we should be thinking about deal size would be helpful. Thank you very much.
Yeah, thank you. Maybe I can address this question. Uliledlimab is globally competitive, as we discussed. It's a high-value asset in our pipeline. We're in active discussion with potential global partners. It's important to mention that we are seeking a significant global partnership deal that matches the expected value of uliledlimab. Having said that, you know, the CD73 field is still new, and there's not much data available at this point. Such a deal-making takes time through all the due diligence and things like that. And we hope that the recent progress and the new data of uliledlimab as presented today, and you know, as the data further mature, that may facilitate our BD discussions. We will share more information as we progress. Thank you.
Thank you, Dr. Zhu, and thank you, Andres, for your question. We'll have time for one last question, and I see Gil Blum from Needham, if you wanted to go ahead and ask your question.
Hello, everyone, and thanks for taking our questions. Maybe a quick one. What is the likelihood of you guys moving forward with a biomarker-driven phase 3, considering you're developing a CD73 diagnostic? Is anyone else in the therapeutic space is looking at a CD73-centric development pathway, meaning also using a diagnostic? Thank you.
John, please.
Thank you. As I mentioned, the work is ongoing to validate. But our plan and what we expect to achieve is to move forward and evaluate the potential for a predictive biomarker targeting CD73. The data are admittedly early, but to date, as I mentioned earlier, we are 8 for 8, with 4 patients having partial response and 4 stable disease. None of the 5 patients who had progressive disease had the higher expression. I'd also mention the patients with the higher expression, I'm not aware of any of them who have yet come off of a treatment administration, so they remain on trial. I'm very encouraged by what we're seeing in this cohort of patients.
If we can identify patients who reliably benefit when receiving our treatment, that would be a you know a very appropriate goal. This is the direction that we're aiming. I think your other question was, are others? I can't speak for everyone, but I do believe our approach is at this point unique and furthest advanced.
Thank you, Dr. Hayslip. Thank you again, Dr. Hayslip and Dr. Zhu for your insights and Dr. Zhu for your insights as well. Thank you everyone for asking your questions. If there are any follow-up questions, please contact your local IR representative and we'll be able to answer any questions you might have.