Nektar Therapeutics (NKTR) Earnings Call Transcript & Summary

Good day, and thank you for standing by. Welcome to the Nektar Therapeutics ESMO-IO Event Conference Call. [Operator Instructions] Please be advised that today's conference may be recorded. [Operator Instructions] I would now like to hand the conference over to your speaker today, Vivian Wu from the IR team. Please go ahead.

Thank you, Crystal. Good afternoon, everyone, and thank you for joining us on today's analyst call to review Nektar's presentation at the 2021 European Society for Medical Oncology Immuno-Oncology Congress. Before we start, I'll remind you that this presentation includes forward-looking statements regarding Nektar's drug candidate BEMPEG and other potential drug candidates, clinical trial results, the timing of the startup and plans for ongoing or planned clinical trials with partners, the therapeutic potential of our drug candidates, the timing outcome of regulatory decisions and future availability of clinical trial data. Because these statements relate to the future, they are subject to inherent uncertainties and risks that are difficult to predict and many of which are outside of our control. Actual results could differ materially, and these statements are subject to important risks and uncertainties set forth in our Form 10-Q that we filed on November 5, 2021, which is available at sec.gov. We undertake no obligation to update any of these statements, whether as a result of new information, future developments or otherwise. A webcast of this call will be available on the IR page at Nektar's website at nektar.com. On our call today are Nektar management team members, Dr. Jonathan Zalevsky, our Head of Research and Development; and Dr. Dimitry Nuyten, our Chief Medical Officer at Nektar Therapeutics. In addition, we are privileged to have here today with us 2 guest speakers: Dr. Daniel Johnson from the Ochsner Medical Center; and Dr. Mehmet Altan from MD Anderson Cancer Center. Dr. Johnson is a medical oncologist and Deputy Director of the Precision Cancer Therapies Phase I research program at Ochsner Medical Center in Louisiana. He specializes in the treatment of patients with melanoma, lung cancer and head and neck cancer. His specific research interests include developing strategies to overcome immunotherapy resistance and prevent immunotherapy-related toxicities. Dr. Altan is an assistant professor in the Department of Thoracic-Head and Neck Medical Oncology, Division of Cancer Medicine at The University of Texas MD Anderson Cancer Center. He is one of the lead investigators in the Phase I/II study of NKTR-255 in combination with cetuximab in patients with solid tumors as well as an author of the data presented today. So here's our agenda for today. To start off, JZ will provide introductory remarks on the BEMPEG program, and then Dimitry will follow with a review of the data from the poster presentation at ESMO-IO. Then he will turn it over to Dr. Johnson for a review of 6 patient cases from the PROPEL study. Finally, we will ask both Drs. Altan and Johnson to provide their takeaways on the data presented today. So with that, I will hand the call over to JZ.

Thank you, Vivian. I'd like to start today with a brief reminder on the mechanism of BEMPEG. And I think it's important to remember what our objectives were when we made the decision to pursue this program. Essentially, we took a validated target for the treatment of cancer, the IL-2 pathway, and we set out to make it more accessible as a medicine with a different pharmacodynamic profile that harness its true potential. Now IL-2 is the central immuno-stimulatory cytokine. And as you can see, as shown here, its discovery was in 1976 by Bob Gallo. And then through pioneering studies by Steve Rosenberg and others over the course of a few decades, we were able to show that it had the potential to be used to treat cancer. And this ultimately led to its first approval in 1992 for the treatment of kidney cancer, and again in 1998 for the treatment of skin cancer. So what was the promise of IL-2? Well this excitement came from an agent that harness the immune system and when given in high doses was notably associated with complete responses in solid tumors and the phrase "durable cancer regression" became associated with IL-2 since its first approvals. And in fact, IL-2 was historically one of the few treatments for adults with Stage 4 solid tumors that was capable of producing complete responses or CRs and these were often very durable, lasting for decades without the need for further therapy. The literature states that, in fact, a majority of complete responders with metastatic renal carcinoma and metastatic melanoma could probably be classified as cures. So what did the original IL-2 data show? And there is a theme here that you will hear throughout our presentation today about this mechanism, but you can see it here in the early data with high-dose IL-2. The complete response is achieved at the time in the response rate compared to other chemotherapy treatments available in the 1990s, the complete response rate was truly notable. But what was beyond that was how those patients with complete responses did over time. And it was remarkable, because it really is the first data that shows that IL-2 has the potential to create not only the depth of response we want in cancer, but also the durability, with an estimated 59 months and 80-month durability of complete response. And bear in mind that these numbers are only estimates from the product insert because, in fact, the actual meeting and duration of complete response was not reached when high-dose IL-2 was approved. However, this observation of durability with IL-2 really opened up the whole field of immunotherapy as the promise of durability by harnessing the immune system became a reality and which was shown ultimately to improve and extend survival. And this will be another theme that we will come back to discuss again later in today's presentation. So in summary, we believe that this pathway and the biology that it controls is extremely important, and this belief is validated by the role of IL-2 therapy in immune-sensitive tumors, such as melanoma and RCC. However, we knew that IL-2 as a medicine also has limitations and that our chemistry platform is very well suited to provide the solution, and so we set out to do just that. So what were IL-2s's limitations? Well, we know from the early development that it was not until IL-2 was given at very high doses did we see the levels of anticancer activity that we wanted to see. And it was later discovered that this was because IL-2 itself has dual functional properties, and they can act on both Tregs as well as effector T cells. And this is because Tregs constitutively expressed the trimeric IL-2 receptor and are the most sensitive lymphocyte to IL-2 signaling. Much higher dose levels of IL-2 are required to dose through the Tregs to stimulate the cytotoxic or tumor killing T cells, which primarily express the dimeric IL-2 receptor. And it's the stimulation of the cytotoxic T cell and the cascade of effects that result from the stimulation which led to the anticancer effects seen with IL-2. And it was not possible to stimulate cytotoxic T cells sufficiently unless you delivered very high doses of IL-2. And to complicate things, IL-2 short half-life meant that in order to achieve an optimal immune-stimulatory effect and overcome short half-life, IL-2 was given in a high dose in an 8-hour interval in a 5-day cycle, with 2 cycles constituting the course. This resulted in severe toxicities of high-dose IL-2 therapy and a requirement for hospitalization. And in fact, most patients are not even able to complete a single cycle due to severe toxicities. And these high-dose IL-2-induced severe toxicities include vascular leak syndrome, pulmonary edema, hypotension and even heart toxicities. So to say it another way, the problem statement is that the combination of cell-type IL-2 pathway signaling sensitivity and that difference between tumor killing and non-tumor killing T cells, the IL-2 receptor complex expression differences is among the cell types. Poor PK of IL-2 requiring frequent administration of very high dose levels, and then consequently, the over activation of the immune system leading to unwanted toxicities, all together makes the use of high-dose IL-2 very challenging as a medicine. And in fact, most patients, as I mentioned, can only tolerate a portion of a 5-day administration cycle, making long-term dosing impossible and very challenging to reach a larger number of patients with this potential medicine. So we would need a solution to make an improved IL-2 agonist that overcomes the limitations of high-dose IL-2 and enables long-term outpatient dosing. So one thing we needed to solve these inherent IL-2 problems was a way to shift the balance of cell type specificity by targeting the right IL-2 subreceptors, the beta gamma or the dimeric receptors, to stimulate proliferation of cytotoxic T cells, but at much lower doses that were not going to cause the toxicities you see with high-dose IL-2 and also to overcome the short half-life, which requires such frequent dose administration. So we used our chemistry platform to achieve these things and really had the perfect chemistry with our polymer engineering to address these IL-2 limitations. First, we use PEGylation to mask the interaction with the alpha receptor so we can buy a signaling to favor the IL-2 receptor beta gamma or the dimeric complex. And this allows us to increase the CD8 T cells over the Tregs in the tumor microenvironment. Second, we use our PEGylation to deliver a controlled, sustained signal to the IL-2 pathway by designing a prodrug, which releases active species over time. And this also has the added benefit to mitigate overactivation of the IL-2 pathway. And essentially, when you take it together, we saw it to widen the therapeutic window between efficacy and safety because it's quite narrow as is used with high-dose IL-2. So we could allow significant tumor exposure to T cells after a single dose to achieve an antibody-like dosing schedule on an every 3-week interval, and we could achieve the goal of turning IL-2 into an outpatient medicine. And in fact, we've done that. Here is the schematic of our molecule design and its mechanistic properties. As many of you know, NKTR-214, also known as BEMPEG or BEMPEG is an investigational IL-2 pathway agonist that is preferential to the beta-gamma IL-2 sub receptors. In vivo, BEMPEG induces increased proliferation and infiltration of CD8 T and natural killer cells without expansion of unwanted T regulatory or Tregs in the tumor microenvironment. BEMPEG plus checkpoint inhibitor treatment increases infiltration of CD8 T cells into the tumor and enhances tumor PD-L1 expression, providing very strong scientific rationale for the combination. So one of the things that you can think about is how is this combination therapy? Why does it make sense? And in fact, there's really a ton of rationale about this. And I think a very easy way to think of it is that the checkpoint inhibitor essentially removes the brakes on the immune system. This is PD-1 blockade. And then by adding BEMPEG, we hit the gas on the immune system. This is the T cell expansion and proliferation. And in fact, as I said earlier, BEMPEG plus checkpoint inhibitor treatment actually increases infiltration of CD8 T cells into the tumor and enhances PD-1 expression on the T cells and PD-L1 expression on the tumor. So the rationale is very straightforward to combine the 2. One important additional takeaway when you think about PD-1 expression on the circulating lymphocytes is that BEMPEG actually increases the expression of the target of the checkpoint inhibitor. Again, a very desirable feature for the combination. So knowing that T cells and PD-L1 expression are prognostic in cancer and that IL-2 is clinically validated target in multiple solid tumors with approvals in RCC and melanoma, we recognize in BEMPEG the key attributes that provide us with the rationale for combination with an anti-PD-1. Notably, BEMPEG induces strong T cell proliferation and activation after each treatment cycle, increases activated T cells and PD-L1 expression in the tumor tissue, synergizes with anti-PD-1 in other tumors that are sensitive to T cells and to anti-PD-1, such as metastatic melanoma, urothelial carcinoma, renal cell carcinoma and non-small cell lung cancer, as we'll discuss today; converts tumor tissue from PD-L1 negative to positive as a single agent treatment and also in combination with the checkpoint. It increases favorable antitumor gene expression, including activating the interferon gamma gene signature. And I'll walk you through these points in a few minutes. But collectively, these attributes of BEMPEG when combined with nivolumab resulted in clinical data where BEMPEG yielded an unprecedented complete response rate in metastatic melanoma and has a well-tolerated safety profile, wherein BEMPEG's AE profile does not overlap with checkpoint inhibitors. So I'll review each of these attributes in more detail and review collectively the evidence that we have on the mechanism. First, we know BEMPEG induces strong T cell proliferation and activation after each treatment cycle. We'll discuss that in the next 2 slides. And to start, shown here, you can see in orange, the PK curve of the activated species that come from BEMPEG's prodrug. And in black, you can see the absolute lymphocyte count in blood. Results from the BEMPEG monotherapy study are shown on the left and in combination with nivolumab on the right. Now we see 2 important BEMPEG effects in these charts. As the levels of activated species rise post dose, we see blood lymphocyte levels transiently reduced because these cells are being activated by BEMPEG and marginating out of the vasculature, and that leads to this transient lymphopenia you see a couple of days post dose. By 1 week post dose, we see lymphocytosis. In fact, a great expansion of lymphocytes above baseline. And by this point post dose, that 1 week interval, these cells are highly proliferative. And the cells -- the levels of cells in the blood are really greatly elevated. And you can see that by comparing the difference in the black line between day 1 and day 8. So what happens when we look at this longitudinally? And some of the key features of the BEMPEG pharmacological response are shown in this next slide. BEMPEG administration can continually provide lymphocyte expansion through successive cycles of dose administration without loss of biological activity. In other words, without tachyphylaxis. Shown on the left are patients treated with up to 20 cycles of the combination of BEMPEG plus nivolumab. So in the first cycle, we conducted intensive sampling, as was shown on the previous slide. But in cycles 2 through 20, we sampled at baseline in 1 week post dose. Importantly, you can continually see the sawtooth pattern of lymphocyte expansion. This feature of BEMPEG, namely its ability to continually expand lymphocyte subsets, is a very desirable property of an immuno-oncology agent because it provides the patient with continual T cell expansion to drive an unending antitumor immune response, which can lead to deepening of responses, durability of responses and inhibit traditional tumor escape mechanisms. We can translate the effects of this peripheral blood lymphocyte expansion to direct action in the tumor microenvironment. As shown on the right side are 2 examples of on-treatment effects seen after the start of treatment with BEMPEG plus nivo. Matched biopsies were collected at baseline in week 3, which is the end of the first dose interval. With IHC analysis, we observed substantial CD8 T cell infiltration into the tumor microenvironment, shown here for all 8 patients in the first-line melanoma cohort for whom we had matching baseline and on-treatment biopsies. You can see CD8 T cell elevation in 7 out of 8 patients, with almost all patients ending week 3 with CD8 T cell levels above the median baseline value for the cohort. We also see PD-L1 conversion from negative to positive, and I'll cover that in more detail in a moment. But when you take all this together, the on-treatment effects including the systemic activation of the immune system leading to peripheral blood expansion of lymphocytes and the local effects in the tumor ranging from increased T cell infiltration and cell signaling leading to PD-L1 conversion, all reflect the positive immunological features of the BEMPEG mechanism of action. Now one additional element of rationale of the combination with checkpoint inhibitors also comes from preclinical testing. Now through the years, we've evaluated BEMPEG plus checkpoint inhibitors in dozens of syngeneic tumor models in mice. And shown here is one exemplary of that. This is the CT26 colon carcinoma. And here, we're showing the effect of BEMPEG combined with multiple checkpoint inhibitors, such as CTLA-4 and PD-1 antibodies. And what you can see is the combinability with both the mechanisms is very high and it's also very durable. And we've also shown in additional studies that even if you rechallenge these animals, they remain protected from additional challenge. Overall, we've also shown that this effect not only leads to tumor control, but leads to infiltration of CD8 T cells into the tumor as well as a very, very favorable CD8 to Treg ratio, namely where the infiltrate is almost heavily dominated almost solely by CD8 T cells. On Slide 17, you can see one of the key features of the cellular infiltration into the tumor microenvironment is how the tumor adapts when the T cells infiltrate. Now one of the phenomenon that we've seen over multiple tumor types is that as the CD8 T cells infiltrate, we can see a remodeling of the tumor microenvironment. And one of the things that we've observed is the conversion of the PD-L1 status, specifically on the tumor. Shown here are results from urothelial carcinoma, where we had 13 patients with paired biopsies; 3 of the patients were PD-L1 positive and 10 of the patients were PD-L1 negative. And we saw 7 out of 10 patients convert from PD-L1 negative to PD-L1 positive. And the range of positivity was really quite remarkable, in some cases getting up to 80%, and the IHC inside of the right shows an example of that. And one of the important mechanistic features of this is that when you kind of take it all together, you have PD-1 elevations that can be induced on the lymphocytes by BEMPEG treatment and PD-L1 elevations that can be induced on the tumor tissue, really making this a very desirable combination partner in combining BEMPEG with a checkpoint inhibitor, such as an anti-PD-1. Now on Slide 18, there's a very important additional feature of the tumor microenvironment that potentiates response to checkpoint inhibition. Now this is data from the study of Ayers, where they showed that an interferon gamma gene signature, whether it's defined by either a 10- or a 28-gene network definition, is highly associated with predicting the ability of patients to respond to treatment with a checkpoint inhibitor. So basically, if the tumor microenvironment has an interferon gene signature inflamed kind of a phenotype, then this is a very favorable phenotype for the patient and very favorable to treat with a checkpoint inhibitor. So if we had an agent that could induce such a genetic remodeling in the tumor microenvironment, that too would be a very beneficial feature for combination with checkpoint inhibition. On Slide 19, we're showing an example from our publication of the BEMPEG monotherapy clinical trial results. And in the bottom panel, you can see the genetic analysis that we conducted on RNA expression levels, pretreatment shown in teal or on-treatment shown in orange. Now you can see a number of genes are elevated when we induce this treatment, including PD-L1, which is indicated in the box, but also genes that are highly attributable to the activation status or what we call an effector T cell phenotype are elevated here, including the interferon gene signature itself. And on Slide 20, we extended those observations, this is data from another publication that we had where here we're evaluating the combination of BEMPEG plus the checkpoint inhibitor nivolumab. And we can see, again, in the tumor microenvironment, numerous genes that are elevated on treatment, notice the BL for baseline in the week 3 for on-treatment showing multiple genes associated with T cell function, including interferon gamma itself, shown in this panel in the red box as well as the gene for PD-1 indicating that the T cells and cells are also elevated. So we've now been able to demonstrate that both from an mRNA perspective as well as from an IHC perspective. Again, really cementing home an enormous amount of scientific rationale for combining BEMPEG with the checkpoint inhibitor. And on Slide 21, to summarize, these features of BEMPEG really all work together to support its combination with an anti-PD-1. And as I said earlier, this translated into clinical data in melanoma which led to a breakthrough designation for the combination of BEMPEG plus nivo. So here is our data in melanoma shown here, reported in JCO manuscript earlier this year. And here is the waterfall plot, with a median follow-up of 29 months for patients treated at the recommended Phase II doses of BEMPEG at 0.006 mg per kg combined with nivolumab. All responses were evaluated by blinded central independent radiology. We illustrate here with arrows 2 more patients who deepened their response since we last presented these data at SITC 2019, both patients which achieved a 100% reduction in RECIST target lesions. Almost 50% of all patients or 90% of patients who responded in the study achieved complete clearance of their target lesions, illustrating the depth of response achieved with this combination over time. In fact, the rate of confirmed RECIST complete response was 34%. With the deepening of response, we reported an updated median reduction from baseline to RECIST target lesions of 78.5% compared to the 61.5% reduction that we presented at SITC 2019. Again indicating that with time, you see these effects further deepen and the efficacy further improve. And we see that patients responded to treatment early on, with a median time to response of 2 months and median time to complete response of 7.9 months. Additionally, you can see that even patients with poor prognostic markers, such as those in PD-L1 negative tumors, high LDH or liver metastases, achieved clinically meaningful responses. And notably, all 5 responses in patients with liver metastases went on to become complete responses, a remarkable outcome for these poor prognosis patients. The median duration of response is still not reached and patients remain on treatment because the maximum treatment duration per protocol was 2 years. And overall, the patients in this cohort had a median of 9 cycles of therapy with the doublet. With the meeting and follow-up of 29 months, we reached median PFS, which was 30.9 months. The PFS rate for 12 months was 56%, and the 24 months is 53% and the 36 months was 46%. So let's put this into context. The median PFS with approved PD-1 inhibitors in this setting ranges from 4.1 to 6.9 months. And with the PD-1 CTLA-4 combination, median PFS is about 11.5 months. And you will recall our comparator arm in the Phase III melanoma study, a single agent nivolumab. So we are very encouraged by the PFS we observed here in this data. Despite having a median follow-up of 29 months, the median overall survival has not been reached. The OS rate at 12 months was 82%, at 24 months was 77% and at 36 months was 71%. All patients who experienced CRs or 100% reduction in target lesions are still alive. So as you saw in our data set, I will also remind you that the depth of response has been well documented in the setting of melanoma. At this point, I'd like to direct your attention to a retrospective meta analysis conducted by the FDA and presented at ASCO 2019. The analysis analyzed 4,826 melanoma patients with a total of 10 randomized trials in order to explore the relationship between depth of response OS with various therapies used to treat previously untreated metastatic melanoma patients. Depth of response in RECIST target lesions clearly correlates with longer OS. I draw your attention to the yellow line on the right of this slide, which are the 76% to 99% decrease, and the blue line which are 100% decrease. This clearly shows that deep responses of greater than 75% are associated with a high rate of estimated overall survival of 24 months. And their analysis clearly indicates that in these patients with melanoma treated with immunotherapies, achievement of a deep response in their target lesions early on after the start of treatment with the IO could be an excellent hallmark for OS. We believe this phenomenon that occurs with IO therapy, correlation of depth of response to OS is what we are also observing in the small cohort of metastatic melanoma patients from PIVOT-02. So in addition to the key attributes I highlighted earlier that we're seeing in the translational data, we know that in combination with nivolumab, BEMPEG yielded an unprecedented CR rate in metastatic melanoma and high CR rate in bladder cancer. And this is critically important because we know that in melanoma and in non-small cell lung cancer, which Dimitry will cover in a moment, depth of response is also associated with longer OS with checkpoint inhibitors and other agents. So before I hand the call to Dimitry just a quick review of the study we have ongoing with BEMPEG. Here are the 5 studies with BMS, which are registrational and 1 that is in Phase II in RCC with a TKI. We also have a head and neck cancer study ongoing in pembrolizumab and additional work for PROPEL. Importantly, our first 2 studies to read out will be metastatic melanoma and renal cell carcinoma, which just so happen to be the first 2 approvals for IL-2, as I mentioned earlier. We are looking forward to those readouts in the first half of 2022. This group of registrational studies underway allow us to be poised for multiple late-stage readouts but potential BLA filings over the next several years and what our very large frontline tumor settings where checkpoint inhibitors make up the backbone of the standard of care. So we are greatly looking forward to these readouts. And with that, I will turn the call over to Dimitry to review the data presentation from ESMO-IO.

Thank you, JZ. Shown here is the study schematic for PROPEL. The study is designed to evaluate the combination of BEMPEG with pembrolizumab in patients with non-small cell lung cancer, both with and without chemotherapy. At ESMO-IO we are reporting on the doublet data with chemotherapy generated to date in the first-line patient setting with previously untreated and advanced -- sorry, previously untreated advanced or metastatic non-small cell lung cancer. PROPEL is a global open-label Phase I/II trial, which is being conducted at approximately 40 centers across the United States, Europe and Australia. As you can see on the schematic here of the PROPEL study design, there was a dose escalation phase, which is detailed on the left-hand side, and there are several dose expansion cohorts as shown on the right-hand side. All patients on the study received pembrolizumab 200 milligrams intravenously every 3 weeks along with BEMPEG every 3 weeks and with chemotherapy in 2 expansion cohorts. The objective of the dose escalation cohort was to evaluate various doses of BEMPEG and evaluate the safety and tolerability. The dose expansion cohort of PROPEL evaluate the addition of BEMPEG plus pembrolizumab with or without chemotherapy for patients with newly diagnosed advanced or metastatic non-small cell lung cancer lacking actionable genetic mutations. We are looking at BEMPEG with and without chemotherapy across different PD-L1 expression cohorts. Dose expansion cohort 2 is fully enrolled. And in this cohort, we evaluated our standard BEMPEG dose of 6 micrograms with pembrolizumab across all 3 established PD-L1 expression groups. These are the data that we are presenting here today. Cohort 3 is evaluating the addition of BEMPEG of 10 micrograms to pembrolizumab across all PD-L1 expression levels to further evaluate a higher dose level. We had patients in the dose escalation cohort who had to be dose reduced from the 10-microgram dose level down as we will show you in a moment. And we have patients in the 10 microgram cohort for non-small cell lung cancer patients as well who had to be dose reduced. We see overall treatment-related AEs that are about double at this dose level versus the 6-microgram dose level, which we don't think is acceptable. These are on-mechanism toxicities we see at the 10-microgram dose level, again, about double of what we saw at the 6-microgram dose level. So this gives us the confidence that the 6-microgram dose level is the right dose for checkpoint inhibitor combinations and also for the checkpoint plus chemotherapy combinations, which we are currently enrolling in Cohorts 4 and 5 in the PROPEL study. These cohorts are evaluating the addition of BEMPEG 6 micrograms to pembrolizumab and platinum-based chemotherapy in patients with tumors that are PD-L1 negative or have low expression, meaning 1% to 49%. Here are the study procedures and assessments. We measured objective response per RECIST 1.1, and we utilized best overall response by blinded independent central radiology review to accomplish this. Scans were conducted every 9 weeks with a 1-week window allowed per protocol. Our efficacy evaluable population is defined as patients with one or more post baseline scans. We also measured baseline PD-L1 expression by central lab. A total of 18 patients were treated in the dose escalation cohorts of the study with BEMPEG doses ranging from 8 to 10 to 12 micrograms in a 3x3 design. This dose escalation cohort included patients with newly diagnosed and previously treated advanced or metastatic solid tumors, including, first, the third-line melanoma; first, the third line non-small cell lung cancer; second-line urothelial carcinoma and first-line head and neck squamous cell carcinoma. This is a table showing patient demographics and disease characteristics in the dose escalation cohorts which enrolled a total of 18 patients. Four patients were treated at the 8-microgram dose level, 7 at the 10-microgram dose level and 7 at the 12-microgram dose level. The 10-microgram dose level had a higher ratio of patients with ECOG Performance Status 1 when compared to the other dose levels. A substantial majority of patients had melanoma or non-small cell lung cancer across all dose levels. However, 2 patients with urothelial carcinoma and 1 patient with head and neck squamous cell carcinoma were also treated in the dose escalation cohorts. This is a table of the treatment-related AEs in the dose escalation cohort. No grade 5 treatment-related AEs were reported. However, 1 dose-limiting toxicity, a grade 3 hypotension was reported at 10 micrograms within the first treatment cycle. It is important to note that 28% of patients or 4 out of 14 at the highest 2 dose levels required a dose reduction due to treatment-related AEs. Although assessments of efficacy was not the primary goal of this cohort, we did observe several bigger confirmed responses for RECIST 1.1. Three patients with BEMPEG at 10 micrograms pembrolizumab cohort responded, a first-line head and neck cancer patient experienced a complete response, a first-line melanoma patient had a PR and a second-line melanoma patient [indiscernible] pembrolizumab monotherapy at a PR. We did need to dose reduce in several of these patients, as I just stated. So again, the dose expansion data we reported at ESMO-IO is just for the doublet data at 6 micrograms of BEMPEG with pembrolizumab in the PD-L1 cohort. The chemotherapy cohorts are currently enrolling. This is a table showing patient demographics and disease characteristics in the dose expansion cohort for the 75 patients that were treated with BEMPEG at 6 micrograms plus pembrolizumab. In this group, 37% of patients had PD-L1 negative tumors, 37% had tumors with a PD-L1 TPS of 1 to 49 and approximately 25% of patients had a tumor with a PD-L1 TPS of 50% or higher. Approximately 2/3 of patients were enrolled in the EU. Likewise, about 2/3 were male, 56% were ECOG Performance Status 1 patient and the median age was 65. Looking across the PD-L1 negative, low versus the high sub groups. The patients with PD-L1 TPS of 1 to 49 were noted to have a higher proportion of ECOG Performance Status 1, meaning about 64% versus 50%. A higher proportion had been initially diagnosed with stage 3 disease, 21% versus 0% to 7%. And most notably, a higher proportion of patients in the 1% to 49% cohort were treated with prior chemotherapy for non-small cell lung cancer, about 25% versus 5%. Dr. Altan will be sharing his insights on the potential clinical significance of differences between subgroups later in the program. This slide shows treatment-related AEs in the dose expansion cohort for patients dosed with BEMPEG at 6 micrograms plus pembrolizumab. 9 patients or 12% experienced at least one serious treatment-related AE and 40% of patients experienced at least one treatment-related grade 3 AE. As you can see, for almost 10% of patients, the related grade 3 AE was lymphocyte count decrease. This is a known effect of BEMPEG and tends to happen in the first 72 hours of treatment, followed by lymphocytosis from the . There was 1 grade 5 treatment-related AE, which was myasthenic syndrome and was assessed by the investigator to be related to pembrolizumab. The most common related AEs were pyrexia, fatigue, asthenia, flu-like symptoms and pruritus. And the most common grade 3/4 treatment-related AE was lymphocyte count decrease, which, as I mentioned, is an on-target effect of IL-2 treatment and one -- and the other one is fatigue. 70 of 75 patients with non-small cell lung cancer treated with BEMPEG at 6 micrograms plus pembrolizumab had at least one post-baseline scan available for assessment and thus were evaluable for efficacy per protocol. This table notes the objective response rates for these efficacy evaluable patients as assessed by RECIST 1.1 for blinded independent central radiology review. Overall, 12 patients, or 17%, were noted to have either a complete or partial objective response and an additional 24 patients experienced stable disease as best overall response in this cohort for a total disease control rate across all cohorts of 51%. Diving deeper into the data, we observed an 18% response rate among patients with PD-L1 negative tumors which represents a substantial improvement of the historically reported objective response rate of approximately 8% for pembrolizumab monotherapy in patients with PD-L1 negative tumors. Additionally, we noted 13% or 2 out of 15 patients with tumors with a high TPS of at least 50% experiencing a complete response. This 13% complete response rate by BICR in high-expressing disease is a substantial improvement over our benchmark studies for pembrolizumab alone, including those evaluating pembrolizumab with or without chemotherapy, the KEYNOTE-042 for monotherapy, the KEYNOTE-189 and 407 for the chemotherapy combinations. The overall response rate for monotherapy in the KEYNOTE-042 in the over 50% and higher cohort was 39% with a CR rate of 2 out of 299 patients or 0.7%. In the KEYNOTE-189, the chemotherapy combination in non-squamous disease, the CR rate was 2 out of 410 or 0.5%. And in squamous disease, the CR rate with chemotherapy was 1.4%. That's 4 out of 278 patients in the KEYNOTE-407 trial. We did observe a lower objective response rate for patients with PD-L1 TPS of 1% to 49% compared to the benchmark studies for pembrolizumab alone. In PROPEL, we saw a 4% overall response rate by BICR and a 12% overall response rate per investigator assessment, where in the KEYNOTE-001 and 042 studies in this subgroup, approximately 17% of patients experienced a response. We believe that the demographic difference is seen for the subgroup including a substantial higher proportion of patients with prior chemotherapy, approximately 25% in our study versus about 4% to 6% in the pembrolizumab trials, a lower ECOG performance status and a higher tumor burden may have contributed to this observation. Dr. Altan and Dr. Johnson will be sharing their thoughts on this later in the program. This slide shows a waterfall plot separating each patient population by PD-L1 status and with annotation for the best overall response. Each bar in the waterfall plot represents an individual patient, and the Y axis is the best percentage change from baseline in target lesions by RECIST. The green dots denote patients who continue on treatment as of the data cutoff of October 29, 2021. Also noted are the mean and median baseline sum of target lesions per RECIST. The 12 patients with an objective response are noted here and experienced a 75% median tumor volume reduction in baseline target lesion. Additionally, 2 out of 5 patients with an objective response in the PD-L1 negative subgroup had a 100% reduction in their target lesions. As I said earlier, we had an 18% overall response rate in the PD-L1 negative cohort and we reported this earlier this month. We are pleased that this is about double the historical rate of pembrolizumab single agent. And of course, the deep responses here, including the 100% tumor volume reduction of target lesion in 2 patients are even more encouraging. I'm also very pleased to report today that from the 3 patients with stable disease on this waterfall plot, we are now reporting a deepening of response in one of these patients. This was just confirmed per BICR. The patient now has a 46% decrease in their target lesions at their second scan, down from 20% at the first scan, making this now an unconfirmed partial response. If this is confirmed at the third scan, we will have a 21% confirmed overall response rate in this population of PD-L1 negative patients. So here are the key takeaways from our data. BEMPEG at the 6-microgram per kilogram dose level with pembrolizumab was well tolerated in the first-line non-small cell lung cancer setting. We observed a compelling overall response rate of 18% by BICR for patients with PD-L1 negative disease compared to historical data for the pembrolizumab monotherapy treatment schedule. Two patients with a 100% reduction in target lesions were seen in this cohort and 1 patient has an unconfirmed PR and 2 patients with stable disease remain on treatment. Notably, a CR rate of 13% by BICR for patients with PD-L1 high disease compared to the historical data with or without chemotherapy and 2 patients with a PR remain on treatment. A 75% median reduction in baseline target lesion was observed in patients with a response and a deepening of reduction in target lesions over time was observed as well. Median duration of response has not been reached for the patients with an objective response. And the assessment of BEMPEG plus pembrolizumab at 6 micrograms per kilogram with chemotherapy is currently ongoing in the first-line setting in patients with a PD-L1 status of below 50. Before I hand the call over to Danny, I'd like to comment briefly on a topic that JZ mentioned earlier on our call. As we know in melanoma, depth of response with checkpoint inhibitors also correlates with a longer overall survival. And shown here is a study of 355 patients with second-line non-small cell lung cancer, where this phenomenon is also noted. For this reason, we are particularly excited about the depth of response that we see with BEMPEG which we attribute to the IL-2 mechanism network depth of response because it's correlation to overall survival, which is the primary endpoint in most registrational trials in oncology, we think is an important and powerful metric for the assessment in early-stage clinical trials. And of course, as I said earlier, with the standard of care, including chemotherapy now in the frontline setting for most patients with PD-L1 status of less than 50% and still many patients with a PD-L1 status of more than 50%, we have opened our chemotherapy combination expansion cohorts. As with checkpoint therapy alone, the depth of response for chemotherapy in non-small cell lung cancer is also associated with longer OS. So the depth and durability of responses we see with BEMPEG is particularly notable in this context and is something we will be using as we develop our clinical strategy moving forward. With that, I would like to hand it over to Danny, who will spend some time on several patients trial, including case studies for 3 patients that were treated at his center. Danny?

Thank you, Dimitry. I'm privileged to be here, and I appreciate the invitation to speak. I've been involved with the development of BEMPEG since my time at MD Anderson in fellowship actually. And in fact, I treated some of the first melanoma patients with the doublet of BEMPEG plus nivolumab. And it's been exciting to be a part of BEMPEG's development over the years. I took that work with me to Ochsner. Ochsner is an investigator site on the PROPEL study, and I'm the principal investigator at Ochsner for the study. I'm going to share a series of patient case studies with you, some of which are my patients and some of which are from other sites in the U.S. and elsewhere. One thing you may notice is that I enrolled a number of patients to the study whose tumors had low or negative PD-L1 expression. And you might ask the question as to why this study would be open to this patient population and why I would enroll a patient with negative PD-L1 expression to an I-O doublet without chemotherapy when the universally accepted standard of care for these patients is pembro plus chemotherapy. And the reason why is because my knowledge of the drug through prior work with BEMPEG. And as Jonathan mentioned earlier, we have observed mechanistic rationale for BEMPEG in low-expressing and negative PD-L1 patients and that BEMPEG can convert these patients into positive expression. And so this gives the patients the ability to have response to immune checkpoint inhibitors when their typical response rate to single-agent pembrolizumab is very low. In fact, this additive benefit is what I personally have observed in patients, and I'm going to show you some of these patients today. So here is a list of what I'll share. Case study #1 is a 61-year-old with stage IV nonsquamous non-small cell lung cancer, who achieved a confirmed partial response with a deep response decrease in 77% of target lesions, tumor burden. Case study #2 is a 55-year-old with Stage 4 widespread nonsquamous non-small cell lung cancer who achieved a confirmed partial response as well, also a deep response of decrease in 76% of tumor burden. Case study #3 is a 62-year-old with stage I nonsquamous non-small cell lung cancer who achieved a confirmed complete response. #4 is a 64-year-old with Stage IV squamous non-small cell lung cancer achieved a partial response, minus 48%. And then #5 is going to be a 59-year-old stage IV widespread nonsquamous non-small cell lung cancer achieved a confirmed PR of minus 73%. And Case #6 is a little different. We're going to talk about a 76-year-old patient with Stage IV head and neck squamous cell carcinoma who achieved a confirmed complete response. So here's the first patient. This is a 61-year-old patient with metastatic nonsquamous non-small cell lung cancer who is treated by one of our co-investigators at his site in Berlin. The patient entered the study and her tumor PD-L1 expression was negative under 1%. Her baseline tumor burden totaled 65 millimeters. Given that the percentage of patients who respond to single-agent pembro in this situation with negative PD-L1 is only 8%, this patient was unlikely to respond to single agent pembro. Overall, this patient did very well on this IO doublet. Her response was substantial, with a 77% reduction in target lesions at week 64 and the patient is still on treatment. You can see she benefited from a deepening response over time from the spider plot. So here are the scans from the same patient. You can see that at day 281 on study, we have a complete response of mediastinal lymph node, which was 18 millimeters at baseline. In addition, the 47-millimeter abdominal lymph node shrunk considerably to 16 millimeters at day 281 and 15 millimeters at day 319. So for the next case, the second case is a 55-year-old patient that was treated by my colleague Dr. Ganti at University of Nebraska. This patient with PD-L1 positive disease had a large tumor burden. At baseline, the diameters totaled 166 millimeters. Of note, the patient had an extremely large liver lesion, which both this location in the liver and size of the lesion usually predicts a poor prognosis with current available treatments. The confirmed response in the target lesions was substantial with BEMPEG plus pembro with a 76% reduction in target lesions through week 48. After completing treatment with the IO doublet, the patient continued to reap benefits with ongoing tumor shrinkage of the liver lesion even while not on treatment and without the need for any new anticancer therapy. So the response did continue when the patient was off treatment as well. On the next slide, you could see the patient had a large liver lesion at baseline, measuring 109 millimeters. In addition, the patient had 3 other target lesions, including a right lung nodule, a mediastinal lymph node and an adrenal nodule, not shown here. Patient also had a sizable poorly circumscribed and ill-defined left lung module that was previously radiated and then hence was not a target lesion. So a high tumor burden. Patient achieved complete response of the adrenal nodule, the mediastinal lymph node and the right lung lesion. And I was very impressed to see that the liver lesion reduced from 109 millimeters to 40.6 millimeters by day 437. With I-O therapies, we do see patients continue to experience tumor reductions off treatment. And what's remarkable about this case is the patient, even after stopping treatment, had additional shrinkage of that liver lesion. Without any new treatments, the patient's response is deepening and is durable. At day 520, the patient had an ongoing durable objective response with a 78% reduction in target lesions. So this case is an example of another patient that traditionally might not -- might be thought to benefit from addition of chemotherapy since there was high tumor burden even though it's PD-L1 positive. But this chemo-sparing I-O regimen of BEMPEG plus pembro led to a significant reduction in tumor burden and ongoing tumor shrinkage even after the therapy was stopped long after administration. In the next case, the third case of a 62-year-old patient with non-small cell lung cancer who was treated by my colleague Dr. Lerner in Minnesota. This patient presented with multiple sites of disease, including liver metastases with the baseline tumor burden of 26 millimeters. Patient was nearing 2 years of study therapy when we completed the data snapshot. You can see that the patient had a 100% reduction of RECIST target lesions at third scan and then continued therapy with a very durable response. On the next slide, the patient's baseline disease burden is substantially larger than the single hepatic target lesion measuring 26 millimeters. Patient had a second nontarget lesion in the liver and a lung nodule at baseline. This patient achieved complete reduction of the nontarget liver lesion by day 120, resolution of the target liver lesion by day 190 and complete response of all target and nontarget lesions by day 505. So this case demonstrates that this phenomenon of BEMPEG plus pembro deepening responses over time that are durable, which we consistently have seen as a pattern of response with this combination. There is also another example of the ability to shrink tumors in the liver, which tend to be the ones more difficult to treat with our current available therapies. The fourth case involves my patient at Ochsner who is a 64-year-old with non-small cell lung cancer and PD-L1 negative disease. It's well established that patients with PD-L1 negative tumors are not likely to respond to pembro monotherapy. As we've said multiple times, this presentation of a response rate of about 8%. And then single agent checkpoint inhibitors are not standard of care for this patient. In my clinic, I do see many patients who are adamantly opposed to treatment with chemotherapy. And this may be from prior bad experiences with family members receiving chemo or potentially just from general negative reputation that chemotherapy has. But that's why a chemo-sparing IO doublet such as BEMPEG and pembro is very attractive to these patients. And as I mentioned, by adding BEMPEG to a checkpoint inhibitor, patients with PD-L1 negative disease can convert to PD-L1 positive disease and, therefore, have a higher chance of response to immune checkpoint inhibitors. So while I would not recommend these patients that with PD-L1 negative disease receive pembro monotherapy alone as the biology supports for BEMPEG and the responses I've observed with the IO doublet population, I felt very comfortable in rolling these PD-L1 nonexpressers to the PROPEL trial. In fact, I've not enrolled a patient with PD-L1 negative disease who hasn't benefited from this IO doublet. And it's very promising and encouraging for patients who oftentimes in my clinic strongly object to treatment with chemotherapy. At baseline, my patient had 2 target lesions, including a 43-millimeter left lung nodule and a 32-millimeter mediastinal lymph node. Patient also had multiple smaller bilateral lung nodules that were identified as non-target. You can see my patient had a good response while on BEMPEG plus pembro with the left lung nodule shrinking from 43 to 25 millimeters and the mediastinal lymph node shrinking from 32 to 14 millimeters by day 358. The fifth case is also my patient. She's a 59-year-old woman with non-small cell lung cancer who presented with widespread metastatic disease that was also PD-L1 negative. My patient had brain metastases at initial presentation, unfortunately, and therefore decided to treat her brain lesions with stereotactic radiosurgery prior to enrolling on the study. But another example of a patient with baseline poor prognostic features and trying to come up with what I think is the best regimen for this patient. She's now tolerating treatment well and she has received 32 -- or she had a 30-second cycle after Thanksgiving. So she's nearing 2 years on this treatment. While the blinded independent radiologist only recorded 2 target lesions at baseline, a 37-millimeter right lung nodule and a 19-millimeter mediastinal lymph node, my patient had a substantially larger tumor burden, including disease in the right lung, the mediastinal lymph node and brain, as I said. These additional lesions were identified as non-target lesions. With the exception of the right lung nodule, the patient's other sites of disease have completely resolved by the second on-treatment scan, performed on day 126. The remaining right lung target lesion has continued to shrink over time and was noted to be 15 millimeters on day 627, resulting in an overall 73% reduction in target lesions. Before moving on to my final case, I want to share one last comment about the frontline lung cohort. Dimitry mentioned earlier in his presentation that we now have an additional responder in the PD-L1 negative cohort on PROPEL. This is also my patient. And we're very pleased to find out that after the data cut for our ESMO-IO presentation, we discovered that this patient who did experience tumor shrinkage with stable disease at the first scan, minus 20%, and our second scan has now gone on to achieve a 46% reduction in our target lesions. And this reduction, an unconfirmed partial response was assessed by BICR. We will have confirmatory scans next month. And this is hot off the press. Actually, the patient doesn't even know I'm about to see here this morning to discuss the scans. I want to highlight that this patient is 83 years old. She was able to tolerate this doublet treatment quite well and we're now seeing notable clinical benefit at second scan. If this patient does confirm at next scan and the objective response rate for PD-L1 negative cohort in PROPEL will be over 20%, which is very impressive when you consider the response rate for single-agent pembro for PD-L1 negative patients is typically in the single digits. So now let me switch gears for a moment. As I participated in both the dose escalation and the dose expansion parts of PROPEL, I had the opportunity to enroll a head and neck cancer patient to the study. And this patient followed the same theme that my other enrolled patients did. She is 76 with recurrent head and neck -- metastatic head and neck squamous cell carcinoma with PD-L1 negative disease. Her tumor was negative for HPV and negative for PD-L1 expression. So both of these tumor characteristics, the negative HPV and negative PD-L1 for head and neck cancers are poor prognostic factors for responding to pembro single agent. In fact, the objective response rate for head and neck cancer for these patients that are PD-L1 negative is less than 5% with single agent pembro. That said, my patient was one of those patients that is very reluctant to be treated with chemo. In 2016, she was -- when she initially presented with the laryngeal tumor that was treated with partial laryngectomy. She had a local recurrence later that year, unfortunately, and was treated with radiotherapy. The standard of care is to treat this with definitive chemoradiation. But again, she was adamant she did not want chemotherapy and opted to just get radiation. She has second local recurrence in 2 subsequent sites of metastatic disease, all of which were surgically managed due to the patient's preference to avoid chemotherapy. And then this patient was -- and continues to be adamantly opposed to treatment with chemo, which would be the standard of care for her, chemo plus IO, similar with lung cancer. But after multiple sites of recurrent disease were found on CT scan, systemic therapy was indicated and the patient was understanding. I had discussed the PROPEL study with her and because there was no chemotherapy in this study, we made the decision to start her on PROPEL in August 2020. So my patient was initially treated with BEMPEG 10 micrograms per kilogram. And after the third cycle, I reduced her dose to 6 mcg per kilogram due to ongoing Grade 3 fatigue. You can see that she experienced a complete response on treatment, which was durable and continues today. I've treated a number of patients at MD Anderson and more recently at Ochsner with the 6-microgram per kilogram dose, and I firmly believe that the 6 micrograms is the correct dose for BEMPEG in combination with checkpoint inhibitor. Patients can tolerate the BEMPEG 6 micrograms per kilogram in combination with checkpoint inhibitor very well, and we're seeing very compelling efficacy at this dose with deepening responses over time. The first-line melanoma data from PIVOT-02 that includes patients treated with BEMPEG plus nivo are the most mature data we have. The objective response rate, the CR rate and the median PFS were all very impressive compared to single-agent nivolumab, which points to that synergistic effect with this IO doublet. And I do treat melanoma, so I'm very eager and excited to see the melanoma Phase III readout early next year. So now turning back to my patient with head and neck cancer. Here are her scans. As you can see, she had disease in the cervical lymph node -- in one of her cervical ones nodes, in both the left and right lung. Patient also had recurrence of disease at the site of the left parotidectomy bed. Her total tumor burden was 65 millimeters. And amazingly, my patient had a substantial response at week 18 and achieved a complete response, by my assessment, at week 27. The blinded independent radiologist confirmed the patient achieved a CR by week 36. This case is remarkable for a couple of reasons. First, we don't tend to see responses in patients with head and neck cancer who are PD-L1 negative and treated with single agent pembro. Second, even patients with PD-L1 positive disease who are treated with pembro rarely receive -- achieve a complete response with head and neck cancer, only about 5%. So seeing a patient with extensive widespread disease respond to BEMPEG plus pembro was very impressive and I look forward to participating on the steering committee and serving as an active recruiter to the new Phase III study in head and neck cancer comparing BEMPEG plus pembro to pembro in PD-L1 positive patients. The only reason PD-L1 negative can't be included is because you can't randomize PD-L1 negative patients to pembro alone. So in closing, I want to underscore what Dimitry presented. With BEMPEG plus a checkpoint inhibitor, we are seeing deep and durable responses. This is critically important because we know that depth of response is associated with a longer overall survival in non-small cell lung cancer. This pattern is observed in patients treated with both checkpoint inhibitors and chemo. So we're looking forward to see the effects of BEMPEG plus pembro plus chemo in the first-line setting for non-small cell lung cancers those cohorts are enrolling. So I'd like to thank all of you for listening this morning. And with that, I'll turn it over to JZ.

Thank you, Danny. Now I'd like to turn the floor over to Dr. Mehmet Altan to share his impressions of the data. Mehmet?

Good morning. I'm a Thoracic and Head and Neck Medical Oncologist, Clinical Investigator at MD Anderson Cancer Center in Houston, and my primary focus is immuno-oncology in the therapy of non-small cell lung cancer. And I am also an investigator in the PROPEL study, therefore, similar to my colleague Dr. Johnson, I have the firsthand experience with BEMPEG. While it has been already reviewed and the people in this call are quite familiar with this, but I feel it's important to note, over time, we learn different solid tumors display different response by terms to immune checkpoint inhibitor therapies, and for thoracic malignancies will approved checkpoint inhibitors provided a significant benefit in a subset of patients, there is an unmet need for better therapy regimens to improve response registration of response patients, quality of life and, at the end of the day, for patients' overall survival. And today, I'm excited to see and review the clinical data from PROPEL study. While the numbers are small, hard to make a clear statement. In my opinion, there are very exciting and promising observations from this data set. Things that I would like to highlight is at 6-microgram dose, the safety profile looks very encouraging with -- given nonoverlapping toxicities with immune checkpoint inhibitors as well as chemotherapy with BEMPEG. These toxicities are generally seen within the first 1 or 2 cycles and particularly majority of these toxicities are predictable and manageable with risk mitigation strategies. But my main enthusiasm about the current work is coming from what we know about IL-2 and findings in this study that's been aligning with the prior observations from earlier studies with BEMPEG which is supporting its class effect in non-small cell lung cancer as well. Particular responses that are seen in PD-L1 negative group, deepening responses over time, and again, limited but encouraging data with complete responses in a number of patients, which is particularly encouraging since these observations are similar with the clinical findings that's been observed in PIVOT-02. And last note is that responses observed in some challenging populations, some of these cases reviewed by Dr. Johnson earlier, such as patients with liver metastasis, brain mets at the time of initial diagnosis, high disease burden, squamous cell histology, encouraging. And one thing during this presentation, I know that also in the study, we do see some elderly patients, if you see a patient cohort age 80, 83. And those are the patients that are generally hard to treat with chemotherapy. And it's important to highlight that median age group was around 65, but there were some outliers with quite old age. So thank you.

Great. Thank you, Dr. Altan. And with that, I will open it up to questions.

[Operator Instructions] And our first question comes from Peter Lawson from Barclays.

Just maybe initially for Dr. Johnson. On the BEMPEG you've seen -- the responses you've seen, do you think it's more efficacious in squamous versus non-squamous lung cancer? And then kind of how difficult do you think is in treating the squamous versus the non-squamous lung cancers? And then I guess another question, a follow-up around that would just be around, you mentioned the unwillingness of patients to have chemo regimens. What are the options that you provide in your patients for chemo-free regimens?

So 2 very good questions. As far as the differences in histology, I haven't noticed much of a difference. I think both the non-squamous and the squamous cells are responding to this doublet. And the squamous cells are, as Dr. Altan had said, typically a poor prognosis, more difficult to treat. And to answer the chemotherapy question, it is very often that we -- that I encounter patients. And to be honest, not necessarily even rationally so, but just a very against chemotherapy probably because of the negative reputation. And the problem for the PD-L1 negative patients, really anybody with a PD-L1, less than 20% in my mind, is there's not good chemo-sparing options. You can -- there are I-O alone options you can give a patient, but the chances of benefit are very small. So I would not recommend these patients I enroll to PROPEL to get I-O alone if I could avoid it. And if I didn't think there was a true contraindication to chemo and I didn't have PROPEL as an option, I would have really tried to push giving them chemotherapy just because single-agent immune checkpoint inhibitors just don't seem to have very good efficacy in that population.

Our next question comes from Chris Shibutani from Goldman Sachs.

This is CJ on for Chris this morning. Curious, Dr. Johnson and Dr. Altan's thoughts on what a meaningful duration of response would be for the pembro monotherapy? And then if you could give us a sense of sort of benchmark expectations coming up for chemotherapy. And perhaps, could you comment a little more on the discontinuations with the Lambert-Eaton syndrome and the death from ] myasthenia? How is that attributed only to the pembrolizumab and kind of ruling out that BEMPEG was exacerbating that potential?

Dr. Altan, would you like to comment first? Danny, can we have Dr. Altan comment first, and then we'll have you talk through some of the SAEs that you saw.

Absolutely.

So if you think about the duration of response, particularly in patient population with combination immunotherapy and chemotherapy, as you know, if the benefit is not getting really driven by the immunotherapy, the duration of response is quite short. And we do see this in particularly landmark studies with KEYNOTE-407, 189, CheckMate 9LA. So it really depends on what the patient is deriving the benefit from. And in terms of the duration of response, what is going to be the ideal time line, I guess we would like to expand patients who are reaching a 2, 3-year mark, and that is going to be an important time point. We do have competitors from single agent immunotherapy. So I think the important thing is to beat those landmarks and try to make this disease more of a chronic disease. So those are the benchmark, I guess, for different patient subsets these studies should be aimed to overcome and improve. And in terms of immunotherapy side effects, I think Dr. Johnson may have more clear knowledge about that specific patient. But I think we kind of know the cytokine-related toxicities, and that is why it's been mainly attributed to pembrolizumab in that specific study. Again, generally, our observation in clinic, and this is similar in my clinical experience, there are not significant overlapping toxicities, and we are not necessarily seeing immune-related adverse events increasing with the combination of BEMPEG plus anti-PD-L1 or PD-1 in terms of typical immune-related adverse events that are seen with combination immune checkpoint inhibitors or single-agent immune checkpoint inhibitors.

I agree with Dr. Altan. The myasthenia gravis is an unfortunate, well-described side effect from pembrolizumab and other immune checkpoint inhibitors. It's rare, but the more we give, the more we're seeing, and there's some good retrospective literature on how to manage this, but it is a difficult side effect to treat. But I don't think necessarily, if you look at the entire population treated with the combination of BEMPEG plus PD-1 inhibitors that side effect or any other immune-related adverse event that's known to be caused by checkpoint inhibitors has increased much. It really is what Dr. Altan said, the cytokine-like side effects that we see, that flu like temporary side effects that aren't the severe ones. But yes, myasthenia gravis is well described as a side effect from immune checkpoint inhibitors, is difficult to treat but, thankfully, relatively rare.

Our next question comes from Jay Olson from Oppenheimer.

Did you observe an increase in PD-L1 expression levels in the PD-L1 low patients who responded to treatment with BEMPEG plus pembro? And also, did you observe any increases in CD8-positive T cells or NK cells in responders? And are there any other biomarkers that might help predict response to the doublet? And then separately, can you talk about the onset of response in patients who responded and how that compares to the onset of response in single-agent pembro?

Jay, this is JZ. So I'm happy to take your question. So we did have the opportunity to evaluate some of the biomarkers from the patients in the study. So we only had 5 patients for whom we had matching baseline and on-treatment biopsies. There were just fewer biopsies that were collected from the study. However, one of the very exciting things that we observed is that in those 5 patients, 4 of them were PD-L1 negative, and 3 out of 4 of them converted to PD-L1 positive. These patients were all in the dose optimization cohort and they span a range of different tumor types, including non-small cell lung cancer as well as melanoma. So one thing that's very important is that we did see those same kind of PD-L1 conversions that I described earlier in the presentation also in the PROPEL study. There were just fewer samples that we were able to collect in this trial. We're looking forward to actually publishing all of this work into the future. And that will include also the CD8 T cell responses, additional analyses of the tumor type itself, right, as well as the gene expression data. We did also look at a number of additional biomarkers, including on-treatment biomarkers, that we'd also include in that publication to come. One of the things that we looked at in this particular population was the relationship with different cell types in the blood. For example, lymphocytes and nonlymphocyte cell populations. And we know that BEMPEG can affect those in different ways. We also looked at that correlation as also being related to kind of a marker on treatment related to the potential for response. So we'll be looking to put all that into a publication to come in the future. Now I'm going to turn it over to Dimitry to discuss the time to response element in the data set. Dimitry?

Thanks, JZ. So the median time to respond in our data set is 2.1 months, which means it's about the first scan of 9 weeks, which is similar to pembrolizumab, the median time, the response reported in the KEYNOTE-024 study. So that's only in high expressers, but that's, let's say, the longest follow-up data and the most extensive data available is 2 months. So same as the first scan. I think the differentiation is not so much to speed of response. And of course, to really model speed of response theoretically, you would have to do scans earlier. From my previous life, I can tell you with avelumab or BAVENCIO, we did do scans many times at 6 weeks and we did see responses at 6 weeks. So if you do your first scan at 9 weeks, you might miss very early responses. So that's hard to model, but I would say it's similar. I think the notable differences that we see is the CR rate, as I gave you the benchmark numbers and the depth of response. So not so much the time, but the depth of response. Let's say, the meaningful long-term correlation with survival is what gets us really excited as a potential differentiator here.

Danny, maybe if I could ask you, since you had a number of patients that you treated in the study, maybe you can also comment about the onset of effect. I mean, scans are one thing, but even just the additional patient's journey in terms of how quickly the effect comes on.

Yes, absolutely. Actually, my patient with head and neck cancer who we treat on the dose optimization cohort, she did have clinically apparent disease of the cervical lymph node that we are monitoring. And that -- we did see that lymph node soften and there was a clinical response just a couple of weeks after starting, 2 to 3 weeks. I don't know if the size had changed much by that point, but we did see effects very early on. And then as far as the difference between PD-1 and the doublet as far as the timing of response, as Dimitry said, it's not necessarily the time to response that's much different, but it's that deepening over time, and it seems pretty consistent on the patients that I've had with deep responses, that each scan, the response continues and the response deepens. And I think the median time to CR for this doublet is something like 8 months. So that's an interesting kind of unique aspect to this combination is that it deepens over time.

Our next question comes from Jessica Fye from JPMorgan.

This is [indiscernible] for Jess. So on the BEMPEG plus pembro and chemo triple combo, I guess chemo alone have been shown to increase PD-L1 expression. So we just wondering what additional benefit of BEMPEG that you think could bring in this triple combo and that [indiscernible] the addition of chemo, right, is no longer like a chemo-free option? And then following on that, if the addition of BEMPEG is to drive deeper response, as they have been shown in the initial data from the double combo, like PFS or OS based endpoint be needed in the future [ PIVOT-02 ] study? And then lastly, when could we expect to see data from the triple combo cohort?

Thanks. Dimitry, would you like to comment on the question, please?

Yes, there's multiple parts. So let me start, let's say, with the PD-L1 status, we do believe, based on the biopsy data we've shown, based on the IL-2 mechanism that we can further enhance the PD-L1 in, let's say, expression or more importantly, getting tumor inflammation and thereby driving deeper and durable responses and that would be through with or without chemotherapy. Whether or not there would be an additional enhancement of chemotherapy with that particular aspect in the doublet, we will have to see with additional data. I'm hopeful that it could be the case. When it comes to the timing of the chemotherapy, data that we are currently, let's say, accruing, we expect to present by mid of 2022 the first data on safety and preliminary efficacy. And then those cohorts will be further enrolling throughout the year. So there will be updated presentations later.

Great. Thanks, Dimitry. And Mehmet, do you mind if I ask you to also provide additional commentary on the relationship of the data that we're seeing in the chemo-sparing setting and how that could reflect in the setting of a triple combination with chemo?

Certainly. So I think we do see increased responses with addition of chemotherapy, as you all know. But the problem is the duration of response in many patients that are treated with that approach, either abbreviated course of chemotherapy or standard 4 to 6 cycles. So I think the major thing is, while we know chemotherapy can cause cytotoxic T cell killing -- sorry, cytotoxic killing and can inflame the tumors, in reality, we are not seeing deepening responses with that approach. And I think we particularly see this with large-scale studies with chemo I-O combinations that complete responses are less than 1%. So the duration of response, deepening responses and CR will be the things to look for with this approach. The other thing is, while we do know that we are capturing probably a majority of these patients in 9-week scans, if we look at PIVOT-02 study, we do see the responses range between 1.5 to 4.1 month and CR was achieved median of 8 months. So we might be still not seeing the full impact of this combination in combination immunotherapy plus PEGylated IL-2. And with addition of chemotherapy, you may actually benefit because you can get earlier responses and with addition of BEMPEG, you can deepening the responses and you can make them more durable. And I think there is one thing that when I look at this data that is notable, is when you look at different PD-L1 populations in this PROPEL study, you do see that there are a number of patients continues to be on treatment. So if what we saw in PIVOT-02, will continue to be something we are going to observe in PROPEL, you may end up seeing a larger number of patients seeing partial responses or CR, which is more exciting for me as an immuno-oncologist that this treatment may provide. So addition of chemotherapy may be able to capture those patients in the setting of a clinical trial. Hopefully, that's going to reflect in real life.

And JZ, if I can add a few more things, right? If you think about, generally speaking, chemotherapy combinations. So when we think about ahead, as we've said on this call, we see definitely a number of promising observations in the chemo-sparing approach. Danny gave a couple of examples of patients who are very much, let's say, against chemotherapy. They really want a chemo-sparing approach. When we think holistically about a registrational trial strategy, we have to think about what are regulatory requirements, what is the standard of care, what is approved, what are patients getting? And below 50%, a vast majority of patients, are getting a chemotherapy inclusive approach. Despite the approval in the 1 to 49, the approval for 1 to 49 and pembro monotherapy is limited to the U.S., it's not available in European countries. And then lastly, even in high expressers, there is some clinical selection of patients who have a higher volume of disease, who have faster proliferating disease based on clinical symptoms. And even in the subgroup of high expressers, almost half of the patients are not getting pembro alone. They're getting a chemotherapy inclusive approach or something else. So with all of that, we think a chemotherapy-inclusive approach for a further registrational strategy is the best way forward despite the very strong signals we see for the chemo-sparing data presented today.

Our next question comes from Greg Harrison from Bank of America.

We're seeing that your -- some of the responses are deepening. We're seeing PRs turn into CRs. How long do you think that it would make sense to continue the therapy in the real world in hopes of continuing to see those responses deepen?

Thanks, Greg. Danny, can I ask you to comment first and then, Mehmet, if you'd like to add as well.

Sure. That's a tough question to answer just based off of what we have now. But from my experience, where I had experience in stopping drug earlier are patients who either have to stop drug because of tolerance or just don't want to continue with BEMPEG because of fatigue and some of those cytokine-like side effects. And those patients who have a CR that I've treated that have then decided to stop that treatment and just continue the checkpoint inhibitor monotherapy, they seem to have very durable responses, too. So I think the question is interesting of how long, especially in those patients that eventually achieve a CR or DPR, how long do you need to keep the BEMPEG on? I don't know the answer to that. I'm sure every patient is going to be different. And with more patients, we may be able to answer that better.

I do agree that there may not be a very fixed duration, and it may change depending on the tumor and patient population. But I think what we've been learning is that with nonoverlapping toxicities and with the given safety profile and hopefully the data coming from PIVOT-02 is suggesting that this can be a treatment similar to at least, in my opinion, at least 1 to 2 years and beyond if it's going to be providing in a benefit that will depend on the risks and benefits. But it's encouraging that the BEMPEG-related toxicities generally weans off after the first couple of cycles. I think serial biopsies' longitudinal follow-up is going to be very valuable to see. But I would imagine that it will be a combination. Probably, they will recommend outpatients until there's a better data for at least 2 years.

Yes. And Greg, one of the things that I'll add just in incline to closing this question is, remember, what did we do to IL-2. We created a drug that you can dose repeatedly, and that can be actually scalable and given to patients over multiple cycles, right, which is just now possible with high-dose IL-2. So for the first time, we did also have the option. And when we see that kind of constellation of effects that leads to deepening of responses over time, getting a patient to maximal benefit, again, it's just one of those options. So we, for the first time, kind of have that as an opportunity. And as Dr. Johnson and Dr. Alton said, as more patients come in, we look at the full kind of risk/benefit of the duration of treatment. We now at least have that option available for the first time.

Next question comes from Boris Peaker from Cowen.

And I think this one is probably for JZ. In the BEMPEG responders, we see a number of patients that were classified as PR, but we saw 100% tumor reduction. Can you comment why they weren't classified as CRs and what that means in terms of their longer-term outcome?

Sure. Great question, Boris. So basically, there are target lesions and nontarget lesions, right, that are both sort of included in the overall assessment of disease. But the RECIST scoring classically is really focused on the target lesions. Nontarget lesions are not classified as target for a variety of reasons, they may be difficult to measure or other kinds of elements. So for the patients that had a minus 100% reduction in their target lesions, these are patients that no longer have target lesions present. So they're minus 100%. We call them a PR. And the only reason why they're not classified as a CR is because they may have some small amount or some residual nontarget disease that's remaining. But I'd like to ask one of our physicians to comment on the impact, right, of complete clearance of the target lesions, which are the more -- the most serious elements, right, in those patients? And maybe Dr. Alton, you could sort of help add color by just giving more insight into what a 100% reduction in target lesion PR means from the patient?

Thank you. I think this information is coming from what we learned from the prior clinical trials, and it seems like the depth of response correlates with duration of response, PFS as well as OS. Of course, it's quite early to comment if this is going to translate the same benefit in BEMPEG. But it is quite encouraging that we are seeing those complete responses. Of course, the other side is the patient perspective, I think, psychologically, physically having low disease burden or no disease burden really impacts the quality of life. That's why when we balance things about the toxicity profile of the drugs and the impact of the treatment to the patients seeing complete response, I guess, gives a relief to everyone.

And our next question comes from Arlinda Lee from Canaccord.

I guess my questions are more for the clinicians. Dr. Johnson, I think you mentioned that had you not PROPEL as an option you would have pressed for chemo, curious whether there were particular characteristics with the patients that you know that you thought of PROPEL to kind of help us predict for which patients might have efficacy, just looking at the responses have been kind of across the board. And I was wondering for both of you guys, if BEMPEG were available, how might you -- which kind of patient populations would you think to look in to use this first.

Sure. Thank you for the question. Yes. I think it wasn't necessarily something about these patients that made me think that they were going to respond on the PROPEL trial. It was more the fact that I knew that they had a very little likelihood of responding to standard treatments. And again, either the patient didn't want chemotherapy where I wanted to be able to offer a chemotherapy-sparing option to them. And I mean, to be honest, the patients I enrolled, all have very poor prognostic factors. I mean PD-L1 negative liver mets, brain mets, I'm sure in the back of everyone's plans, like, what are you doing enrolling these patients on the trial. But it was -- it's because I know the biology of BEMPEG can convert PD-L1 negative to positive. And so I knew that was a good option for these patients that were in a hard spot and did not want chemotherapy. And then as far as how I would foresee this being used, I think exactly how we're talking about it. Even though this is on clinical trial, every time you're treating a patient and you have a patient in front of you, you're trying to come up with what you think is the best treatment option for that patient individually. And having a chemo-sparing option that can convert PD-L1 negative to PD-L1 positive is really enticing for that situation.

I can also comment on this question. So I think there are patient preferences, but also there are clinical contraindications for chemotherapy and particularly patients previously treated for their locally advanced disease or their bone marrow suppression or cumulative toxicities related with the treatment sometimes [ and miss ] our ability to offer them chemotherapy. And we know as currently single agent or combination immune checkpoint inhibitor therapies, despite how much it changes our therapy landscape is just in every patient population is not as effective as we would like to see. So in that case, I think particularly for certain patient populations, there's just really unmet need. That's why, in my opinion, this data about low PD-L1 or PD-L1 negative patient population or with high disease burden as well as some squamous cell carcinoma brain metastases, these are the patient populations we always feel a little bit nervous about single-agent immunotherapies in concentration. So it's great to see that signal coming, and I hope to see it's going to translate into a more statistically significant data in larger scale patient cohorts. But there is definitely an unmet need in some of those patient populations. And regardless how in high PD-L1 population, immunotherapy works well. Still we lose majority of our patients within 2 to 3 years. So there is still that part is quite low, and we need to do better on that.

And that does conclude our question-and-answer session for today's conference. I'd now like to turn the conference back over to Vivian Wu for any closing remarks.

Thank you, Crystal. Thank you very much for joining us today. Have a wonderful afternoon.

This concludes today's conference call. Thank you for participating, and you may now disconnect.