Nektar Therapeutics (NKTR) Earnings Call Transcript & Summary

Great. Good afternoon, everyone. My name is Jess Fye. I'm one of the biotech analysts at JPMorgan, and we're continuing the 2020 Health Care Conference this afternoon with Nektar. Just a quick housekeeping note. We're going to do a Q&A session with the management team right after this ends. It's down the hall in the Olympic room. You can follow me down there. But with that, I'll turn it over to the company's President and CEO, Howard Robin.

Thank you, Jessica. I'm going to be making some forward-looking statements today. So if you take a look at the therapeutic areas we're working in, in the pipeline, I think if you look around this conference, you would be hard-pressed to find a company our size that has such a diverse and important pipeline in areas of pain, immuno-oncology, immunology and also virology. And I think this says a lot about what we've done over the years with our programs. Now I'll start with our pain program, NKTR-181, and I'll talk about that briefly. Our ADCOM is tomorrow so I don't think we have to spend a lot of time on it today. We'll see what tomorrow brings. I'm very hopeful that we have an approval. And interestingly enough, we've had this drug in over 2,200 patients, 15 different clinical trials. And we have human abuse potential data that demonstrates clearly that at therapeutic doses, NKTR-181 is as likable and causes as much euphoria as placebo. So yes, you do see some euphoria that looks like low doses of oxycodone, but only at 10 -- 3 to 10 or 12x the therapeutic dose. So if you stay within the labeling and you use the therapeutic dose, you should look like placebo when it comes to euphoria and likability. And remember, NKTR-181 is a new molecule. It's a new opioid molecule. There is no formulation. This isn't a packaged drug, if you will. It's the first new opioid in probably 30 years. And we're very hopeful that it will make great inroads in helping the U.S. with its massive opioid crisis. So let's move on to immunology, oncology and immunology. We have a new collaboration that we've entered into, an expanded collaboration with BMS. I'm going to talk about that a little bit later. But I'd first like to start with what we have done in harnessing the immuno-oncology cycle. So we decided that we wanted to catch every piece of the oncology I-O cycle, starting with NKTR-262, which is a small molecule that induces antigens; moving on to bempeg, which causes effector T cell proliferation; and ultimately, NKTR-255, IL-15, which causes a proliferation of NK cells and memory T cells. So to start, let's focus on, at the moment, the crown jewel and the most advanced program which is bempeg, which causes a proliferation of effector T cells. Now what is so special about bempeg? First of all, I think we all agree that the IL-2 pathway is exceptionally important. And when we went to harness an agonist of the IL-2 pathway, we could have easily targeted, as other companies have done, a selective approach where we only bind to beta-gamma receptors and not bind to alpha at all. And we purposely didn't do that. We didn't make a selective agonist. We did work on that in our laboratories. It's actually much easier to do than what we've done, which is a biased agonist. And we felt that it was very important to get some transient binding to the alpha receptor, and that's so you can enhance priming in the lymph nodes. And that means that you get T cell proliferation when that new tumor antigen is presented. So we thought a biased approach where you get some transient alpha binding and full beta-gamma binding was very important. If you just do a selective agonist, then we saw a very much diminished effect. The other benefit of bempeg is it's a prodrug. And because it's a prodrug, you don't have the risk of cytokine storm. You don't have the risk of an overactivation of the IL-2 pathway, which can cause, as you know, some very serious problems. Lastly, that gives it antibody-like dosing, once every 3 weeks in an outpatient setting, so we're very pleased with that. I want to show you some data. This is the data we showed at SITC 2019, and this is in metastatic melanoma. And I think this is a real good indicator of how important this drug can be. So if you look at this data which we presented at SITC, you can see that 34% of the patients had a complete response. And by the way, this is all central-read, blinded data. So this is all independently read, centrally read, not Nektar-read. 34% complete response: 42% had a 100% reduction in target lesions, and 47%, almost half, had a greater-than-75% reduction in target lesions. Now why is that so important? If you remember at SITC, the FDA presented metadata where they looked at melanoma and they looked at I-O therapies in melanoma. And they found that if you could get to a 75% or greater response, you had a very high likelihood of having good results in PFS and overall survival. So the fact that we've had half the patients had a 75% or greater response is very important. And interestingly enough, of the patients who responded, of anybody who responded, so if you respond to the drug, 90% of the patients who had a response went on to achieve 75% or greater reduction in target lesions. And if you look at those down arrows, those are the deepening of responses that you see since the last SITC -- since the last data cut before the SITC 2019 data cut. We'll have another data cut this year, but this is what we presented at SITC. Now this is important because -- there's another important thing here that you should note. Every patient who had either stable disease or a response did not have any further tumor progression. So if you were stable or you had a response, you either stayed stable or you continued to deepen your response. No one showed tumor progression after they had stable disease or response. Very important distinction. And if you look at some of the disease indicators that are -- that have bad prognosis, like liver metastases, which we all know is a serious problem here, all 5 responders that had liver mets experienced complete responses. So we're very pleased with this data. I think it led to a breakthrough designation as a first-line agent. I don't really know of any other cancer drug that got breakthrough designation in a first-line setting. So I think there might be one, but I don't know of any. So we're very pleased with this. Now this is also important. If you look at the PFS, nivo as a single agent has a median PFS of about 6.9 months. This is historical medians, of course. Nivo plus ipi, a fairly toxic regimen, has a median PFS of 11.5 months. We are now at 18.6 months, and we have not reached PFS yet. I would expect it to exceed 20 months. And that is a dramatic benefit of using a nivo/bempeg combination. So what does our new joint development plan look like? It's bempeg and nivo on 7 different indications, 4 tumor types, over 3,000 patients. First-line metastatic melanoma is, of course, running. That study, by the way, we should have response rate data by the end of this year. And we should have -- we will have our PFS data, which is the primary endpoint of the study, by the middle of next year. We're starting an adjuvant melanoma study, which is bempeg plus nivo versus nivo in adjuvant melanoma, a very, very large market, as you know. We have a study running in bempeg and nivo in a TKI sparing regimen in metastatic renal cell carcinoma. We're starting a first-line metastatic RCC study with a triplet, bempeg nivo and axi, versus nivo and axi, and we think that has a lot of potential. In bladder, we are running a first-line, cis-ineligible cancer study in PD-L1 negative patients, and that has the potential for accelerated approval. And we're now starting a muscle-invasive bladder cancer study as a supportive study to that in the event we're successful with accelerated approval. And we're also starting a study in first-line non-small cell lung cancer, bempeg plus nivo, in a dose escalation and then expansion program. And BMS is running that program. BMS is paying for 100% of the cost of that program. Now just to give you a flavor of how committed they are because I know that's been a concern for everybody. If you look at their share of the clinical costs associated with what's on this screen, it's $1.2 billion. That's what they have committed to. That should give you some comfort that bempeg is real. Economics, some modifications to the economics, generally all in our favor. Additional near-term milestones: so a $25 million milestone payment on the start of MIBC; an acceleration -- a $25 million acceleration of a milestone payment in the adjuvant melanoma study, 75 -- and those 2 will come this year, so that's an additional $50 million this year; and then a $75 million accelerated milestone payment at the start of the Phase III in triple -- in first-line non-small cell lung cancer. The rest of the economics are unchanged. Development costs are shared 1/3 Nektar, 2/3 BMS. Nektar books global revenues. Profit split is 65% Nektar, 35% BMS. And first milestone on the first approval is $650 million, and then on the next 3 approvals, $260 million each. Now in addition to this, we're running the PROPEL study, which is a combination of bempeg and pembro, in first-line non-small cell lung cancer. We'll have the initial data from that study at the end of this year. We're doing 40 patients dose optimization, 58 patients in expansion, and what we're looking at is increasing the dose. So every -- all the data that I've showed you so far is in -- with 0.006 mg/kg. And we decided that because lung cancer is less immunogenic, we wanted to see if we could increase the dose of bempeg, and we're going from 0.006 to 0.008 and to 0.010 mg/kg. And 1 -- as in parallel, we're running a combination with pembro in PD-L1 less than 1, 1 to 49 and greater than 50. And we'll have readout on about 20 patients of that by the end of this year. In addition to everything we're doing in those programs, we also have a program that has started with Pfizer, combining Nektar and Pfizer checkpoint inhibitor and other programs in prostate cancer and head and neck cancer. And we've started those clinical trials with Pfizer. With Vaccibody, we've started a study in head and neck cancer, combining bempeg with their neoadjuvant vaccine. And we will be starting a study with BioXcel, combining their molecule bempeg and Pfizer's avelumab in pancreatic cancer. So I think that's a fairly broad clinical program for bempeg. Now let's move back in the immune cycle and talk about what we're doing with NKTR-262. NKTR-262 is a TLR 7 agonist, creates the antigen, and the important aspect of this molecule is that it has abscopal effects. So obviously, you're not going to inject every single tumor. Wouldn't develop a drug that requires the injection of every tumor with a TLR agonist. Here, you can inject 1 or 2 or 3 primary tumors, and you should see abscopal effects, and we are seeing abscopal effects, across all the tumors. So that antigen is then proliferated in all the tumors, and then you combine it with bempeg and you get that expansion of CD8+ T cells, and you should get a very significant tumor-killing effect. We -- that is the REVEAL study. Phase I/II is ongoing. Dose escalation is -- we haven't hit maximum tolerable dose yet. We expect to have that dose escalation to be completed this year. Now I'm going to move onto something that we haven't talked a lot about in the past but it's exceptionally important, just as important as bempeg actually, and the mechanism is just as important in I-O, and that's IL-15. And why is IL-15 so important? Well, it causes the proliferation of NK cells and the expansion of CD8 memory cells. Now I think everybody is coming around to the belief that NK cells are just as important as T cells in the I-O cascade. And what we're doing with NKTR-255 is very interesting. So it does expand NK cell number and function and it does expand memory T cell function. So if you think about it, if you can -- one of the problems associated with these great drugs, these ADCC antibodies, dara, RITUXAN, et cetera, fantastic drugs, but they deplete NK cells. And as they deplete NK cells, they stop working. So if you can combine a drug like IL-15, NKTR-255, that enhances the proliferation of NK cells, combine that with ADCC antibodies, you should have a very, very profound effect. I'm going to talk about what we're doing there. And in the area of CAR-T, CAR-T is also very effective but for a relatively short period of time. So by adding IL-15 and causing a proliferation of memory T cells, you should be able to get a much better durable duration of response associated with CAR-T therapy. I want to show you some preclinical data that led us to the combination with dara. So this is a model where dara doesn't work very well at all, and that's why we picked it. And as you can see, if you look at the NK cell count in bone marrow, that's where you need to look, you can see that when -- dara doesn't work very well in this model at all, but when you combine dara and NKTR-255, you get a significant increase in NK cell, in natural killer cell count in the bone marrow. Then in the middle box -- chart, you can see we're measuring human lymphoma cell count in bone marrow. And there, while dara doesn't have very much an effect in knocking down that lymphoma cell count, when you combine dara and NKTR 255, again you get a very pronounced reduction in lymphoma cell count. And the same thing trends -- goes right over to survival in the right box, where if you're looking at lymphoma model of survival, dara doesn't look that much better than the control. And again, that's why we picked this model because this is a model that dara doesn't have good results in, and you could see when you combine it with NKTR-255, you have a dramatic increase in survival. Now when you look at CAR-T, we did these experiments at the Fred Hutchinson Cancer Center, if you look at that first box, these are human CAR-T cells in mouse blood. If you look at CAR-T only, you could see over time the effect diminishes. When you add NKTR-255, IL-15 to it, you see a dramatic difference in terms of the duration of effect. And if you look at the right box where you're looking at tumor growth, you can see that with CAR-T only, the tumors eventually return. It's sort of flat line when you combine it with NKTR-255. So this was done by some of the pioneers in CAR-T therapy at Fred Hutch, and we're very excited about starting studies here as well. I want to show you something else. I'm going to jump back to NK cells for a minute. This is really, I think, extremely important. You have to induce intracellular granzyme B to support that apoptosis by the NK cells. So if you look at native IL-15, you get -- the purple boxes, you see what that looks like in terms of granzyme B secretion. And if you look at the mutein proteins that are being developed by other companies approaching IL-15, because they're muteins, they only bind to the beta-gamma receptors. They don't bind to all the receptors. And you get a very low level of granzyme B, which is actually critical to NK cell function. Now if you look at NKTR-255, which we've engineered to bind to all forms of the IL-15 receptor, you see actually maximal genzyme B production and even more so than the native protein. So this is really -- this is in human blood, and it's very exciting. I want to show you something else because IL-18 is just as important. You have to create IL-18 to activate NK cells. And if you look at the native protein, it has -- in the purple, that's the level of IL-18 expression you would see. In the mutein proteins, which only binds the alpha to the beta-gamma receptor, you see a much more limited effect on terms of IL-18. And if you look at NKTR-255, which again engages all forms of the beta -- of the IL-15 receptors, you see even more proliferation of IL-18 even compared to the native molecule. So also -- this was really exciting, and this tells us that if NK cell biology is critical, we have a way of activating it in a very important way. So what are we doing in the clinic? We started these studies. The first study is a dose escalation study of NKTR-255 in multiple myeloma and NHL patients in the refractory salvage setting. We have to find safety, of course, to get the recommended Phase II dose and then we will start. So that study has started. And then we move on to 3 cohorts with multiple myeloma and NHL as a single agent in salvage therapy. We move on to cohort -- we will also have cohort B, which is NKTR-255 plus dara, in multiple myeloma. And then we will move -- we will also initiate cohort C, which is non-Hodgkin's lymphoma, a combination of 255 and rituximab. So very exciting program. This should tell us whether we have an important molecule here, and I think we do. Another important program that we have running now is a collaboration with Janssen in a -- a preclinical research collaboration with 255 being studied in combination with a number of the programs in their pipeline. They're paying the entire cost of these preclinical studies. And I think it's another important way for us to see the potential for 255, especially with a company as successful as Janssen in the field of oncology. Now something else that's very interesting, and I talked about virology before, this is very different. We have a collaboration with Gilead in studying NKTR-255 in their antiviral models. And they're conducting -- at the moment, they're in the process of conducting nonhuman primate studies. They're paying for 100% of the cost of this. And this is to see if we can use NKTR-255, IL-15 against latent viruses, which can then be attacked by Gilead's antiviral programs. So again another very, very important application for an IL-15 molecule. If this works, it just could possibly change a number of things in virology. And I said Gilead is paying for 100% of the cost of this program. We retain rights to this molecule. So our NKTR-255 has not been licensed out, just to make it clear. Now I'm going to switch gears. I'm going to go into the polar opposite of NKTR-214 or bempeg. This is the opposite. While bempeg binds -- principally binds to beta-gamma to induce effector T cells, NKTR-358 stimulates regulatory T cells. And we're using -- we have a collaboration with Lilly to study NKTR-358 in various autoimmune diseases. Now this is human data that we presented at EULAR 2019. And as you can see, we were able to achieve a peak 15-fold increase in Tregs and the Treg/Tcon ratio. And what's important is it's also highly dose-dependent. So it gives you a great deal of comfort when you see this level of dose dependence. And I know Lilly is very, very excited about this program. We have running now a number of programs. So we will have data from a Phase I multiple ascending dose study in lupus patients to be presented this year at a major medical meeting. Lilly is initiating a Phase II study this year in lupus. There, Lilly has already initiated 2 additional Phase Ib studies in psoriasis and atopic dermatitis. And Lilly will start, they haven't said which one yet, but Lilly will start an additional Phase II study in a new autoimmune disease indication this year as well. They're running these programs all the way through registrational trials. Just to remind you of the economics, it was $150 million upfront. There's $250 million in development milestones. Maximum cost share to us is 25% Nektar, 75%. Lilly. And it has significant and I do mean significant double-digit royalties, and we have the option to co-promote. So lastly, I'd like to just share with you our milestones for this year. We ended 2019 with $1.6 billion in cash. For NKTR-181, we hopefully have approval and launch and through our subsidiary that we created, Inheris. And like I said, the ADCOM will be tomorrow, so stay tuned. In bempeg, we have -- we will start 2 new registrational trials with bempeg plus nivo. There'll be additional PIVOT-02 data updates. The first potential data from the PIVOT IO 001 study, that's the Phase III study in metastatic melanoma, should be -- we should -- as I said, we should have response rate data at the end of this year, with PFS data middle of next year. And the initial PROPEL data, which is the combination of bempeg with pembro in first-line non-small cell lung cancer, we should have the initial 20 patients of data this year. With NKTR-358, start of the Phase II study in moderate to severe lupus patients, start of the second Phase II study in a new autoimmune disease, and data from the multiple ascending dose study will be presented. Regarding NKTR-262, we should have data from the dose escalation portion of the REVEAL trial, which is 262 plus bempeg. And lastly, with NKTR 255, we should have the first clinical data in the Phase I study in patients with non-Hodgkin's lymphoma and multiple myeloma and then, of course, potential preclinical data that comes from the Gilead and Janssen collaboration. So that is a -- I think that's a very broad portfolio. I think we're in great shape to succeed. I will tell you that, of course, all the issues that were surrounding the Bristol-Myers collaboration and the concerns about bempeg have been put behind us. It's all wrapped up. And I think overall, we're in a great position to succeed. And with $1.6 billion in cash, we certainly have the funds to get there. So thank you very much for joining me today. Appreciate it.

Do you want -- Howard Robin, CEO of Nektar. Gil, do you want to introduce...

Gil Labrucherie, the COO and CFO of Nektar.

Wei Lin, Head of Development at Nektar Therapeutics.

Good. We have some other guests here too, which hopefully will collaborate. Okay.

Maybe just heading into the ADCOM tomorrow. What was your kind of reaction to the FDA briefing documents released last week?

Well, I think, look, the briefing documents are always somewhat sterile in terms of how they approach things. And of course, I think one of the -- I think the only question, of course, is should this be approved for chronic lower back pain? Should it be approved for all pain? Or should it not be approved? That's the question in front of the ADCOM. I think, of course, human abuse potential will become a major part of that. And as I said in my presentation, look, we've had this in 2,200 patients and healthy volunteers. And while at doses that are considerably higher than the therapeutic dose, you do see some evidence of euphoria. After all, it is an opioid. It's a potent opioid. You'd have to go 3 to 12x the therapeutic dose to start seeing euphoria. And that would be euphoria that's equivalent to a low dose or a middle dose of oxycodone. At the therapeutic dose range for this drug, 100 to 400 milligrams, it didn't look any different than placebo in terms of euphoria or likability. So yes, if you want to take a half bottle of pills or something, yes, I think you might start to see some euphoria. But remember, there's another component here. The onset of action is 2 to 3 hours. This is for chronic pain. People typically don't take opioids to get high 3 hours down the road. And of course, it is a new molecule, and it is a molecule, it's not a formulation, which means you can't do anything to it. You can't alter it chemically to free it up and make it a more potent opioid. You could denature the molecule, but the bonds between the polymer conjugates in the opioid is stronger than the opioid bonds themselves. So you could ruin the opioid, but you're not going to free it up from the polymer conjugate chemistry that's involved with it. So look, I don't know how the FDA is going to go. I think they do see it as -- I do think they want to approve it. I do think they see it as a very important stop -- start in dealing with the opioid crisis. Look, many people become addicted -- they're not trying to abuse the drug. They take the opioid, they take their therapeutic -- their honest therapeutic dose because they have back pain or some other pain. They take their therapeutic dose and they find that it causes euphoria and it starts to become very likable. And you don't see that with the -- you don't see that at all with the therapeutic doses of NKTR-181. Like I said, if you want to take 5x, 8x the therapeutic dose, you want to take 10 pills instead of 2 pills, yes, I guess, you could start to become euphoric. It is an opioid after all. But that's not what's causing the opioid crisis in this country. The opioid crisis started with patients who need drugs for their pain, and there are no other -- listen, there's nothing else coming. There's no other pain drugs in development that are going to change the opioid crisis. This is the only one. So patients have chronic lower back pain, they need something, they want to take the fair -- they want to take an honest therapeutic dose and they don't want to experience that euphoria. They want to go to work and not be euphoric or dysphoric. So I think we have a reasonable chance. I think it's a very important program. I think the FDA does want to see it through. And we'll find out tomorrow what the ADCOM thinks. Must be some other questions. I covered everything that well, huh?

Can you talk about the background of the talks with Bristol leading up to your update last week on what the [ civil ] trials could be?

Sure. Look, I think the program got a little bogged down through the last year for a number of reasons, basically 2 reasons. One is Bristol acquired Celgene, and that certainly had a tremendous expense associated with it. And they're looking at how they want to apply their funding to various programs, which they now have a number of additional programs that they have to fund. And I think that certainly gave them some pause. It wasn't -- I wasn't really too happy about that, but I also understand it. In terms of -- and then you combine that with the manufacturing problem that we had. It certainly gave BMS reason to think about how they should proceed. Now the manufacturing problem, as you all know, was back in 2016. It was a -- just for reference, it was a single bad lot -- or bad batch of an intermediate that got into 2 lots of the drug. We made 22 lots of bempeg. Two lots got this single bad batch of intermediate. And at that time, we didn't know it because the assays weren't developed then to be able to identify and characterize the molecule as well as it is now. And as we were getting ready for commercial production, we put a lot of work into developing new assays for this very, very unique molecule. Now those 2 -- those -- that batch of intermediate that passed assays back in 2016 would never pass today. And BMS put a lot of time into auditing Nektar and doing -- in essence, redoing diligence on our programs, both in terms of manufacturing and in the clinical, to make sure that they were completely comfortable. Now you know we've met with the -- as soon as we found this, we took it to the FDA. The FDA has been very comfortable with it. The FDA is comfortable with now our new assay methodology and our new manufacturing processes, and so is BMS. So I think that issue is completely behind us now. And you could -- as I said, if you look at that -- I said in my presentation, if you look at the new clinical plan, their share of that clinical plan in terms of development expense is $1.2 billion. You don't jump into that unless you feel comfortable.

And what about the decision to do more work in lung cancer but not start a pivotal? I mean how did that [ come up ]?

Well, look, the -- to a certain extent, some of the manufacturing differences confounded things. But lung cancer is also a much less immunogenic type cancer. So where you see -- where you can use doses of 0.006 with things like metastatic melanoma or bladder, in lung cancer, the results weren't as -- as we saw in REVEAL from -- and I shouldn't say REVEAL, that's a different thing, as we saw -- as we observed in PROPEL weren't as compelling. And we said, look, rather than get it wrong, let's reexplore. And we're doing the same thing in the PROPEL study with pembro. Let's explore higher doses. Let's go to 0.008. Let's go to 0.010 mg/kg. And let's see if the side effects are tolerable. I think they will be. There'll be some more side effects, but I think they'll all be acceptable within a normal range for I-O drugs. And let's see if we can get a more profound effect on tumors in that setting. So we're doing that dose expansion and dose escalation in the -- with nivo, and they're paying for that 100 -- they're doing that study 100%. And then we are also doing the same kind of thing in our study with bempeg and pembro. And I think, in the end, there's not a lot -- there's very few -- there's no other opportunities for I-O in lung cancer. So while this might be a little bit further delayed than we would like, if it works, it should be a much better effect than you would see with the combination of checkpoint and chemo. I mean what else is there? Checkpoint and chemo. So I think in lung cancer, let's see if we could eliminate the chemo, and let's see if higher doses of bempeg made from batches of drug that were not problematic, let's -- and let's also combine it with pembro because pembro has a great track record in lung cancer. And let's see if there's a difference there as well, pembro versus nivo. So I think we have both studies running, and we're very excited about that. And clearly, if we see good results with nivo, I mean, BMS is our partner. That's how we're going to move forward.

I think -- and the other thing I would like to add is we certainly have dosed 400 or 500 patients with bempeg. And along the way, we've learned a lot about how to manage the safety profile of the drug. I think while we have full confidence that the 6-microgram dose that's going through registration in combination with nivolumab across multiple indications, we also think that with the improved safety management, we could actually probably dose potentially bempeg at a higher dose and perhaps potentially improve the efficacy. That's another reason why we thought it's worthwhile at this stage in the program to see if we can actually increase the dose.

It seems like in PIVOT, some of those first-line lung cancer cohorts were taking a while to enroll. So what's going to change with this Phase I/II that Bristol is running as it relates to their ability to enroll first-line patients [indiscernible] combination?

For the -- so the first-line lung small cell in combination with nivolumab, that will be operationalized by Bristol. BMS have a depth of experience in lung cancer. They're the first one to get a label in second-line lung cancer. So they certainly have conducted numerous trials, registration or otherwise, in non-small cell lung cancer. So I think we have full confidence they have the experience to do that. And then I think for us, on our part, our PROPEL study, the backbones goes down on top of pembrolizumab. That's the accepted standard of care as monotherapy in lung cancer. So we also think that engaging investigators and talking through with a lot of sites, there's greater enthusiasm for doing that combination as well. So we think both studies will be able to enroll quite well.

And look, I think there's a general belief that IL-2 is a very important pathway here because no one wants to stick it out just with chemo. And I think they believe IL-2 is a very important pathway. I think we all believe that. And the fact that we've had now breakthrough designation in metastatic melanoma and such particularly good results, I think people -- I think investigators feel that if we up the dose and we go to the 0.008 and 0.010, we could hopefully see those same kind of results in lung cancer. So that -- I think there's been a pretty -- I don't think we're having any difficulty enrolling the pembro combination study. And we'll start the nivo combination -- BMS will start the nivo combination study.

I think you reiterated the time line for the first efficacy look from the melanoma Phase III coming in the fourth quarter. That's it for interim, right? What's the potential to turn that into a CR? And would you need to get the FDA's feedback that they will be comfortable using CR as a basis for an accelerated approval before you make a switch like that?

Yes. So we do plan to engage with the agency regarding the endpoint. And I think this is something that the team is actually working on. I think some of the people in the audience, [ interestingly, we -- ] many of us will go to the FDA with the breakthrough designation. We do have an open door policy with the FDA to talk about that. I think -- we certainly think that's a worthwhile discussion to have given the FDA analysis. I think also, it's also the probably the -- from our point of view, looking at the FDA data, that's probably more translatable to survival than any other response criteria. And obviously, in the end, it's going to be FDA's feedback that's going to give us guidance about changing the endpoint. As it stands today, it's still ORR. And based on the FDA feedback, then we may consider changing to a CR rate as the first. And so that study really has 3 approval endpoints built in: right now a response, followed by a PFS about 6 months later and followed by overall survival. So we have 3 shots at goal. And if any of these criteria crosses, then obviously, we have an opportunity to file for approval.

Interestingly enough, the data that we're presenting on metastatic melanoma, all the data that I've shown you in PIVOT-02, that's a combination also with lots 2 and 5, which were problematic. The response rate go -- the complete response rate goes up into the 40s if you pull out lot 2 and 5 and just look at lot 1 and 3. So I think there's a lot of potential here to get to an overall response rate, or even as Wei said, talk to the FDA about using complete response as an endpoint. Because what's the complete response with nivo? It's -- I think it's like 7%, 8%, 9%. And we've seen 34%. And if you pull out the bad lots, it's 42% or 43%. That's a tremendous difference. So we'll see. That's -- it's something we're seriously considering.

If for the interim you did switch to one based on CR, how do you estimate the time line for when you might hit that kind of evaluation point?

Yes. Right now I think the time line may not change significantly, but some of that have to be based on the FDA discussion. Because in addition to the endpoint changing, we also have to assess how long a duration of follow-up should we have afterwards. Right now the median time to CR is about 7 months. And then through discussion with the FDA, we can have our understanding how long a follow-up is that meaningful in delivering a data package to the FDA to say that, okay, now we can assess the duration of response for the patient who experienced CR as well as you're capturing enough patients who have achieved CR to really have a meaningful look to the comparator, which is nivolumab in the study. So there are a number of criteria that's going to be involved in the FDA discussion. And what the time line moves like, will really depend on those discussions.

But we spent very little alpha on that. So most of the study is powered to get you a PFS response by the middle of next year. We did allocate some alpha spend to getting the ORR or even CR, but the primary endpoint of the trial was PFS, right.

So the trial doesn't stop on a CR interim. It's not necessarily a bad sign for PFS.

No. I mean if a CR interim can get you an accelerated approval, of course, that would be great. But we're not going to stop -- that's not going to stop the trial because the primary endpoint is PFS.

You say the study is only powered for PFS?

The study is actually powered for all 3 endpoints. So there's alpha spend for ORR or CR, followed by PFS, followed by OS. So with those statistical powering, we can actually make claims and hence file for approval for any of those endpoints.

And this is why we started the adjuvant melanoma study. We're going to be starting adjuvant this year. That's a much, much larger market than metastatic. I think if it works well in metastatic, there's no reason to believe it's not going to work well as adjuvant therapy. And that dramatically expands the market, so we're excited about that also in combination with nivo.

What's the latest thinking on time lines for readouts for the pivotal study in bladder cancer and the first-line RCC you filed?

Yes. So for the bladder study, that's also going on. We have just recently converted the study design from a 2-arm study to a single-arm study. So that's also response-based. And then there's a conversion strategy, which is using muscle-invasive bladder cancer, which is going to be a randomized study to convert that from accelerated approval in the first-line setting to a full approval with both MIBC as well as the first-line metastatic. So that study, because of change in the study design and also increase in the sample size, it used to be 165, now it's 205-patient study, so time line has been extended. We expect there will probably be some readout sometime in 2021.

And what's the response rate that you need to rule out [indiscernible]?

Yes. So it's not going to be purely based on just some risk response benchmarking against something else. Because if you keep in mind that historically, when both atezol as well as pembrolizumab were approved in the first-line setting based on accelerated approval criteria in exactly the same indication we're trying to pursue, they numerically did not improve upon the 36% response rate of gem/carbo. But because it's a combination of response as well as durability of response, that's what basis for the FDA approval as well. So I'm not sure there's a absolute benchmark. And in our interaction with the FDA, what their feedback was they're going to look at the totality of data, which a combination of safety profile of our combination versus the chemo as well as the response rate and as well as the durability of the response. I think at least for a number of the criteria, the safety profile as well as the durability, we have pretty good confidence that our combination should be superior. And then if we deliver a fairly reasonable response rate, comparable to what will be actually presented publicly, I think that's a reasonable [ sure thing ] approach to get accelerated approval in the first-line indication.

Can you share a strategy or your aspiration for 255 or IL-15 trials, especially from the perspective of research collaboration and development collaboration?

Well, look, I think -- yes, I think 255 is potentially just as important as bempeg in terms of IL-15 being an important mechanism in the I-O cycle. I have no plans to out-license that or partner that at this point, no plans at all. We have enough cash to take that program forward in the way we're doing it. The collaboration that we have with Janssen is a preclinical research collaboration to explore where the possibilities are, how it works with a number of their programs that are in development. And they own their molecules, we own our molecule. The collaboration we had with Gilead as an antiviral agent, which is really exciting, and we'll see what comes out of the nonhuman primate studies. And only they can do those studies. Obviously, we can't do those studies. They're funding 100% of that. That has -- we did say to them, if we ever partner it out as an antiviral agent, they get first right in negotiation, but I really don't see us doing that. I think they have their compounds. We have our compounds. In the field of immuno-oncology where it has the most relevance in terms of NK cell proliferation as well as CAR-T enhancement, I mean, strategically, there's no reason to partner it. I mean quite frankly, it's very easy for us to do studies in combination with ADCC antibodies. I mean that is -- that's not that complex, and those studies are pretty straightforward. And you think about dara and RITUXAN, I mean, there's so many really good ADCC antibodies. There's a range of programs there that could all benefit from NK cell proliferation. And in CAR-T, the same thing. I mean it's very simple to combine it with CAR-T, and we've been doing a lot of work at the Hutch to do that. And I don't see at this point a BMS collaboration for 255. I don't see it. I think it's something we can do ourselves and take it to the next level for sure. It doesn't mean there's no interest.

I'm sorry, what's the timing for the Phase III RCC trial?

The RCC trial is actually a survival endpoint, although we're going to look at the response rate early on. But really, it's a survival endpoint. So it will be -- the study is actually enrolling. So it's event-driven. So right now as we get closer, we'll be -- project that accurately. So that's probably at least 3 years out.

But what's very exciting is metastatic melanoma with PFS data in the middle of next year. With breakthrough designation, it's a potential approval at the end of '21. That's what's exciting.

Yes. And with the initiation of an adjuvant study, I think then we really hope and have confidence that bempeg plus nivolumab can be established as a standard of care across both early as well as metastatic disease melanoma.

Okay. There are no other questions? Thank you very much for coming. Appreciate it.

Thank you.