Zai Lab Limited (ZLAB) Earnings Call Transcript & Summary

All right. Okay. Great. Thank you. I'm Yigal Nochomovitz, biotech analyst here at Citi in New York. This is the second day of our Virtual Oncology Summit. It's my pleasure to have with me from Zai Lab, Rafael Amado, who is the Head of Research and Development at the company. So welcome, Rafael. Thank you for doing this. For those listening, if you have questions for the company, just e-mail me and then I can relay those over to Rafael. So maybe just to start out, if you could kind of just give a bit of a high-level overview of the Zai Lab and how you're transitioning from a local China business to more of a global enterprise as you build out your pipeline.

Yes. Thank you for the opportunity to chat with you. The company was founded in 2014 to bring medicines to patients in China, and that engine continues and is commercially profitable. But more and more, we are focused on bringing in global products, and we have our own engine to discover either protein therapeutics, translational medicine and everything that's required to do drug discovery as well as business development. And in the recent past, we've actually moved the global pipeline together with the local regional very effectively. Our lead product is Zoci, which is a DLL3 ADC, which we can talk about, is in Phase III right now. And we've had INDs that have moved really fast to the clinic like ZL-1503 in atopic dermatitis; LRRC15, which is an ADC, which is in Phase I; immunocytokines, PD-1, IL-12; T-cell engagers. And so what we're doing is take advantage a bit of our global capabilities that we have had to build, but on the back of a very efficient engine in China that was delivering products for the local regional approvals, and they are now a strong component of a way for us to screen these global products. And also if we see that there's activity for China to participate and gives us access to patients. The local regional business continues and VYVGART is a long-term opportunity with multiple indications. We're launching COBENFY this year, TIVDAK as well, and we are filing TTF or Tumor Treating Fields in pancreatic cancer. So it is really a dual engine. The last one has been built more recently, but it's progressing quite fast. And much of it is at the expense of having formed a really strong team and also having the China capabilities for access to patients.

Okay. Well, let's start with the lead asset, Zoci, the DLL3 ADC, which is getting a lot of attention, as you know. So maybe if you could just kind of characterize the activity that you've seen with this asset so far. What is -- what are the areas that you'd like to comment on in terms of where it's got the most potential, in small cell lung cancer and perhaps in some other settings as well where DLL3 is widely expressed.

Yes. So DLL3 antibody drug conjugate utilizes a cleavable linker. And of course, we started in the tumor -- in the neuroendocrine tumor where DLL3 is most expressed, which is small cell lung cancer. And we have a very ambitious program there. We started our Phase III study, and that was on the heels of a very large Phase I/II study that now has over 150 patients that showed us that the response rate at the dose that we chose for development is in the order of 68% with durability across all the doses and including doses above 2 milligrams per kilogram of over 6 months. So this was in multiple lines of therapy, second, third line, et cetera. We then -- we did a lot of careful dose finding under Project Optimus, comparing various doses, and we landed on 1.6 because it was the best balance of efficacy and safety. So it's probably one of the safest ADCs. There's about 13% grade 3 toxicity. There's no drug discontinuation. And then some of the properties to highlight is that it's incredibly active in the brain. We are looking at this very carefully, but patients tend to respond after the first or second dose at an 80% rate, which is a big problem, brain metastasis in small cell lung cancer. So in addition to that, because of this, we want to move the molecule to first-line in neuroendocrine carcinoma, where we're seeing encouraging data as well and also to combine it with other products like T-cell engagers in addition to checkpoint inhibitors. So those are some of the properties of this product, and we're already enrolling in second line after platinum and after tarlatamab, and we plan to start the frontline studies and the NEC study this year.

Okay. So the first study would be for an accelerated approval. Is that in the second line? Is that how you're thinking about it?

The way it's designed is it's an overall survival study with an interim analysis for response which could lead to an accelerated approval. So all patients need to be accrued before we do that analysis, and we think that they will be accrued by March or so, about a year from now. And then we will do the response comparison. It's not just response, it's also durability. And then if the accelerated approval occurs, then it will continue on for overall survival for confirmation. So the interim analysis is positive. We'll file in 2027 with the PDUFA time line somewhere in 2028.

This would be -- you said you got about a 68% overall response rate in the data you've seen so far. So this would be an accelerated approval based on an ORR response rate. Is that how you're thinking about it?

That's correct. So it would be based on an ORR response rate compared to the standards that are available worldwide. This is a global study, just like our initial Phase I study was a global study. And we will include chemotherapy amrubicin because that's something that's used in Japan, lurbinectedin and topotecan. And we're including, again, patients that have failed tarlatamab. So that's why we're not including tarlatamab in the control arm. So it's a comparison to second line and the patients -- I'm sorry, to chemotherapy and the patients could be second line or may have received tarlatamab as well.

What would you be expecting for the control arm, blending all the topotecan and lurbinectedin and amrubicin, what would be expected there from response...

So response rate in the mid-20s or so.

Okay. So it sounds like you have a pretty good delta already built in if you reproduce what you've seen in the dose escalation part.

Yes. As a matter of fact, we wouldn't need that many patients. It's just that we have to have all the patients accrued. Otherwise, there will be no [indiscernible] after the approval to continue the study. And also, we need to build a sufficiently large database to be able to file. So the study is, as you allude to, overpower.

Okay. And then assuming you get the accelerated approval, then you would -- then it would be fully approved on OS?

Yes, the study will continue blinded and we will follow patients for overall survival. Of course, we don't control post-progression therapy. So patients that have not received tarlatamab would receive tarlatamab or lurbinectedin or any other therapy. But we think that it would be distributed randomly. The only thing we're doing is stratifying for tarlatamab or no tarlatamab in the patients that are enrolled so that we have an even number in both arms.

Okay. And are there other considerations, for example, to have another endpoint that would look at just the brain mets because you mentioned that the response there was also very, very high. Is that something that you would -- has that been built into the study?

Yes. It's been built into the study. It's not a specific endpoint because the response rate in the brain is difficult to categorize as a regulatory endpoint. But I must say that it's something that investigators really have -- really capitalized on because unlike other studies, we decided to enroll patients that have brain metastases that were untreated because a lot of these patients present with very bulky disease and they progress very fast. So, say, if they have to stop for brachytherapy or some other -- Gamma Knife or something else, their systemic disease would just move quite fast and the patient would be debilitated to receive chemotherapy. So we actually have higher response rates in patients that have untreated brain mets. Now we will look at that in the statistical analysis plan, and we will report on that. And there is a lot of interest in this, but it won't be an "approvable" endpoint. The endpoint will be overall response by RECIST.

Okay. And then you mentioned -- so then the frontline study, how would that work? You're going to go up against like tarlatamab? Or how would you do that? What would be the plan?

Yes. So tarlatamab has some studies in first line in combination with chemotherapy, also study in so-called maintenance. There are some examples of ADCs that combined with checkpoint inhibitors have been able to supplant chemotherapy. And we hope that this is one of those examples. So to that end, in the Phase I/II study that we started and we've added several cohorts. One of the cohorts that we added was first-line patients treated with 1310 plus atezolizumab or that doublet plus carboplatin. And then we yet need to analyze that data for durability and then make a decision as to what the regimen will be. The control will be very obvious. It will be the current standard, which is platinum etoposide and a checkpoint inhibitor. We don't think that by the time that we start this field -- this study, the field would have changed so dramatically. So at the moment, our goal is to see enough durability and high enough response rate to be able to supplant chemotherapy in the front line.

Okay. So that you said you would combine with like -- did you say atezolizumab, a PD-1?

A PD-1. It will probably be the durvalumab or atezolizumab. They are both used at the preference of the investigators.

Okay. And when could you foresee that once -- that first-line study starting?

So thankfully, we don't have good durability yet because patients are staying on study. So that data is going to come sometime -- mature data is going to come sometime in the second half of the year. So we plan to present other data along the year, but this will come later on. But we obviously will know what the data tells us in the beginning of the second half and then make a decision then as to whether to proceed. There are other possibilities, which is combinations with T-cell engagers and so on, but there is insufficient data to really be able to launch a Phase III study at the moment. So if we do more than one, our first one would be chemo-sparing starting in the second half of the year.

Okay. And these studies, the second-line study as well as the contemplated first-line trial, these are global studies, right? Where are they enrolling, in which geographies?

Yes. So we decided to do global studies even from the inception because we think that, that represents real-world data. I mean we've seen other products do trials in China. And we know by experience that there are differences between regions. So the study will take place in Japan, in Korea, in China, in Europe and obviously, in the United States, in Turkey and some other countries. So it will be a truly global trial.

Okay. That makes sense. Okay. And in terms of the competitive landscape because there are a few other DLL3 ADCs out there, as you know, and there are some that are not ADCs, but they're different like T-cell engagers. How do you see your asset in this competitive landscape? You mentioned you have a good dose. You picked the 1.6, which we felt was a good mix of balance, safety and tolerability with efficacy.

Yes, exactly. I mean we've seen other payload linkers that are similar to this one that we are employing. And the doses are slightly higher. Of course, it has to do also with the affinity of the antibody, whether it is internalized or not, depending on the target, the half-life, et cetera. But 1.6 is a relatively modest dose as opposed to 2, 2.4, et cetera. So as a consequence, we don't have a lot of myelosuppression. We don't have grade 3 or above ILD and the combinability should be easier. I don't see these products competing with T-cell engagers. T-cell engagers tend to require higher target expression, and the ADCs tend to work really well in bulky disease in spite of low expression because the payload and the linker gets cleaved in the tumor microenvironment. So even cells that are bystander can be killed by the payload, which in our case, is a TOPO1 inhibitor. Doing that will expose immunogenic death and new epitopes that having been attracted the T cells to the tumor bed, they could potentially react to. So we view this as orthogonal mechanisms of action. With regards to other ADCs, there are some that are more advanced, perhaps the [ IL-1 ] is the most advanced one, but we have at least a year ahead of them. And so far, their data comes from China is -- the follow-up is slightly lower. And we'll just have to see how all these drugs play in the treatment of small cell lung cancer, a lot of it is going to have to do with how they're developed, how they're combined and how they're positioned, what line of therapy. And so we are very focused on our product and moving it as fast as possible.

Okay. So let me see. So there are a few other assets that are worth discussing. I'm not sure which one you want to mention next, but sticking with the ADCs, you have the LRRC15 ADC. This is the 6201. What -- tell us more about that target and what -- where you would go with that molecule?

Yes. So this is a Leucine-containing molecule that is expressed in tumors that have mesenchymal origin, namely sarcomas, some neuroblastomas, some melanomas and it's highly expressed in osteosarcoma as well. But it so happened that it's also expressed in fibrous tumors that contain fibroblasts. And we have preclinical data with several models that have no expression of the target in the tumor cells, but it's expressed in the so-called CAFs or cancer-associated fibroblast. And because the payload and linker get cleaved -- that tumor microenvironment, there can be influx and efflux of the payload into cancer cells that don't have the target. So in PDX models, we see activity not only in tumor -- marker-positive tumor, but also in marker-negative tumors. And of those breast cancer, pancreatic cancer, head and neck, lung cancer, those are the ones that tend to have the highest expression in fibroblast. So our study is dividing the cohorts into patients that are tumor negative but positive and also a group of sarcoma patients. So I think it's a pretty exciting drug. I mean this target has already been used. There was another company that developed -- that discovered an ADC and there were -- there was some activity, but it was just too toxic and the development of it was abandoned. So there is a chance that it may work on doxorubicin failures in sarcoma and ifosfamide. And if it works in tumor negative, I think it will prove a very sort of pioneering event in the field of ADC, which is that you don't actually need to have the target in the tumor for these drugs to actually have activity.

Okay. And that one, I understand that just started or is starting this quarter.

It just started. Yes, we got in January, the May Proceed letter from FDA, and we are opening sites and recruiting patients. And we filed in China, the CTA. And once we get sites open there, we obviously limit the number of patients, but we start getting accrual relatively fast. I must say there's been a great interest in the sarcoma community, both from Europe and the United States on this asset. So I'm looking forward to seeing the results.

Okay. And then the other oncology assets that are coming up that are going to enter the clinic this year are the ZL-1222, this is the PD-1, IL-12 and then the ZL-1311, which is the MUC17, CD3. What can you -- which one of those would you like to highlight as you're most excited about?

Yes. So in oncology, we're working on ADCs and innovation of ADCs. So we will proceed with dual payloads, biparatopics and some other modalities. We also want to enrich for tumor -- for T-cell engagers as well. And so we started with MUC17. That's a pretty exciting molecule, at least preclinically. It targets two epitopes. It also has silencing of -- not complete silencing, but attenuation of the CD3 moiety so that perhaps CRS is not as pronounced as it can be with the typical bispecifics. So this is for gastric and gastroesophageal junction and other GI tumors. So that will be an IND this year. And then we have this immunocytokines like PD-1, IL-12. This has been a long time coming being able to give IL-12. IL-12 is a very potent CD8 cell stimulant. And we have data that PD-1-resistant tumors actually can recover sensitivity when exposed to IL-12. So what we did is we have dual chain PD-1 in the N-terminals and then we constructed IL-12 [ mutant, ] which is obviously alter the amino acids to signal in an attenuated way. And so it doesn't signal in the periphery and doesn't cause all the toxicity of IL-12. It only signals in the tumor microenvironment in CD8 cells. And we have done some elegant studies where tumors regress, but if we actually deplete a CD8, the tumors grow and it seems to be mediated by CD8 as well as NK cells. So it will be a product that we'll probably test in PD-1 failures. And then from there, it could move into other tumors where checkpoint inhibitors are the mainstay of therapy and see whether they're better, just like the whole field of PD-1, VEGF.

Okay. And this one has -- for both 1222 and 1311, you decided to do the Fc silencing. Is that right?

Yes. So this is -- we mutated Fc to increase the half-life in this product, yes.

Okay. And I'm not familiar with other assets that have these combos like the PD-1, IL-12 and the MUC17, CD3. Are these unique as far as you know, in the landscape?

At least they have innovative properties. I mean the other changes that are being done to bispecifics is introducing costimulatory molecules. They may stimulate CD8s with more potency and perhaps derive a more juvenile kind of population because T-cell exhaustion is one of the mechanisms of resistance to bispecifics. We perhaps may get into this area with either 4-1BB or CD28. And there is a lot of innovation in this space. And we can access companies, not just in the U.S., but also in China, where there's a lot of know-how -- technical know-how in both small molecules and protein sciences. So the same thing with ADCs where now dual payloads are coming into play biparatopics, more than one epitope and then payloads that are small molecules that are not even chemotherapy that can be targeted to specific tumors. So these are some of the innovation in this field that we would like to be a part of and at least be able to test them and see whether we have go, no-go decisions in patients potentially in Asia where we have much more access than in the U.S.

I just remembered, though, there is another PD-1, IL-12 from Innovent, right? How is yours different? Or is it different?

Yes, it's a larger molecule. As far as we have been able to tell, the IL-12 signaling is really, really attenuated. There's hardly any IL-12 stimulation. Time will tell in terms of how it behaves in the clinic. But dialing this requires very careful mutation of amino acids because you don't want it to be too promiscuous, but at the same time, it has to simulate the CD8 cell. So our impression is that the IL-12 simulation of our product is more adequate than the Innovent one, but we will have to prove this in the clinic.

Okay. And then moving away from oncology, you have the IL-13, IL-31 receptor bispecific for atopic derm. And I think you're going to have some data in the healthy volunteers coming up this year. So what do you need to see on that readout to move forward and prepare for a Phase II?

Yes. We're very excited about this product. In preclinical nonhuman primates, we see inhibition of pruritus immediately and it tracks with a single dose, it tracks with the half-life. The same with phospho STAT, it gets abrogated for the duration of the PK of the product. So we are testing now the same thing in the PK and PD in healthy volunteers in single doses. We launched that IND last year, and we've already accrued two of the cohorts of the healthy volunteers. And then we're moving into multiple ascending doses in patients with atopic dermatitis. In terms of what we need to see is abrogation of scratching, which may lead to a faster healing of the lesions mediated by IL-13 because of mechanical damage of the skin, and also a half-life that allows extended dosing by patients. And of course, if we can get lack of conjunctivitis or some of the side effects that are seen with Dupi, that would obviously be a plus. We continue to generate preclinical data in other Th2-mediated diseases like asthma, rhinitis, prurigo nodularis and eosinophilic esophagitis, et cetera. We will be publishing those data, but we're starting in atopic derm and the patient cohort, the multiple ascending dose, we should have actually efficacy data in addition to PK/PD in the second half of this year.

Okay. And then just, I guess, more generally, as you think about the global development pipeline across oncology, the assets we talked about and then the autoimmune, what is Zai Lab's plans as far as development and maybe even commercialization? You're going to do this as an independent organization? Or would you maybe think that you need partners in other territories to help market the products? Or how do you think about that?

Yes, that's an excellent question. Obviously, for Zoci, we want to have a very broad development plan, and we are pursuing this, as I described before. Obviously, it will depend on what kind of activity we see with the other products. And we will file at least one IND every year and perhaps bring one product from business development. I think in oncology, we can prosecute these programs. In non-oncology, it may be a little bit harder for us to do it alone. And we will be open to collaborate with companies that have been in this space and have done multiple indications in the past. But we think that before doing that, even though we do have conversations, we ought to characterize the product better, not only just in terms of getting INDs, but also getting subcutaneous formulations that are going to allow for self-injection and as well as good PK, good PD and efficacy. So I think we will probably take this to Phase IIb, 1503 and then take it from there. But yes, we see ourselves as strong R&D engine that has a commercial branch in China. And eventually, we'll launch products in the United States and we will be a global company with multiple assets, mostly in oncology and immunology.

So if you launch -- I mean, we're talking far in the future here, I guess, maybe not so far in the future. But if you launch DLL3 and you launch the PD-1, IL-2 (sic) [ IL-12 ] and the MUC17, if those all launch in the United States, get the atopic derm product to the market, would you still -- you'd have them in Europe and the United States. Would they be in China, too, or you wouldn't launch them in China, or you don't know. It depends?

I mean it would be logical to retain rights in China, but I think every product will be treated on its own merits and depending on the resources that we have. I mean, they are finite, both in terms of people and personnel and our ability to fund all these programs. But at the moment, every product that we are developing, we have a development plan for China and now Japan as well. So there's a group in China that is taking care of Asia Pacific. But if a product gets out-licensed, then those would be discussions to be had, whether this is global rights or whether anything gets carved out. I mean I think that will be done on a one-to-one basis.

You mentioned -- I just wanted to ask one specific question. On Optune, you said you're going to be filing for the Tumor Treating Fields in pancreatic, yes?

That's correct. Yes. I mean the file is ready. We obtain Innovative Medical Device, which allows us to have a 9-month review. And as you know, Novocure obtained FDA approval. So we have the dossier ready and we expect approval to be happening this year.

And then -- but you still are selling it for GBM, right?

Yes. Yes, we are selling it for GBM. That's correct.

And -- but as far as I understand, for non-small cell lung cancer, you decided to not pursue that, correct? Is that right?

Yes, that's correct. I think for non-small cell lung cancer, there was insufficient interest from the technical perspective, business perspective and opportunity costs that we didn't launch LUNAR-2. Lung cancer is a difficult -- more difficult indication and pancreatic cancer is a stronger unmet medical need. We also have constraints commercially and we -- and development-wise. So we decided to wait for PANOVA and likely, it was a positive trial. So we're pretty excited about this.

We didn't talk about an important asset with in kidney disease. So let's -- in the last few minutes, few minutes, I guess, pove, povetacicept. So that's an important product. Can you briefly just describe the market for IgAN there in China? And how could that asset be sold in China?

Yes. So povetacicept is a BAFF-APRIL targeting agent. So it essentially inhibits the secretion of antibodies including IgA. The data -- the Phase II data is actually quite impressive. The market is pretty large. It's over 3 million patients with IgAN. They get treated normally with blood pressure medication like renin-angiotensin-aldosterone system sort of inhibitors, diuretics. And the problem is when they become nephrotic and eventually 20% to 30 or so patients with time develop a decrease in renal function to the point that they can end up in dialysis. So it is an important medical problem. And our partner has moved pretty fast to try to get approval based on an interim analysis, and we are having discussions with CDE to do likewise and then continue to follow the patients for a longer time for a full approval. So the endpoints are proteinuria and maintenance of renal function. And those discussions are taking place now with our partner, Vertex and FDA and us and CDE.

Okay. Well, thanks, Rafael. Appreciate it. A lot of interesting assets, a lot of novel biology, and we're looking forward to seeing how it all shakes out. So thank you.

Thank you so much, Yigal.

Appreciate it.

Take care.