MacroGenics, Inc. (MGNX) Earnings Call Transcript & Summary

Hello, everyone. Thanks for joining us. My name is Jonathan Chang. I'm part of the SVB Leerink equity research team. It's my pleasure to host the management team of MacroGenics. And we have with us today, President and CEO, Scott Koenig. For the investors joining us, feel free to submit questions or e-mail them to me at [email protected]. So let's get started. Scott, would you like to briefly introduce the company?

Thanks, Jonathan. Thanks for the opportunity to present today. This is an exciting time for MacroGenics. As you know, we had a great 2020, and we expect a very exciting 2021 as well. The company is about 20 years old. We're focused on immune-based therapeutics. We are a fully integrated company that does everything from early discovery, and we're about to launch our first molecule. The company has developed 3 proprietary technologies, one around the Fc, end of an antibody molecule. Our engineering is incorporated in 2 of our molecules, including the newly marketed MARGENZA molecule. We have a very robust bispecific, multispecific technology called DART that is incorporated in multiple molecules in our portfolio. And then we have a number of ADC molecules using our antibody technology. We currently have 8 molecules in clinical development and a very large and robust preclinical portfolio. 7 of the 8 molecules have now shown very significant clinical activity. The eighth one which is still in dose escalation. So we're very excited about the prospects going forward of our portfolio and future molecules.

Great. Thanks for the introduction. Let's start with your B7-H3 efforts. First, can you talk about why B7-H3 is an attractive target in oncology?

So we were one of the first companies to highlight the value of B7-H3. We've actually been working on this molecule for over 12 years now. It, as you know, is a ligand in the so-called B7 family. Just to put it in perspective, people who know PD-L1, that's basically B7-H1, and this is B7-H3. The level of expression of B7-H3 is very high on tumors. In fact, a level higher than that of PD-L1 on most tumors. It's broadly expressed on most solid tumors and has a very relatively low level of expression on normal tissues. What we had done at very early of our efforts is we developed large amount of antibodies to this target, and we defined various epitopes of that molecule and ultimately selected particularly variable domains that had the best binding characteristics for tumor versus normal tissues. When we explored the expression of B7-H3, we found it not only on the dividing tumor cells, but it's found also on the aligning endothelial cells of vasculature of tumors as well as on tumor initiating or cancer stem cells. And it's also upregulated on the so-called Treg cells that have an upregulation of FoxP3. It's been associated with high expression on poor clinical outcome. It's been found to associate with metastasis. There's been literature that shows, for instance, an inverse correlation of B7-H3 expression and responsiveness to checkpoint molecules, where those with high B7-H3 respond poorly in lung cancer to checkpoint molecules. And so given all those parameters there, we went forward with developing multiple molecules using different mechanisms of action. Right now, we have 2 active molecules in development: one around an ADC molecule; and one around our Fc engineering technology. And historically, we've also used the redirected killing bispecific DART mechanism and have future efforts going there on a next-generation molecule as well.

Great. So just following up on that. So you have had multiple programs and do have multiple programs targeting B7-H3 in different approaches. How do you think about whether it makes sense to use, whether it's an ADC approach or bispecific or Fc engineered antibody approach?

So again, based on both our preclinical data and experience so far in clinical, we see these mechanistically orthogonal approaches to treating disease. As you know, given the broad expression of the target here, we are able to use both approaches in treating tumors. And in fact, we have examples, where, for example, historically, in our dose escalation study of enoblituzumab of Fc engineering, we saw responses in patients with GU tumors, including tumor shrinkage of prostate cancer. As you know, more recently, at ASCO last year, we presented some of some very exciting data on our ADC in prostate cancer. So there is 2 mechanisms -- 2 distinct mechanisms that are hitting a similar tumor type. The same thing with regard to, for instance, lung cancer. About 2.5 years ago, we presented data of enoblituzumab in combination with pembro in lung cancer in late-line patients seeing responses of over 35% their objective responses. In our dose escalation of the 2 patients we had in our ADC, both patients had evidence of tumor shrinkage with regard to lung cancer. So we think that this is an opportunity to actually exploit both mechanisms of action given that one is more immunologically based in terms of activating the immune system. The other is more traditionally cytotoxic based on the warhead associated with the ADC.

Got it. On MGC018, can you talk about the current status -- and that's your B7-H3 ADT program. Talk about the current status and progress of that Phase I/II study?

Yes. So the data that we presented at ASCO, we were in the middle of dose escalation at 3 mg per kg. After ASCO last year, based on analysis of the pharmacokinetics data, we decided to move forward at the 3 mg per kg dose. However, we continued in that dose escalation study to enroll patients at 4 mg per kg in a 3 plus 3 design. When we decided to move forward at 3 mg per kg, we selected 3 populations to pursue a further study on. One was in castration-resistant prostate cancer. The plan was to enroll 40 patients there. Second was in non-small cell lung cancer. The plan was to enroll 20 patients there. And the third was in triple-negative breast cancer, again, 20 patients there. We put in our amendments to the various sites for these expansions. We got approval to go forward initially in October. And we started enrolling our first patients in November for prostate and now continuing to enroll in all 3 cohorts in the early -- in the first 2 months of this year. So what we imagine is by midyear, we should have the 40 patients fully enrolled for prostate, the 20 patients we're targeting to have enrolled for lung. We think maybe the triple negatives may lag a little behind in terms of enrollment, just given that there are new agents available for triple-negative breast, and obviously, you need to go through that therapy. What we have said is that we want to provide updates midyear. We did put in an abstract placeholder for ASCO this past week. Obviously, you don't have much data on any of these new patients. But what we would do is if the abstract is accepted, we would update with a data cut in the spring and provide whatever data on new patients plus some of the patients -- some -- the patients that were enrolled at the 4 mg per kg, which are actually still -- a number of those patients are still being treated. So if for some reason, it doesn't get accepted, then what we would do is just provide outside that venue an update midyear. And then obviously, as these additional scans come in and analysis come in later in the year, we will provide additional updates on more patients.

Got it. That's helpful. As you've mentioned, you have several expansion cohorts for 018 in different solid tumor types. How should we be thinking about the potential for this program across different tumor types, especially given the broad expression profile of B7-H3? Are there tumor types that you think would be more amenable to 018?

Well, as I said, we're selecting those 2, if you actually look at the -- we've spent a lot of time, obviously, in the analysis of expression patterns. We've looked at over 1,500 tumor specimens by ourselves, plus obviously, what's also in the literature. Right now, we don't see any major limitation in terms of the opportunity here, given that the -- we're seeing evidence of anti-tumor effects, both with the ADC, the historical data we had with our bispecific DART molecule, the Fc engineered enoblituzumab molecule. There are tumors, for instance, that have even higher levels of expression than prostate. A good example, for instance, is head and neck cancer. In our earlier studies with enoblituzumab and pembro in late-line patients, we saw about a 33% response rate in that population. What we are about to start literally in the next week or 2 is enrollment in a frontline head and neck study of enoblituzumab, our Fc engineered B7-H3, plus 2 of our checkpoint molecules, retifanlimab and PD-L1 positives. That's the molecule we have licensed to Incyte. And then for the PD-L1 negative head and neck patients, we will treat them with our PD-1 LAG-3 bispecific tebotelimab. So that's going to start very shortly. But that one, as we just discussed before, could also be a very good tumor indication for the ADC as well. So that's another example going forward. But there are a number of other tumor types that we have interest in. And it is likely that we will expand the ADC into other tumor indications, and we'll discuss that later this year.

Got it. What has the safety experience with 018 been to date?

So as I described, it has been quite manageable in a way predictable based on the toxin conjugate here. So what we have seen most notably in the dose escalation that we reported at ASCO last year, we would see some hematological effects, most prominently neutropenia, which would respond to either a holding a dose or reduction in dose or treatment with G-CSF, so that was one of the more prevalent ones. We would see cases, again, based on the profile of the patients and toxicities of so-called hand-foot syndrome that you see associated with a lot of chemotherapeutics, particularly in patients, who've had similar reactions in their earlier lines of therapy to taxanes and other chemotherapeutics. I would say those are the 2 most prevailing with some maculopapular type rashes, again manageable. So we'll have to see our experience as we expand into a larger number of patients.

Got it. So switching over to MARGENZA or margetuximab. Can you talk about the gastric cancer opportunity for this drug and help set expectations ahead of the first half MAHOGANY Module A?

Yes. So as you know, we're about to launch in breast cancer. We had -- based on studies that we conducted and published on that looking at a combination of MARGENZA and an anti-PD-1 in second-line therapy, these are patients that had already progressed on HERCEPTIN plus chemotherapy, we saw a very nice objective response rate of 44% at the end in the intense street population. And particularly the PD-L1 positive HER2 3+ positive population, we were seeing overall survival benefit of over 20 months, which compared favorably to second-line survival of about 9 months and compared to even the front-line data of about 13 months. So given those parameters, we launched into the MAHOGANY study, which were essentially 2 studies into one. The first one which you alluded to, which is called the Module A of MAHOGANY is looking at this HER2 3+ PD-L1 positive population, where we're looking at a combination of margetuximab plus retifanlimab, our anti-PD-1 that out-licensed to Incyte. The first cut of this data will be -- we've now enrolled 40 patients in that first cohort that was designed in the study. We obviously have to have a couple of -- we need to have 2 scan follow-ups at least to declare that data. Our plan is to provide update midyear on where we are on the analysis of those 40 patients. Just to give you a historical perspective, the TOGA regimen of HERCEPTIN plus chemo in that front-line population had an objective response rate of 47% and with obviously a very high-grade 3 toxicity profile. So if we're obviously exceeding that and based on whatever internal parameters, we think we'll -- if we meet those objectives, we would then move forward to complete the enrollment in that single-arm study. And based on conversations with the FDA, if successful, we would then file for accelerated approval on that arm of the study and then would continue to enroll in a more traditionally defined HER2-positive population in combination with chemotherapy and 1 of 2 checkpoint molecules.

Got it. Maybe for the sake of time, and I doubt we're going to get to every one of your programs in our allotted time. Let's switch over to MGD019, your PD-1 × CTLA-4 DART molecule. First question, how do you see this -- your program differentiated from other PD-1 × CTLA-4 bispecifics in development?

So we have put in a lot of thought on the design of this molecule. And as you're probably aware, we published in December the preclinical as well as some of the translational data and the initial clinical data from our dose escalation study in cell reports in December, which got a very favorable review. We're excited about this program because we knew that a combination of targeting PD-1 × CTLA-4 had additive and, in some cases, one could even argue, synergistic activity in particular tumor types. But that combination was limited by the toxicities when observed when one typically went at 3 mg per kg of ipi and higher doses of nivo. And so what has always limited that combination for Bristol is the toxicity associated with us moving forward. So when we designed our molecule, we looked at various configurations of this bispecific, where we made a bispecific that had monovalent interaction with PD-1 and CTLA-4 or a 2 to 1 of binding to PD-1 and 1 to CTLA-4 or 2 to 2 PD-1 and CTLA-4. We ultimately arrived at this tetravalent bispecific design, and we also decided from the beginning, instead of putting a IgG1 backbone, which would then allow for the promotion of ADCC activity, we would make this as an IgG4. So you essentially eliminate that, but still have half-life advantages there. The rationale is, is that we believe that, number one, by making it an IgG4, you would not eliminate cells that are CTLA-4 positive alone, which are a large subset of so-called Treg cells, which we believe was one of the major contributors of the toxicity observed with Ipi/Nivo combination, particularly GI toxicity. In fact, that was borne out both in our dose escalation study in -- where we went up to 100 mg per kg in monkeys, do not see the GI toxicity, seen with Ipi/Nivo and similarly, when we went up to 10 mg per kg in patients, again, not hitting dose-limiting toxicity there. By doing that and additionally showing the activity that had all the salutary effects of activation of T cells, expansion of T cells at both CD4 and CD8 the upregulation of ICOS on CD4s and then seeing the therapeutic effects in tumors that are not typically checkpoint responsive, that will give us a lot of pause that this was a valuable drug. I should also note that the blockade effect of this tetravalent bispecific if you look at a tumor site, the cells -- the T cells that infiltrate that have a high expression of both PD-1 and CTLA-4 and because you now have higher avidity of such a molecule, you get more selective binding, which we also demonstrated in the paper that we published. And we found that the blockade, for instance, of the interaction between CTLA-4 and B7-1, which is a natural ligand, was up to hundredfold greater than that of ipilimumab or CTLA-4 alone. So all those reasons, we believe, contribute to the superior potential of our molecule.

Got it. Maybe at this point, we'll switch over to some submitted questions, and we got a couple on this program as it relates to why you chose the indications you chose to develop this drug? And why you didn't choose certain indications? So maybe if you could just first refresh our memory as to what the status of this program is, and then discuss the rationale behind choosing the indications they chose to expand this in?

Yes. So again, we -- as I was commenting before without specificity, we began to see objective responses in tumors that traditionally were not responsive to checkpoint molecules. So we had a patient, for instance, with -- who had been in very late-line multiple lines of prostate cancer, and we saw a complete response in that patient with castration-resistant prostate cancer. We saw a patient with a stable colorectal cancer and, as you know, checkpoints do not work very well there. That patient still has a very prolonged PR response even though that patient is now off therapy. He's ended up having some immune-related reactions there. And then we had a patient with thymoma, again, not traditionally seen as a treatment with checkpoint, saw a very nice, almost complete remission there as well. Patient with fallopian tube saw unconfirmed PR there. So what we decided was expand out initially in the MS-stable colorectal. We also -- just to get some head-to-head comparison to Ipi/Nivo, we decided to go into PD-L1-naive non-small cell lung cancer. So those are the 2 initial tumor types we have. We are going to add on additional tumor types as well, but those were the ones out of the starting gate. So just view that as the entry point.

Understood. And maybe that latter part answered this question, but why not prostate cancer?

Well, it could be prostate as well. So stay tuned for some discussion about that in the future. As you know, we have the investment we discussed on B7-H3 on prostate with the ADC. So there is obviously opportunity for us to expand with other molecules there as well. So stay tuned for that.

Got it. Now I believe we're expecting an update from this program at midyear. Can you help set expectations ahead of that?

Yes. I mean right now, we have no submitted abstracts for this program. We think that we're just about finishing out the enrollment on the colorectal cohort. There was up to 20 patients in that. We're just getting started on the lung. Obviously, we had to go to Europe for the PD-L1-naive, given that standard of care is anti-PD-1 in the U.S. So that obviously fall behind. So our goal is to provide some form in the middle of the year, the updates most likely on -- initially on the colorectal. And also then talk about just as we just discussed, prospects for additional cohorts going forward.

Got it. I'm getting the warning sign that our time is nearly up. So maybe if you could just wrap up with other catalysts and milestones that we should expect for the remainder of the year. I know we've already touched on a couple. But if you could, in our remaining time, touch on those, that would be great.

Yes. I would say -- I would add on the 4 distinct ones. One is, we have 2 out-licensed molecules that have upcoming PDUFA dates. So the anti-PD-1 with Incyte has a PDUFA date for the anal cancer, that obviously comes with some significant milestone payments if it's successful. That's in July. Provention Bio, which we out-licensed teplizumab for type 1 diabetes. If successful, again, on a July PDUFA date there, would have what's been announced a $60 million milestone there. On the other programs on flotetuzumab, we're enrolling in a registration study in the primary induction failure early relapsed patients. We should have updates on the first sets of cohorts there as a single-arm study in the registration study, probably by the time of ASH. And then we will provide some additional updates on tebotelimab. We're next steps going forward on the PD-1 x LAG-3 program, which, as you know, we are pursuing as -- in patients with DLBCL, patients with very solid tumors that are in combination, with marge. I think that sort of -- one final thing is, is that ImmunoGen has started developing a second -- our second ADC molecule to ADAM9. And it is likely that, that will move through dose escalation by the latter part of this year. So there may be ability to update that either late this year or early next year.

Got it. Let me -- sorry. Let me sneak in one final submitted question. And that is for 018, what is the duration of PSA 50 responses that you hope to see with a follow-up of the patients treated last year?

As long as possible. No. So again, what we had shown at the time of ASCO last year, we had one patient at that point was at least 6 months. And so truly, we certainly want to be north of 3, 4 months, we'd like to be 6 or more months or longer in regard to that in terms of both PSA reductions as well as progression-free survival.

Got it. Well, that's all the time we have. So thank you very much, Scott, for taking the time. We appreciate it.

Thank you, Jonathan. Nice to see you.

Have a good day. Bye.