Amgen Inc. (AMGN) Management Presents at Goldman Sachs 40th Annual Global Healthcare Conference (Transcript)

|
About: Amgen Inc. (AMGN)
by: SA Transcripts
Subscribers Only
Earning Call Audio

Amgen Inc. (NASDAQ:AMGN) Goldman Sachs 40th Annual Global Healthcare Conference June 12, 2019 2:20 PM ET

Company Participants

David Reese - Executive Vice President of R&D

Conference Call Participants

Terence Flynn - Goldman Sachs

Terence Flynn

Okay, great. I think we are going to get started. Thanks for joining us for our session with Amgen. I am Terence Flynn, and I am the Biotech Analyst here at Goldman and we are very pleased to have Dave Reese, who is Executive Vice President of R&D for Amgen.

And with that, I am going to turn it over to Dave for some opening remarks and then I will dive into questions.

David Reese

Well, thanks, Terrence. So, and it’s pleasure to be here today. Let me just make a few comments about our approach to R&D. I’ve been the Head of R&D for just about a year now. And how do I conceptualize R&D in our industry as it exists now and projecting out ten or fifteen years.

So, to me, there are two core existential problems that we need to solve. One of them is, how do we improve the success rates. Currently, they still run around 8% and it’s been that way for some time that varies by therapeutic areas. But we must do better.

And then, two, how do we reduce cycle times and drug development. Still 12 to 14 years on average from idea in the laboratory to achieving marketing authorizations for a drug. And so, to tackle both of those problems, we are creating and have created, what I would call, sort of an integrated triple threat.

We believe, we strongly believe that genetics will be foundational for drug development going forward, particularly outside of oncology. We are not talking tumor genomics here, but everything else and we’ve made through the acquisition of deCODE genetics and then enhancement of their work in the last several years, huge investments in genetics.

And just this morning, in fact, we announced the large collaboration with Intermountain Health on integrated delivery network in the – of course, mountain region of the United States. They are about roughly 3 million patients in their system. We are hoping over the next five years or so to sequence 500,000 individuals.

This will be the largest single project, focused project of its kind in the United States and I think is a tremendous collaboration of the leadership that Intermountain was visionary and one thing to be able to, both on an individual patient level begin using genetics to guide how they deliver healthcare to learn across their populations. How they can improve the delivery of healthcare.

And then of course, from our end, to take these insights and say, can we make new highly effective medicines. So that genetics piece is critical. I think, we are indisputably the world’s leaders in that right now and I feel very good about where we are. But that’s a starting point and coupled with that, you need strength in biology.

Genetics gives you ultimately link to a target or a pathway. There is a lot unique to figure out. I think that’s Amgen’s historic strength and we continue to buttress that.

And then finally, you need to engineer molecules based on those insights and so, I think about R&D, we’ve created really an integrated unit that is thinking from, okay, we’ve identified a genetic target through, okay, what kind of molecule are we going to bring to bear here. And I feel very good about where we are in that respect.

And then finally, to touch briefly on speed of development, I led from that first within Amgen starting several years ago to really greatly accelerate our development timeline from early preclinical work through clinical development that AMG 510. KRAS program is an example of where we applied what we’ve learned and again in an integrated fashion across the spectrum of drug development.

I think we’ve probably taken on average three years off our development timelines right now and I think we can do better than that going forward.

So, I pause there and open it up for broader questions.

Terence Flynn

Great. Well I think that’s a great framework to start with. Maybe the one question, just as you think about measuring productivity of the R&D organization, you mentioned the 8% rate is a cycle time. Are those the key metrics that you and management are looking at? Are there other metrics in terms of timeline?

David Reese

We have intermediate metrics. To me the only important metric ultimately is, how many drugs are approved and are successful and by successful I mean, you have both access to the patients need them and reimbursement. So that there is an appropriate return on investment. Now of course, along the way, in research and development we have various metrics.

The number of new product teams that we form each year for example is a metric coming out of our discovery research lab. And when we form a product team, that means, we have a clinical candidate molecule that we are intending to take into human. In the last few years, we’ve had record numbers of product teams formed and I think we are reaching a new equilibrium in terms of productivity at Amgen.

Terence Flynn

Okay. And in terms of the Intermountain Health collaboration, maybe just help us think about, what does that bring in that deCODE doesn’t have? Or what does that add to deCODE in terms of capabilities?

David Reese

I think there are a couple things. One of the secrets which no secret in genetics is power of numbers. deCODE has roughly 1.6 million participants in its databases now. This will add another 0.5 million. That puts us almost a log ahead of where almost anyone else is.

So, sheer power of numbers is important as you are trying to associate genetic variations with disease states or protection from disease states. In addition, this is a more, when you want genetics that caused more the outbred population, so it’s a little more heterogeneous and allows us to then broaden a mix of populations that we are looking at.

Terence Flynn

Okay. And when do you think you can start seeing some efforts in terms of new targets, et cetera coming out of that collaboration?

David Reese

Yes, we started this more an – probably a little early to speculate on timelines. But our goal is to be pretty aggressive and ambitious and we have, again, I think the best genetics unit in the world, okay.

Terence Flynn

Great. Maybe, we will move on to ASCO. I know this was an ASCO for you guys. A lot of new exciting data. Maybe we will start with 510. You mentioned that in your opening remarks. But maybe just give us a little bit of the background here around the target, why you decide – why did you decided to pursue KRAS? And then, kind of the genesis of your program? And then we can talk through some of the more detailed data.

David Reese

Sure. And I am happy to talk about KRAS. So, let me start actually with our framework in oncology.

Terence Flynn

Sure.

David Reese

Because that’s what led us to KRAS and our oncology strategy is predicated on the belief that the future of oncology meaning the next 15, 20 years is the marriage of precision oncology or targeted therapy with immune oncology. You are starting to see a resurgence actually of precision oncology and targeted therapy which have been quiet for a little while.

Now, within Amgen, we said, okay, within those spaces, where are we going to make our strategic bet. Within on precision oncology, we said, we are going to go after very high value targets, KRAS, MCL-1, Holy Grail type targets. We are not a chemistry organization of 3,000 individuals, but we have, what I characterize as a – team there.

They are an international class group of chemists and I think the KRAS program is an example of that. So we said, okay, we are going after high value targets. The KRAS field opened up from a – with a paper from UCSF in the last three, four, five years that gave some structural clues about how you may go after that target or why do we care about KRAS, where RAS is one of the first oncogenes identified mutations in the different forms of RAS occurring up to 40% of all human cancers.

It is the most frequently mutated oncogene that’s often a driver. Oncogene occurs early in the pathogenesis of disease. But it has been undruggable for almost four decades and the reason it’s been undruggable is that, it’s – the RAS is a globular protein. I have likened it to a golf ball with little dimple.

So, when you design small molecule inhibitors that as a protein usually there is a pocket and you figure out a way to bind in that pocket and block the function of the protein.

Here, there wasn’t much about pocket. Now, we were able to understand building on some of that structural biology and through, I think some very elegant structural biology and then ultimately medicinal chemistry of our own way to slot into a shallow groove that’s created on the surface on the molecule attached to the assisting – that’s the C and the G12C mutation covalently, so, permanently.

And then, there is also another shallow groove that we figured out opens up and we could reach in and grab that, as well. So it was a tremendous feat of medicinal chemistry on the part of the team.

Terence Flynn

Okay, great. Maybe as we talk through the data, there are couple of key takeaways. The first was just obviously activity in a number of tumor types. But maybe a little bit more striking in non-small cell lung cancer relative to colorectal which are the two predominant types where G12C is seen.

So, maybe one of the debates is again, dose versus biology and gain, maybe help us think about the relative efficacy in those different populations?

David Reese

Yes. So, in – what Terence is alluding to is that, in non-small cell lung cancer, five of ten patients for whom we had data as of nine days ago, I think when it was presented had responses. We hadn’t seen any formal responses in colorectal cancer although roughly half the patients it had some shrinkage in their tumors on scans.

And so, one question is why is there this differential response? I’d say a few things. Number one, we had only treated one patient with colorectal cancer at the top target dose. And so, we are enrolling a number of patients very rapidly at that dose. And so, I think in the coming months, we’ll actually have much deeper insight into that particular question.

A second question is, these tumors are complex in terms of their molecular wiring. And so, one thing that we are working very hard to understand as we accumulate data is, are there other suites of mutations that help predict response in resistance. Does that drive you towards specific combination therapies? And how does that make you position this drug as part of the sort of broader pallet of therapies that’s going to offer for these patients.

So, I feel very good about where we are. I can report that, we just very recently opened the expansion phase of that first in human study and we expect to complete enrollment within the next few weeks give or take. It’s a reflection of the tremendous interest of our investigators and what they are seeing in the clinics.

Terence Flynn

And that’s both for the lung and colorectal in terms of the expansion?

David Reese

Yes, it’s open and then we are moving on to open combination trials as well we presented data at AACR suggesting and we got the manuscript under review now. But we’ll report these data more completely, but suggesting that KRAS in addition can inflame tumors impart probably through up regulations class 1 and may see expression.

And so the – it was quite a potent interaction between checkpoint inhibitors and the AMG 510. So we are opening in days – that was going to matter of days or so of the first patients in that combination.

Terence Flynn

Okay. And is that by tumor type or that all tumor types in terms of combination?

David Reese

We’ll share a little more later in the year as we move that along.

Terence Flynn

Okay. And then, one – in terms of thinking about other tumor types, were there any other patients that were enrolled? Can you just remind us what did you see in terms of activity?

David Reese

There were just – there was less than a handful of patients and you will tend to see endometrial cancer. One of our real focus is going forward will be to make sure we look at pancreatic cancer where KRAS G12C mutations occur in couple percent of patients, endometrial cancer, appendiceal carcinomas, that’s strangely enough have a relatively high prevalence of this particular mutation.

And then, there are other handful of others. So, you know we put a lot of emphasis on determining safety and efficacy in those populations.

Terence Flynn

Okay. And I think the other question was, I mean, any of these targeted therapies, whether it was EGFR, ALC or RET, now KRAS’s durability of response. And so, as you think through that, is there anything unique about KRAS that we should consider as we think about how durable some of these responses might be over the longer term?

David Reese

Yes, it’s a challenging question. One way, and that’s – it’s a great question and I am asked at all the time and I point out that it took 40 years to get a molecule that’s only been in the clinic seven months. So there is still couple unanswered questions that we need to broach.

But, as I said, RAS mutations are often occur relatively early in the genesis of many of these tumors in adolescents as various clones develop. They will generally harbor those mutations and that gives you some optimism that you could see real duration of response. But, I think in this case, when you are sailing near the edge of the years, some, the clinical data will be determinative, I think the correlative signs in the biomarker it will also be critical here.

Terence Flynn

Okay. And then, one more just obviously people are highly focused on kind of past the market here. We’ve seen other companies employ kind of, Phase 2 single-arm studies. It would seem like that’s a probably a pathway that you guys could follow. Have you had any conversations yet with regulators in terms of how many patients kind of duration that you could address that?

David Reese

Yes, what I can say about that, we normally don’t comment in detail on regulatory interactions. But I would say, we’ve had very positive interactions and as the program moves along, we’ll talk more about that.

As you pointed out, in these settings where there really is no standard therapy available or patients who have exhausted all standard therapies or can be expedited ways towards registration and that’s certainly something we are keenly interested in.

Terence Flynn

Okay. And you guys, I think you mentioned that you feel like you have an optimized dose here at the 960 for both the colorectal and lung?

David Reese

We did very extensive modeling pre-clinically. And I think there are couple things to keep in mind here. So this is a covalent inhibitor. This is not a traditional kinase inhibitor where you want to shut it down for 24 hours and in fact, RAS itself, like the kinetics of RAS production it’s half life is about a day.

And we have , I think a very extensive preclinical evidence that if you achieve the appropriate exposure for a couple hours, two to four hours, you basically poison RAS for the duration of that dosing interval and in fact the exposures we are achieving in the clinics are well beyond what we were required for tumor regression. So, I feel good about the dosing and we feel quite good about the quite benign adverse event profile we are seeing so far.

Terence Flynn

Okay. Maybe, just last question on this front, before I move to the bispecific platform is, just talk to us about differentiation versus the competitor program already that’s out there. I know there are some questions in terms of how similar or different these programs are. And then anything at this point in terms of how to think about resistance, overlapping resistance?

David Reese

Sure. What – so, I can't comment on other programs, because we haven't seen any data.

Terence Flynn

Okay.

David Reese

And so I am sort of repeatedly asked to comment on the mirage which is difficult.

Terence Flynn

Okay.

David Reese

And so, in the fullness of time, we will see the data as they unfold in the field. Might resistance develop, of course, I mean, I wouldn’t be so naïve to think that after everything we’ve learned about targeted therapy oncology that resistance will not emerge at least in some tumors, that’s a – I think a critical question that we need to address in terms of development going forward.

It’s one reason I actually like very much the combination of the checkpoint inhibitors to really bring an orthogonal immunologic approach to bear on the tumor at the same time and try to prevent that sort of escape.

Terence Flynn

Okay, great. Well, the other program, you had a couple of updates on your BiTE platform at ASCO with respect to BCMA and the PSMA specifically, and so maybe first on the BCMA, we’d just be curious, kind of, how to think about that data in the context of the competitive landscape and then where does your half life extended version really fit in and it seems like, to me, that’s probably the more commercially viable of the two approaches.

David Reese

So, we presented, as you know, it’s an update of the first in human study from AMG 420, the first-generation continuous infusion BiTE at ASCO. And there – it was a sort of a natural extension, longer follow-up in these patients. I think, the takeaway was that the response rate was enhanced a little bit with longer duration of exposure in these – in some of these patients.

A number of them are continuing to receive therapy. So those data will evolve as we go forward. I would say the adverse event profile, there were no real changes from what we had reported at ASH. And our investigators remain incredibly enthusiastic. As you know that the BCMA landscape is incredibly dense and with multiple molecules and mechanisms of action at last, kind of for 38 something give or take.

Our approach to development here will be to say, okay, where do we believe the BiTE can really have the biggest impact and so we will be looking to move to earlier lines of therapy. For example, where you’ve got a lower tumor burden, where our goal really will be to drive patients into a minimal residual disease remission that hopefully in the last four years.

I think that, one of our goals with the BiTE programs is also to be able to deliver an attractive economic package for molecules that are dosed in micrograms, at microgram levels and that we think, in terms of things like cost of goods, we’ll be extremely competitive. So, we like where we sit there. Are we going to certainly blanket the entire landscape, I am not sure.

But, we want to focus our resources where we think we can deliver the most value for patients and again generating appropriate return on investments.

Terence Flynn

And is it, I know you guys have previously said that, the half-life extended version is about a year behind.

David Reese

Yes, so, that – so, AMG 701 is a half-life extended version of BCMA BiTE. It is moving through dose escalation. I think we’ll have our initial clinical data either later this year, first part of next year from that program.

Terence Flynn

Okay. And do you think it’s reasonable to assume that, based on pre-clinical efficacy, we should see a comparable level of efficacy between the half-life extended and the continuous infusion?

David Reese

Based on everything that we’ve seen pre-clinically, we are quite optimistic about the half-life extended program. Of course, we have to demonstrate that.

Terence Flynn

And then the infection rates, again, that’s the other thing that to me, it looks like maybe be able to optimize with the half-life extended version. Is that, all assumptions?

David Reese

I think there is a potential advantage in that domain, because the patients don’t require an intraputaminal catheter.

Terence Flynn

Yes. Okay.

David Reese

And these patients are profoundly immune-suppressed. The background infection rate in patients with myeloma who had multiple prior lines of therapy are actually quite high.

Terence Flynn

Okay. And then the other program was the PSMA BiTE for prostate cancer. And again I think that was the first solid tumor data that you guys have shown with the BiTE platform. And I guess, one of the takeaways was, because of the way that trial transitioned the trial, you didn’t reach really top dose. So, maybe just kind of help us frame what you saw and the confidence in again the half-life extended version that I think is again already in the clinic?

David Reese

Yes, that is a great question. So, these programs target PSMA, which is prostate-specific membraning – a perfectly misnamed proteins target in certain ways. But nevertheless a very good target in prostate cancer and there is evidence that PSMA expression actually tends to go up as the disease progresses. So, another sponsor initially had this program that was through Micromet collaboration that preexisted our acquisition of Micromet. We reobtained the rights to the program.

And as you were noting presented from somewhat truncated first in human trial data at ASCO that reached the top dose of 80 micrograms. There were no dose-limiting toxicities when the other sponsor for strategic reasons decided not to move forward. I think what was notable in that program were a couple of things, one it was extremely well tolerated.

Number two, there were a number of patients who had PSA declines and then there were a couple patients who had what we characterize is real clinical responses and single one patient in particular who had dramatic decline in all of the tumor markers and a new resolution of skeletal lesions which were dense on various scans, PET scans and bone scans.

That patient also had quite significant symptomatic improvement and went from filling up paperwork for hospice before entering the trial to actually skiing five or six months later and had quite a prolonged response. So I think that gives us proof of principle and now we need to broaden that experience and as you noted, we’ve introduced a half-life extended PSMA targeting BiTE into the clinic and that molecule marching through dose escalation actions pretty quickly.

And so, we will see, whether it’s later this year or more likely sometime next year we will have initial clinical data from that program.

Terence Flynn

Okay. What – and then in terms of any of the other BiTEs that you wanted to highlight that you are looking for to the, call it, 12 to 18 months in terms of some of the key readouts, because I know it’s been a big push for you guys that’s giving some of the BiTE data out there?

David Reese

No, no, no, that’s a great question. We have roughly a dozen either in the clinic or just about to be in the clinic across hematologic malignancies and solid tumors. So we’ve got several BiTEs targeting acute leukemia AML. And so, those are ones that we are looking for over the next year or 18 months as you pointed out in terms of the timeline.

And then, in solid tumors, we’ve discussed a couple of the programs. Another I would highlight is DLL3, AMG 757 which is a target that is expressed in small cell lung carcinoma widely in other neuro endocrine tumors in which small cell lung cancer is a form of neuro endocrine tumor.

And that actually we believe in terms of solid tumors is one of our best targets, because the differential expression between that, the target in tumor cells and in normal cells is quite extreme and that’s what you want for a BiTE target. So, that’s marching through dose escalation as well and that’s one we’ll be looking for read out for the coming years.

Terence Flynn

And so, that one even you are still optimistic even given the Rova-T data. So the ADC DLL3?

David Reese

So, what I would say is that, based on what we know on the Rova-T program, I think the challenges there were related to the modality in the warhead and I haven’t seen anything that that tells me that there is an issue with the target. And so, I would say, we remain quite optimistic for DLL3 as a BiTE target.

Terence Flynn

Okay, great. Well, maybe if I am going to pass to breast cancer just in interest of time, Tezepelumab is another late-stage program that you guys have talked a lot about in Phase III for asthma. Maybe just walk us through kind of where we stand with enrollment, potential timelines on data and then the other big question I would get is just differentiation versus whether it’s depicts into the I/O five, so where so where we will talk to tell you that kind of ultimately fit in the landscape given the number of new biologics?

David Reese

And so, Tezepelumab is one of my favorite molecules. I have some biases there, because I was running our early development programs and we put that into humans and we helped design the inhaled allergen challenge study that was ultimately published in the New England Journal of Medicine. And that really gave us the sense that – it’s something interesting here. The Phase III program is moving along with our partners AstraZeneca on track.

We will complete enrollment on schedule in the coming months with the data readout next year. So I think it’s all systems go in the Phase III program. And, one of the reasons we really like the molecule is that in the Phase II program and based on the underlying and immunology of the target, we believe that this could have a role in both high eosinophilic or the allergic forms of asthma, as well as the non-eosinophilic or non-allergic forms of asthma.

For the latter, there are not any real biologic therapies or good therapies now for patients with severe disease. So this will potentially enable you to go across a very broad population. And in the target TSLP, is an upstream modulator or instigator of the inflammatory cascade. It’s also released from the epithelial side.

So it’s not – so we are not attacking the immune side here, but actually one of the nodes that we believe really is at the crux of the inflammatory response that occurs in asthma.

Terence Flynn

Okay. And given it is more upstream, are there any safety concerns we should think about an infection risk, things like that?

David Reese

One of the – our hope would actually be the reverse.

Terence Flynn

Okay.

David Reese

Because TSLP is not really present in non-inflammatory. So it’s not like a cytokine, which is part of the normal balance that your immune system carries forward. It’s really released in inflammation setting and in fact, TSLP levels in tissues or that you can measure systemically are quite, quite low in a in non-inflammatory state. So, the Phase II data was quite good in terms of the adverse event profile and we’ll see what the Phase III data shows.

Terence Flynn

Great. And I know you guys, atopic dermatitis is kind of the other indication that you are focused on. You had run a previous trial there. You may had some learnings from that, you made some changes conducting a new Phase II trial. But maybe just give us a little bit more detail on kind of the key changes why you are confident in atopic dermatitis you have given the prior readout?

David Reese

Yes, I would say a couple things. So, number one the evidence preclinical and translational evidence linking TSLP to the disease stage is actually a stronger, stronger in atopic dermatitis as it is in asthma. I think we have learned – and the first Phase II trial was launched sometime ago. And I think that’s a field that’s rapidly evolving.

We’ve learned a lot about background therapies, how to control for them, the natural history of the disease which naturally waxes and wanes in the absence of therapy. And I think we’ve designed what quite comfortable is a quite rigorous Phase II trial to really test the hypothesis in atopic dermatitis.

Terence Flynn

And would we – could we also see data next year from that study?

David Reese

I am probably not ready to speculate on what the timelines might be for that trial.

David Reese

Okay. Maybe Omecamtiv and other late-stage programs, I guess, for heart failure here. You did start a new Phase III trial. Maybe just help us think about the rationale behind that study and how you think about kind of the what’s the differentiated feature of this drug? What are the risks in that?

David Reese

So, Omecamtiv is a novel mechanism of action. The heart failure field hasn’t really seen a novel mechanism of action going on two decades now. So I think, that’s one thing that makes us optimistic. The second thing is the mechanism of action by really enhancing the contractility apparatus in the heart is orthogonal. So, almost all of the other therapies that currently exists which in one way or another affect the afterload that the heart sees or the preload burden that the heart sees.

So that it can be used in combination with therapies or layered on therapies and in fact, that’s the design of the Phase III trial. The Phase III program itself is also on track with our partners, cytokinetics and we anticipate wrapping up enrollment in that Phase III program within the next coming weeks to few months on schedule.

That is over 8,000 patients trial. So it will be one of the largest studies ever conducted in heart failure and we hope also one of the most rigorously conducted trials in heart failure.

David Reese

Okay. Great. Maybe just in the last couple minutes, as we talk through your different therapeutic areas, cancer, immunology, cardio, we did dip in CNS, you obviously have aim a big anywhere that you feel like you have a need to kind of build out your early mid-stage pipeline in terms of thinking about other areas that you maybe need to bolster up a little better. Do you feel like you’ve got a pretty adequate opportunity set?

David Reese

Yes, we like the strategic focus in the four therapeutic areas that we’ve got right now. Of course, we are always open to scientific serendipity. But our largest emphasis by far will be on those therapeutic areas which are large.

When you think about our aspiration to really affect diseases that have a large effect on the public health, cardiovascular disease, cancer, neuro degeneration, these constitute 80% of the public health burden going forward and I would anticipate that in most of those areas we will have a significant focus going forward.

David Reese

Okay. Great. Well, I think we are out of time. Thank you, very much Dave.

David Reese

Thank you. It’s been a pleasure.

David Reese

Thanks.

Question-and-Answer Session

Operator