Alkermes plc (ALKS) Earnings Call Transcript & Summary

All right, all right. Good morning, everyone. My name is Cory Kasimov, senior biotech analyst at JPMorgan, and it's my pleasure to introduce our next company, which is Alkermes. Here to present for Alkermes is Chairman and CEO, Richard Pops and following Richard's presentation, there's a breakout down the hall to the [ left ] Olympic room. So with that, turn it over to Rich.

Thank you, Cory, and good morning, everybody. This is a period of intense activity at Alkermes. We've built a $1 billion top line business through developing a series of really important medicines that we're really proud of. But we're now focused on actively shaping the future of this business through an intense assessment of our products, our pipeline and even our corporate structure. So if you haven't been watching closely, you probably haven't seen a number of the exciting changes that have been occurring, and that's what we're going to focus on this morning. I will make forward-looking statements. We always ask you to take a look at the disclosures that we make in our periodic filings because we try to capture those risks in the most accurate way that we possibly can. So one thing that hasn't changed in this company is our mission. And this for us, at this stage, is far more than just a theory or an aspirational statement. This is what we do every day. We're out there every day through our science and our medicines and our advocacy, working to change the expectations of what a successful treatment means to patients with serious conditions. Alkermes is a positive force for change out in the community and we're making a real-world impact and we're very proud of that. With that said, we're actively shaping the future of the business in these 3 principal domains. First, expanding and driving the growth of our product portfolio. Second is advancing a diversified pipeline in 2 major therapeutic areas, CNS, where we've been for a long time and oncology, where our pipeline and our capabilities are growing remarkably. Third is positioning the business to deliver long-term growth and profitability through our cost structure and our organization. So let's take a quick look at these in turn, starting with the commercial portfolio. VIVITROL and ARISTADA are our 2 proprietary commercial products for the treatment of addiction and schizophrenia. VIVITROL is a once-monthly, extended-release opioid antagonist approved for the prevention of relapsed opioid dependence following detoxification. It's the only drug that's approved with that indication. There are really only 3 FDA-approved medicines for the treatment of opioid dependence: methadone and buprenorphine, which are both replacement therapies and then VIVITROL, which requires detoxification and monthly injections to prevent relapsed opioid dependence. This is a unique medicine and we've had to work state by state and county by county to build awareness and facilitate and promote access to it. This week, between 10,000 and 11,000 people will most likely go on VIVITROL injections. So 10,000 or 11,000 families and individuals and even communities will be impacted by that medicine. We're really proud of the fact that VIVITROL is used this extensively. ARISTADA is our long-acting injectable atypical antipsychotic, a new molecule invented by Alkermes scientists and is the first and only long-acting injectable that can be fully dosed on day 1 for up to 2 months. For schizophrenia patients, having to make the decision and remember to take a medication 365 days a year is a tough adherent schedule. Once every 2-month dosing means that patients and healthcare providers can reduce that to 6 decisions a year for therapeutic levels to be maintained for patients. That's very powerful and allows patients to focus on other elements of their lives. Yet long-acting therapies remain completely underutilized in the U.S. These medicines go to the core essence of what Alkermes is as a company. We're helping patients in disease areas that many pharmaceutical companies have abandoned. We deal with government payers. We work in ecosystems to try to drive systematic changes that will enable access to new medicines and we built a specialized and distinctive set of commercial capabilities in order to bring these medicines to patients. These are unique products with long patent lives ahead of them and we're proud of them because they present not just an opportunity to make a new exciting scientific contribution but to make a meaningful difference in the lives of patients, their families and even society. The 2 products are growing and they comprise an increasing percentage of our revenue. We have $1 billion top line business, as I mentioned at the outset. VIVITROL in the light blue and ARISTADA in the dark blue are growing on top of a solid foundation of royalty and manufacturing revenues. And the revenue engine continues to evolve. In 2012 -- I'm sorry, in 2020, we expect a new stream of royalties from Biogen's launch of VUMERITY and we're planning for potential approval of ALKS 3831 at the end of the year, which will contribute to and grow our psychiatry franchise. I'll spend a second on VUMERITY, which was just approved. It's a novel next-generation oral fumarate. It was discovered and developed by Alkermes and it received approval last quarter. Biogen will commercialize it and is launching it into the large U.S. MS market. It's interesting to note in the box on the right just how many patients are treated each year, approximately 325,000 patients and how many patients change therapy each year, which is approximately 60,000. Alkermes receives a mid-teens royalty from Biogen and this product is profitable for us from the first dose that has been sold. Further, VUMERITY's IP is distinct from TECFIDERA's and it bestows a long patent life, which may open additional life cycle management opportunities for this drug. This represents an exciting opportunity for a new, efficacious, well-tolerated oral agent, and we're looking forward to tracking Biogen's launch progress. In contrast to VUMERITY, we intend to commercial 3831 ourselves. ALKS 3831 is a potential new oral treatment for adults with schizophrenia and bipolar 1 disorder. 3831 offers the robust efficacy of olanzapine, with the addition of a new molecular entity called samidorphan to mitigate the inherent weight gain liability of olanzapine. We filed the NDA for 3831 in November for both schizophrenia and bipolar 1. The picture on the bottom shows what the dosage form actually looks like. It's a solid dose bilayer tablet comprised of olanzapine and samidorphan and available in all the standard olanzapine doses combined with 10 milligrams of samidorphan in each of those presentations. Our launch preparations are focused in 2 principal areas shown on the right, disease state awareness with physicians focusing on the total health implications of schizophrenia and bipolar disease; and engagement with payers to build the case for access to this important new medicine. 3831, as I said, will build and extend our presence in schizophrenia, which began in earnest with ARISTADA. Olanzapine is interesting because it continues to have a huge, important presence in these indications despite the significant weight gain associated with its use. Today, 20% of the prescriptions in schizophrenia are for olanzapine despite its weight limitations because of its efficacy. A similar story in bipolar where olanzapine's efficacy is very well established. This provides a foundation for the introduction of 3831 and we're looking forward to launching it next year. Now let's turn to the pipeline, which is focused in CNS and in oncology. I'll spend a second on how we got there because Alkermes R&D capabilities are quite strong and they've evolved significantly over the last several years. Our origins in formulation and drug delivery drove these strong capabilities in modifying existing drugs, applying sophisticated drug delivery technologies and formation technologies to create new molecular entities, right? I'm sorry, new medicines with new values for patients. This was quite successful and it led to our long-acting atypical antipsychotics, RISPERDAL CONSTA, INVEGA SUSTENNA, which are small molecules as well as BYDUREON, which is a peptide. The next phase we shifted to, including the logic, the sensibility of the drug delivery technology, but building it into new molecular entities, prodrugs and other new chemical entities and new biological entities that incorporated but didn't depend on external manufacturing or other technologies. And that was similarly successful. That led to ARISTADA and VUMERITY and ALKS 3831. But we didn't stop there because the universe, the ecosystems are demanding even more innovation and new targets, new biology and new value-added for patients. And we've been making the shift for the past several years with a focus in CNS and in oncology, embodied in new small molecules and new complex protein fusion products and with a focus in psychiatry, where we have a long clinical experience as well as in neurodegeneration, which is an immediate adjacency and this burgeoning work that we've been doing in oncology, which began with 4230, extends now with IL-10 and other engineered cytokines. We have a fully elaborated and proven R&D and commercial capability in CNS and that provides a good platform for us to move into oncology. We've built a serious set of capabilities in oncology now. And the focus in 2020 is really to expand our clinical trial network, primarily in the U.S., outside the U.S. and with plans for strategic collaboration as well as maximize the development program and anticipate commercialization for 4230. So this is the pipeline with some important new additions for you to note. 3831 for schizophrenia and bipolar, NDA filed in November of last year. 4230, our immuno-oncology agent currently enrolling patients in Phase I/II studies. Our IL-10 program, which is in final candidate selection, our HDAC inhibitors for neurodegeneration and potentially oncology. So let's focus on a couple of those, starting with the cytokine platform, which is based on IL-2 and IL-10. So 4230 is a stable single polypeptide designed to exploit well-established IL-2 biology. It works by stimulating the production of tumor-killing CD8+ cells and natural killer cells. We have a large clinical program ongoing called ARTISTRY. And we most recently showed data from the program at the SITC meeting in November, and the data continues to accumulate, although I want to review the specific data here today. First principles, IL-2 activates and expands cancer-fighting lineages, these CD8+ NK cells that I described. But it also does the same for immunosuppressive regulatory T cells called Treg, they're on the left. It does so because native IL-2 binds to 2 different receptor types. On the left, the high affinity receptor, which is comprised of 3 subunits: the alpha, the beta, and the gamma. This interaction drives that expansion to Tregs, the regulatory T cells. Off right is the intermediate affinity receptor, which is comprised only of the 2 subunits, the beta and the gamma. This interaction drives the expansion of the CD8+ and the NK cells. And that's the interaction that we wanted to design 4230 to exploit. We accomplished this by some elegant molecular biology, which enabled us to create a new protein comprised of IL-2 linked to the alpha receptor as a single molecule. So you can see from the cartoon, this new molecule is unable to bind to the high affinity receptor because the alpha chain is taking up that space but binds easily to the intermediate affinity receptor on the right. So 4230 so far is designed to do a very specific thing in the body and that's exactly what it's demonstrating in the clinic, dose-dependent expansions of CD8+ and NK cells without a corresponding increase in expansion of regulatory T cells. The clinical development program has expanded as the data have accumulated. ARTISTRY-1 test 4230 via intravenous dosing and has 3 parts. Part A, we escalated doses to demonstrate the first evidence of the pharmacodynamics and to establish a recommended Phase II dose. With that dose established, we launched Part B, which is seeking monotherapy efficacy signals in renal cell carcinoma and in melanoma. Part C tests 4230 in combination with pembrolizumab, KEYTRUDA in multiple tumor types. ARTISTRY-2 is focused on subcutaneous dosing and follows a similar pathway as ARTISTRY-1. We'll be testing in dose escalation, looking for a recommended Phase II dose for the subcu route. And if we're successful with that, the program will expand into monotherapy and combination cohorts as well. ION is a collaboration with the Hutch, focused on head and neck cancer, which is run through a consortium of oncology testing centers around the country. So the program is really gaining a tremendous amount of momentum. So 4230 was our first demonstration of an engineered cytokine conferring new benefits, potentially for patients and for the treatment of multiple types of tumors. IL-10 is the second one. IL-2 is an interesting active immunoregulatory cytokine and has distinct concentration-dependent effects. At low concentrations, you see on the left, it actually activates innate immune cells and provides anti-inflammatory as well as immunosuppressive activities. But at higher concentrations, it activates potent antitumor cytotoxic activity. The problem with IL-10 as a native molecule is its brief half-life. So it makes it very difficult to achieve these high concentrations with that as the molecule. So it requires some engineering. Our IL-10 molecules are complex fusion proteins and they leverage this protein engineering expertise at Alkermes. They present 2 IL-10 dimers in the green. So 2 dimers on each arm of an antibody framework. This allows us to achieve higher systemic concentrations with a longer half-life and the inclusion of the Fc portion of the antibody also enables us to engineer an Fc-functioning -- Fc-mediated functionality. So we have a number of constructs in preclinical development, and we plan to select our lead compound this year. We presented data from one of those leads last month at a European conference. These data showed that of fusion proteins activated the targeted cell types demonstrated antitumor activity with weekly dosing in a murine model of colorectal cancer. And this poster is available to you if you'd like to see the data in detail. So that's it for the engineered cytokines. Let's talk now about the most recent acquisition to the portfolio, which are our HDAC inhibitors for epigenetic regulation of synaptogenesis in the brain. It's a mouthful, but I'll explain it a little bit more now. Synapsis play a vital role in brain function. They are the points of communication between neurons. In the picture on the right, you see a picture of 2 neurons at the top of the bottom and the space between them. That's a synapse, and there are more than 100 trillion of them in your brain. Synaptic function is critical to brain development, learning and memory. And synaptogenesis, the creation of new synapses occurs throughout a healthy person's lifespan. Many diseases are characterized by synaptic dysfunction. These are called synaptopathies and include the psychiatric conditions like schizophrenia and depression as well as neurodegenerative disease like Alzheimer's and Huntington's and other developmental conditions. What's so interesting about targeting the synaptopathies is they occur across a wide range of pathologies, independent of the condition, the hypothesis of that increased synetic density and function should lead to clinically relevant benefits. And here's an example in Alzheimer's disease. Showing that the loss of synapses is what correlates with the cognitive decline that you observe in patients. On the left, in the gray boxes, you see 3 images of single neurons from a far left a control brain, in the middle, a patient with Alzheimer's pathology, and on the right, an Alzheimer's patient with clinical dementia. The bumps are called spines and they're the site of the synapses, the points of connection with other neurons. The cognitively normal patient in the middle has more spines than the patient with clinical dementia. And on the right is another experiment or another set of data that quantify that in a different way. So this is looking at synapses in postmortem brains of patients with Alzheimer's disease. And as you see, more synapse loss correlated with more severe disease progression in Alzheimer's disease. So it's a really exciting area for targeting for drug development. Synaptogenesis is under epigenetic control, meaning the transcription of genes associated with the formation of synapses regulated within the cell. DNA is tightly wrapped around histones in this picture, those light blue disks. And that's how DNA is packed in the cell. In order for a gene to be transcribed, it has to be accessible and this is accomplished in a number of ways, one of which is via what's called -- sorry, acetylation. And the picture, the acetyl groups are shown as the yellow dots, acetylate histones open up the DNA and allow for transcription. Conversely, deacetylation causes tight coiling of the DNA and stops transcription. HDAC increase acetylation and drive prosynaptic gene expression in synaptogenesis. So there's an extensive literature on this, on the prosynaptic effects of the HDAC inhibitors. HDAC inhibitors have clear neurobiological effects at the molecular, structural and functional level, and we've been following this biology and this chemistry for a number of years. What got us excited was actually the evolution of the chemistry because the biological foundation was quite strong. HDACs have been approved as oncology agents but the approval of the first generation HDACs are limited by hematological toxicities. It was not clear that the hematological toxicities could be separated from the prosynaptic effects. The cartoon on the right shows 2 HDACs, HDAC 1 and HDAC 2. And typically, your chemistry effort will be directed toward developing inhibitors of these particular isoforms. But HDACs actually function in association with multiprotein complexes. Our chemistry targets these complexes. And specifically, one called CoREST. CoREST is directly involved in the repression of prosynaptic genes in neurons. There are a number of other HDAC complexes and there's much more to explore and understand as we try to harness the potential of this epigenetic platform. So the focus of the program has been in these 3 principal areas. Developing brain penetrant chemotypes that have great pharmacokinetics and good pharmaceutical properties that are selective for this CoREST complex and in so doing, have significantly improved hematological toxicity. And we've made important progress in all 3 of these domains. Data from the program was recently published in Chemical Neuroscience, showing that these novel HDAC inhibitor compounds that selectively inhibit the CoREST complex increase synaptic proteins and spine density and confer functional changes while minimizing hematological side effects. So the program is advancing to preclinical research and IND enabling activities. The primary focus is on these HDAC CoREST inhibitors for synaptopathies. This includes highly prevalent conditions like Alzheimer's disease as well as orphan indications. It's important to restate here that our therapeutic approach conceptually independent -- is conceptually independent of the underlying pathology. Whether Alzheimer's is driven by microglia or by tau or by beta amyloid, it results in synaptopathies, and which is what we're seeking to address. So we're interested in expanding this program to oncology and other disease areas where the biology is compelling. And we're developing, as we anticipate our move into the clinic, biomarkers and translational tools to give us early insight in clinical development. So we're quite excited about the biology and the chemistry around this new platform for us. So I'm going to finish just by summarizing a tremendous amount of work that's been going on within Alkermes to prepare for long-term growth and profitability. And that's taking place in 5 domains, focused on revenue growth; enriching and focusing the pipeline; efficient allocation of capital; our cost structure; and even our governance. So the revenue growth is going to come from execution of the commercial plans for VIVITROL and ARISTADA. These are important medicines in complex markets that require real expertise and specialization. This year, we'll turn on a new stream of cash flows with VUMERITY and prepare for the potential approval and launch of 3831, which will build our presence in psychiatry and leverage our existing psychiatry infrastructure. We talked today about the pipe was evolved away from advanced formulations toward new biology and chemistry, anchored by the 4230 oncology program and now including the IL-10 program and our HDAC inhibitor program. We've been focused on the allocation of capital toward high value, high return settings. Following the receipt of the $150 million payment upon the approval of VUMERITY, we acquired Rodin Therapeutics and the HDAC platform. We went through a process then of prioritizing and focusing in the R&D investment on the highest value programs. And that was part of an overall exercise in evaluating our entire cost structure. In the fourth quarter, we implemented a strategic restructuring in order to reduce our cost structure and in so doing, accelerate and improve sustained non-GAAP profitability. And we've given you some of the metrics around the numbers that affect our forward-looking guidance that you can refer to. And finally, we strengthened our Board of Directors with 2 new independent experts, one, a public company CFO focused on value creation across the business; the other a true oncology expert and thought leader in the field. So it's been a busy and productive time in the company. We're really looking forward to our accomplishments ahead of us in 2020. And I'll finish there. We're actively shaping the future of an exciting business focused in these areas: driving the growth of our proprietary commercial products, advancing a sophisticated pipeline, and positioning the company for long-term growth and for profitability. So thanks very much.Okay, good. It's 9:30. I'm Richard Pops, the CEO of Alkermes. I'm joined by Jim Frates, our CFO; and Cory is up here as well from JPMorgan. So it's your breakout, so we're happy to take questions from the audience.

I can start. Hopefully, we'll have some questions from the audience though, too. I guess where I wanted to start is just kind of getting a better understanding of the commercial side of the business before we get to the pipeline and maybe we can go product-by-product to understand what kind of the pushes and the pulls are as we look out at 2020 and in gaining traction with your existing products and some of the newer ones that are coming on. So maybe we start with ARISTADA. I mean we're surprised it hasn't had a more profound impact on the market. What is being done and can be done to kind of gain that traction to sort of accelerate growth here?

So ARISTADA is our long-acting injectable atypical antipsychotic. And it's got a really excellent product profile and essentially embodies all the learnings that we gained through developing RISPERDAL CONSTA and INVEGA SUSTENNA with J&J. It's growing nicely. It's -- but I agree, Cory, I wish it were growing faster, but it also is a market that is not prone to rapid change. If you look at the history of INVEGA SUSTENNA, RISPERDAL CONSTA, many, many years of continuous growth as they build the use of a long-acting injectables in psychiatry and also just changing physician behavior with these long-acting injectables. The product has a new profile in many ways, year-by-year, following its first approval to where it stands today, and that's important. It was first approved with a range of doses and a range of durations, and that was the principal attribute of the molecule on top of its safety and efficacy. Then we had the 2-month dose approved, which was a unique dosage form in the product offering. And then we also had what's called INITIO, approved, which is allows physicians to initiate treatment with a long-acting injectable in the hospital without the need for oral supplementation for multiple weeks. So now it's interesting, the marketing message for the product in 2020 has moved away from this idea of flexibility. But the doctors choose what they want to do to, this very clear offering of INITIO plus 2 months, 6 injections a year and you have patients that are stable. That was made real with the results of a study we called ALPINE, which came in the spring of last year. And that, recall, was a large randomized study, where patients were randomized to receive either INVEGA SUSTENNA given monthly or the regimen that I just described. And those data showed equivalent efficacy with superior tolerability for ARISTADA. That's a really important finding. INVEGA SUSTENNA is a $2 billion drug in the U.S. and we're just getting going with ARISTADA. So this year, 2020 is about execution of the commercial plan, keeping focused on that message and driving more utilization of long-acting injectables.

So when you look at like the Abilify competitive landscape here, the long -- from a long-acting point of view, the Maintena profile seems suboptimal relative to ARISTADA. So how has that product been able to maintain market share? Maybe what's the reluctance on the -- from the physician point of view to switch and use ARISTADA more broadly.

I think it's less of an affirmative statement that they're reluctant to go to ARISTADA. It's just this complacency, this behavior in psychiatry, particularly in the LIS patient population, in Medicaid patient population, where there's just not a lot of impetus for change. So when we first started and people say, "Well, look, we have a long-acting aripiprazole formulation already on formulary and we'll stick with that." But as the space between our offering and the competitive offering changes with Initio with 2-month, it's harder to say that you can just stick with what you had at the outset. But this is the challenge in schizophrenia in general, and we'll face it with 3831 as well. I mean these are not markets where they are actively seeking new medicines, new branded medicines and jumping on innovative new technologies. They tend to be reluctant to introduce new branded medications into genericized markets. But we say the countervailing force is that there are literally millions of patients who need better treatment. And the churn of these medicines, the churn of patients coming off of medicines is so significant. There's many opportunities to pick up new starts. So we like the profile of ARISTADA very much. We think it's very competitive, and I think the -- we see continued growth for this front for a long time to come.

All right. And then to stay within schizophrenia, given the comments you just made, how should we be thinking about 3831 given some of the hurdles that you've been talking about?

Well, 3831 is so interesting because 3831 is about unleashing the potential of olanzapine for use in the front line. Once again, olanzapine, Zyprexa, it was 40% market share in the go-go days of when both risperidone and olanzapine were launching the first atypicals, and these were the biggest drugs in pharma. And it was only after a period of time in the marketplace, became obvious, evident that the significant weight limitation associated with olanzapine. So when we set out to develop 3831, it wasn't the idea of creating a new antipsychotic. We knew that olanzapine and clozapine were considered to be the 2 most efficacious antipsychotics. The question was, could we attenuate the weight gain? But interestingly, when you get through the development program of having shown that we can do that, the offering is not a drug that doesn't make a gain in weight. The offering is the efficacy. The efficacy is what's going to drive people to this medicine. So in a situation that I just described, where you can stipulate that essentially no patient will get access to a branded medication until they failed multiple oral generic products. When they failed multiple times, what we find from our market researches, what the doctors are excited about is not necessarily another safe, well-tolerated drug, they want efficacy. And they know there's a lot of knowledge in the field of the efficacy of olanzapine. So that's -- I think we're really excited about that. With the overlying caveat that we started, which is change happens slowly in the LIS and in Medicaid populations, particularly disadvantaged patient populations. You don't have teams of patient advocates burning down the building to get access to new medications. It's driven by the physicians with their understanding of the profile. They're using the medicine themselves, getting experience with it and seeing the value of it.

So this -- the launch here isn't about going to the physicians treating those 20% of schizophrenia patients who are getting olanzapine, getting them to switch from a generic to a brand?

Yes. I think it will be a mistake to think that there's a patient population out there, 20% that are on olanzapine. No one stays on olanzapine very long. A lot of that is in the acute phase when people present in the emergency room in the hospital an acute break schizophrenia that get put on olanzapine because it's so efficacious, it's mildly sedating. It's easy to use in the short term. But people just can't stay on it for the long term. And that's why it's not even indicated frontline for the treatment of schizophrenia anymore. So when we talk to physicians and say, "No, this is -- if this is real, if you really have attenuated the weight associated with it, that opens up the use of our using this medicine in schizophrenia and also in bipolar." And so we will get approval in both schizophrenia and bipolar, and they're actually very different because the bipolar component is not a government pay as much, it's more of a commercial. It's a different patient. It's actually a different doctor. So we're going through the process now of figuring out how best to allocate those commercial resources across those 2 different payer groups.

And then on the regulatory front, you've announced that you submitted your NDA back in November. I guess first, has that been accepted yet in the 60-day window? And then also, is there anything you expect with -- from a regulatory point of view? Any nuances to this application to think about?

We should be accepted very soon. I actually don't even know what specific date, but it should be close now. And in this case, we agree with FDA at the outset, the pivotal design would be the 2 pivotal studies, one focused on efficacy, which was a classic efficacy study with positive-negative symptom scales as the primary endpoint. You've seen those data. We ran an olanzapine comparator in that, although the primary comparison was not to olanzapine, it was to placebo. We also numerically separated -- we were numerically superior than olanzapine. But you should think of -- we maintain the comparable efficacy of olanzapine. The second pivotal study was focused on weight. And that was a 6-month head-to-head study versus olanzapine that essentially recapitulated the data from the large randomized Phase II study that we presented as well, showing that we attenuate that weight gain associated with olanzapine. Those are -- that's the basis of the filing. And we hope for a smooth regulatory review.

And from a launch prep point of view, can you just talk about what you're doing there to get this ready and how much you're able to leverage the infrastructure you've established for ARISTADA?

Yes. It's an important point because we're in that market right now with ARISTADA. We're calling on the subset of physicians who really write injectables, which is a subset of the overall physician basically, who write a drug like 3831. But importantly, with the national accounts and the health systems and the policy and all the other things that are necessary, medical affairs, we have that presence already. So the pre-launch activities are focused on 2 major domains. One is what you call disease state awareness, which is not brand specific but the focus of that is really getting to doctors and talking about the overall health of these patients. And it's not just about getting on over antipsychotic, it's about what is the holistic picture of these patients' health. It's amazing. Patients with schizophrenia have 2x greater all-cause mortality. They die twice as fast. And the average life expectancy of a patient with schizophrenia is 25 years less than somebody who doesn't. And that's -- I mean, it's intolerable. And so it's not just about getting an antipsychotic, it's keeping on the antipsychotic. The antipsychotic doesn't make them become obese or get diabetes but -- so still addressing the unmet needs in schizophrenia, that's number one. That's a physician-focused activity. The second is focus on payers because payers -- we know payers are going to put up impediments to using this medicine. So you need to build the dossier and the logic and understand the space between 3831 and other drugs that are currently in the marketplace.

Okay. So then on VIVITROL and kind of going to the traction question there. You've talked for a long time about how the traction is going to be nonlinear with this product, and we know about all the kind of the grassroots efforts there. Is still your view of this nonlinear traction and that you -- things are kind of percolating state levels in towns and stuff like that? I mean your expectations for this product what they've always been? Or has it changed?

It's -- I always say without any hint of facetiousness, there is no product like VIVITROL. So there is no analog to VIVITROL. VIVITROL is not -- it's almost like a social service, it's its own phenomenon. And it doesn't yield to reps calling on doctors, its ultimate commercial potential will be driven by state and federal policy. 10,000 patients will get VIVITROL this week approximately. So it's gone from being a theoretical construct. It's in a really important, yet the dominant form of therapy in the country is replacement medicines with SUBOXONE and methadone. With that said, 60,000 people are going to die of opioid overdose this year. And so policymakers at the local level and the federal level are still intensely focused on trying to change this treatment system to drive better outcomes. And VIVITROL is an important part of that. I say it's stochastic because imagine not just 50 states but imagine 3,000 counties, all with different approaches to dealing with a bunch of federal money flowing in from the top because the federal government has responded to the crisis by allocating money to the states. But they say the states are laboratories of democracy. They're also the laboratories of care. Half of the treatment centers in America don't use medicine. Most of them don't even measure outcomes. So the whole idea is, in all these 3,000 different ecosystems, some of them are using more VIVITROL and getting better. Other ones are running out of money and using less VIVITROL. Then we have states that are activating, we have -- but the encouraging thing as we go into 2020, is that I think we have more than 20 states that are growing at more than 25% now. And that's from a very low base, so you don't see it in the carton sales so much because when they started 0, it's not hard to grow at 25%. But these are places that were -- they were inactive before. They were not even motivated to change the treatment system. So you can see -- so when we model, when we project, when we guide in 2020, for example, we'll try to make a different determination, which will generally be some type of linear extrapolation from the growth rate that comes into the time that we guide, recognizing things can change over the course of the year in many different ways.

Do you get color from the states that are using VIVITROL and growing VIVITROL that are -- remain well funded, that they continue to use more of the product?

Yes. And what's so vexing is the way that public funding can occur. You can have these brilliant programs that are doing really well that run out of funding appropriations in September or in June, depending on the fiscal year. And it's just until they can go through another legislative cycle, get another appropriation. I mean that's why I say it's not within the medical treatment system. This is -- there's no analogy in oncology or in virology or anything, where so much of this is driven by states allocating funds not just for the purchase of VIVITROL, but to create the social systems that enable care to be delivered to patients with serious opioid addiction rather than incarcerating them.

Okay. And then has there been any impact from the noise around the marketing campaign for VIVITROL?

Well that -- we market in a highly compliant way. And what people have -- there are people on the other side who are interested in the use only of replacement medicines, who really want to see, particularly in the criminal justice system, access in the criminal justice system of methadone and SUBOXONE. And as you've seen, there's a lot of interest from the sheriffs and police chiefs and drug courts on the idea of detoxification and monthly VIVITROL injections. So we support those efforts. But we also -- when we advocate, we advocate for patient access to all 3 forms of medicine because if we can get people just getting care, medication-assisted care, we know that the standard of care is going to improve and more patients ultimately wind up choosing VIVITROL over time.

Okay. Again, feel free to raise your hand and ask a question.

Can you talk a little bit about the cost reduction program you're doing in terms of how long you're going to get there? Is there more room to go, et cetera?

Yes. I'll let Jim answer the question. The question was about the cost reduction programs and the specifics on it.

Yes, thank you. Good morning, everybody. I think we started last year with a particular target for the -- really from a bottoms-up approach of our business. So we really did a zero-based approach and looked at a few areas that we needed to do and continue to do well. Rich has touched on them this morning, marketing VIVITROL and ARISTADA well. That takes a certain infrastructure across the country in these medicines. It's very different from marketing, very high-priced drugs in orphan diseases. And so if we're going to play in psychiatry and addiction, we need a certain infrastructure there. So maintaining that sales infrastructure was very important. We then went through our R&D infrastructure and made sure there were a number of programs, including early-stage research into the basis of addiction and healing the addictive brain that we decided we would not -- no longer invest in. So we've concentrated our areas of investment in R&D, specifically in the neuroscience programs that we have, that we've talked a little bit about it on the oncology side with IL-2 and the recently announced IL-10 program. And then on the G&A side and the infrastructure side, we really tried to pare back and prune the company to be appropriately staged for growth for the next, say 3 to 5 years, with the appropriate infrastructure. So we targeted $150 million in savings based on what would have been our 2020 expectations for growth. So when we went through this, when we announced it on our Q3 earnings call, if you look at our -- where The Street expectations were for 2020, we think we'll be able to save $150 million roughly in that sense. Now after that, a few days or weeks later, we announced the Rodin acquisition and that's going to include a roughly $20 million investment in their R&D programs into next year. So you need to think about kind of a net $130 million savings off of where we started with the restructuring program. But we think we're in the right place. We've taken this seriously. Obviously, we look at it every year when we do our budgeting, but I think we've tightened the organization so we can execute on the goals we have for 2020 and 2021 and beyond.

So maybe just candidly on R&D on oncology. Obviously, you guys are a neurology company, you've done quite well. You're getting into oncology, right? So it is the graveyard companies that try to get into oncology when they're sort of focused on something else. Why would you go that route versus...

So the question was about why oncology, why did we get into oncology. And actually, it derived from our protein engineering capabilities that originated from our formulation and drug delivery capabilities. So we actually have a great facility in modifying and creating different mutants of proteins. Originally, for the idea of improving stability, circulating half-life, targeting, whatever. But we got very excited about this idea of being able to use what's called circular permutation, where we can take a polypeptide and actually instead of it having 2 open ends, close those ends and create a circle and then open up a new set of termini on that, and fuse new things on to that. That led to the 4230 program in IL-2. So IL-2 is actually, it was driven not by our saying we want to go into oncology, it was driven by we knew we had a hypothesis that we could engineer IL-2 in a very specific way to engineer out. It's still arteriosus binding in favor of shifting that affinity toward the preferential binding that we were seeking, which would expand certainly in terms of T cells for treating cancer. So we went into the clinic, and we said again, we're not sure whether -- this is several years ago, whether we're going to be an oncology company. We can put this in the clinic and see whether we're having intended pharmacodynamic effect. And from the early experiments we were seeing, yes. Now we began then to build this immunology and oncology group to be able to prosecute this a little bit. But what's interesting is that a lot of the people who work at Alkermes that came there to work on the CNS programs had backgrounds in oncology drug development. Our Head of R&D for example, our Head of Regulatory, a number of people within the company and so we've attracted now this cadre of folks. The weakness we've had up until this point has been scale in terms of clinical operations, particularly OUS because so much of the enrollment, there's so much competitive enrollment in oncology. And that's a major piece of what we're doing this year is expanding our OUS clinical trial sites. But we have a large clinical operations team that we had in place. So building off that was not particularly difficult. With all that said, this is a product that cries for collaboration. Because if the biology plays out the way that we expect it to, this can be a very promiscuous counterparty for a number of different combinations, and you can spend a lot of money on that. And so we do see, building into this now, a shared approach to it with expertise that we would bring in from the outside. I'm not sure if that deal will happen in 2020. I think the data that will support such a deal will reveal themselves in 2020 or not, but that's definitely part of the plan. Now with IL-10 coming behind that, we don't think that we're going to be a dilatant in the oncology phase. What's happening in the labs at the company is seriously focused both on the small molecule side and on the protein side in oncology. So we're going to keep going there.

I just have one more on that. Can you give any type of breakdown in terms of the SG&A is pretty high? But as you said, there's a new [indiscernible] when you take the margin of products. Can you give any guide again in terms of how much you've spent on sales infrastructure and [indiscernible] products? Just looking to get a sense of where the spending is going.

Well, much of that information is in each of our 10-Q filings, in the MD&A. So I'd point you to that directly. But VIVITROL has been a profitable project -- product for us. ARISTADA is getting there with its growth rate. And so now I think the other thing that's important to note is when 3831 is coming along, we ought to see a great deal of leverage from that infrastructure in the CNS space and psychiatry in particular, with the launch of 3831. So...

Question on VIVITROL. Your initial data, the funding issue you're talking about, now you have real-world experience. What is sort of the magic duration in your view for the patient to be on VIVITROL to get maximum time for clearance from addiction? Or these folks life time dependent. Some point in between [indiscernible]?

So the question was about, with the experience that we have in the real world, about what's the optimal duration for a patient to be on VIVITROL. And it's a complicated answer, I'll make it brief. The neuroscience, the neurobiology would say that patients should be on the antagonist for 6 months, definitely more than 6 months to 1 year. I mean it takes time for the addicted brain to heal. Sadly, many patients are not on VIVITROL that long. What happens in the real world is that patients often fall into one of several categories. We have a distressingly large number of patients who get one shot of VIVITROL. And then because of the treatment system inadequacy, it's very difficult to get them to their next provider. Their next -- to maintain that continuity in the community. And so that's just a tragedy. There are many patients who have a number of VIVITROL injections and then say, "You know what, I've got this under control. I don't need to take VIVITROL anymore." And that's probably -- that's a bridge to a different life. Sadly, there's a large number of patients who take VIVITROL for a while and say, "Look, I can control my opioid dependence through this tool, so I can go back to using opioids." And then there are people who stay on VIVITROL forever because their life was chaotic before and then they found VIVITROL and they stay on it. So there's not a single answer. But the hardcore science would say, you need to be on the antagonist medication for a number of months, if not years.

Rich, on the oncology front, when you say something like 4230 cries out for collaboration, is that clinical trial collaboration you're talking about? You're talking about licensing partnership type of [ trialing ]?

I mean financial and operational because there's -- I mean, right now, we're starting to get inundated with people and investigators saying, we'd like to try 4230 in combination with x. And every one of those, the team comes back and say, "This is a great idea and it only costs $5 million." But then you multiply that by 20 and if the program is successful, then the $5 million becomes $15 million. And you can happily take yourself into bankruptcy with all the exciting science that you can do. So this is about first, principles, elaborating the medical and clinical value of the medicine and then picking your spots. If you are by yourself, you pick your spots very carefully. If you have a partnership, you could pick more spots and drive the expected NPV of the medicine.

Okay. With that, our time is up, thank you very much.

Thank you all.