Amarin Corporation PLC (AMRN) 2017 Q2 法說會逐字稿

Welcome to Amarin Corporation conference call to discuss its second quarter 2017 financial and operating results. This conference is being recorded today, August 2, 2017.

I would now like to turn the conference over to Elisabeth Schwartz, Senior Director of Investor Relations for Amarin.

Thank you all for joining us today. Please be aware that this conference call will contain forward-looking statements that are intended to be covered under the safe harbor provided by the Private Securities Litigation Reform Act. Examples of such statements include, but are not limited to, our current expectations regarding our commercial and financial performance, including levels of Vascepa prescriptions, Vascepa product and licensing revenues; trends and wholesaler inventories; costs and other commercial metrics; gross margin, expenditures and the adequacy of our financial resources; our current expectations regarding our cardiovascular outcome study, such as timing of interim analysis, study completion, regulatory review and likelihood of success; our plans and preparation for expanded promotion of Vascepa and related market positioning and potential; our plans to purchase additional supply of Vascepa; our goals regarding the timing and scope of international expansion; and our current expectations regarding the effect of our co-promotion agreement on our business.

These statements are based on information available to us today, August 2, 2017. We may not actually achieve our goals, carry out our plans or intentions or meet the expectations disclosed in our forward-looking statements. Actual results or events could differ materially, so you should not place undue reliance on these statements. We assume no obligation to update these statements as circumstances change. Our forward-looking statements do not reflect the potential impact of significant transactions we may enter into, such as mergers, acquisitions, dispositions, joint ventures or any material agreements that we may enter into, amend or terminate.

For additional information concerning the factors that could cause actual results to differ materially, please see the Forward-looking Statement section in today's press release and the Risk Factors section of our quarterly report on Form 10-Q for the quarter ended June 30, 2017. These documents have been filed with the SEC and are available through Investor Relations section of our website at www.amarincorp.com. We encourage everyone to read these documents. This call is intended for investors in Amarin and is not intended to promote the use of Vascepa outside of its approved indication.

Please note that we are also providing slides to accompany this morning's call. These slides, which can be found on our website, www.amarincorp.com, in the Investor Relations section, under the subcategory Events and Presentations, summarize some of the key updates discussed on today's call. Finally, an archive of this call will be posted on the Amarin website, also, in the Investor Relations section.

I'll now turn the call over to John Thero, President and Chief Executive Officer of Amarin.

Good morning. As referenced in our press release, it's an exciting time in Amarin, with record reported revenues and the completion of our landmark cardiovascular outcome study getting closer. We appreciate that you're joining us for this call. We recognize that many companies are reporting this morning. We will try to keep this call succinct.

As in the past, we will start with prepared remarks regarding Amarin's recent commercial, operational and financial performance and then open the line to questions.

Q2 historically has been a good quarter for Vascepa growth. And Q2 2017 was not an exception. We achieved record net product revenues of $44.9 million in Q2 2017, bringing net product revenues for the first half of 2017 to $79.3 million. Unlike in Q2 2016, revenue growth in Q2 2017 does not appear to have been significantly impacted by changes in Vascepa inventory levels at independent wholesalers. Such inventory levels, which we don't control, began and ended Q2 2017 with little change and within a typical industry range based upon days of sales.

Vascepa product revenues in Q1 and Q2 2017 exceeded our expectations coming into 2017. We are bullish that we will continue to see meaningful revenue growth for the balance of 2017. Our net product revenue growth has been led by continued productivity improvements by our commercial team and the positive effect of expanded managed care coverage implemented at the start of this year. Accordingly, we are increasing our guidance for full year 2017 Vascepa net product revenues to an estimated range of $165 million to $175 million.

Mike Kalb, our CFO, will provide additional comments regarding our financial performance. I first want to make a few comments regarding our landmark cardiovascular outcome study, REDUCE-IT.

Recall that this 8,175-patient study started in late 2011. It is a pleasure to have reached proximity to the completion of the study that currently we're able to talk about likely having results from the study approximately a year from now. When we started the REDUCE-IT study, industry focus was on lowering LDL, known as bad cholesterol, and raising HDL, known as good cholesterol. Since that time, the industry has increasingly recognized that lowering LDL cholesterol is important to lowering cardiovascular risk. But that even with lowering LDL cholesterol, many patients still had significant residual risk. During this time period, clinical results have led many to question the benefit that can be achieved by raising HDL or good cholesterol. Clinically, the approach of focusing on raising good cholesterol is being marginalized. Importantly, the need to lower cardiovascular risk after statin therapy has also led many to revisit how improving other biomarkers of cardiovascular risk may help to address the very significant residual cardiovascular risk remaining after statin therapy, including lowering persistent high triglycerides, Apo C-III , and other measures of atherogenic particles as well as showing benefit in inflammatory markers in reducing coronary plaque.

Vascepa's current clinical results from the MARINE and ANCHOR studies as well as other supportive data on these parameters have led opinion leaders to consider how Vascepa fits this need and has brought increased attention to, and interest in, the ongoing REDUCE-IT study. Some of these trends in thought have not yet extended to most general physician practices. But key opinion leaders now are increasingly turning to consider Vascepa. This trend has enhanced in the context of recent cardiovascular outcome studies like CANTOS where the focus was on an inflammation marker and against the backdrop of large genetic studies presented since the start of REDUCE-IT, which suggest that triglycerides and LDL cholesterol levels are similarly effective as predictors of coronary heart disease. The attention is further supported by evidence of EPA's various potential cardiovascular protective mechanisms beyond triglyceride lowering and, of course, the successful JELIS cardiovascular outcome studies of EPA in Japan. This evolution of thought is gaining momentum. There is growing recognition that understanding the significance of lipids, lipoproteins and inflammatory markers on cardiovascular disease is a complex undertaking. It is often not just improving a particular biomarker that may matter, but rather what drug effects may be at play when improvements are consistently shown among relevant biomarkers. That is, it may matter how the markers are improved, with what drug and what are the biomarker improvements signaling about potential cardiovascular risk reduction.

That is the perspective thought leaders have applied to Vascepa, which has consistently shown improvements in the classically studied lipids, lipoproteins and inflammatory markers and has other supportive evidence of cardiovascular risk reduction. An obvious example comes from the results of the JELIS study, in which the effects of EPA were evaluated. But as we have outlined previously, the supportive evidence, which gives us confidence in REDUCE-IT, goes well beyond the JELIS study. Furthermore, there is greater [convergence] on the needs to manage associated risks of not only cardiovascular disease but also diabetes and obesity, which is drawing increased attention of cardiologists, lipidologists, endocrinologists and other thought leaders. In July, a group of physicians, independent of Amarin, formed and attended a medical conference, where the intersection of treating these very disease conditions was the central focus of attention. I believe that the results of the REDUCE-IT study, assuming the results are consistent with our expectations, will help accelerate this dialogue and provide physicians with improved pathway for addressing this pressing medical challenge.

The incidence of cardiovascular death in the U.S. is increasing and the cost of care is huge, demonstrating that Vascepa can reduce the occurrence of major cardiovascular events, should be a significant advance for preventive medical care and lower the cost of treating such events. In the past couple of years, including recently, we've seen reports of other therapies that have been advanced as potential add-on therapies to statin treatment. These therapies highlight the importance of residual risk to be addressed beyond LDL cholesterol lowering with statin therapy. Thus far, none of these studies have reported the level of relative risk reduction as reported in the JELIS study, the study in Japan of the outcomes benefit derived from adding EPA to statin therapy. And while none of these other studies have evaluated the effects of therapies in the patient population being studied in REDUCE-IT, it is notable that Vascepa is unique from the therapies being studied by others, not only in its mechanisms of action but also with regard to such practical factors as product affordability, route of administration, meaning oral versus injectable, safety profile and tolerability.

We are seeing an increasingly greater importance being placed on pragmatic treatment. Pragmatic treatment has 5 key elements: efficacy; safety; ease of use, including tolerability; low cost; and fit for the treatment paradigm.

Vascepa already has existing clinical data and years of [post-approval] market experience that support its favorable efficacy, safety and tolerability profile. It also has a convenient oral dose form and affordable price and broad managed care coverage. These attributes have positioned Vascepa favorably in its current market. We believe the fact that Vascepa is, today, a currently available and pragmatic treatment will help Vascepa potentially become a standard of care therapy following positive REDUCE-IT results.

Collectively, these recent studies by other companies provide us with added confidence in the potential opportunity for Vascepa. Individually, results suggested by top line reporting of the CANTOS study provides us with added encouragement regarding the potential for positive results in the REDUCE-IT study. The CANTOS study suggested that there is value to lowering inflammatory markers in statin-treated patients and emphasized the high level of cardiovascular risk in secondary prevention patients.

Effects of Vascepa in Phase III studies included that Vascepa lowered various inflammatory markers and secondary prevention will represent an important subgroup of patients in the REDUCE-IT study, as it is one of the inclusion criteria for assessing risk for enrollment into the study. A complete list of inclusion criteria for the REDUCE-IT study is available on our website at www.amarincorp.com.

Amarin, of course, remains blinded to the results of the REDUCE-IT study. As expressed previously, we expect that the onset of the target 1,612 primary MACE events in the study, which is the target for study completion, will be reached in early 2018, with results of the study reported in Q2 or Q3 2018.

We do not have an update to provide you today regarding the scheduled interim efficacy and safety review by the independent data monitoring committee based upon approximately 80% of the target events in the study, except to say that we're informed that we should expect to get a report from the committee this quarter, after which we will provide a public update regarding the committee's recommendation. We emphasize that Amarin will not see the data as part of this review, and moreover, for the reasons that we have previously described, we do not anticipate that the independent committee will recommend shortening the duration of this first-ever study in this important patient population. We have been conducting this study for over 5 years. We too want a robust result, which we believe to be in the best interest of patient care, Vascepa's commercial opportunity and, of course, our shareholders.

At this point, I'd like to turn the call over to Mike Kalb, our Senior Vice President and Chief Financial Officer. Mike?

Thanks, John. As John mentioned earlier in the call, second quarter of 2017 showed significant product revenue and script growth over the corresponding quarter in 2016. Our Q2 net product revenue of $44.9 million was $12.1 million, or 37%, above the $32.8 million reported for the 3 months ended June 30, 2016. We recorded net product revenue of $79.3 million and $58.1 million during the 6 months ended June 30, 2017, and 2016, respectively, an increase of $21.2 million, or 36%. Our rise in Q2 net product revenue was driven primarily by an increase in normalized total Vascepa prescriptions led by continued productivity improvements by our commercial team and the effect of expanded managed care coverage implemented at the start of 2017. These positive factors were partially offset by a slightly lower average net selling price for Vascepa during the quarter, driven primarily by a higher proportion of revenues derived from Vascepa prescriptions filled by patients covered by Medicare insurance, the rebate levels for which are higher on average than the rebates offered under commercial managed care insurance plans. Based on data provided by Symphony Health Solutions and IMS Health, estimated normalized total Vascepa prescriptions during Q2 2017 increased by approximately 114,000 and 127,000, respectively, to 344,000 and 372,000, respectively, over the 3 months ended June 30, 2016. This calculates to associated growth of approximately 50% and 52%, respectively, over Q2 2016 and 13% and 11%, respectively, over the first quarter of this year.

As reported previously during Q2, 2016, an increase in the level of inventories held by our distributors calculated based on estimated days of Vascepa sales on hand resulted in the net overall increase in net product revenues of approximately $2.9 million to $3.2 million. During Q2 2017, the level of inventories held by our distributors decreased modestly as compared to inventories held at the beginning of the quarter, as calculated on the same basis. Accordingly, on a pro forma basis, without the impact of the changes in the channel inventory levels, our percentage of net product revenue growth in Q2 2017 was consistent with the Q2 2017 growth in prescriptions.

As we have commented previously, prescription levels do not absolutely correlate to product revenue as channel inventory levels on a days sales basis vary based on a number of factors. We also believe based on information available to us that channel inventory levels at the end of the second quarter of 2017 are within ordinary ranges. As John mentioned, we are increasing our guidance for 2017 Vascepa net product revenues to an estimated range of $165 million to $175 million from the $155 million to $165 million guidance range we provided earlier this year.

In addition to product revenue, we recognized licensing revenue of $300,000 in Q2 2017 and $600,000 in the first half of 2017. Such licensing revenues relate to agreements for the commercialization of Vascepa outside the United States. Most of this licensing revenue came from amortization of the initial $15 million we received from our agreement with Eddingpharm for China.

Gross margin from product revenues was 75% in the second quarter of 2017 as compared to 73% for the second quarter of 2016. This improvement was driven primarily by lower costs, especially reduced API production costs.

Selling, general and administrative expenses for the 6 months ended June 30, 2017, and 2016 were $65.7 million and $54.1 million, respectively, an increase of $11.6 million, or 22%. The increase is due primarily to increased co-promotion fees resulting from increased sales levels, increased promotional activities and increased legal costs, which are subject to quarterly variability.

Research and development expense for the 6 months ended June 30, 2017, and 2016 was $24.5 million and $26.3 million, respectively, a decrease of $1.8 million, or 7%. This decrease is mainly due to timing of REDUCE-IT and related costs.

Amarin reported cash and cash equivalents of $85.5 million at June 30, 2017. Net cash flows from operations, excluding the Q1 debt restructuring, interest and royalties and R&D costs in the 3 months and 6 months ended June 30, 2017, were modestly positive. We are anticipating that net cash flows for 2017 as a whole on this basis will be positive. However, we expect continued variability to the timing of certain items, including API purchases and payments to our Vascepa co-promotion partner. In particular, we expect our cash outflows to be somewhat higher due to the timing of payments in Q3 2017 than Q2 2017.

While we are approaching the end of the REDUCE-IT study, we have not yet reached a phase where costs decline meaningfully and savings we realize from no longer recruiting patients for the study offset costs associated with preparing the study for completion, including conducting the upcoming interim look. For the 6 months ended June 30, 2017, cash outflows related to research and development were approximately $20.8 million and cash paid for interest and royalties in aggregate was approximately $7.5 million.

We anticipate that our SG&A expenses, excluding noncash costs, in the second half of 2017 will approximate expense levels in the first half of the year. We expect R&D expenses, excluding noncash costs, in the second half of 2017 will remain in a range which is consistent with R&D expense levels in the first half of 2017. The majority of these costs are REDUCE-IT related. We expect our operating expense levels to continue to vary from quarter to quarter.

Internationally, our partners in China and the Middle East are making progress. Our partners incur the vast majority of costs related to developing these markets for Vascepa. We anticipate a clinical study at Vascepa in China to be commenced by our partner later this year. We anticipate that late in 2017 or early in 2018, we will receive word from our partner in the Middle East that they have secured the first of what we hope is a series of approvals from countries in the Middle East, allowing for promotion of Vascepa in those countries by our partner.

As of June 30, 2017, we had $37.5 million in net accounts receivable, which are current, and $24.8 million in inventory. As of quarter-end, we had accounts payable and accrued expenses of $65.6 million, which increased from $43.8 million at December 31, 2016, primarily due to the timing of rebates, co-promotion fees and supplier payments. For those who are modeling our financials beyond 2017, we want to highlight that under the co-agreement, there are differences between expenses and cash flow as they relate to pending tail payments. Assuming Kowa continues to effectively co-promote Vascepa for at least the balance of 2017 and 2018 as planned, Kowa is entitled to a tail payment based off the level of co-promotion fee they earn in 2018. We anticipate recording such tail payments as part of SG&A expense under U.S. GAAP in 2018 despite the payment becoming due from a cash perspective in installments over 3 years beginning in 2019. This anticipated 2018 accrual of noncash expense for the tail payments is not part of our 2017 expense levels.

This tail payment provision is not new. We draw more attention to it now because we are getting closer to 2018, and we wanted to highlight this upcoming difference between expense accrual and later cash payment under the agreement. What is new is that we recently amended the Kowa agreement. The amendment was made to tighten physician targeting by Kowa sales representatives, without lowering the overall number of sales calls made by Kowa sales representatives. We do not expect cash outflows to Kowa under the amended agreement to exceed percentages in the teens of gross margin in 2017 or 2018, as the amended co-promotion fee is 18% in 2017, with a modest increase for 2018, partially offset by certain other refinements. As a reminder, our relationship with Kowa is expected to conclude at the end of 2018 after the anticipated REDUCE-IT results. As of June 30, 2017, Amarin had approximately 270.8 million American depository shares, or ADSs, and ordinary shares outstanding; 32.8 million share equivalents of Series A convertible preferred shares outstanding; and approximately 23.7 million equivalent shares underlying stock options at a weighted average exercise price of $3.25 as well as $12.1 million equivalent shares underlying restricted or deferred stock units.

I will now turn the call back over to John Thero for closing remarks. John?

Thank you, Mike. We look forward to continuing to provide you with updates regarding Amarin's progress. We are scheduled to attend 3 investor conferences in September, including Citi's Annual Biotech Conference, Rodman & Renshaw's Global Investment Conference and Cantor Fitzgerald's Healthcare Conference. We will webcast any podium presentations we make at these conferences.

As promised, we will provide an update on the recommendations of the DMC based upon their interim look at REDUCE-IT efficacy and safety results. In speaking with investors, it sounds like you appreciate that it's unlikely that the recommendation of the DMC will differ from a recommendation of continuing to completion. Hopefully, Amarin has made it clear that our expectation is that the study will continue to completion. Just as continue-to-completion recommendation was not a major news issue at the 60% interim look last year, the same will, hopefully, be the result of the recommendation from the 80% look, although a key difference is that the study is now a year closer to completion and that in the expected continue-to-completion scenario, we anticipate having the results from the study to report approximately a year from now.

With that, we conclude our prepared remarks. We would like to open the line to some questions. Operator?

(Operator Instructions) Our first question is from Louise Chen with Cantor Fitzgerald.

So I had a few questions here. First question I had is, in terms of the REDUCE-IT study, when you report the final results in 2018, what type of outcome do you think that you need to see to drive further uptake of the product? And then second study -- second question I have is, if the REDUCE-IT study is positive, would you increase your marketing to consumers? And what type of programs would you pursue? And how should we think about the costs for that? And then the last thing is, if REDUCE-IT is positive, do you think that you will continue to market this drug on your own? Or would you seek a primary care partner? How do you think about the expanded market opportunity?

Louise, this is John. Thanks for the questions. I'm here with a variety of teammates, but let me dive into those responses. And I think I'll get them covered, but they'll jump in if I'm missing anything major. With respect to the results, your first question, results of the study and what's meaningful, that's actually a very interesting question. We had done some market analysis several years ago that suggested that anything in the high-double -- high-single digits rather in terms of relative risk reduction would be meaningful, particularly given a product that has the affordable cost and safety profile of Vascepa. Since that time, other studies have reported out that have had in some cases like in IMPROVE-IT, some receptivity despite even lower relative risk reductions than that. Other studies have reported out like FOURIER's on PCSK9s that, I think, also somewhat lowered the bar of not outstanding relative risk reduction and that relative risk reduction at a very high cost. So from a commercial perspective, I think that, in some sense, the bar is lower than what there may be from a regulatory perspective. From a regulatory perspective, of course, we need to hit a statistically significant relative risk reduction. We're, in parallel, continuing to look at real world economics data, all of which is continuing to support our thesis here that addressing lipids, lipoproteins and other markers of risk beyond LDL reduction provide value in that -- in patients who have high triglycerides, that the risk is significantly higher than those without. So this will all become much clearer, obviously, after we have REDUCE-IT results. But I think that the need for what we strive (inaudible) pragmatic treatment that hits those 5 characteristics we think -- which we think Vascepa touches on, across the board, will position Vascepa uniquely and positively. With regard to the marketing of the product post REDUCE-IT, we've done a lot of planning on that. We have intentionally kept the current size of our sales force steady. We have evaluated and continue to evaluate whether -- certainly, the argument can be made -- and our analysis shows, that if we were to add more sales reps today, we would grow revenues faster. We've been harboring our cash to, as we say in our 10-Qs, our current resources on the cash side get us through the REDUCE-IT results, we think that the right spending of our cash is to continue to invest that in the completion of the REDUCE-IT study. Coming out of REDUCE-IT, we think that the right size for our U.S.-based sales team is between 400 and 500 sales representatives that would, not necessarily initially, but, hopefully, shortly thereafter also begin to evolve into consumer marketing. We have, historically, been a predominantly direct-to-physician in our promotion of this product. But given the nature of the product and the number of people affected and with outcomes results, we think this should become a consumer story. We recently added a head, a very experienced consumer marketing person. We are doing analysis as to what the right messaging would be, what the right timing would be. And after we've done some further market analysis, we'll have better answers, internally at least. We'll really need to see the REDUCE-IT results before pulling the trigger on it as to what scope of consumer marketing should be executed on. But that's expanded sales force in that range, but also a consumer piece. But details will be better illuminated after we see the REDUCE-IT results. In terms of the last question about, do we do this on our own, do we continue to go forward with our existing co-promotion partner or do we do something with somebody else? The piece that we can control here is our own execution. So we think that this is an opportunity that should be measured in the billions in terms of dollars. We have progressed it to this point and have terrific support amongst KOLs, terrific managed care coverage, good strong supply. But 5 years now of positive history in the market, we think we are positioned that we can do this ourselves. But we work for our shareholders. So we're always looking for how can you maximize value. We're not relying on others. We don't -- we can't control that piece. We're making sure that we can do it on our own, and we think we can. But I hope those questions are -- I hope I've answered your questions.

Our next question is from Joel Beatty with Citi.

This is Shawn calling in for Joel. With REDUCE-IT outcomes getting close, can you talk to us a little bit about key secondary endpoints and individual MACE subcomponents that, do you think, can be meaningful for your message and marketing strategy? And could you frame that answer in the context of CANTOS and how that readout may have changed the way that you're approaching this?

Shawn, I'll jump in a bit on the first piece of it and then I'll ask our Chief Medical Officer, Craig Granowitz, to talk a little bit about CANTOS. With regard to secondary endpoints, for our ANCHOR study and our MARINE study, as you've probably seen, we've had a strong series of publications over extended period of time. The science in lipid management is deep and evolving, pretty complex, actually. And the science behind the (inaudible) disease is equally deep and complex. We think that there's been value, because different key opinion leaders look at different lipid measures in different ways to having broad data for people who want to dig deeper. That being said, what really matters here is the primary endpoints of this study. And in the primary endpoint, we think we're off to the races relative to the commercial marketing. And it is really getting to that primary endpoint that's key. Will we look at the underlying data from the start? Absolutely, because we want to make sure that we're not conveying information which is in any way misleading. But what matters most is the primary endpoint: are we across a composite MACE endpoint, reducing residual risk in a significant way versus placebo and statin therapy? I think that's really where the primary focus should be. Relative to CANTOS, Craig, if you want to jump in on that.

Sure. Shawn, thank you for your question. And we're very excited by the results of CANTOS and that they're going to be presented at the European Society of Cardiology meeting later this month. We believe that it's an important study for a number of reasons. The first and most importantly is, it highlights, again, the residual risk and the need for treatment beyond just lowering LDL. And that's significant risk remains despite managing patients with statins. And in that regard, it continues to highlight that the number of deaths due to cardiovascular disease in the United States, actually, has reversed that negative trend. It's actually increasing again and remains the #1 cause of death in the United States. The second issue around CANTOS, which is so exciting is it validates that there are new mechanisms of action that are out there and that new mechanisms of actions can have a clinical effect on top of all the existing treatments that are available. Specifically to Vascepa, many of the same markers that are important in the CANTOS trial are also positively affected by Vascepa, particularly CRP. And as you know, CRP is an entry criteria for the CANTOS study and we've demonstrated in a number of our trials and published that CRP levels are affected by Vascepa. So we believe one of the mechanisms of action, or one of the many mechanisms of action of Vascepa, is related to its anti-inflammatory effects, the nature of which exactly have not yet been characterized, but certainly, it's another positive indication on the connection of inflammation and outcomes, and we know that Vascepa is an important anti-inflammatory. I think the last area that is important and John touched on this in his comments is the importance of having a pragmatic therapy. And I think there's been quite a bit written about the costs and administration of Ilaris, which is the drug studied in CANTOS. And a very high per dose cost and an injectable route of administration, I think, will make that product far more difficult to be used across the tens of millions of people that are at risk. And in that regard, we also see it again as a validation of Vascepa -- our lower acquisition costs, our great tolerability profile and over 150,000 patients on treatment today and the ease of plotting an oral dosage form into the current modes of care today. So I think in that regard, we see CANTOS as being important in and of its own right, but further validating the Vascepa and REDUCE-IT hypothesis.

And as a quick follow-up, are there any co-indications currently in the patients in the 500-plus indication that clinicians are using Vascepa more in? And do you think those trends could forecast how the market could play out after REDUCE-IT?

The broad spectrum of benefit provided by Vascepa has been published and is available to those clinicians who are paying attention. As a reminder, we are only calling on -- with our sales force -- a very small sliver of the population of physicians who prescribe lipid-modifying therapies. We don't have the resources to go beyond that. Of those knowledgeable physicians, there is a range as to what those physicians view to be the benefits coming from Vascepa. Some of them are on the triglyceride side of things, some are on more specific atherogenic measures like Apo C-III or apo B. Others are on the reductions in inflammation. There's some use that's noncardiovascular that's going on. But predominantly, particularly, in the population of patients with very high triglycerides, the risk there is pancreatitis, in particular, and whether the physicians' key focus is on the triglyceride reduction or the atherogenic particle reduction or the inflammation reduction, they're all pointing towards their view that there is risk reduction coming from the product. I think our opportunity continues to be significant relative to treatment of patients who have triglycerides greater than 500. In some ways, they're not as recognizable in the system sometimes, as patients with triglycerides, say, between 150 and 500, many of whom are already on statin therapy, for example. And as we get down below the 500, the diversity of use and the value as to what different clinicians, particularly endocrinologists versus cardiologists versus GPs versus endos look at, can vary. And the broad spectrum of benefit that Vascepa provides is interesting in that respect. Everybody is looking for outcomes data. So moving from biomarkers to outcomes-based results, we believe, will be dramatic and probably provide a very different answer to this question than what we have today.

Our next question is from Matthew Andrews with Jefferies.

Mike, a couple for you. Gross net appears to be variable quarter-to-quarter and looks to be roughly 50% to 55% in Q2. So is that reasonable moving forward through the back half of '17? Second, can you discuss expectations on cash levels in the runway? And I had a couple clinical after that.

Thanks, Matthew. Yes, as we mentioned on the call, the rebates are greater on the Med Part D, and I think given where we are with the [access to lives], probably it's a good indication of where we'll be for the second half of the year.

Go ahead. (inaudible) second piece of the question.

Expectations on cash, as we've said and continue to say, we believe we have enough cash to get through REDUCE-IT results, which we believe will be Q2 or Q3 2018. As mentioned, we are cash flow positive when we back out the R&D, Q1 debt restructuring and then interest, royalties and various finance [constructions], and we think that trend will continue.

Okay. And then at what point do you consider resuming clinical development of the statin combo AMR-102? And then do you have any sense how broadly Vascepa is used -- being used in the mixed dyslipidemia market and switches from generic Vascepa? And that's it.

I think in that question, you mean generic Lovaza at the end there, but I'll...

Oh, yes, yes. Sorry, I apologize. You're right, Lovaza.

There's no generic Vascepa, and we're not expecting a generic Vascepa for quite a while.

Yes, apologies there. My mistake.

No, no. I just wanted to point that one out. In terms of the combo therapy for those who maybe haven't focused on or thought about that opportunity for a while, we had, before the FDA reversed its position following results from other therapies (inaudible) relative to approval of the mixed dyslipidemia indication coming out of our successful -- I mean, FDA agreed, successful results of the ANCHOR study, we had done feasibility work with regard to combo products, looking at the combination of Vascepa with statin therapy, actually, we picked rosuvastatin, which is Crestor, as our model for that. Why? Because technically actually that was a more difficult statin to work with relative to the combination (inaudible) we could prove feasibility with that particular statin. If we wanted to do something like atorvastatin, LIPITOR, that would be technically easier. We reached a stage of proving technical feasibility. And then after the FDA surprise, now 4 years back, we put that product -- project on hold. It continues to be on hold after we see REDUCE-IT results. We anticipate that we will have R&D expenditures coming down dramatically versus where they are today, and we'll be reviewing a host of different potential development opportunities, including the potential advancement of a combination product with statin therapy. We will want to, of course, do market analysis at that point in time as to the convenience advantage of having sort of 2 therapies delivered together. But we were going to hold off on that decision until after we see the REDUCE-IT results. In terms of your second question of -- I'm sorry -- who is using Vascepa today and are they above or below 500 mg per deciliter in terms of their triglyceride levels and are we getting those new starts versus switches from fenofibrates or from Lovaza -- at least that was my interpretation of the question. So a few things, one is remembering that until about 2 years ago, we were only marketing to -- or only promoting to use above 500. Following an agreement that we reached with the FDA about 1.5 years ago, we've been promoting, as allowed, the results of the ANCHOR study. The lipids are -- and in parallel, the FDA had removed data from the label of Lovaza relative to its effect on patients in the 200 to 500 range, as its label is the same as ours and also the precautionary language and removed from the labels of fenofibrates and niacin for that matter, the use of (inaudible) therapies on top of statins, because in their view, the risks outweighed the benefits based upon the available clinical evidence. I think, in general, trying to put everybody on the same footing with respect to labeling for what's known. It is very difficult to quantify patients who are at various lipid levels and where they start from. Trying to figure out where the baseline lipids are is tricky in part because, yes, these patients are often on a prior therapy, whether it be Lovaza, which is predominantly generic these days, or fenofibrates, which is predominantly generic these days. And switches occur. We have various anecdotes that got published, anecdotes of 7 patients, 10 patients, 12 patients published over the years by various clinicians, showing that patients switched from these other therapies to Vascepa predominantly showed improvement to their lipid profile. But these are not broad, well-controlled, high-in-number studies. We do remind people that the labels for these earlier generation products were all arrived at based upon registrational studies done in a pre-statin era, meaning they were studied without any statins in the study. We think that the more real world study of Vascepa has been appropriate and valuable. And the fact that Vascepa has been shown to work well both on and off of statin therapy, we think is a benefit (inaudible) safety profile of our drug benefit as well. It's -- most of our starts -- most of our interacts, are new starts. Some of those are patients who probably were on Lovaza 6 months or 8 months ago. It's not uncommon to roll off of prior product. But most of our patient starts are new starts. We are increasingly getting switches, but it's still predominantly that we're getting patients who are new starts, and it tends to be the riskier patients in physician practices that are getting Vascepa as the treatment. So I don't have specific quantification for you, but I'm hoping that some of those comments are useful.

Actually, we'll take one more call. And that will be John Boris with SunTrust.

First question for Mike. Just the number of selling days in the quarter versus 2Q of '16 and then the number of selling days you'll have in 3Q versus the year-over-year comparable, Mike. That's the first question.

Yes. Thanks, John. So first, I will say that we think we were and continue to be within normal industry ranges. Based on our calculations, we're probably in and around the 2-week range now. At the end of Q2 '16, we were probably in or around 4 weeks based on day sales on hand. I think the second part of that question, where do we think that will go for the rest of the year, look, we don't control, obviously, how the wholesalers buy, but if we're at around 2 weeks right now, I don't think there's a whole lot of room for them to go down.

Secondly, on the co-agreement, are they just-- how did the agreement actually change? Are they just focusing with the same level of detail effort on more high-decile physicians? But how would you qualify the changes that were made to that agreement? And then I have one last question that's clinically driven.

John, this is John Thero. With regard to the Kowa agreement, we've been working with them now for 3 years. They've been contributing positively. We and they have accumulated a lot of experience together. And as we looked at our targeting, we felt as though the tightening of targeting would be valuable. So they're going to continue to make the same number of sales calls. But we think that rather than calling on a large, large number of physicians but only getting to some of them once per year, that tightening it up and going just higher value of physicians makes the most sense. There were some economic changes made to reflect realities of where we are today, relative basis, but the key change in terms of driving revenues, we believe, will be the heightened focus on key targets and that should drive increased frequency to those targets and, we hope, increased prescriptions based upon that targeting.

So they are concentrating on more high-decile, higher-prescribing physicians, with the same level of detail effort?

That's correct.

Great. And then last question for Steve. If -- one thing that Merck clearly indicated was that one of the findings out of the IMPROVE-IT study was that the longer that it went, the data clearly at the tail end of the curve matured and separated clearly from the active statin arm. They used that finding to allow REVEAL to go to completion, but again, it was at the tail end. Although the results haven't been presented, but it was more at the tail end, where there was separation that led to a significantly -- statistically significant effect. Just your thoughts about the read-through from those 2 studies, and most notably REVEAL, which is just recently top lined to that of the IMPROVE-IT study.

Yes. Well, I think one phenomenon you're seeking to is ensuring that you have sufficient patient follow-up. And as you know from our study design, it was designed to have a median follow-up of about 4 to 5 years. And as published in our design paper, there will be some patients followed up for as long as 6.5 years. So -- and then, of course, each molecule is going to have its own potential time frame across which it might separate from the placebo arm. And of course, we have the JELIS results, which at least in that study, the exact same active ingredient separated early and continued to maintain its separation. So these things can vary depending on the mechanism of action. Obviously, for the REVEAL study, you're talking about a CETP inhibition, and in the IMPROVE-IT study, you're talking about a different class of agent being added onto a statin therapy, ezetimibe. So we feel confident in our study design in terms of our follow-up and in terms of the other design parameters. And I hope that addresses your question.

It does.

This conclude today's conference. You may disconnect your lines at this time. And thank you for your participation.