ABIOMED Inc (ABMD) 2019 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Q3 2019 Abiomed, Inc. Earnings Conference Call. (Operator Instructions) As a reminder, this conference call may be recorded.

I would now like to introduce your host for today's conference, Mr. Todd Trapp, Vice President, Chief Financial Officer. Sir, you may begin.

Thank you, Crystal, and good morning, everyone, and welcome to Abiomed's Third Quarter Fiscal 2019 Earnings Conference Call. This is Todd Trapp, Vice President and Chief Financial Officer for Abiomed, and here with me is Mike Minogue, Abiomed's Chairman, President and Chief Executive Officer.

The format for today's call will be as follows: first, Mike will discuss third quarter business and operational highlights and then I'll review our financial results, which were outlined in today's press release. After that, we'll open the call to your questions.

Before we begin, I would like to remind you that our presentation today includes forward-looking statements as it relates to our discussion of our outlook. The company cautions investors that any forward-looking statement involves risks and uncertainties and is not a guarantee of future performance. Actual results may differ materially from those expressed or implied in the forward-looking statements due to a variety of factors. These factors are described under forward-looking statements in our earnings press release and our most recent 10-Q and 10-K filed with the SEC. We do not undertake any obligation to update forward-looking statements.

With that, let me turn the call over to Abiomed's Chairman, President and Chief Executive Officer, Mike Minogue.

Thank you, Todd, and good morning, everyone. This quarter, Abiomed celebrated a historic milestone, as we treated more than 100,000 patients in the U.S., Europe and Japan. We've invested well over $0.5 billion in research and compiled 20 years of Impella development to create the technology for the field of heart recovery with 597 patents and 442 patents pending. We are impacting more patients than ever before due to our commitment to sustainable growth while improving clinical outcomes.

In Q3, our disciplined approach delivered our best ever quarterly results, with $201 million in revenue, up 30% year-over-year and a record 31.1% operating margin. We have strong growth and execution in our 3 key countries: U.S., Germany and Japan.

Abiomed remains one of the fastest-growing GAAP profitable med-tech companies and maintains a stellar balance sheet with $458 million in cash, but no debt. This allows us to remain focused on our mission of recovering hearts and saving lives, while we continue to invest in innovation.

On this call, I will highlight our successful launch in Japan and our expanding 24/7 capabilities in the United States to improve patient outcomes for clinical -- for today's clinical crisis. So first on Impella adoption in Japan. Japan had another strong quarter, delivering $5.4 million in revenue, and we continue to maintain the rigor around training and education.

In Q3, we opened 13 sites, bringing our total to 48 Impella hospitals out of the 103 already approved by the Government Physician Committee. We now have 1% penetration of the 50,000 potential patients. Japan is an ideal match with our mission of recovering hearts and saving lives because there's a cultural aversion to heart transplants and open chest procedures with a strong desire for native heart recovery.

The Impella launch is going well in both clinical outcomes and revenue. I returned from Japan on Monday night where I enjoyed my time with the physicians and our field team. It was rewarding to see firsthand the impact Impella can have on patients in cardiogenic shock. And here, the interest from physicians on the science of unloading the heart. On January 15, we hosted a press conference in Tokyo with [Dr. June Nakata] and [Dr. Yashikisava]. They discussed the benefits of heart recovery with Impella and highlighted the recovery story of a [Mr. Rydel Temavankai] a professional fisherman, who at 28 years old survived cardiogenic shock from myocarditis and went home with his own heart. He is a perfect example of the quality of life benefits and cost savings of heart recovery in Japan.

As previously stated, Abiomed is focused on launching our Japanese business in a controlled manner with comprehensive field training to ensure we achieve the best patient outcomes. We're excited about the progress to date, but have a lot more to do in order to become the standard-of-care in Japan.

Additionally, we are investing in the commercial distribution to prepare for the expansion of products in the short and midterm, such as Impella CP, Impella Connect, Impella 5.5 and Impella RP.

Our second highlight for today's call is our expanding 24/7 capabilities to improve patient outcomes for a clinical crisis. Today, heart failure is a clinical crisis in the United States, with 47% of women and 36% of men over 45 years old dying within 5 years of their first heart attack.

In addition, cardiogenic shock carries one of the highest mortality rates in healthcare. The mortality rate for cardiogenic shock has been 50% for the last 20 years without Impella. Abiomed is investing to improve clinical outcomes around both survival and native heart recovery with on-site and on-call support 24/7 and tracking outcomes on nearly all commercial patients. No other company in this space provides this level of support to help improve patient outcomes.

Our new technology called Impella Connect, launched at 36 U.S. hospitals, provides real-time in the cloud live monitoring to patients on Impella support. This allows Abiomed to provide a new level of exclusive support and offers our customers another tool to deliver better outcomes for patients.

And today, we're proud to announce that Impella Connect has received CE Mark approval in Europe.

On the clinical research side, we're utilizing both Impella Quality Database and the cVAD study to complement our 6 FDA studies and conduct extensive post-market surveillance to identify and validate best practices.

In total, we now have IQ clinical data in the U.S. from over 90,000 patients at more than 1,300 sites. Best practices identified within the IQ Database are further validated in our cVAD study. The cVAD study is an IRB approved, prospective and audited study utilizing FDA definitions on more than 5,000 patients.

For your reference, we included 2 summary slides on the IQ, cVAD and Impella Connect in our quarterly update.

Abiomed's clinical experience and research has stimulated a collaboration on education and led to physician-driven initiatives, such as the National Cardiogenic Shock Initiative, or NCSI. NCSI continues to improve awareness of the benefits of a shock protocol of best practices, which includes inserting Impella before PCI to unload the heart, stabilizing the patient and minimizing inotropic therapy. NCSI has now expanded to 63 hospitals and is having an impact on patient survival with native heart recovery.

A recent patient story shows the life-changing power of best practice protocols. Todd Bester, a 41-year-old man from Rockwall, Texas, suddenly collapsed at home. His girlfriend, a nurse, immediately began CPR. While in cardiac arrest, Todd received multiple defibrillator shocks by paramedics in route to a local hospital and was life-flighted to Presbyterian Hospital in Dallas. Todd was in cardiogenic shock with organ failure due to 90% blockage of his major coronary arteries and had an ejection fraction of only 10%.

In the Cath Lab, Dr. James Park and his team followed the NCSI protocol as a new participating hospital and immediately implanted the Impella CP pump to support his heart before they placed multiple stents.

After 2 full days on Impella support, Todd's heart recovered enough to explant the Impella CP, and he was discharged home 6 days later to spend Christmas with his family. Todd's story highlights how protocols with Impella support before PCI can optimize and achieve the ultimate goal of survival and native heart recovery.

Before concluding, I'd like to remind those who missed our recent investor presentation at J.P. Morgan that we've posted our presentation with an automatic slide advance and animations on the Investor Section of our website and included a link in our quarter's press release.

The presentation includes further details into our pipeline of new products and the science around our pending pivotal FDA STEMI DTU study.

In summary, we're proud of our 100,000 patient milestone, and we'll continue to grow the field of heart recovery and improve patient outcomes by partnering with our customers to use real-world studies to identify and validate best practice protocols.

Abiomed remains focused on our disciplined execution and sustainable growth so that even more patients around the world can benefit from heart recovery.

I want to thank our employees and customers who put patients first every day or as we say in Japan, (foreign language).

I would like to thank our shareholders for their continued support as well.

I'll now turn the call back over to Todd Trapp.

Thank you, Mike. As Mike mentioned, we delivered another record quarter, with 30% top line growth and a 31% operating margin. Revenue in the third quarter was $201 million, an increase of $47 million or 30% versus last year, with solid performance across all geographies. This represents the 17th consecutive quarter of greater than 25% revenue growth for Abiomed.

In the U.S., revenue rose 27% to $173 million, driven by a 24% increase in overall patient utilization. Outside the U.S., revenue was $28 million, up 59% versus prior year, due to strength in both Germany and Japan.

On a constant-currency basis, Germany saw revenues increase 41%, due to patient utilization. And we continue to be very pleased with our controlled rollout in Japan, as we opened 13 sites and delivered $5.4 million of revenue in the quarter.

In the U.S., at the end of our fiscal Q3, the Impella 2.5 and CP have been placed at approximately 1,300 sites. The Impella 5.0 has been placed at 578 sites, while the RP is now at roughly 400 sites in total.

Reorder performance in Q3 was also strong, with a rate slightly above 100%, consistent with the prior quarter. Average combined inventory at the hospitals for the Impella 2.5 and CP rose slightly from 4.2 to 4.3 units per site during the quarter, as higher volume hospitals carried more inventory to support their cases.

In the third quarter, gross margin was 83%, down 80 basis points versus last year, due to sales mix and manufacturing investments, primarily more operators to support our sales growth in both Danvers and Aachen.

R&D expense for the quarter totaled $24 million, up 36% versus Q3 of last year. The variance was driven by new product development for the Impella 5.5 and Impella ECP as well as for data collection related to our ongoing prospective cVAD study.

These investments will lay the groundwork for improving clinical outcomes and sustaining long-term growth.

SG&A expense was $80 million in the quarter, up 20% versus prior year. The increase was due to expansion of our commercial team, additional marketing programs, higher incentive compensation in support of our Japanese launch.

In the third quarter, we delivered $62 million of operating income, an increase of 39% versus prior year. This translated into an operating margin of 31.1%, a new record for the company. Our margin expanded 200 basis points over prior year due to higher volume, which more than offset our growth investments.

GAAP net income for the quarter increased 235% to $45 million or $0.97 per diluted share. The year-over-year increase was due to strong operating performance and a lower effective tax rate. As a reminder, third quarter fiscal 2018 net income included a tax charge of $22 million or $0.48 per share as a result of the U.S. tax reform. Normalizing for this adjustment, EPS increased 26% in the quarter.

Our balance sheet remains debt free, and we ended the quarter with $458 million of cash and marketable securities. Our top priority for cash is to support organic growth initiatives and to continue to build on our intellectual property advantage.

Now turning to guidance. We've increased our fiscal year 2019 revenue guidance to $780 million, an increase of approximately 31% from the prior year. This compares to our prior fiscal 2019 guidance range of $765 million to $770 million, an increase of 29% to 30%. This guidance projects Q4 revenue to be approximately $280 million, up 25% of a tough comparison, wherein the prior year, we reported 40% growth. We also expect 2019 full year operating margin to be in the 29% to 29.5% range compared to our prior guide of 28% to 30%.

In summary, Abiomed had another record quarter. Our razor-blade business model enables us to recover hearts and save people's lives as we deliver top-tier revenue growth, greater than 80% gross margins and an operating margin now approaching 30%.

Keep in mind that the 30% was our 5-year operating margin target that we communicated 3.5 years ago. So we're ahead of that projection, and we will continue to drive for best-in-class performance.

What makes Abiomed really unique is that we've been able to deliver consistent top-tier financial results, while at the same time making significant strategic investments in the business.

We're investing in new products and enhancements like SmartAssist, Impella Connect, Impella 5.5 and Impella ECP. We're focused on clinical research and potential new indications like STEMI. We've expanded our manufacturing capacity in both Danvers and Aachen to support future demand and we also continue to add to our industry-leading commercial team and provide education and 24/7 hospital support, all with a focus on improving patient outcomes.

It's a great time to be at Abiomed, and we're very proud of what we are able to accomplish for our patients.

Operator, please now open the line for questions.

(Operator Instructions) And our first question comes from Chris Pasquale from Guggenheim.

Mike, I want to start with some of the new technology you guys are rolling out. Can you talk about the early experience with Impella Connect in the U.S.? How that changes? How hospitals use Impella when they have access to that 24/7 support? And maybe what the plans are to expand that limited launch to more of a full rollout?

Sure, Chris. Thanks for the question. The Impella Connect is a new tool and for everyone's benefit. It allows us to use a one-way video transmission out of the hospital utilizing their WiFi system. So it's HIPAA compliant and it's secure. And it replicates on whatever tool that you're visualizing, whether it's an iPad or your phone, that console, of the patient on support. So it's real-time as if you were standing in the ICU room looking at the console. The second thing is that you're not able to make any changes and it is secure. So it does allow for the productivity, but also the security of the patient. Now the way we've used it is to ensure that the device is being used correctly, that we can see alarms, we can use it for training. If at any moment there's an alarm that goes off, you can have it set up that you can get a text or an e-mail that the critical alarm has gone off. If anything is changed bedside, the flow rates, it records it, so you can go back and look at when and how the flow rates were changed. And in combination with the new software that's on the consoles, the SmartAssist, it gives us the ability to start looking at algorithms to see how the patient is doing with the intent to wean people off. So if you combined the SmartAssist capability, look at pressures in the ventricle and cardiac output and cardiac power, you now have really a navigation set of tools that you didn't have before. And I equate it to flying a plane and landing the plane. And what you wanted to do is you want to maintain the proper hemodynamics, the optimal unloading, but then if you have the ability to reload a patient and you can only do that with an Impella forward-flow pump because you're not bypassing the heart. That's where we believe we can increase recovery, whether it's for an acute patient or acute on chronic. And I draw the investors' attention to a paper that came out last quarter in ASAIO, in the journal by -- Richard Cheng is the first author. And it talks about how at Cedars Hospital in L.A., they're utilizing the 5.0 to take severely decompensated shock patients and basically salvage them back either to baseline recovery or onto a bridge or either durable LVAD or cardiac transplantation. So there's a lot of information that is going to be coming out of the consoles itself, but then Impella Connect gives a real-time ability for everyone to see it. Now to give one last practical example is, there are times where on weekends or late nights, some of the nurses on-call are not necessarily certified experts in managing these patients. And as a routine procedure, they'll sometimes change sheets or shift the patients. And when that happens, you have the ability to shift and move the Impella pump. So a couple of months ago, on a Friday night at 10:45, a nurse was changing sheets on the bed and turned our patient, which caused the device to move, and the nurse silenced the alarm. And within a few minutes, the nurse received a call from our on-call site people that were watching and monitoring to let them know that she just silenced an alarm that we did not want silenced, to get the physician on-call, which she did. They made the adjustment using the console and the pressures from the device to reposition the Impella properly, which they did, and the device ran flawlessly through the weekend. And on Monday night, the patient was weaned off with a successful outcome.

So the fact that we have all this information on the console, that we can track it 24/7 makes the physicians more productive, makes us more productive. And it really allows us, again, to focus on the proper care in the ICU. And with the algorithms that we have currently and that are coming, we're going to be able to measure things like contractility, weanability, cardiac power output to really provide a new service that doesn't exist and has never existed in hemodynamic support.

And can you just remind us about the economics of this for you guys? There's an investment in service and support on your end. Are customers paying for this as a subscription? Is it something you're providing to them for free to drive utilization of pumps?

So the specific part of SmartAssist, which gives you all the information in the algorithms on the console, that's provided at no cost. That's part of our service contract. For the Impella Connect 24/7, we have a pilot phase of sites that are receiving it and giving us input. Once we hit a certain number, around 100 sites in the U.S., we will be looking at some type of service model, software model where we'll tier the pricing based on the outcomes and productivity of their team versus the amount of requirement that we're utilizing. So we'll have some type of shared expense, but again, what -- the main intent is to improve outcomes for these patients in the ICU. And just to remind all our investors, these are the sickest patients in the hospital. And when you can recover someone, they'll likely leave the hospital earlier with better outcomes, they'll have less readmissions. And these are also some of the most costly patients in the system. So improving outcomes has a financial benefit for the hospital. And just having the ability to see and monitor patients real-time essentially improves the hospital employee productivity and it certainly improves our productivity as well.

Our next question comes from Jayson Bedford from Raymond James.

Just a couple on Japan. It seems like Japan is tracking ahead of your initial expectations. Just wondering if you have an updated number as to what you think revenue can be from Japan in '19?

Sure, Jason. Appreciate your question. So our initial guidance in Japan was about $10 million of revenue. So year-to-date right now after the $5.4 million in Q3, there're just shy of about $12 million. And so we're forecasting right now for fiscal year 2019 for Japan to be somewhere in that $17 million to $18 million range for the year.

Okay. And just as a follow-up to that. Mike, are you accelerating your efforts there? I realize that it's a controlled launch, but from a headcount perspective, are you adding more resources given the momentum?

We are and we're also preparing for some of the new products that I've talked about, so Impella CP, Impella Connect, Impella RP and then after that Impella 5.5. So we plan to be the standard-of-care in Japan for the next 10-plus years. To remind our investors, last April, the intra-aortic balloon pump was deemed Class III for cryogenic shock, which means it's harmful and not to be used. And we have a lot of growing momentum, because in Japan, recovery is the ideal alternative and option, and it also has significant cost savings for the system.

And just I'll ask one more and then drop. Do you expect any new countries to come on in fiscal '20?

We currently have been seeding new countries. So we have patients being done in Australia, in India, other parts of Europe. Italy is a dedicated distributor, that's been going well. But we also are very disciplined that we need to see a commitment from the government, along with the physician societies that they recognize Impella is a new unloading therapy for -- to focus on heart recovery. And what that means is, it's identified as its own DRG. And that's what's happened in the U.S., in Germany and Japan, and we want to maintain that. So all our patients in the future can get the care required. We can build out the infrastructure and heart recovery can become the standard of care.

Our next question comes from Bruce Nudell from SunTrust.

I had a couple trial design questions. With regards to the DTU trial, could you comment briefly about the enrollment criteria to ensure that the kind of heart attack size or the infarct size will be big enough? And also, any point estimates for expected infarct sizes in treatment and control arms? And then, secondly, with regards to evidence development in AKI prevention in high-risk PCI, any initial thoughts about how you might use the cVAD Registry to get approved statement that Impella is, in fact, protective of the kidneys during high-risk PCI?

Sure. So Bruce, thanks for the question. I'll try to summarize it. Obviously, we can have follow-up as well. First, on the trial design, we spent a lot of time at J.P. Morgan talking about the results from the STEMI study. We're very pleased that we've shown that it's safe and feasible to delay for 30 minutes the reperfusion while you unload. And we've met the endpoints and we're moving on to the pivotal. As of this moment, we do not have confirmation on the primary endpoint in details of the sample size purely because, not the government shutdown, but because of the storms in D.C. We're meeting again with the FDA next week. So we're happy to give kind of our overall concept of what we believe, but we can't commit to anything yet because we don't have formal confirmation. So as we stated at JPMorgan, we're happy that the pilot revealed that it's feasible, but also showed signal that the Impella unloading has a potential to reduce your infarct size as compared to area at risk. We used a greater than 4 ST sum in the study. And you see that same trend in that, but you only had a handful of patients that were at 4, the majority were either 5 or 6. But to remind everyone, we actually had a P value when the ST sum was 6 or greater. So that's pretty impressive for only 50 patients. We still believe that infarct size measurement at 3 to 5 days with an MRI is the gold standard. And this reduction of infarct size or an infarct there likely leads to heart failure in the long term. So our intent for this study will be to have a shorter-term endpoint around the infarct size and a longer term endpoint with follow-up on a clinical heart failure measurement. This study will answer important questions. Go ahead if you have a follow-up.

Yes. The only other question then was just about the protect AKI prevention. Any idea for evidence at all there?

Sure. So on the AKI, we actually have data from PROTECT II, the RCT that we conducted back in 2012. And PROTECT II showed for the first time that you could decouple the amount of contrast to renal failure. So the Impella arm had more -- statistically more contrast and had numerically lower renal failure. And woman actually had more AKI than men. So we're doing an analysis on that as well. We do currently have multiple studies that have been done in Europe and the U.S. And we presented data that will be published using the cVAD Registry demonstrating the continued reduction of AKI and showing a little bit of a benefit to the higher flow CP in reducing AKI compared to the Impella 2.5. So there'll be more science coming on that this year, there will be more publications, there will also be some more studies as well. It does appear that there is a connection between unloading the heart. There appears to be a hormonal connection between the kidneys and the production of urine. And for a long period of time, many physicians used to refer to Impella as a urine pump and we'll continue to publish that. But we believe that AKI, acute kidney injury, is an Achilles heel, is a problem for all PCI, for CABG as well as for CAVR. And there'll be a tremendous amount of science on this rolling out in the future.

Our next question comes from Isaac Ro from Goldman Sachs.

So I wanted to start with the expense side actually. The operating leverage came in a little better than we expected. Just trying to think through some of the things that might have gone right there. And if you could kind of help us frame out what you're assuming for the rest of the year sequentially on any kind of unusual expense items that help us kind of think through operating leverage?

Sure. Thanks, Isaac, for the question. So I think, what drove margins in the quarter, I mean, 31% which were a record for the company is, it's the first time we broke the 30% barrier. The 200 base points of margin improvement was really driven by the higher volume, which offset some of our growth investments in the quarter. So we continued to invest a good portion of our heads that we added on the manufacturing side, operators in both Danvers and Aachen to really support our top line growth. We also continued to add to our commercial team in the quarter as well. In Q3, we spent a little bit more on R&D to support some of the new product pipelines like the 5.5 and the expandable sheath. And as I mentioned, these investors will continue to lay the groundwork for this longer-term growth of the business. As I look at Q4 and as you know, we don't give formal quarterly guidance, but I guess, some of the puts and takes, I'd say, is we will have higher revenue in Q4 versus Q3, as we do typically do in our business. In Q4, we do have higher payroll taxes, primarily social security as the calendar year begins in the clock and those taxes reset. We are going to continue to rollout Impella Connect and CP Optical, and that will be -- put a little bit of pressure on our gross margins, but obviously, something we're really excited about to roll out that technology. I think, in Q4, we're also forecasted to spend a little bit more on R&D, primarily around third-project party spend, again, to support some of our projects like the 5.5 and ECP and our expandable sheath. And I think, in Q4, we'll also continue to add to our commercial team at a similar level in Q3. So I expect to see some nice margin expansion year-over-year. If you think about how we progressed during the course of the year, in Q1, we started off with 26% operating margins. Q2, we moved up to 27.7%, 31% at Q3 and we'll be a little bit north of that in Q4. So exiting the year at the right rate.

Great. If I can just ask for a quick follow up on key pipeline events. Just 5.5, when that comes to the U.S. ECP first in man, and then what comes next is STEMI, that would be great?

Sure, Isaac, this is Mike. So 5.5 is approved in Europe. We're doing patients now under a limited launch. As we said, we expect to have the 5.5 as a supplement to the 5.0 by the end of this coming calendar year. For ECP, we expect to do our first in man this year as well. And for the STEMI study, we expect to start this year, we just need to make sure that we get everything organized, we get the final confirmation from the FDA and then roll it out with the proper site selection, but we're really excited. We have some additional products and other indications we're working on top of all of that.

Our next question comes from Raj Denhoy from Jefferies.

Wonder if I could just follow up on a couple of things. So on the STEMI trial, have you confirmed that it's going to be a 2-arm study as opposed to a 3-arm study at this point? That there won't just be an Impella in immediate arm in the trial?

Yes. As far as the 100% confirmation, we feel very strong that that's probably -- that's what we're recommending, that's what the discussions. To remind everyone, we did the pilot so that we can confirm and measure out the specific differences of the 30-minute delay because the Impella was in both arms, was in both -- in the patients during the PCI and after the PCI. And really, what we're trying to show is that this 30-minute preconditioning has the ability to activate a cardioprotective program that the hypothesis was from animals would reduce the reperfusion injury. We believe that we've confirmed that. We've shown that not only is there no harm done to the patient, there appears to be a benefit when the ST sum is 6 or greater and that was the intent. So now moving forward, we're going to do standard of care, which is just a Door to Unload -- which is -- I mean, sorry Door to Balloon, which is today's standard of the angioplasty balloon to opening up. And so we're going to be comparing plumbing only to a heart pump with plumbing. And that's what we're excited about.

Great. Great. And then just recently over the quarter, the PHP trial restarted, the Abbott trial. Any thoughts on that trial? Any of the updates that were provided? Have you heard anything from the field yet in terms of physician interest in that trial? So any thoughts would be helpful.

Sure. I mean Abbott, we will let Abbott give you all the details of their own study, but we like the added support and focus on recruiting high-risk PCI patients that have advanced coronary disease. As I said, they're a good company and they've got a large platform for education and they create a lot of clinical awareness. We all agree there's an unmet need for patients that are high risk that are turned down for surgery. We've referenced the publications that show the majority of these patients never make it to the cath lab or get referred in. And there's 1 paper estimating it's over 300,000 patients per year in the U.S. alone. That being said, it's still a 13.8 French catheter. They did have to stop the study. We're very confident in our innovation and our IP, and we have a 20-year head start. So we look forward to having them join along and -- the CMO of Abbott is a great guy and he used to be a customer. So we wish them the best of luck.

Great. And just maybe 1 last one just, I think you had alluded earlier that you might be doing an Analyst Meeting later this year to sort of give us some updates in terms of your assumptions for market size and the like? Is that still the plan for later this year?

That is Raj, and essentially, we work on behalf of our investors and our analysts. And so there's a lot of new information coming on the STEMI, on the 5.5, on some of the other capabilities we have. So -- and then also updating the total addressable market versus our current addressable market. So we do plan to do that this calendar year. We want to get a couple of these other milestones behind us so we can give more detail and have an Investor Day of substance, which means it's probably later this calendar year.

Our next question comes from Margaret Kaczor from William Blair.

First off, I wanted to follow-up on the Impella Connect in the U.S. You kind of referenced 100 sites that you're trying to get to initially. Is there a backlog of demand where the hospitals or clinicians are asking for this in the hospital right now that you guys can continue to launch into? And then are the hospitals used to that shared expense model that you described? Kind of what's been the feedback so far from those you have described it to?

So Margaret, that's a good question. So the first component of it is, there's definitely an amazing cool factor that the physicians when they have their phone in their hand and they can see this and they can look at the way the patient was tracked or monitored or any change in the night, I mean, that element of technology is -- it doesn't exist anywhere else. The second thing is, on top of the fact that they can see the console, we've put new tools on the console itself, things around cardiac power output, cardiac output, looking at pressures in the left ventricle. Things that they also haven't had the ability to monitor and understand. And so we're giving them the ability to see how the heart is doing, the contractility, and what's happening, with the goal of weaning the patient off. So that part is in conjunction. And then, from a demand, I think, that the people that are focused on the outcomes, the people that are putting the best practice protocols in place, this is where the evolution is going to go, because if you can manage these patients properly, it's not necessarily just 1 less day in the ICU. It's life or death, and it means the patient may go home alive with their own heart. So these are massive changes in the outcomes for their patients. These are huge benefits to the patients and their families. And I think that doesn't sink in at times is the fact that we're managing people that have the highest mortality risk in the hospital and we're giving them 24/7 information to save the patient's life and transform the standard of care. These physicians and nurses have watched these patients die for 20 years or watched them go on to more invasive, more expensive procedures. So the demand out there is a sincere expectation and a quest to improve the standard of care. So we're giving them these tools. As far as the shared revenue of the model, I think that you have to prove that out, which is why we're going to pilot it at the certain sites. But certainly, having this information is going to have a difference and what we're looking at doing is rewarding the site that have the protocols, that have the expertise. And we understand other sites may require a little bit more handholding and more intense support from us 24/7, which includes coming to the ICU. But because of that, that will be over time to wean them back to get their hospital up and get their hospital to where these other best practice hospitals are that are achieving these 70% and 80% survival rates with 95%-plus native heart recovery. So that's the model, and we're super excited about it.

Okay. That's great. And in terms of the P&L, just to kind of hit on 2 things. So you mentioned hitting kind of the prior long-term guidance, I assume we'll get an update on that at the Analyst Day. But Todd or Mike, as you guys look at your P&L and your go-to-market strategy, how do you -- or who do you view as your top peers for operating margin longer term?

The company that we typically benchmark ourselves would be Intuitive. It's one of the companies that has best-in-class revenue as well as margin. So that's a company that we kind of model ourselves at here as well. So again, based on what we're seeing right now, I mean, for us, it's always trying to find that focus between really that short-term margins and that longer-term growth initiatives and I think that's one of the great things about our business models that we can do both, right. We can grow top line faster than most. And we can drive further margin expansion while continuing to invest back in the business. And that's what we're going to do. I mean, if you think about where we were just a few years back in 2015, we were a 12.5% margin business. And over that time period we spent close to $300 million of R&D. And now, we're approaching 30% and we're not done yet. And so we're going to strive to be best-in-class on both the top line as well as on the bottom line, Margaret.

Our next question comes from David Lewis from Morgan Stanley.

Mike, just a couple of quick knits for Todd and then 1 strategic for Mike. Just on the financial side, Todd, I just want to confirm this number in the deck. The 95 CP -- I am sorry, 2.5 placements this quarter, that seems like a very big number. I want to make sure I have that right. What drove that particular number? And just the R&D spend that was the real disconnect on our margin math this quarter. That R&D number continues to scale. Where is that incremental investment going? And then I have 1 strategic for Mike.

David, can you just repeat your first question on the sites?

Sure. I think in the deck -- the 2.5 site number, that looked like a very big number. I just want to understand what are the drivers of that incremental 2.5 site add was in the quarter?

So if you're doing new sites, David, we did 104 new sites, 40 of which were CP. And that compares to 101 total sites prior year. So the breakdown of new sites for 2.5 CP, 5.0 and RP went from 101 to 104. Is that the number you're referring to?

Pretty consistent, we had a different number. So we'll come take that off-line. And then -- that's helpful. And then, Todd, just R&D. R&D investment continues to scale the last 3 quarters. Where is that $7 million, I think, up year-on-year? Where is that incremental investment going?

It's going in the pipeline of our new products like the Impella 5.5, Impella ECP and expandabilities as well we continued to build out our cVAD study and our IQ database. And as you know, that data has been a great foundation for us to attain expanded indications for our products. So I would say between the products as well as our ability to build out our cVAD study and our IQ database.

Okay. And then on the -- just the outlook for 2020. Mike, if you think about the growth, could you just give us a sense of where you're thinking about sort of geographic growth? Obviously, the U.S. coming off harder comps, international could scale with new countries and Japan is ramping. Just give us a flavor for sort of drivers of geographic growth next year as well as new product growth.

So David, relative to your question on geographic, I think what you'll see is continue to be -- maintain focus in the U.S. and Germany. Potentially, based on some additional reimbursement in countries in Germany. We could expand potentially a little bit more resources in France or the U.K. or other parts of Europe. We also expect it now to move a little bit faster. And what I would consider in Japan to be the full launch. We've had kind of this limited launch to ensure we get the infrastructure. So you'll see us try to ramp it a little bit faster in Japan. We always give each quarter the number of sites that are already approved. So we're only at 48, but 103 sites are ready and approved by the government. So where -- we could pick that up, we could be twice the rate now there. And that's going to go to 350 hospitals. But there are certainly more hospitals than the 350, and the balloon pumps are at more than 350 hospitals in Japan. So we'll really see a push in Japan. And we'll continue to build out and plant seeds around the rest of Asia for areas to -- that are focused on heart recovery and just more execution, more discipline. The new tool of the Impella Connect really gives us the ability to audit and check as we open new countries to be able to ensure that they're managing the patient's the way we want them managed and that they're utilizing the best practices. And I think the Impella Connect gives us a little bit more flexibility to experiment and audit in some of these countries that before we might have been hesitant to launch in.

And our next question comes from Danielle Antalffy from Leerink.

Todd, first question for you on the Q4 guidance. I appreciate that it's a tough year-over-year comparable. But it feels like momentum is actually accelerating or at the very least stable. Can you remind us what drove strength in the Q4 last year and the puts and takes of guidance this year given that you're clearly doing better in Japan? Can you talk a little bit about why you're expecting a step down in growth beyond the fact that it's a tough comp?

Yes, well, I mean, listen, consistent with past practices, in Q3, we took up the reign for the third time this year. It was $15 million off the low end and $10 million off the high end. And so again, we're focusing on $780 million for the year, up 31%. But Danielle, it does come down to really bigger numbers and tough comps. I mean, as you mentioned, last year, in Q4, we grew 40% from a top line perspective. And so we're lapping some really difficult quarters. And so we're going to continue to grow at a pace that we think is sustainable with a focus on patient outcomes. And again, as Mike mentioned, our goal remains to be one of the fastest-growing, most profitable med-tech companies in the market. And so I think it's more comps than anything else.

Okay. That's helpful. And just from a cash perspective, Mike, I appreciate you said the focus is investing organically into the business, but you continue to generate strong cash flow. You're now at over $450 million in cash, no debt. At what point do you start to deploy capital elsewhere because it feels like you're generating cash faster than you can spend it. So just curious how to think about your view over the next few years on deploying that capital.

Danielle, we're very focused on our organic growth and our execution. We like to have -- we like having a very big cash position that allows us to continue to invest in innovation as well as create a stronger IP position. And we plan to continue to enforce our IP position. We are placing investments though, on things that are adjacent. So we placed $15 million in Shockwave. I think it's clever technology that helps people with vascular disease and allows us potentially to place more Impellas through the femoral arteries. We've also made upwards of $30 million, $40 million in other investments in new technologies that might be 5 to 7 years out that have a fit into what we're doing or our patient path. And we'll continue to be selective and invest in areas where we see a benefit where we can leverage our core competency of recovering hearts and saving lives.

And I'm showing no further questions from our phone line. I'd now like to turn the conference back over to Mike Minogue for any closing remarks.

Great, I want to thank everyone for their time today and if you have any follow-up questions, please feel free to reach out. Have a great day.

Ladies and gentlemen, thank you for participating in today's conference. This does conclude the program. You may all disconnect. Everyone, have a wonderful day.