ABIOMED Inc (ABMD) 2008 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the third quarter 2008 Abiomed, Inc. earnings conference call. My name is Stacy, and I will be your moderator for today.

(OPERATOR INSTRUCTIONS)

I would now like to turn the presentation over to your host for today, Mr. Dan Sutherby, Chief Financial Officer. Please proceed, sir.

Good morning, and welcome to Abiomed's fiscal third quarter of 2008 earnings conference call. This is Dan Sutherby, Abiomed's Chief Financial Officer. I'm here with Mike Minogue, Abiomed's Chairman, President and Chief Executive Officer, and Dr. Karim Benali, our Chief Medical Officer.

The format for the call will be as follows. First, Mike will provide you with strategic highlights for the quarter. Karim will then provide a clinical and regulatory update. I'll provide a summary of the financial results and then we'll open the call for your questions.

Before we begin, it is necessary to remind you that during the course of this call, we will be making forward-looking statements, including statements regarding future financial performance, product development efforts, Abiomed's strategic operational initiatives, market response to our new products, our progress towards commercial growth and future opportunities. And Abiomed's actual results may differ materially from those anticipated in these forward-looking statements based upon a number of factors including uncertainties associated with development, testing and related regulatory approvals, competition, technological change, anticipated future losses, complex manufacturing, high quality requirements, dependence on limited sources of supply, government regulation, future capital needs and other risks detailed in our SEC filings.

Investors are cautioned not to place undue reliance on any forward-looking statements, which speak only as of the date of today's conference call. The Company undertakes no obligation to publically release the results of any revisions to these forward-looking statements that may be made to reflect events or circumstances that occur after the date of this conference call or to reflect the occurrence of unanticipated events.

Comparative reference made today financially in this call to revenue, gross margin or other increases or decreases refer to third quarter of fiscal 2008 as compared to the third quarter of fiscal 2007.

In addition, today we will be discussing our Impella products that are not yet FDA approved.

I am now pleased to introduce Mike Minogue.

Thanks, Dan.

Good morning, everyone. Thank you for taking the time to join us today. It has been a very productive quarter on several levels.

So there's two topics for today's call -- Q3 financial results, and we're also going to do an extensive review of our trials and regulatory status on multiple products.

So, first, on Q3 performance, revenue is up 24%, to $16 million. It's the highest ever in the Company history. Revenue in Europe and the U.S. was the highest ever. Gross margins continue to be strong, at 76%, and Impella disposable revenue was up 192%. Excluding intra-aortic balloons, we shipped 676 disposables this quarter.

So overall the trends on sales, gross margin, regulatory approvals are positive. These results were achieved without 510(k) clearance on the Impella 2.5 and with one week of the iPulse PMA approval.

Our second topic is on our regulatory status, and I would like to first summarize the many updates since last quarter. And after that Dr. Karim Benali, our Chief Medical Officer, will give the descriptions of the primary and secondary endpoints of the current Impella 2.5 high-risk PCI study, the pending Impella 2.5 AMI study and also update you on the Impella 5.0 study.

So let me first summarize the U.S. regulatory path since our last call, and there are seven main updates.

First, on the Impella 2.5 510(k) update, we're working to resolve and complete the bench testing for our Impella 2.5 510(k) submission. As our January 24 press release stated, the three other questions from the FDA on labeling and clinical data have been resolved. While there are no guarantees for a 510(k) clearance, current informal feedback is from the FDA that the Impella 2.5 is on a substantially equivalent path. Our timeline remains unchanged, with expectations of on or before March 31, 2008 for such a clearance.

Second update is today we announced that we've submitted officially for the IDE approval to start the Impella 2.5 AMI pivotal study, after much review on the protocol with the FDA, and Dr. Benali will discuss this in more detail.

On our Impella 5.0, we're continuing on our 5.0 pilot study and have disclosed this month at the Society of Thoracic Surgeons very positive results for survival in this at-risk patient population.

Fourth, we announced our new AB Portable Driver, and today we announced that we have submitted for IDE approval to begin a U.S. study for patient use and hospital discharge at 20 transplant hospitals for 30 patients. We also received approval from the FDA for the AB5000 VAD with one-year reliability. In Europe this quarter we anticipate CE Mark approval for the AB5000 portable console, and we're actively recruiting the initial hospitals now.

On our iPulse platform, we received PMA supplement approval on December 19 on our unique combination intra-aortic balloon VAD console.

On AbioCor, We received HDE supplement approval January 24. We've selected four hospitals with the clinical expertise and long-term dedication to this technology. They are The Johns Hopkins Hospital in Baltimore, Maryland; Robert Wood Johnson University Hospital in New Brunswick, New Jersey, home of the Giants; Texas Heart Institute at St. Luke's Episcopal Hospital in Houston, Texas; and St. Vincent's Hospital in Indianapolis, Indiana. The device sells for $250,000, and we now have a $1.25 million backlog in orders. As we have stated in the past, this will be a very controlled rollout to ensure success for the patient, the caregivers, the hospital and Abiomed.

CMS recently proposed reimbursing the hospital for the AbioCor procedure, as did three other health insurance providers. We now have validation from both the FDA and CMS on this unique technology, and we are very pleased with our progress.

Number seven is relative to the new products and new U.S. studies such as the Impella implantable, the Impella right-side percutaneous pump and the Impella pediatric. And we will wait until our other submissions are completed before we move forward on these products.

Pertaining to the rationale of completing a PMA path, with the Impella 2.5 510(k) clearance, we believe a PMA approval would lay the foundation for Impella to become the global standard of care in the cath lab for high-risk PCI and AMI, as compared with today's standard of care, the balloon pump, which only has a 510(k).

Now I'd like to turn the call over to Dr. Benali.

Thank you, Mike, and good morning, everyone.

As Mike mentioned I would like to provide an update on the high-risk PCI pivotal study with Impella 2.5, the pending AMI pivotal study with Impella 2.5 and finally the Impella 5.0 pilot study.

Today, interventional cardiologists are attacking more complex and high-risk cases during percutaneous coronary interventions, or PCIs. Some solutions have been attempts to provide hemodynamic support during these high-risk PCI cases. For instance, if the left main coronary artery dissects during PCI, the heart can stop, and the outcome can be disastrous if the patient is not supported with a pump that can provide enough blood to the general circulation to maintain brain, kidney, heart and other organ perfusion.

We believe that the Impella can help cardiologists improve patient outcomes for these high-risk patients. PROTECT I, the first U.S. safety pilot trial, showed excellent results in terms of safety and ease of use of the technology. I would like to remind everyone that the safety results of the PROTECT I pilot study were presented last PCT in October 2007. The results were also accepted by FDA, which allowed continued enrollment after the first 20 patients.

Dr. William O'Neill, professor at the University of Miami, is a leader in the field of interventional cardiology. He is also an author of more than 600 publications. Dr. O'Neill is the principal investigator of the PROTECT I and actually also PROTECT II trial. Dr. O'Neill stated, referring to the Impella 2.5, "This is by far the most exciting, impressive and effective hemodynamic support system I have ever seen, and I think it's going to have a huge role in the cath lab."

In Q3, we started patient enrollment in PROTECT II pivotal trial. PROTECT II is a randomized trial that is designed to assess the superiority of Impella 2.5 over the intra-aortic balloon pump. We have expanded the inclusion criteria to include patients with triple vessel disease, and we have increased the number of hospitals from 8 to 150 for PROTECT II. Therefore, we expect higher enrollment rate at significantly more hospitals as compared to PROTECT I.

Patients that are enrolled in PROTECT II trial present with poor cardiac function with an ejection fraction of 35% or less and are scheduled for an intervention on either the last remaining patent vessel or the left main coronary artery that supplies blood to more than 75% of the left heart. Or the patient can present with poor cardiac function with an ejection fraction of 30% or less and triple vessel disease.

Based on the registry of the National Heart, Lung and Blood Institute in the United States, referenced in the paper published in The American Journal of Cardiology in 2003, about 75% of patients with poor cardiac function have multivessel disease. For example, to date, in PROTECT II, about 70% of enrolled patients have triple vessel disease.

Three hundred twenty-seven patients will be enrolled in each arm of the study, for a total of 654. Again, 327 patients will be enrolled in each arm of the study, for a total of 654 patients. To give some perspective on the number, each year market reports show about 23,000 intra-aortic balloon pumps are used for high-risk PCI in the United States. The primary objective of the study is to demonstrate evidence that the prophylactic use of Impella 2.5 is safe and more effective in preventing major events as compared to intra-aortic balloon pump in patients undergoing high-risk PCI.

The primary endpoint will be a composite endpoint of major events assessed at three days post-PCI. These major events include death, myocardial infarction, or heart attack, stroke, repeat intervention on the same vessel, need for cardiac or vascular surgical operation, acute renal dysfunction, increasing aortic insufficiency, severe hypertension requiring immediate treatment, CPR, or severe arrhythmia requiring emergent treatment, and, finally, failure to eliminate or reduce significantly the blockage in the targeted vessel.

The outcome of the Impella arm will be compared to the control group treated with intra-aortic balloon pump. Improvement of renal perfusion and cardiac power output will be tested as secondary endpoints. We believe that the Impella will provide the better perfusion for the kidneys and better cardiac power output as compared to intra-aortic balloon pump. The cardiac power output is a combination of the cardiac output and the blood pressure, and has been shown to be the best hemodynamic predictor for mortality, as demonstrated in two recent publications in 2004 and 2007 by Finke and Mendoza in the Journal of American College of Cardiology and the American Heart Journal, respectively.

We are very pleased to see the enthusiasm and the overwhelming positive response for the interventional cardiologist community with respect to PROTECT II. We believe that PROTECT II will help better understand the risk profile and treatment pathways for this patient population. At the same time, it will help assess the role of the prophylactic and active circulatory support during these high-risk PCI procedures as compared to the [IAB based on] (inaudible) guidelines recommended by the ACC and the AHA and published in the Journal of American College of Cardiology and Circulation in 2006.

With respect to our second pivotal trial, we are pleased to announce that we have formally submitted the IDE for approval of the AMI study protocol after multiple discussions and consultation with the FDA. Because the protocol is now under formal review at the FDA, we cannot confirm all the details of the submission until their final acceptance. However, we would like to provide an overview of what we submitted for the trial design.

The object of this study is to demonstrate the superiority in terms of safety and effectiveness of the Impella 2.5 as compared to the intra-aortic balloon pump in the treatment of patients undergoing PCI during a heart attack with hemodynamic instability. The study will include about 192 patients in each arm, for a total of 384 patients at up to 150 hospitals. Again, it's 192 patients in each arm, for a total of 384 patients at up to 150 hospitals. To give some perspective on the number, each year market reports show approximately 21,000 intra-aortic balloon pumps are used for AMI and another 18,000 are used for general hemodynamic instability.

We believe based on our European experience that Impella will assist the failing heart in pumping more blood to the heart, the brain and the kidneys during heart attack as compared to the intra-aortic balloon pump. At the same time, because of its unloading effect, it may reduce the oxygen consumption of the heart muscle during this ischemic period. As a result, this may translate into better outcomes for the patients, with the potential preservation of the heart muscle and reduction of the infarct size, as suggested by the MACH 2 trial recently conduct in Europe by Dr. Henriques from the Academic Medical Center in Amsterdam and presented at last ACC conference in March 2007.

Therefore, the primary endpoint will be the observed reduction in major adverse events in the Impella 2.5 group through 30 days as compared to the control group treated with the intra-aortic balloon pump. Such events include death, reinfarction, stroke, target vessel revascularization, acute renal failure, acute hepatic failure, acute (inaudible) ischemia, increase in aortic insufficiency or a need for major cardiovascular operation. Early unloading will have an impact on the time to recovery, inotropic dose usage and reduction of infarct size as assessed by the improvement of cardiac function. Therefore, these parameters have been proposed as part of the secondary efficacy endpoint.

As we ramp up, we will be able to more accurately predict the institutional review board, or IRB, process, training and patient recruitment. On both studies we will give further guidance in the future for completion time. The process of approving the study protocol at each hospital usually takes about 60 to 120 days, which includes the IRB approval process.

To share more insight into the process, we have as of today eight centers actively enrolling, a total of 31 centers with IRB approval, 25 centers are pending IRB approvals for this quarter and 124 hospitals have signed nondisclosure agreements with the Company to participate in the high-risk PCI trial. Again, eight centers actively enrolling, a total of 31 centers with IRB approval. In addition to that, 25 centers are pending IRB approvals this quarter and a total of 124 hospitals that have signed nondisclosure agreements with the Company to participate in the high-risk PCI trial.

Finally, I would like to give an update on the RECOVER I pilot study with the Impella 5.0. Last week, as Mike mentioned, at the Society of Thoracic Surgery meeting in Florida, Dr. Bartley Griffith, from the University of Maryland, who is also the national principal investigator of the trial, presented the preliminary results of the interim analysis, including the first 13 patients enrolled in the 5.0 trial. As a reminder, the RECOVER I study aims to assess the safety and potential effectiveness of the Impella 5.0 in cardiac surgery. The patients have to present with cardiogenic shock or low cardiac output syndrome after separation from heart-lung machine despite inotropic support.

The preliminary results are very encouraging. Of the 13 patients that were enrolled at the time of the interim analysis, 13 -- 10 of the 13 recovered and were discharged home with their own heart. Again, 10 out of the 13 recovered and were discharged at home with their own heart. One is still in the hospital recovering. And two patients expired.

Although this is preliminary analysis, 83% survival rate with heart recovery for all survivors is very positive. The expected mortality rate for this particular patient population using standard of care intra-aortic balloon pump, inotropes and sometimes surgical VADs, is 41%, based on the validated (inaudible) model.

The preliminary results show also an excellent safety profile, with low adverse event rates for the Impella 5.0 in the surgical settings. The patients were supported for up to 7.8 days, with an average flow of 4.3 liters per minute. The Impella 5.0 was easy to use, improved significantly the hemodynamics of the patients and enabled a reduction of inotropes usage. We remain confident that Impella 5.0 will play a significant role in the care of acute patients in both cardiac surgery and in the cath lab.

In summary, we are very pleased with the clinical experience and the progress we have gained in the United States with the Impella product portfolio. The U.S. results continue to confirm the European experience, with over 1,500 patients in over 40 publications on the Impella platform. We will continue to work with our key opinion leaders, our clinical partners, the FDA and other agencies to offer the best solutions for our patients.

Thank you for attention, and now I would like to turn the call over to Dan Sutherby, our Chief Financial Officer.

Dan?

Thanks, Karim.

If you would please turn to the P&L attached to our press release, I'll provide some details on our financial results for the fiscal third quarter.

First to revenues. On our product mix for the quarter, total disposables, service and other revenue -- meaning the non-console revenue -- comprised approximately 85% of revenue for the quarter. This drove strong gross margins of 76% in fiscal Q3 '08. Included in cost of goods sold during the third fiscal quarter of '08 were approximately $600,000 of expenses related to our strategic capacity ramp-up for Impella, iPulse and AbioCor, as we planned for these respective product launches and regulatory approvals. These costs were not absorbed into inventory per GAAP and negatively impacted our reported gross margin in this quarter by approximately 400 basis points.

Continuing down the income statement, our fiscal Q3 R&D, or research and development, expenses were $6.9 million, up $1.3 million compared to the third fiscal quarter of 2007, reflecting our increased investments in new products, and also include site startup costs for our Impella pivotal trial. R&D expense for the third fiscal quarter of '08 also included stock option expense of approximately $300,000.

Now to discuss selling, general and administrative, or SG&A expenses. Fiscal Q3 '08 SG&A expenses were $13.5 million, compared to $10.9 million for Q3 of fiscal '07, reflecting our investments and our global distribution in cath lab clinical experts. Total sales and clinical head count globally was just over 70, up roughly 40% from a year ago. SG&A expenses for this quarter also included stock option expense of approximately $1 million.

Included in the P&L section entitled "Other Income, Other Expenses," you can see we have a separate line item for the change in fair value of the WorldHeart note receivable and warrant. As we outlined in our Form 8-K filed in December of 2007, we entered into a strategic investment with WorldHeart whereby we issued a secured convertible note for $5 million that is convertible at our option into shares of WorldHeart. $1 million of this loan was funded in cash during December 2007, and the remaining $4 million of cash was funded in January 2008.

Under the agreement, we received a warrant for the purchase of up to 3.4 million shares of WorldHeart common stock. Under derivative accounting, we mark to market the fair value of the conversion feature on the note receivable and warrant associated with the WorldHeart transaction. During the three months ended December 31, 2007, as you can see, we recorded a gain of $600,000 from this investment. In future quarters we expect to see increases or decreases in the fair value of these underlying instruments, and therefore you should expect to see fluctuations in this particular line item of our P&L in future quarters. I would also remind you that this is a noncash item.

The net loss for Q3 of fiscal 2008 was $8.3 million, or $0.26 per share, compared to a net loss of Q3 of fiscal 2007 of $6.7 million, or $0.25 per share. The net loss for this quarter includes $1.4 million related to stock option expense and $400,000 related to intangibles amortization. Excluding the $1.8 million aggregate effect of these items, the net loss for fiscal Q3 '08 was approximately $6.5 million, or approximately $0.20 per share.

If you would now turn to the balance sheet, as of December 31, 2007, our cash and investments totaled approximately $50.5 million, compared to $59.5 million at the end of our fiscal Q2. During Q3 and included in the cash burn for the quarter we invested approximately $3 million for capacity ramp-up and inventory build for Impella. We paid $2.5 million for a litigation settlement during the quarter. And we paid $1 million for the first tranche of our investment in WorldHeart, as I discussed.

We will now open the call to your questions.

(OPERATOR INSTRUCTIONS)

Your first question comes from the line of Greg Simpson, with Stifel. Please proceed.

Good morning. Impressive quarter all the way around, and certainly for a company that's been fairly tight with information, I mean, a lot of information given out today, so I appreciate that, as well. First of all, Dan, if I could -- let's dig into the quarter first. There's a lot of ground to cover here. But can you maybe help give us a little more detail with the iPulse approval coming very late in the quarter? Can you give us some sense as to whether there were iPulse shipments or some degree of color on that front? And the same thing, I guess, in tandem with that, any balloons shipped in the quarter?

Sure, Greg. We did receive the approval, as you mentioned, late December '07. It is CE Marked in Europe, as you probably know, so we actually shipped a handful of units outside the U.S. during the quarter. For the balloons, we did start shipping balloons actually at a modest level in our fiscal Q4 '07 and continued to ship just north of 200 in fiscal Q3 of '08.

Okay, but any iPulse shipments in the U.S. in Q3?

We have some folks that are getting it, but were unable to get them trained and ship it and collect revenue for the quarter.

Okay, thanks. That answers the question. Secondly, for either Mike or Karim, on the Impella 2.5 and the required bench testing to satisfy the FDA, any elaboration there? Not specifically what they're looking for, but any elaboration as to how detailed that might be?

You know, bench testing historically is looking at some of the reliability. We have actual clinical data, but they wanted us to do a bench test as compared to one of the predicate devices, and as we stated in the 8-K, none of the predicate devices had actual patient clinical data.

Okay.

It's more of an apples to apples.

Right. Okay. Thank you. And then, on the Impella AMI study, a lot of detail there on the potential for the number of patients and the number of centers. I'm curious, Mike, if you can give us some sense of how much overlap do you expect between the high-risk PCI pivotal and the AMI trial, in terms of centers?

We do expect some overlap. You'll see that some of the centers will have both. However, we do find there are certain centers that they're primarily driven by angioplasty, and they've got incredibly high volumes, versus some of the other centers which might be more -- not so much of an academic center, but big systems, where they pull in from six or seven other hospitals, and they do more AMIs, and that's their primary focus.

We like to do them in conjunction, because, you know, it's the same process. You put Impella in and then you open up the vessel. But one patient is emergent and one is elective and scheduled, and it's a good way to train. And then if we look at the lessons learned from Europe, at Dr. Jose Henriques' center at the Academic Medical Center in Amsterdam, he did patients he did publish on high-risk PCI. His whole team is proficient and well trained.

And then they went and did a study, the MACH 2, which we presented at ACC, which also will be published here soon, showing again similar results, that there's an improvement with the increased hemodynamic support. And now they have patients that when they come in at 2:00 a.m. their facility is already trained on them, because they have done also high-risk patients. So it's good to have both together. There will be some that are different, though.

On the overlapping centers, if I can ask a naive question, how much overlap can you get on the IRB process, and does that speed it up at all?

It's a good point. It does help and it does speed it up, as well.

Okay. All right. Thanks. I'll let others have a chance and I'll get back in line.

And just to comment, there is as much interest in the AMI as there is with the high-risk PCI. Both have great interest, because they're both issues that have been clinical dilemmas for the last 10-plus years. If you look at cardiogenic shock and the results, you'll see that really not a lot of that has changed as far as the mortality rates in the last 15 years.

Hey, one last one.

But there -- just to comment, one other clarification, there is no overlap in the patients. The high-risk patients are scheduled. AMI patients will be coming in with an active heart attack or AMI.

Okay, thanks. And just one last follow-up, actually on PROTECT II, you gave us a lot of information there on the status of the trial. At any point in the process do you think you'll give us some update on the number of patients? You told us how many centers are currently enrolling, but will you give us any kind of update on the number of patients?

We may, and that's as we move forward we'll look to see what the folks are interested in and we'll give as much information to be clear. That's why today we're disclosing the number of centers, the number of IRBs, the number of centers that have signed to participate, etc., and we'll continue to update as we go.

Okay. Thanks very much.

Your next question comes from the line of David Lewis, with Morgan Stanley. Please proceed.

Good morning, guys.

Good morning, David.

A few questions. I guess, Karim, just starting off on the clinical front, can you give us a sense of the rationale for going after a composite essentially on safety versus other sort of binary endpoints on hemodynamic support, reduction in infarct size or mortality proper? Just give us a sense of was that a just talking to collaborators? And could you walk us through that?

It's a good question. You know, when you look, for instance, at the high-risk PCI, those patients are at the high risk not only because of the (inaudible) that they have but because of the comorbidity that they have. For instance, someone that has like a chronic kidney dysfunction will probably suffer from -- after the PCI -- from the kidneys, because of the contrast given, if he doesn't have a good perfusion of the kidneys.

So for that reason we were looking at all the different impacts that the pump can have on the outcome of the patients that can be meaningful from the clinical point of view. And in discussion with the FDA we came up with this composite endpoint that we believe is going to have an impact just beyond the hemodynamic support. It really has an impact on the outcome of the patient. And those are the composite endpoints that we believe the pump will have an advantage over the intra-aortic balloon pump to address the clinical need.

Also, David, the FDA will not accept the hemodynamic indicator as the primary, even though it's the strongest correlate to mortality. And that has been published. That's why it's secondary. And if it was as simple as hemodynamic support, you essentially -- that we don't really need to study. That's been well documented. This is a pump, and a balloon is an enhancement of blood flow in the descending aorta. So that's why you'll see it's a secondary, not a primary.

I guess my point is I view this as positive (inaudible) you have -- you're taking a -- it's a more powerful clinical message to the physician. You're not taking any binary risk on a single endpoint, essentially. Is that part of the strategy?

No, I think it's more on the clinical settings that matters. I mean, it's really a combination of the different endpoints. It's not only one binary effect on mortality. The patient probably may not die, but if the patient is kept in the hospital for an extended hospitalization or has other adverse events, we want to be able to capture that, because that has an impact on the health of the patient.

Okay. And should we assume that the AMI trial is, generally speaking, going to enroll slower than PROTECT II given maybe issues with informed consent, given the emerging status of the patient? And can you give us an update on how long you're thinking for PROTECT II enrollment?

We haven't given any guidance, and we are early in the process. We have expanded considerably the inclusion criteria to include three vessel disease for the high-risk PCI patients. So we believe that the enrollment rate will be much higher than what we observed in the PROTECT I. But we are early in the game, so we don't have a good assessment of what that will be. We will be -- I mean, in the future -- able to communicate this when we have more accurate assessment of the enrollment rate.

And, David, on the question of AMI versus the high-risk, I think it depends on the center. Some centers are telling us they'll enroll a lot quicker for the AMI because that's what their -- that's what they do. That's their focus for the community. And some centers are -- selectively, they do a lot of PCIs and they have a lot of high risk, and they think they'll do that faster. So we're going to keep updating it as we go for that specific reason.

Okay. Just a couple more here. I apologize. It sounds like there's a significant CapEx inventory build here, or separate issues, but maybe a combination of the two overseas. Could you give us a sense of, given that inventory build that you're working on, where we are in terms of sales and marketing to push Impella upon potential approval and where we are in capacity?

As far as the distribution, we've given it. We have over 100 folks out there in Europe in the primary countries, as well as the U.S., and those are people that are trained. We also have a bullpen of people who are cath lab experts, and our surgery nurses have all been cross-trained, as well. So we feel like we have the potential now to what we've been calling Phase I, and last quarter we gave an update of our increased investment in the manufacturing facility, and we haven't disclosed any more than that. But we are preparing for an increased amount of shipments, as well as having the people there to train them.

Okay, and just last, two-part question for Dan. Dan, R&D was higher in the quarter. Obviously it's because the clinical trials were initiated, but you knew that was going to happen. So I'm wondering did enrollment simply go faster than you were anticipating? And can you give us an Impella growth number ex-U.S. sales?

First, on the R&D, David, I did mention we have some site startup costs to get the sites rolling, whether it's the IRB costs or other costs we incur to get them rolling. On a going forward basis, we do have some modest clinical reporting patient costs. So, as I mentioned, that did contribute to the R&D increase. On the Impella numbers, at this point we haven't broken out the details of dollar revenue by product line. Keep in mind, too, the R&D number increase also included R&D relative to the Portable Driver that we announced recently, and certainly some final R&D costs on AbioCor, that of course is now approved, as well as iPulse, which is now approved. So it reflects the strategy to continue to invest in new products and broaden the portfolio.

As well as AbioCor, which is now approved.

Okay. Well, thank you very much.

Thanks, David.

Your next question comes from the line of Bruce Nudell , with UBS. Please

Good morning, guys. Thanks for the question. My first question pertains to what background rate of MACE, as you've defined it, do you expect in the control arms of the PROTECT II and the AMI trials?

We haven't really disclosed that particular number, but, I mean, we are trying to plan for the relative difference between the Impella and the IABP to be 30% less in the Impella group. Again, this is not only mortality. This is the composite endpoint. When you take the average in the control group, we believe that the Impella will be 30% less MACE, or major adverse event, not necessarily the classical MACE, the composite endpoint that we defined, so 30% less than the IABP in relative difference.

Sure. And when you designed that trial, how did you arrive at that calculation? What was the kind of experience that drove it? I mean, what was the background set of studies that gave you that idea that you could beat it by that much?

Yes, I mean, it's a very good question. Actually, we based our assessment based on two different sources. One was based on what has been reported in different registries for this targeted patient population based on the literature. And number two is really based on what we have observed in Europe on the Impella as well as what we have observed in PROTECT I trial. So that got us to the assessment of what will be the expected major adverse event rate with the Impella -- in the Impella group for the PROTECT II.

And then my final question --

Bruce, this is Mike.

Yes, go ahead, Mike.

Just to clarify that, is people think of MACE as a classic. This is really major events as defined by the list of those that we gave out that we specifically studied. We also had the benefit of when we did the pilot evaluating the mortality rates and the adverse events at the centers in the pilot on their current procedures with intra-aortic balloon pumps. So we did look at what the current standard was and we did select those events that we thought the Impella would improve on with its hemodynamic support.

And, Mike, my final question is I certainly hope that it goes the right way, but as a risk manager, when you look at it, what happens if there's a strong trend and you don't quite hit the endpoint? Do you think it has dire consequences with regards to falling in the most favorable reimbursement bucket?

The -- well, there's two questions there, Bruce. One is the reimbursement and the other is the clinical, and I think our company has a core competency in both, and we know that they're different approaches. However, if you look at the FDA, the way they're measuring this is major events. If you look at CMS, what they look at is a cost analysis. And cost analysis can be anything from less stay in the hospital to mortality.

The one thing that we know for sure because it is published and it is a pump is the Impella 2.5 pumps blood, actively unloads the heart. That's published, that we have that documented. And it is a pump. You can put it on a loop. Whereas the intra-aortic balloon pump merely enhances the blood flow, and it relies on other elements, like inotropes, which are known, they do have side effects, and those are published, as well.

So there's more to this than just a balloon versus Impella. This is a complete paradigm shift in talking about unloading the heart and what that means when you improve the hemodynamic support, which again has been documented that it increases blood flow back to the heart. It increases blood flow to the brain. It increases blood flow to the kidneys. And all those have beneficial effects that we're measuring and looking at in our major events.

No, sure, I mean, it has rationale. But you don't think just missing the endpoint could in any way jeopardize reimbursement, or you think that, rather, just a positive endpoint could ensure continued great reimbursement. How would you guide us to think about that?

Yes, I would say that the Impella is a VAD, a ventricle assist device, that does not require to have a sternotomy or a heart-lung machine. It can be put in within minutes. It pumps blood. It actively unloads. And it happens to be more cost-effective, because you don't have those other side effects and adverse events that you do when you go to heart surgery. And the CMS system looks at costs. And I don't think you can compare Impella to a balloon, just like you can't compare a balloon to aspirin. They're different treatment paths, and the way the CMS will evaluate it is based on the current standard of care as well as what the technology does.

And just, finally, if you have -- if you're successful, the PMA works well, convincing outcomes, do you think -- does that actually have implications for any follow-on Impella sort of products in terms of the regulatory path they have to take?

Bruce, I'm not sure I understand the question.

In other words, if you get a PMA approval for Impella, does that raise the regulatory burden for any potential competitors?

Oh, I understand. So if Impella has a PMA, that means it no longer would be a 510(k), and that means it no longer could use Impella as a predicate device in the future. And just to also comment, intra-aortic balloon pumps, to your question about the PMA path, only have a 510(k) clearance, and in all the studies consistently it does not show a reduction in mortality, but yet it is reimbursed and it is the standard of care and it is recommended guidelines for high-risk patients by ACC and AHA.

Thanks so much. Sorry to take your time.

Thanks, Bruce.

Your next question comes from the line of [Sean Omad], with Lehman Brothers. Please proceed.

Hey, guys. This is Sean Omad, filling in for Bob Hopkins at Lehman Brothers. How are you guys?

Good.

Good morning, Sean.

Good morning. Great quarter. Congrats. Just a couple of quick ones for you. In terms of trial timelines, I know that you haven't really been able to break out enrollment rates, but any guidance on what we should be expecting about completion of these trials, like how long, basically, are they going to take?

Sean, we know why you want to know that, and what we've said is we're giving you all of the insight into the numbers, centers, the enrollment rate, how many centers are going -- looking at IRB approval this quarter, as well as the number of folks that have signed our NDA, which means they're going through their process and continuing it, and we excluded folks that had signed and had then backed out for whatever reason. So there's 124 hospitals that signed the NDA that are also pursuing this.

Triple vessel disease does increase the ability for us to roll, and so far in PROTECT II 70% of the patients have triple vessel disease. So rather than give you kind of a more wide open range, what we thought we'd do is move further down the path and then start giving some visibility to the number of patients and where we're at with the number of centers enrolled, and then we'll be able to more accurately predict it.

Just to also point out that the 510(k) will allow us -- 510(k) clearance will allow us to have a commercial rollout, and during the trials we also are generating revenue at these centers, and they are fully reimbursed. So we understand your reason for the question, but just to reiterate, there's eight centers now enrolling. There's 31 centers that have IRB approval. There's 25 more centers that have pending IRB approval this quarter. And there's 124 hospitals that are in the process now of going through the IRB approval and lining up their protocols. And that is just for the high-risk PCI. AMI is going to happen here shortly, or we hope it happens here shortly, as we stated, and that will then add to another study at each of these centers.

No, I understand it makes sense to sort of hold off on estimates about that sort of thing until you have a little bit more enrollment data. I guess one more quick one, in terms of full-year revenue guidance, I think you guided to top line growth of about 20%, and just looking at our numbers, you have to post a relatively big number in the fourth quarter to get to that. Any modification on that guidance at this point?

So we had given an update that what we had said before in the past was 20% or greater, so that's 61 or more. And what happened in the first half is the summer months was a little slower, as we stated, but you'll see that this quarter is at $16 million, so we did pick up, I think, some of the -- some of that from Q2. It's our highest quarter ever. And so if we go into the range of $58 million to $62 million, that puts us in a window of 15% percent to 22, and historically Q4 is our highest quarter. So if we did 16.6, that would put us at 58. So that would be a $600,000 increase from this quarter. And if we do anything higher than that, that puts us back right into the range which was -- which had been prior stated.

Right. Right. Makes sense. I mean, it looks like you guys have had to do looks like close to $20 million, $19.5 million, to get all the way to a 20%, though, for the full year.

Correct.

Okay. And then one more quick one in terms of your Impella sales number for the third quarter, does it include $600,000 of deferred trial revenue from the second quarter, which is what we had been modeling?

It does, and then you should assume that each quarter there's this lag of other revenue that is either the IRB or the process or the training will continue, and that's why we gave all those numbers. We're trying to catch up every quarter with the number of centers. Some of it is administrative. We have certain centers that we have the purchase order process done. They don't have the IRB approval. We have other centers where we have the IRB approval and now we're working through the purchasing process.

Sean, this is Dan, just to add to that. That Impella number also includes sales of Impella in Europe and outside the United States, as well.

Understand, understood. Okay. Thanks a lot, guys. Great quarter.

Thanks, Sean.

Your next question comes from the line of Amy Stevens, with Susquehanna Financial. Please proceed.

Yes, good morning. Thank you so much. Actually my questions have been answered at this time. I appreciate it.

Thanks, Amy.

Your final question comes from the line of Eric Schneider, with UBS.

Good morning, guys.

Good morning, Eric.

Hope you don't mind getting two sets of questions from us here at UBS.

Not at all.

Since we have a little bit different focus, just on Impella, did you mention what the total revenue growth rate was? I know you mentioned both console and disposables, but did you mention total?

The Impella disposables grew 192%. The Impella platform grew 180%.

And I noticed on the WorldHeart investment that you'd funded a million of it, but the -- when that was originally announced, the expectation was that the additional funding would happen early in January. Is there something specific that's missing that caused that not to happen, or is that --

No, that has happened already, Eric.

That has happened? Okay.

Yes, so it was a million prior quarter and it's $4 million this quarter, and that has happened.

Okay.

$4 million in January, Eric. This is Dan. $4 million was January '08.

Okay. Just wanted to make sure that actually did happen, or if it didn't that something had -- that there was a reason that it hadn't. What were -- what are the sort of on average -- I know it's going to vary quite a bit -- but the per site startup costs for PROTECT II?

Essentially it's an IRB expense, which is a nominal expense. It's a couple thousand dollars. And then there's a fee per patient for their patient data collection, which is somewhat minimal, because the majority of the information that we are requesting back is standard of care today. But there does take a little time. So it's not a significant amount.

Okay. And I think finally, on those trials, you said 70% of the people enrolling so far in PROTECT II are enrolling -- are included as part of this triple vessel EF less than or equal to 30. Is that indicative of the relative sizes of those populations, or does that just happen to be what the number is now?

What it shows is that there -- in the PROTECT I, we didn't have triple vessel disease.

Right.

In PROTECT II, now that we have triple vessel disease, so far 70% of the patients have been triple vessel. And if you go out, Eric, and you look at all the literature where they're talking about low EF, and you'll find that approximately 70 to 75% of those patients have triple vessel disease. So that's really the most common, and that's why we think that will add to the enrollment rate, and plus we went from eight centers to 150.

Right.

So the two together are very helpful.

Okay. And then finally just a follow-up on the adjustment to guidance, the full-year guidance. So coming out of the last quarter you were still at the greater than 20% for the year. Was there -- what was it that slipped in this quarter or that you're concerned could slip in the next that would cause you to be at the low end of the range?

What the primary was the iPulse took a little longer, so it was actually over the time frame of what the goal of the -- the internal goal of the FDA as far as responding to us on that. And then the second was the AMI trial, we had significant discussion back and forth on the trial design. That took longer, based on the discussions and agreement of what the protocol and trial design would look like.

And so that's the case and those are the two issues, but as far as moving forward, the iPulse is now approved, and because we spent the time on the AMI discussions we think this will be relatively quick and we'll move forward in enrolling. We didn't necessarily change the part of the spread, because it was 61 or greater, so the range is now 58 to 62, so we're in the window, plus or minus $1 million.

Right. The range certainly encompasses what you'd suggested before. I just wondered what it would take that was slipping or that you were concerned could slip to get to the bottom, but I think you answered it.

Correct.

So, again, thank you for the responses, and good luck with the Impella.

Thank you.

We have a follow-up question from the line of Greg Simpson, with Stifel. Please proceed.

Thank you. Just a couple of quick ones, guys. First of all, on that topic, guidance, and specifically the fourth quarter, with the delay in iPulse, can you give us any kind of fuzzy indication of what the iPulse backlog might look like in that fourth quarter? I mean, there clearly should be some pent-up demand. I'm wondering how significant it might be.

Greg, the -- if you look today, you know, in the past we were going to the transplant centers, and more than 50% have the AB5000 console. And if you go to the open heart hospitals, today more than -- it's about 23% have the AB5000 console. But there are several hundred BVS consoles out there. So for the last four quarters, we really have been waiting for the iPulse, because the iPulse is the more cost-effective optimal device for these open heart hospitals. So while we have penetrated the transplant centers, there's 100 of those, we really want to get to the 900 open heart hospitals, because we need the iPulse to get there. And for those that have BVS, this is the natural trade-in for that. So now the iPulse can run the BVS if that's what they're comfortable with, the AB or the balloon, and up until this point you're really not able to market or promote anything that is not FDA approved. We are pretty by the book, and we followed that parameter. And now we're going to be going out talking to all the open heart hospitals that have BVS about aggressive trade-ins to get them up and running with a recovery program.

Okay. Are you -- can you give us any suggestion as to what the trade-in policy might be here, or do you want to get that specific?

It'll be a credit. It could be $10,000, $15,000, and there are certain components and parts and service credits that we could do. But the idea is really to get them up and running and using BVS and AB -- primarily AB, because it now has expanded applications. And based on the reimbursement, they have an incentive for the patients to recover their heart, but from a financial perspective, if they can implant and explant our cannula or our tubes for BVS or AB, they move to DRG 1, which is the highest paying DRG in the country. So they now have a clinical incentive and they also have a financial incentive that every hospital should have a recovery program. And our strategic plan is every cath lab will have Impella, and every ICU and surgery suite will have iPulse, and we think that both are required to have a center of excellence to treat high-risk PCI patients as well as heart attack patients.

Okay. Mike, then, on the AMI trial, just to confirm what should, I guess, be obvious, in that trial will the Impella also be eligible for full reimbursement?

It is.

It is. Okay. And then last one, knowing that you're a Giants fan and your CFO is a Patriots fan, he suggested to me yesterday that you have not gloated yet. Is that going to be the case, or do you want to take an opportunity to do that now?

It's a secret that I'm a Giants fan, but now it's no longer, and I'm going to be very careful starting my car tonight when I leave for home.

All right. Thanks very much.

I refuse to gloat. I'm proud of the Giants, though.

Okay. Thank you.

With no more further questions in the queue, I'd like to turn the presentation back over to Mr. Michael Minogue for closing remarks.

Thanks, everyone, for your time today. You saw we made a lot of great progress on the revenue, on our gross margins and, very significantly, on our regulatory path, and we look forward to our next update. Have a great day.

Thank you for your participation in today's conference. This concludes your presentation. You may now disconnect, and have a good day.