ABIOMED Inc (ABMD) 2003 Q1 法說會逐字稿

Good morning and welcome to the Abiomed investor conference call. All lines will be in a listen-only mode until the question and answer session. At the end of today's presentation, instructions will be given should anyone want to ask a question. At the request of Abiomed, this call is being recorded. If anybody has any objections, you may end at this time. I would now like to introduce today's host, Mr. John Thero; sir, you may begin.

Good morning, everybody. Thanks for joining us. I'm John Thero, Chief Financial Officer of Abiomed, Inc. I'm joined today by Dr. David Lederman, acting as presidential Chief Executive Officer, as well as Dr. Ed Berger, acting as Vice President of strategic planning and corporate communications. Our agenda for this call consists of four primary topics. First, David Lederman will provide comments on our ongoing Abiocor clinical trial. Second, I will make some comments on our recently reported financial results, and third we will outline some efforts that we are undertaking to communicate regularly with investors, and fourth, we will open the call for questions from analysts.

Before turning things over to David Lederman, I need to caution investors that this conference call covers forward-looking statements covering various aspects of Abiomed's plans, objectives, expectations, and intentions. In particular, we will make statements regarding the timing and results of the Abiocor clinical trial and future levels of company revenues, expenses, and cash. These statements speak only as of today. Abiomed's actual results may differ materially and are subject to numerous risks and uncertainties. Many of these risks are described in Abiomed's a filing with the Securities and Exchange Commission, and investors are urged to consider these risks and not place undue reliance on these forward-looking statements.

That being said, let me turn things over to David Lederman for comment.

Thank you and good morning to everybody. I want to update you on the Abiocor program. I want to update you on the Abiocor program, and I want to tell you two general topics; one is, what we have learned, and are we doing now to move forward in both the clinical and commercial paths? But before going into some detail, I think I want to express on behalf of the company and my own to acknowledge our admiration for all of the patients that have been enrolled in the trial as well as the families, and also to express our sadness over the fact that we were unable to provide even longer life and even more opportunities to enjoy life for many of the patients.

I also want to remind us all that today which is approximately 2.7 patient years of Abiocor experience, that we have a patient, Tom Christerson, who is the second patient implanted at Jewish hospital, and that he's right now one week away from reaching his first anniversary on Abiocor, which is a very, very important milestone. We are delighted that Abiocor is able to extend his life and allowed both him and his family to enjoy each other. Beyond this recognition of this incredible achievement and consistent with our policy on patient privacy, I will not address any patient's specific issue nor respond to a specific patient or their medical course. Instead, I will try to update you on what we have learned collectively from all of the patients during this trial.

I also want to say that we are all, at Abiomed, very disappointed but not discouraged that we did not implant any patients this summer. We can do better, and I will try to point out to you some areas of current focus that hopefully will result in patient enrollment soon as we move forward.

But I want to go back to two topics that I mentioned at the beginning. First, what have we learned? We have learned tremendously. And again, 2.7 patient years later, we have learned that the most sophisticated implantable device works extremely well and provides full support for patients even under debilitated multiple organ conditions that would pose an enormous challenge to our healthy, natural heart. That case of research and development sponsored by the national institute of health that pursued this trial with the right technology, and I would embarrassed if I didn't acknowledge the government support of this very, very important effort.

Tom's approach of his first anniversary on Abiocor and his ability to live at home shows that the Abiocor can achieve the desired clinical effects. We learned that when we started the trial, we had hoped to show that we could at least double the life expectancy of a number of our patients, and after enrolling several patients, four of them went beyond four and a half months, and I think that is remarkable. In so doing, we have learned so much. We know so much more today than we did when we started out. So important for our ongoing efforts to move the product into the clinical and commercial path.

We also learned that the incorporation of a four-legged structure we have called a cage used in the animal model to prevent tissue that remains from the chambers of the removed heart from blocking blood flow into the actual Abiocor, the site where a thrombus or clot could form, in addition to that structure, the patient's condition did not allow appropriate levels of medication to prevent clotting, so the combination of those two, we think, pose a potential risk which may explain the reason why three of the patients actually suffered clot -- strokes. Now to eliminate this potential effect, this cage structure has been eliminated, and whenever possible, anti-coagulation assessment has been heightened as we screen patients moving forward.

We also learned that the economic model that we thought we would be able to progress during the trial is in fact starting to show with very limited data the right results. Namely that the cost level to support a patient once the patient is discharged home is very low, and therefore the feasibility for Abiocor for patients that are discharged home is quite high from the economic point of view and this is critical in order to demonstrate that this technology works. The fact that the patient does not require immunosuppression when at home and does not require tremendous amount of medical care, therefore, bringing the cost to a reasonable level is a welcome confirmation of what we thought originally, and of course we need a lot of the data, and part of our effort is to collect the data as we continue to work on the reimbursements aspect for the technology when it becomes available.

I should also say since I'm talking about economics that because of the inherent conditions of initial patients, their hospital stay has been much longer than we would expect once we learn all that we need to learn from the clinical trial and begin to enroll patients that are not in such dire conditions in terms of multiple organ failure. Again, if a patient can be discharged without complications, we have shown feasibility that its support is more than affordable to society.

We, in that sense, have had an accelerated learning curve because of the longer support that many of the patients have received. Because of the multiple exclusions of many of the patients, four out of five patients had multiple out of hospital excursions to go to a restaurant and visit relatives and so forth, and we learn much from it, especially the discharge to the home. We spent a lot of time learning and assessing what needs to be done so the quality of this patient's life at home and eventually, hopefully when they go back to work is maximumized.

And in so -- in giving you some of the details, the message that I want to give to you is the time during the last three or four months when we have not had clinical trials, time has not been unproductive. A tremendous amount of work and the process has been moved dramatically to improve both the trial and the prospects for the patients to be enrolled. For example, the number of alarms designed into the system which our scientists and engineers felt were very, very important during implantation as well as the postoperative in-hospital course, because of all that they learned and corroborated in terms of the performance of the system, but at the same time when the patient had some out of hospital excursions or when the patient goes home, the alarms can undermine quality of life. There are too many alarms that are useful and available valuable to scientists but not to patients, and as a result, we have reduced the number of alarm conditions only to those where the either quality of life or the rife life of the patient can be impacted, and that has a tremendous impact on the quality of life of this patient.

Just to give you an idea, we are talking about alarm conditions that could have been 50 or 60 in one day. Bringing them down to a level where there are a few and that's our goal. And again, they are informational and ought to be, eventually go down to an absolute minimum. Especially when a patient is asleep, and especially when the alarms are not in any way threatening the safety of the patient. By the way, the majority of these alarms have a lot to do with the movement of bringing power into the heart and relative alignment and so forth, and a lot of actions have been taken to reduce them by making it more flexible and easier to handle them. These are all external to the system.

I had mentioned in the last conference when we spoke that the majority that took place in the last 20 to 30 years is implanted into the patient, and those were the most sophisticated and those were the areas that took the longest to develop, and those really we had to do very little to. The external systems that are practical issues, and therefore does not impact the patient because the patient can evaluate them, you remove a plug something in or to remove and replace a battery are simpler from the technology point of view, but those we taken a great deal, and we think more and more the next group of patients are going to go home, and we have raised the bar in terms of what we can provide to the patients. These are all very important things we have learned. Again a lot of this work we are doing right now and have been doing over the last three or four months is work that normally would be done after the FDA summation, not before. So we haven't lost time to commercial entry. We have learned a lot and continue to do that. I want to stress that point. We believe that the path that can deliver the desired quality of life is not lengthening as a result of the inactivity in the past three or four months. But we need to do better, and what are we doing to move forward at a faster rate?

Well, we continue to try to find the right patient, this time the right patient who can go home, yet who have, like the previous patients, a very short lifespan if left without Abiocor support. This is a much higher bar with the home discharge objective, even the patient referral from cardiologists has become more challenging and requires us to rethink the process and the information that we give to cardiologists so they refer to the hospital the right patients. I'll give you one example that is sort of a silly but important example and that is that we would prefer but do not insist on having the next patient live much closer to the hospital than some of the patients we have implanted in the past. This is important because the life, the family support and convenience and quality of life of the patient is in the hands, if they live near the hospital so if there is a medical checkup, they can go -- don't have to travel many, many hours and in some patients it would have meant days if traveling by car. So we are also concentrating our immediate efforts for the remaining eight patients of the trial on those centers in the USA and Europe who have the most experience with artificial hearts. Since our bar of expectation is much higher now than before we started the trial because we have learned so much more, this has strong implications as to where the next implants are going to be done.

I think that I am going to stop here and let John take charge again to give you an overview of our status right now with regards to the BVS and other activities. But I want to repeat that we have accomplished a lot for this trial phase even though we are disappointed that we didn't enroll new patients during the summer. We can and will do better, and we hope with we will start implanting new patients as a result of renewed efforts.

With that I would like John to take charge again.

Thank you, David. On the financial side, Abiomed's form 10-K and 10-Q were filed with the SEC last night, and in those there were discussion and analysis of our financial results as audited both for fiscal year ended March 31, '02, and unaudited for our quarter ended June 30th, '02. I'm not going to try to repeat all the discussion and analysis in those reports here, but I will make a few comments I hope are helpful.

Abiomed's product revenues in all the periods were derived entirely from our BVS product line. Our BVS product revenues increased 25% in our fiscal year ended March 31, '02, and our BVS product revenues then declined 15% in the first quarter of the current year. Our BVS revenues, as you know, have been historically volatile from period to period, and as a policy we do not provide forecasts for future performance, but it's fair to say that this period to period fluctuations are likely to continue; it's also fair to say that we are not predicting a 25% increase on a going-forward basis.

Overall our shipments and revenues from BVS blood pumps have not been -- excuse me, have been increasing. In Q1 last year we reported strong sales of BVS consoles. These strong console sales in Q1 were not repeated in Q1 this year. As a result while our revenues from both pumps increased in Q1 this year, they did not increase in a rate sufficient to offset the decline in console revenues; however, they did contribute to increasing sales margins, and our BVS business remains profitable.

Our expense levels including non-cash costs associated with acquisition of the heart in September of 2001 increased in all categories during the fiscal year ended March 31, '02, compared to our fiscal year ended March 31 '01. However these expenses decreased in all categories in the first quarter of our current fiscal year compared to the same quarter last year.

In the R&D expense category, most of this movement had to do with the timing of expenses related to materials purchased and clinical trial support. In the SG&A category, the expense paralleled the increases and decreases in revenues in the same periods. SG&A costs were also affected by the timing of our public relations activities and legal costs. Note that our losses from operations declined slightly both in our fiscal year ended March 31, '02, and our quarter ended June 30th, '02, compared to same period of last year.

Our cash balance for June 30, '02, was approximately $54 million. We continue to spend money on those things we need to have done. We have no long-term debt.

The third topic we promised to discuss in this conference call, the topic on investor relations. We understand that we've been relatively quiet in recent months as we waited for our new auditors to complete their work. We've heard that certain investors don't feel that we've had enough time to answer their questions or we haven't responded to enough questions in past calls. With the goals improving the frequency and responsiveness of communication of information to your questions, we're going to provide a monthly forum for management to respond to investor questions. We'll be asking you to e-mail your questions to us at IR@Abiomed.com so that we can provide thoughtful and clear answers to more of your questions. We'll be sending out notices of these forums in advance regarding timing and how to access them.

At this time I'd like to open this call to your questions. We ask that you limit your questions so that we can respond to as many as possible. Operator, if you could give instruction on how to do that.

Thank you. At this time we are ready to begin the Q&A session. If you'd like to ask a question, press star 1. To cancel your question, you may press star 2. Once again, that's star 1 to ask a question and star 2 to cancel. Our first question comes from Kurt Kruger of Bank of America.

This is Raj for Kurt; he had to step out. I wonder if you could give us a little bit more color on exactly what the patient recruitment is looking like. It's been a number of months now since you guys mentioned it and I'm curious where you are continuing to look at patients, are they coming up and you're deciding not to do them? It seems a lot of time has passed. We look at -- we always talk about the hundreds of thousands of patients that could potentially benefit from this, it strikes us that you haven't been able to find any in the last months.

Let me try to answer this question. First of all, one cannot draw relationship between the hundred thousand other patients we are enrolling because the patients we are enrolling are a subset that are extremely sick and has to meet some very, very strict inclusion criteria including probably of less than 30 days, and also when one talks about the potential market one is talking about the more acute patients, and as you know, all seven patients that is we have enrolled so far have been really cardiomiopathy, largely ischemic and not AMI patients. So I want to make that distinction clear. Number two, we have been screening patients. I don't know off hand what the exact rate is, but we have had many of the centers send us patients that we have been screening, and not only in the U.S. but also internationally, and there have been very recent why they have not met all of our inclusion and exclusion criteria so that we have not enrolled them. If one was to analyze, we had to devise that we had already far more clinical experience with some of the patients we have screened have been eligible for the device, and the answer is probably yes. But not at this time; the duration has to do with, as I said to you before, sometimes patient that is have come to us that do not meet our Abiocor criteria very, very strictly, and therefore we are continuing to follow them. You understand, we don't think the probability of their death is as high as they were deemed to be from clinical observation, and when we did Abioscore, we found it did not have a high probability of death. We are continuing to monitor them. They have no other alternative. They are typically class 4.

So you do have some 0atients in the queue.

The answer is yes. There are some patients that we are continuing to monitor. New patients are coming in, old patients are being dismissed as not eligible for a consultation for a variety of reasons that are part of our inclusion and exclusion criteria. That is about as much detail as I can give you.

Another point of clarity, you said have you not changed your regulatory clinical timeline; does that mean you still plan to have the additional patients done by the end of this year? Or when do you think you'll have them next?

Look, we were hoping when we started the summer we would be enrolling patients, and right now we're -- it's September, and we certainly hope it's doable to do the remaining eight patients before the end of this year. But we are going to do the right patient at the right time because in the end, having patients that do well and get discharged home and teach us what needs to be done to help us is more important than getting the trial done. If we had done all 15 patients earlier on, say in January and February, and we have not learned a lot of the things that we have learned from some of the patients who, unfortunately, are no longer alive. We would not have done some things like the removal of the structure an other things that would have caused its to do a new trial or request for permission to do a new trial. The important thing is we have learned an enormous amount. Even though we are disappointed because we are ready and have been ready now for several months to get going, we continue to learn and continue to do things that will improve the at-home life of this patient. So the answer is, yes, it could very well be done and could happen, and I couldn't tell you because I really don't know. If I could, I would tell you. We could have three or four patients in one month or one week, you know, but we don't know.

All the centers are ready and trained and waiting?

Ready and trained as I mentioned before. We are now focusing on the next patients at the centers which are the most experienced centers. The number of centers we expect implantations to take place in over the next four months is somewhat smaller than the total number centers that are ready to do.

Thank you.

Thank you. Our next question is Greg Simpson from A.G. Edwards.

David, I'm going to follow up on that, actually. First of all, on the informed consent process, we've had conversations with some of your clinical investigators and the impression we get from them is if the informed consent was modified which is not necessary, but if it was modified there would be no shortage of patients applying. I want to see if I can get you to comment on the possibility of modifying that, at least modestly, to free up more patients and a couple of others.

Yeah, if I understand correctly, your comment on you conversation with the clinical centers, the recommendation has been that if the patients were allowed to give proxy consent instead of -- which means their relatives could sign and give consent, we would have many, many more patients to choose from at this point. We, the company still feels that it's too early for that. We think that we would -- we do want the patient to know and to consent to the implantation, and that, in essence, removes a very large number of patients because a majority of AMI patients are not in a position to have the time or be in a situation without much sedation where they can really understand what is being said. We don't think it is ethically correct to do that yet until we have more patient experience. We agree with them that we would have many more patients, but we want to still hold to doing it only for patients who can sign themselves. Now, there are some AMI patients who conceivably could fall into that category. For example, there are patients who have other conditions that render them ineligible for transplantation or any other therapy because they have other conditions, sometimes malignancies that are not necessarily an exclusion for our patients, and they happen to be young and happen to suffer an AMI or not suffer an AMI, but they young so they are in good health as a patient suffering an AMI. For example, an a malignancy of the heart itself. Since that, we are being very careful to make sure that we do not enroll patients that will undermine the clinical trial. We are being far, far more careful in looking at the history of the patients in every respect than we were before. Again, because our bar is now to discharge patients to go home and to show that, in fact, they can lead normal, reasonably normal lives at fair costs.

Is it fair to say, David, that if you, you do the next three or four implants -- I won't try to hold you to a number -- but then modifying the form of consent could become a possibility as you get a few more implants under your belt?

Yes.

Okay. I'll leave that at that. A couple of others that are all kind of related. Your selection criteria as you try to get more selective and get patients that can get out of the hospital more quickly and things like that, obviously the goal to get better outcomes, to get patients out, but the initial data to date, you mentioned the 2.7 patient years and works out to 900-something days, and again, talking to investigators, that far exceeds anything you probably expected at the beginning of the trial, and our conversations, they put out a rapport on this recently suggest the number of patient days plays into the ability to file for FDA approval, not just the absolute number of implants. Can you talk about that, and if, say, you get the next five implants done, you get the right patients and you have extremely strong results with these patients, does that give you the ability to move quicker and stay on track or maybe shortcut the FDA process a little bit? And I don't you don't want to run afoul of the FDA, so feel free to dance around a bit.

Look, -- I cannot much more sum up [inaudible] except to say that we think we have a number of patients, a good proportion of the eight remaining patients who are able to lead lives like Tom or to go home and lead lives that we think will have a compelling argument to convince ourselves, and if we convince ourselves, to convince the FDA that it is ethically correct to move forward with a process that would allow us to make this technology available to more patients.

Okay. Then this is actually falling together nicely. Going to seem like I planned this. The final question would be, again, you mentioned initially fine tuning the patient selection process, and one of the things to find patients to properly anti-coagulate. I know you don't want to make patient-specific comments, but I'm going to walk a fine line here. My understanding is of the patients done to date, excluding the two that didn't survive the surgery, that Christerson is the only one of the bunch that has been able to withstand a, quote, normal anti-coagulation process. And I know that all comes into what you look for going forward with respect to the cage and things like that. Can you talk about that without violating your pledge not to talk about specific patients? Is that correct, and does that give you increased faith, I guess, that if you identify the right patients, the possibility of having a Christerson-like result is greater.

Let me, I think that first of all, I think that the statement is not correct. Not exactly correct. There are two factors that today potentially cannot -- well more than two but two factors that are device-related or related to the implantation of the Abiocor. One is the cage and the other is the unability to anti-coagulate the patient. So let me tell you what I can tell you. I can tell you that of the five patients that have been supported, there were two patients that tolerated the anti-coagulation medication. All of those hearts have that cage. Okay?

Okay.

Does that answer your question?

Yeah, it sure does.

The other thing I wanted to say is that we, I mean, this is public knowledge so it doesn't violate the patient's privacy issues because it has been already described. In the case of our first patient which is one of the patients who could not tolerate anti-coagulation for a long period of time and he went five months. What we have seen in general, let me say in general for all of these patients that, unfortunately, are no longer alive, but we have seen that the work beyond our expectations in terms of how clean they came out even though for some of these patients, and I don't want to say which ones, the level of anti-coagulation that they tolerated was lower than a patient who has only heart problems. We are remarkably impressed at how well and that gives us tremendous confidence that the device works quite well if one could medicate a patient at the same level as a patient with other heart problem. We are very, very encouraged by that. Again, I want to restate what I said to you before, namely two patients with different methods of assessing the coagulatability of the patient, what the state of coagulability of the patient's blood is. Some are more sophisticated than other methods, and we requiring all our centers to use the sophisticated technology, and when we looked at how much time all the patients spent or tolerated the regime of anticoagulation, we found that the two who tolerated it best were the two patients who have shown no evidence of any stroke or neurological event. So that again is also very encouraging to us. But we have only five supported patients, and we need more patients to be able to prove that unequivocally.

Thanks very much.

Thank you.

Our next question comes from Alex Arrow from Ladenburg Thalmann Company.

Thanks, just one clarifying question about the statistical power of the study as you plan to move forward with the submission. Previously you gave us guidance that 15 to 30 patients would give you enough statistical power to satisfy the FDA, and why you were able to come up with a trial plan that had fewer ends than some other trials out there. I'm wondering if over the course of the last eight months you've had additional conversations with the FDA, can you give us more guidance on if that is still the case? Particularly in light of the fact the FDA has recently advocated what seems to me a new method of statical processing in which they are encouraging companies to increase the size of the end by drawing in data from other studies, something I had not seen them do a lot previously. Given that perhaps the mindset of the FDA is moving toward requiring larger ends for statical power, and how does that coincide with your plan for the 15-30 patients?

First of all let me say we have been in communication with the FDA. We are always in communication with the FDA, but in terms of the actual details of our study, I can't disclose what our current strategies. We are of the opinion of the company and doesn't mean that the FDA -- I don't want to attribute any statement to the FDA, but we are of the opinion of the trial with a smaller number of patients than people think. I think that the 15-30 was based on an earlier analysis that we had done and still think it can be done with a number in that order. I think your statement about statistics which we looked into five years ago, three years ago, two years ago and continue to look at is under the recommendations that have been made, in particular at the summit which took place a couple of weeks ago was really focused on LAVDs. And all of the discussions including comparison with the heart mate 1 as well as patients required to a trial to demonstrate one year survivor benefits, overdose trials and medical therapy, we are all focused on LAVDs, and my understanding is that that point was not made strong enough, that all the discussions really dealt with those devices.

Why would the statistical treatment of an LAVD study differ from the statistical treatment from an Abiocor study?

Because they have to demonstrate effectiveness versus an established device. A small difference between devices requires a large number of patients. When they ask for a large difference, it requires a smaller number of patients, and there is an enormous difference between an artificial heart and an LAVD, and a small different between LAVDs. In fact the numbers, if I can recall -- I wasn't at the meeting -- but it was reported to me that to demonstrate a 5% improvement at one year survival with an LAVD, you require an additional 1,931 days. If the difference in the outcome starts to increase, the number of patients goes dramatically down, and I don't think that anybody will argue there is a major difference between the Abiocor and any LAVD.

Okay. Thank you.

Next question.

Our next question comes from Randy Hoff of A.G. Edwards.

Thank you and congratulations, David, and your whole team, on the success so far. David, I'm going to take a leap here beyond what appears to be the one remaining major issue, and that is thrombus, and assume that it is going to be behind us based on your optimistic comments from earlier today, and go to longevity issues. We have now as you pointed out, Mr. Christerson, who has been out for nearly a year, and you also have running parallel with the clinical experience in human beings bench testing of the device at the system level. Could you talk for a moment about what you've seen in terms of indications from both the bench and the longer lived patients, Mr. Christerson in particular, with respect to the continued optimal functionality of the Abiocor under longer durations of activity?

Okay. First of all I want to remind you and those of you who are participating in this conference that when we started the trial, part of the design and part of the criteria for defining a two-month milestone and then our own expectation at the time that we would be able to go six months, but measuring very clearly morbidity, that we would not enroll patients with a responsibility beyond one month. That remains ethically correct because you cannot implant a device on a patient if you think that the patient could live without the device. You have to do an analysis, and at that time, since we made our informed consent available publicly, at that time we have reliable data that shows that those devices were at that time running less than a year. During the course of the clinical trial we went beyond the one-year duration, and at this point in time, we have devices running longer. If you do the math, obviously it has to be -- Tom has been implanted almost for a year, by now we have to be into the two-year time frame. The equivalence of bench testing to a human testing is not straight forward. In some respects, the bench is more demanding. In other respects, the real human situation is far more demanding. The bench test does not show clots and thrombi, which is why we have to go into clinical trials, but it does show when something is excessively wearing. Now the physiological conditions are assimilated as best as we can, but it still is sometimes more demanding, and we have learned much also during this two years and as part of our program, we are testing new systems and improving the systems because as you know, our ultimate goal is to have a five-year device but we're not there. Our concern about patients if they start approaching one year and longer is real. We look at all the possibilities, new devices will last longer but old devices -- not only but devices implanted in humans, when they start entering the period where we have experienced any failure in the lab, we start looking for strategies as to if we should replace the device in the patient and something that is going on at all times and we consider those things, the relative risks of replacing the device and risk of allowing it to stay. The other thing is that there are some signs when we have learned of this in the last two years, we have learned a great deal from the reliability testing that there are signs of wear we can detect before. Okay? And so that there is certain information that would tell us to prepare if we think it could be something imminent happening. This is not for always but so far in this trial, out of the -- I have another calculation, it's probably we are approaching 200 million bits accumulative among all patients and my understanding is that we have missed three. Three beats out of 300 million so we have no system failure in our clinical trial and remarkable how well that works. But your question is real. We do not have a five-year device. And we do not have one. We didn't go into the trial, we never made a claim when we could not make a claim that we have a two-year device even if it goes three years. Remember, there's a difference between durability and reliability. It's not only how long it will go, but how many of those will go that long and at this point in time, we have more systems that are continuing to be put on test longer and longer and part of the clinical and reliability strategies. There will be and I have mentioned this in the past, will be new clinical trials even as we hope to get approval for the first generation, it will be with that second generation clinical trials and third generation that goes longer and those patients will be less and less sick because we have higher confidence in the clinical as well as reliability of devices, and therefore enrolled faster. This is the nature of the strategy and I think that your concern and your question is very, very valid. We think about that and we said which we can monitor to detect if there is failure and make an evaluation as to the risk to let the device stay in any patient versus scheduling a surgery to replace the device. By the way, we had in the case of one patient, we had to replace a battery as one example of a component that had lost its capacity and was degrading, and we felt at some point or the physician, the team as well as us felt it was the risk of abdominal replacement was lower so we have learned a lot and the batteries we put in and the way we use the batteries have improved dramatically so that won't happen before a year and hopefully go longer and longer and longer. We are learning.

Indeed, you are. If I could just follow up for the point of clarification and one small minor point here, David, you say we don't have a five-year device yet. I've been tagging along for quite some time here, and we have heard at the bench level testing of valves have the equivalent of 50 years and other statistics like that with respect to component parts of the device, but that it has not been tested but a period of two years or so on the bench at the system level. Now when you say that we don't have a five-year device yet, it is it true that you don't have a five year device or that you don't know you have a five-year device simply because we haven't gone on the second and third years with the first generation?

A lot of questions there. First, let me correct one thing that you said. We have not demonstrated 50 years on the valve but I believe we have demonstrated 25 years. Our valves and the components with contact blood, namely the ventricles and the valves and all of those components on a component testing basis have shown severy, very long life which you have to demonstrate. But you cannot accelerate test the whole system. You just cannot go -- With a valve you can go 10-20 times faster than normal and get in one year 20-year equivalence. You cannot do that with the same system. To your question, we know from -- we don't know that the systems that go longer than two years cannot go to five years. Maybe that some of them do. But we know if we had 100 of those, okay, our requirement that a large fraction of those to go five years would not make it. You understand, and theorical considerations that you learn after testing and then you analyze wear properties so you can extrapolate them and project that some components that are showing some signs of wear, if they continue at that rate may not make it five years, and we know that, too. What we have done is sometime a year ago, sometimes longer ago, sometimes six months ago, and sometimes two months ago, we have developed components that can go longer than those that have shown wear, and they are now back on tests. So I don't think that when we achieve the five-year reliability goal that it will be something we have all components changed. Some will be replaced by others by others that show better properties. Okay? Did I answer your question?

Yes, you did. Perfectly. I wondered how you could take what you have today and say kind of emphatically that it's not going to work five years because you haven't gotten there yet and the components tested for more than that, it seemed that absence the wear you are suggesting now in the second part of your answer is that you couldn't say that it won't go five years. Remember, I'm in sales so I take a much more optimistic view of things.

Let me give you an example so you understand. You can take for example, a plastic membrane. You worry when you're designing something, can it flex that many times? Can it go a billion cycles or 500 million cycles without breaking and run those and accelerate those and say, gee, they all work. Then you put it in a system and the system is somewhat different than in addition to flexing, it also touches occasionally some component that is a titanium component. From the flexing point of view, it never failed, but from the abrasion between that material and titanium, it shows that after you measured the thickness after going the equivalent of three years it's thinner and shows wear marks. Now you project how much is it thinner and if it continues at that rate, how long would it last. The answer is maybe it won't last five years. What can you do? Two things, replace the material for one with better wear properties components or modify it so it never touches the titanium, and you can do both to solve the problem. There are all kinds of things that the only way you can learn is by doing it, testing it and going back and analyzing what happens and doing it again. The question is have we seen anything that fundamentally tells us we could not achieve five years and the answer is, no.

Great. And secondly one of my associates just came in with this question and I'll relay it; of the folks that you screened, some speculation in the office for the last several months is that people don't want the Abiocor. Could you address that, David, in terms of the level of interest in the Abiocor of those who have been screened or in the queue or whatever of rate and their state of mind as far as accepting an Abiocor should they be chosen and qualified to accept one.

The answer to the question is I don't know the numbers but I can tell you if ask me has there been a patient who after admitting the criteria said he or she is not interested, the answer is, yes. I don't think that it's a large number, and I don't know what the ratio is, but we have encountered a patient that I know of and why I made that answer that after being considered and not definitively, we have not gone through the process, the patient said that he didn't want to. I don't think this is a pattern more an exception.

Is that patient still alive?

I don't know. I don't know whether that patient is alive but I want you to know we are monitoring the patients we screened to assess how long do they live if not implanted with the Abiocor, and remember, yeah, so I don't know but I'm going to look it up.

Thank you very much, David.

Thank you.

Thank you. We have time for one more question, and our final question comes from Harold Webber of Salomon Smith Barney.

I got here late, I don't know if you talked about this issue. Can you talk about the scenario at this point in the standards in regards to females?

Yes, as you know, the current generation Abiocor is -- the size of the device is such that only approximately one out of five females, it would be appropriate for them. We would like very much, very, very much to enroll more than one female in this trial, but we haven't yet found a person that fits all of the requirements. Our other device under development will address and will likely befit a large fraction of all the female population but at this point in time, the odds are if I were to give you any Todds, the odds are that the majority of the patients remain for this device will be male but we hope we'll get one or two females.

Okay. That's based on the current unit and its current size.

Yes. One out of five women it would fit in. I have to tell you that it hasn't been -- the majority of referrals to us have been male so it hasn't been that we have had a large number of female that have been ruled as not eligible because of fit, they have not referred to us as many females.

Because they feel it's a smaller group to pick from.

You remind me of something we are not doing and going to do and thanks for the reminder; we need to emphasize to the referral cardiologist that they should definitely make an effort to not exclude females, and I don't believe that they are.

Subconsciously they are.

They may be because we keep saying that the device fits only one in five. Right. I think that's a very good point that we will pay attention to that and incorporate it to our message to the cardiologists.

And any kind of a, I don't know, outlook on when the smaller device might be ready for use?

Well, our plan as we have stated before and we're on schedule is to be ready for trials in the calendar year 20.

Okay. Thanks. Hopefully we'll hear something good soon.

I hope so. We're doing everything that we can. As I said before, we are equally frustrated and hope we are doing new things we are learning. We ought to know better and doing everything we can to do better.

Keep up the good work. Thank you.

Thank you, everyone, for joining us. We look forward to communicating with you as we go forward and appreciate your interest and support.

Operator: Thank you for participating in today's conference call and have a nice day.