Aspira Women's Health Inc (AWH) 2004 Q4 法說會逐字稿

Good morning. My name is Bonnie, and I will be your conference facilitator. At this time, I would like to welcome everyone to the Ciphergen announcement to review year-end 2004 financial results and events. All lines have been placed on mute to prevent any background noise. After the speakers' remarks there will be a question-and-answer period. If you would like to ask a question during this time, simply press star, then the number 1 on your telephone keypad. If you would like to withdraw your question, press star, then the number 2 on your telephone keypad. Thank you. Ms. Carruthers, you may begin your conference.

Thank you. Good morning, ladies and gentlemen. With me today are William Rich, President and CEO; Matt Hogan, CFO; Martin Verhoef, President of the Biosystems Division; and Gail Page, President of the Diagnostic Division. Copies of the earnings press release were distributed last night and are available on our website. I would like to remind everyone that this call is for information purposes only. This call is being recorded and is copyrighted. And, therefore, please note, it cannot be recorded, transcribed, or rebroadcast without Ciphergen's permission. Your participation implies consent to our recording this call. If you do not agree with these terms, please drop off the line.

Our discussion today contains some forward-looking statements, including Ciphergen's expectations of future strategic plans and operational results. Various risks may cause Ciphergen's actual results to differ materially from these expectations. For a list and description of some of these risks and uncertainties, please see the report filed by Ciphergen with the Securities and Exchange Commission. The information in this conference call related to projections or other forward-looking statements may be relied upon subject to the previous Safe Harbor statement as of the date of this call, and may continue to be used while the call is maintained on our website. I'd like to [inaudible - background noise].

Thanks, Sue. I'm going to begin with a discussion of our financials, and I'll then turn the call over to Bill for a review of other corporate highlights. On November 30th we completed the sale of our BioSepra business to Pall Corporation for 32 million net of cash DAC. As a result, we now report our financial results reflecting BioSepra as a discontinued operation. Ciphergen reported fourth quarter 2004 revenue of 10.1 million, a decline of 13 percent as compared to 11.6 million in the fourth quarter of 2003, but up 19 percent from 8.5 million in the third quarter of 2004. We reported revenue from the sale of 37 systems, of which 8 represented trade-ins or upgrades of existing systems. 3 of our orders were for our older PBS IIc line, 2 were for ProteinChip interfaces, and 32 orders were for Series 4000, of which 18 were Enterprise Editions and 14 were Personal Editions. Instrument sales, which include upgrades and accessories, generated about 52 percent of total revenue in the fourth quarter, with 5.3 million in sales, compared to 6.8 million in the fourth quarter of 2003, and up 35 percent as compared to 3.9 million in the third quarter of 2004.

Total ProteinChip array sales were 2.8 million in the fourth quarter, or 27 percent of total revenue. Array revenues grew 18 percent on a year-over-year basis, and 19 percent sequentially. The remainder of our revenue was related to service activities. For the quarter, total service revenue was 2.1 million, or 21 percent of total revenue. Collaborative service projects accounted for about 403,000 of the total in the fourth quarter, with the remainder coming from providing maintenance services to our customers and paid training activities. Our gross profit for the fourth quarter was 6.2 million, which represents a 62 percent gross margin. During the quarter there were 2 major events impacting gross margin. First, much of our materials cost in the Series 4000 had previously been expensed to R&D when the system was in development, which helped gross margin. Offsetting that we substantially boosted our excess or obsolescence reserve for PBS IIcs, as we now see a more rapid transition to the new system then was previously forecast.

Sales and marketing expense, excluding deferred comp was 6.3 million in the fourth quarter of 2004 versus 5.8 million in the fourth quarter of 2003. However, our sales and marketing expense was actually about 400,000 less, or 6 percent, in the fourth quarter than compared to the previous quarter, due to the full effect of cost reduction actions taken in July. Research and development expense, excluding deferred comp, was 3.7 million in the fourth quarter as compared to 5.7 million in the fourth quarter of 2003. Our R&D expense was about 600,000, or 14 percent lower in the fourth quarter than in our third quarter, with virtually all the reduction related to our research tools business as opposed to our diagnostics efforts. General and administration, excluding deferred comp, was 3.3 million in the fourth quarter of 2004, versus 2.9 million in the fourth quarter of 2003. Most of that increase was due to the cost of complying with Sarbanes-Oxley. Our G&A expense was about 2 percent lower in the fourth quarter than in the third quarter, despite an increase in costs associated with complying with Sarbanes-Oxley. Due to cost reduction and program cuts taken in July, total operating expenses in the fourth quarter represent a decline by an annualized figure of 12.1 million as compared to the second quarter of 2004. This reduction exceeded our previously stated goals of reducing operating expenses by an $8 million annualized run rate by the fourth quarter. Our operating loss from continuing operations was 7.2 million in the fourth quarter, and our net loss from continuing operations was 7.6 million.

The BioSepra transaction was a significant event for us and bears some discussion. BioSepra was a very fine acquisition for us in 2001. We generated significant R&D synergy by combining BioSepra's chromatography expertise with our chip chemistries. We pioneered the concept of SELDI-assisted process proteomics, and we grew BioSepra's revenues from the time of acquisition. However, the large-scale sorbents business requires a specialized sales force and entails a very long sales cycle to generate orders of significant size. Our judgment was that we're better served to focus our resources, both financial and managerial, on our research tools business and our diagnostics business. We've been talking to Pall Corporation for long time about their strategic interest in acquiring our chromatography product line, and BioSepra's a unique property in that regard. It's a wonderful fit for their membranes business and their large, worldwide sales force can likely grow the sorbent sales at a faster rate than we could on our own.

However, we still have substantial links to the business. Egisto Boschetti and a small research team remain in Cergy remain with Ciphergen, while providing transition consulting to Pall. We achieved a market and technology carve-out, such that we get the right to make, modify, or buy up to an undisclosed quantity of each sorbent per year for use and sale outside the process scale market. And we entered into a license sales and marketing alliance, whereby we'll jointly market process proteomics to the industrial-scale market. Pall sees this as an important means of differentiating themselves from the competition and fueling sorbent growth -- growth, and has already established process proteomics service centers in the U.S. and Europe. Meanwhile, we expect to continue to generate ProteinChip system and array sales in this regard. Financially, the sale of BioSepra resulted in a gain of $18.5 million. This gain excludes 1 million of the sales price, which is being held in escrow until November 2005. The discontinued operation had a net loss of about 600,000 for the 2 months during the quarter that we owned it. As a result of these 2 figures, we recorded a net income of 10.3 million during the quarter. We ended the quarter and year with 37.6 million in cash and investments.

Looking ahead for a moment, the forward-looking statements I'm about to make are subject to risks and uncertainties as described at the beginning of this call, and we don't intend to update these comments or forecasts prior to our next earnings conference call. We expect 2005 revenue of 44 to 48 million, with Q1 revenue of 8 to 9 million, and then revenue growth increasing in subsequent quarters. We expect that our gross margin will be in the 65 to 68 percent range. We anticipate a net loss for the year of 21 to 25 million, with losses trending down throughout the year. I should comment that in that late January we took further steps to reduce our operating costs in our Biosystems Division to better align with revenue and business conditions. In addition to the effect of the BioSepra transaction, our headcount is currently down approximately 12 percent as compared to November 30th, which is to say, reflecting those employees that left the Company due to the BioSepra sale. We don't run our divisions with separate P&Ls, largely because we want them to operating synergistically with an emphasis on divisional teamwork. However, the reality is what we have is a Biosystems Division that's now firmly committed to operating as a profitable enterprise, while we continue to support Diagnostics Division as an earlier-stage enterprise with significant potential. The cost reduction steps taken in January targeted the Biosystems Division, while not affecting our investment in the Diagnostics Division. Now I'd like to turn the call over to Bill Rich for additional comments.

Thank you, Matt. Normally I start these Webcasts with our Biosystem Division results and then talk about Diagnostics Division. Today I'd like to reverse that to ensure that our Diagnostics Division receives the attention that it deserves and to emphasize that progress towards commercialization of our first diagnostic test is actually the strongest proof of the power and enablement of SELDI ProteinChip technology that translates biomarker discovery to high-value assets. We believe this will be a major driving force for rapid expansion of Biosystem product sales, as well as for creating a significant new business opportunity in protein molecular diagnostics.

Let me start by talking about our lead program in ovarian cancer. As many of you are aware, we published an important paper in cancer research in August last year describing a 500 sample multi-site study in which we discovered and independently validated a 3 marker multi-marker panel of biomarkers that showed the promise of improved detection of ovarian cancer. Co-authors on this paper included several thought leaders in the field of ovarian cancer including Dr. Dan Chan at Johns Hopkins Medical School; Dr. Bob Bast, Vice President of Translational Medicine at M.D. Anderson; and Dr. Ian Jacobs, who has one of the largest ovarian cancer studies in Europe, who is currently at the University College of London. Subsequent to that time we've run additional samples that validate the correlation of 2 of the markers with the detection of ovarian cancer. We recorded initial results from a study utilizing 436 patient samples at the International Gynecological Cancer Society meeting in October. This study confirmed findings recorded in the Cancer Research paper regarding the down regulation of 2 specific proteins in the early-stage ovarian cancer study as well. In fact, one of the exciting findings of these studies were that these markers correlate therapy and outcome, and appear useful for recurrence monitoring.

And then in December at the M.D. Anderson Memorial Sloan-Kettering International Conference on Ovarian Cancer we reported validation from yet another sample set for these same indications. We've done considerable assay optimization work to produce a high throughput, precise and highly reproducible assay around these 2 markers and have made good progress on that score. Ciphergen and Johns Hopkins together are currently working with the third biomarker to better optimize its reproducibility in actual clinical lab testing before using up our precious early-stage ovarian samples for further validation studies. The primary focus in optimizing the final assay conditions is understanding the pre-analytical variables including, sample handling, freezing, and thawing the samples prior to analysis. To accelerate and expand this work, Ciphergen and M.D. Anderson Cancer Center signed a collaboration agreement recently involving analysis of clinical samples provided by M.D. Anderson for ovarian cancer using Ciphergen's new Deep Proteome and Pattern Track suite of proteomics tools. Ciphergen has exclusive rights to license discoveries made during the course of this collaboration. This work will address multiple clinical questions including the validation of the markers described in the Cancer Research article, discovery and validation of markers that might distinguish ovarian cancer from other gynecological masses, and prediction of treatment response. We anticipate describing further data on our ovarian cancer program at the Society of Gynecological Oncology meeting in March.

We are also collaborating with the National Cancer Institute's Clinical Proteomics Reference Lab in developing their own ovarian cancer pattern-based multi-marker study using SELDI as a diagnostic assay platform. Their mandate is to develop and validate a proteomic-based clinical diagnostic test and device to file for FDA approval for ovarian cancer diagnostics. Operationally, we achieved a number of milestones recently to prepare Ciphergen for the commercialization of our diagnostic tests. Our chip automation project, with its anticipated benefits in terms of improved chip quality, reproducibility, cost reduction, and scalability represents 1 meaningful step forward to achieve our initiating it this year. In addition, we have completed a GAAP analysis of our operations, assessing what steps will be required to move from our current ISO 9000 status to cGMP status. We have now set up an internal team and are committed to achieving cGMP status during 2005. Also, by the end of the first quarter we anticipate receiving an economic analysis from a group at Northwestern related to reimbursement and value modeling of our ovarian assay. This is clearly an important part of our commercialization planning. Although there can be no assurance that this will be the case, it remains our goal to introduce our first diagnostic test as an ASR during 2005.

Let me move on to discuss several additional programs underway at our Diagnostics Division in addition to the ovarian cancer work. You should note that our collaborators on these programs come largely from our existing install base of customers, which we believe are our, quote, new diagnostic test farm club. Of course, this is a major source of synergy between the Biosystems Division and the Diagnostics Division. Here are just a few examples. First, our customers at the Children's Hospital in Boston, working in collaboration with scientists from Ciphergen, presented results at the American Society For Hematology's 46th Annual Meeting suggesting proteomic analysis of circulating platelets can potentially be useful for early cancer diagnosis. Next, the John Hopkins University School of Medicine and Ciphergen have renewed their agreement to discover and validate biomarkers for detection of early-stage breast, ovarian, pancreatic, and prostate cancer. Collaboration had resulted in publications already in journals such as Cancer Research, Clinical Cancer Research, and Clinical Chemistry. This has been and continues to be a very productive relationship.

Next we are collaborating with another customer at M.D. Anderson Cancer Center to discover and validate markers that can predict a lung cancer patient's response to cisplatin therapy. This is in addition to the collaboration with Dr. Bast on ovarian cancer. We are also collaborating with another important customer at INSERM, which is, of course, the French NIH, to identify biomarkers derived from inflammatory cells that can predict outcome in patients with colon cancer. The inflammatory reaction is an important component of the host response protein amplification cascade that may provide targets for cancer detection monitoring and therapeutics. We're also working with a second group of customers at INSERM to identify biomarkers that are protective against hypertension, a major cause of morbidity and mortality, including myocardial infarction, stroke, and renal disease. Also in France, our customers at the Institute Paoli-Calmettes and Ciphergen are working together to identify proteins that can predict response to anthracycline or Herceptin therapy. Better prediction of response may predict more tailored therapeutic decisions.

Also, Ciphergen is a participant in the ADDNET Consortium in Scandinavia dedicated to discover, characterize, and validate urinary biomarkers for renal impairment in patients with diabetes. Ciphergen technology and expertise in biomarker discovery and assay are key components in this European effort to develop novel diagnostic tests for the early detection of renal failure due to diabetes. Also, our customers at Stanford Medical School are collaborating with Ciphergen to discover novel biomarkers that can be used to diagnose peripheral artery disease, a major complication of diabetes. Preliminary results were presented at the Society For Vascular Medicine and Biology meeting and will be presented in the Young Investor -- the Young Investigator Award Competition at the American College of Cardiology in March. Next, in the field of Alzheimer's disease, Ciphergen and our collaborators have identified a set of biomarkers that appear to be changed in the early stages of Alzheimer's disease. Many of these markers appear to have a pathophysiological basis, suggesting that they may be useful in identifying drug targets or monitoring for drug efficacy. These results were presented in a plenary lecture by Dr. Kaj Blennow at the 9th International Congress for Alzheimer's Disease in 2004.

Next, McGill University and Ciphergen are collaborating to discover, validate, and identify blood biomarkers for parasitic diseases including Chagas disease, toxoplasmosis, and malaria. Early detection of these diseases may reduce morbidity associated with long-term infection. Chagas is a major disease problem in Latin America, which is currently potentially threatening the blood supply now in North America as well. These are just a few of the many collaborations that we have that we think will potentially produce new diagnostic tests and also, of course, fuel the sales of our systems business. Partnership discussions. We continue to advance deal discussions with various parties in the reference laboratory and in vitro diagnostic industries around commercialization of our pipeline of diagnostic tests. We have progressed to detailed term sheets with multiple parties. We remain confident that we will be able to compete -- complete a significant partnership in the first half of 2005.

Let me segue from the diagnostics division back to the Biosystems Division. I'd like to reiterate the strong synergy we see between these 2 divisions. Success on the diagnostic side, discovering and validating biomarkers, identifying them and developing a research assay validates the value of our ProteinChip biomarker proteomics technology. With our customers and collaborators we are further optimizing these multi-marker assays and validating them in multi-site clinical studies using our platform. Success here provides a strong proof statement for the adoption of our technology and products. We strongly believe that success in our Diagnostics Division will create strong market acceptance for our products and services among the large number of proteomics-based translational medicine researchers in academia and in pharma. In addition, we see this synergy in the other direction. Our large base of installed users are currently making biomarker discoveries around various clinical questions. And these represent a further source of potential diagnostic tests for our Diagnostics Division, ergo, our farm club. Of course, successful commercialization of these tests provide our shareholders with major new business opportunity.

In our Biosystems Division, we experienced declining sales for the first time in the second quarter last year, followed by a transition quarter in Q3 as we introduced our Series 4000 ProteinChip system, and then further sequential improvement in Q4. This interruption in our historical multi-year growth rate was caused by introduction of competitive biomarkers proteomics methods, instrumentation, and services, causing our potential customers to evaluate -- stop and evaluate these alternatives. Market reaction to the Series 4000 has been quite positive. We believe that this new product has distinct advantages over alternative proteomic methods, in particular, the speed of getting from discovery to assays. It's all about assays. We also see increasing scientific success and publications coming from prestigious laboratories around the world who are using our ProteinChip technology. This is creating a growing acceptance by the market, and we anticipate renewed revenue growth to result in 2005.

Series 4000 features the Pattern Track trademark biomarker discovery to assay process, and integrates Ciphergen's proprietary ProteinChip arrays, SELDI-TOF-MS detection, and biomarker pattern software in a single optimized system that can enable rapid biomarker discovery, validation, and development of predictive high-throughput SELDI-based assays. The system provides improved system sensitivity, resolving power, and reproducibility for biomarker discovery. It is -- its pattern recognition-based software selects the optimal -- optimal combination of biomarkers from literally hundreds of biomarker candidates that can easily be observed from differential profiling experiments, automated single and multi-market assay optimization and validation, rapid purification, and identification of the biomarkers, and finally, automated and quantitative SELDI-based assays. Series 4000 is an easy-to-use bench-top system with a lower cost of goods, allowing a -- or an introductory, lower price point than previous systems, and has features consistent with our continued strategy for promoting decentralized proteomics and enabling researchers in 50,000 laboratories to perform their own work at their bench top with high performance.

Series 4000 provides, actually, 4 biomarker subsystems in 1. You can discover, purify, ID, and assay in different modes, each are optimized to provide maximum speed and ease of use and performance for implementing the Pattern Track Process. Our Pattern Track Process is the culmination of years of work in development and with our customers to combine SELDI-based differential protein profiling and identification capabilities with pattern recognition software for the purpose of discovering and validating and IDing multiple protein biomarkers that, together, create high predictive value assays. Also, as part of the Pattern Track process, the SELDI system is used as the final quantitative assay platform to run the multi-marker assay. This rapid discovery to validation to assay capability is the powerful and the unique aspect in biomarker proteomics.

One of the most significant things we will be bringing to market this year is an enhancement to our Pattern Track process in the form of new sample preparation tools. We're currently using these tools in house for our proprietary discovery use, and will offer them commercially by Q3 to select collaborators and customers. These new products, which are trademarked Deep Proteome tools, enable detection of low abundant proteins in complex biological fluids such as serum, which is potentially solving a major problem today in biomarker proteomics. We consider these tools to be a major advancement that our potential customers will find compelling. Our new Deep Proteome tools feature Protein Equalizer Beads. Invented in collaboration with the American Red Cross by Ciphergen -- with Ciphergen this powerful combinatorial chemistry-based bead technology allows a deeper search for low abundant proteins in proteomes, biological fluids, by equalizing protein concentration rather than, quote, depleting proteins. This reduces the high-concentration proteins and simultaneously enriches the low-concentration proteins. In effect, this produces a signal-to-noise enhancement analogous to the signal-to-noise enhancement produced by PCR for DNA. But through noise reduction rather than signal enhancement. Current depletion methods suffer from the unwanted loss of many proteins during the depletion process compared to our protein equalizer method. We believe this is a significant scientific advancement, and will help enable both our Biosystems and diagnostic businesses.

We also saw growth in our Biomarker Pathways tools which include, predominantly, protein interaction discovery mapping, or IDM kits and protocols, which we introduced in late 2003. These enable the -- these enabled improved Series 4000 capabilities to discover novel protein interactions and build quantitative SELDI-based immunoassays to study biological pathways, function, and drug development processes. Protein interactions applications are 1 of the fastest growing new applications areas for us and in proteomics. In 2004 interaction difference mapping products represented a primary or secondary application for over 35 percent of our new system sales. Protein interaction applications and their extension to biomarker pathway elucidation represent a major opportunity.

Next, we believe our pattern-based multi-marker discovery approach to biomarker proteomics, combined with the Series 4000's well-engineered system features and specifications that optimize our Pattern Track Process, will be of significant interest to core proteomics labs, offering a complimentary approach to their current so-called top-down and bottom-up biomarker proteomics methods. We plan to extend our marketing programs to the core proteomics labs with a new Deep Proteome and Biomarker Pathway tools, we believe, will also provide expanded applications interest to the core proteomics groups as well as, of course, our traditional clinical research customers.

Pharma programs. Regarding the pharmaceutical and biotech market segment, all of the major pharma companies are seriously adopting various biomarker strategies, and we expect to benefit from that. Our ProteinChip system is currently being evaluated by most of the major pharmaceutical companies to identify biomarkers of drug safety and efficacy to monitor drug treatment and reduce late-stage attrition of compounds. Advances in the analysis and identification of low-abundance proteins through the implementation of Deep Proteome and Biomarker Pathway products should further enable the discovery of the novel biomarkers and elucidation of pathways in drug discovery and development. The Company, Ciphergen, is currently in discussions now with multiple pharmaceutical companies to partner in the field of Alzheimer's disease based on promising discoveries and assays that we have developed in our Copenhagen Biomarker Discovery Center and with other collaborators.

Clinical proteomics research. There is a growing list of users throughout the world doing disease biomarker discovery and assay development applications with now over 600 of our systems sold throughout the world with multiple users or applications for systems. For example, here's some numbers. In the Alzheimer's disease area, we have over 100 users or applications going on currently with our users. Over 100 people -- 100 cases of prostate cancer applications, over 130 applications in ovarian cancer, over 130 in breast cancer, over 40 in colon cancer, and 85 in -- over 85 in other various cancers. In the cardiovascular disease area, over 100; in the HIV/AIDS area, over 135; infectious disease, over 95; diabetes, 55; kidney disease , 45; and arthritis, greater than 15. So you can see that the number and breadth of applications of our users is quite substantial.

Publication progress. This broad use of ProteinChip technology has led to 122 scientific papers in 2004, which represent a 60 percent increase in publications over the number of publications in 2003, and brings our total publications to us now to over 350. Ciphergen scientists and our customers represent -- presented 48 papers at the 2004 AACR meeting and other major disease-based conferences. This growing list of successes continues to feed our systems growth and potential diagnostics pipeline and proof of our Pattern Track Process and our ProteinChip technology. Patents. We filed an additional 27 patent applications during 2004, the most new filings of any year in our history. These applications relate to our products, technologies, and the biomarker discoveries. Our patent portfolio now consists of 23 issued U.S. patents, 102 pending U.S. patent application families. In January 2005 we were granted U.S. patent 6844165, directed to the detection of multiple diagnostic markers by SELDI-TOF-MS for the diagnosis of disease. This patent strongly compliments several existing issued patents covering biomarker proteomics and potential use in diagnostics.

Conclusion. We remain confident and committed to 2005 will be a very different story from a difficult 2004. We believe that our Diagnostics Division is advancing several potentially valuable diagnostic assays towards clinical use and that we will have multiple opportunities to describe that process during the course of the year, as well as partnerships and the potential launch of our first diagnostics ASR product. On the Biosystems side, we believe that the trend towards translational proteomics and the merits of our Pattern Track approach; the introduction of Deep Proteome and advances in our Biomarker Pathway tools; the successes, in fact, of the Diagnostic Division; and the success of our users around the world will result in significant revenue growth of our Biosystems products in 2005. I think I'll stop here and take questions. Operator?

At this time, I would like to remind everyone, if you would like to ask a question, press star, then the number 1 on your telephone keypad. We'll pause for just a moment to compile the Q&A roster. Your first question comes from Eric Schmidt of SG Cowen.

Good morning. I was hoping for a little bit more color on the competitive environment in the Biosystems Division. You had mentioned some weakness in sales in 2004 and others coming into that business. Has anyone left the business, or what's, maybe, the reason that things seem to be improving in the last 3 months?

Martin, why don't you take that?

Okay. Thank you, very much, Eric. The competitive environment, as Bill discussed is actually determined by 2 factors right now. 1 is there's more companies entering this market with mass spec technologies, but also with non-mass spec technologies. And that creates, actually, confusion. And then, on the other hand, you still have this core mass spec facilities who tend to -- who really, massively entered this biomarker market segment and are offering services in competition with buying a SELDI system. What we are seeing, though, and that gives us a little reason of optimism, is that our current own users are actually becoming more and more successful, and that is actually most measurable by the publications growth that we've seen last year. And we are now getting some real feedback from the market that that is starting to get some clarity about the value proposition that we have out there with our Pattern Track Process.

Eric, let me make another comment, if I may. The -- there's the evolving evaluation of the value proposition, in fact, in proteomics. It surrounds the fact of are sequences enough? Is the total, huge list of sequences really useful, or is it, in fact, what am I going to do with those sequences, which, ergo, is produce an assay and the bottleneck, now, is becoming highly recognized. And I think it's something we recognized several years ago. The bottleneck is how you -- not how you get a sequence of something. The bottleneck is how do I make an assay from it that's useful? So I get thousands of protein sequences. So what? Which one or ones do I actually make an assay out of that will help me navigate some kind of a process of understanding?

And that's the power of our technology. As our customers are beginning to recognize and as scientists are beginning to recognize that that is the problem and that we have a very unique and significant solution, I think that's the driving force that we're beginning to see. We developed the 4000 so that, in fact, had all the features of being, not only a good discovery system that could do this, but it also, now, and what differentiates it most strongly from our previous system, is a quantitative assay platform. So its reproducibility and ability to actually, then, do the assays that you discover on it is one of the most compelling new features. And I think as the marketplace is beginning to see that, I think we're going to get a lot of traction.

And as follow-up, could either Matt or Martin comment on the number of placements, instrument placements over the last quarter or last year that were to academic labs as opposed to commercial labs, and what the funding environment looks like in that subsector?

The -- what is actually happening is the majority of our systems go to academia, and I don't have the exact numbers but ratio is probably something like 70/30, or 70 percent of our systems go to the academia and about 30 percent go to the industry, mostly the pharmaceutical industry. However, recently, we've seen a trend of a pickup in the pharma industry regarding more requests out of the pharma industry. And we're very -- we're having active programs in place for the industry to address those requests. So we are actually expecting growth from the industry part.

Okay. And the funding in academia with the NIH budget being flat?

For us, quite frankly it remains -- it remains the same because we haven't really had much traction at the NIH in the past year, so I don't think it's going to go down from there.

Okay. Thanks a lot.

Again, if you would like to ask a question, press star, then the number 1 on your telephone keypad. Your next question comes from Adam Chazan of Pacific Growth.

Hey, guys. A couple of quick questions. First off, can you just remind us how many of the boxes sold in the quarter were to Pall? And then as they set up these centers, how many do you hope to sell into them, or, you know, how many orders do you have in hand?

We sold 3 to Pall to set up these process proteomics centers. And what was the second part of your question?

You know, I guess, what's the outlook? How many of these centers are they setting up? And, then, how many more boxes might go to them over the course of '05?

You know, we really don't have a figure for how many might go to them. They've set up -- they're setting up a center in the U.S. and 1 in Europe. And then, we'll sort of see how it goes.

Okay.

There's nothing more firm than that.

Okay. And then, your thoughts, you know, you had a significant number of trade-ins, you got 600 systems in the installed base soon. How are you thinking about, you know, an upgrade cycle, you know, over the course of '05, or the instrument sales we're looking or the expectations are those upgrades, are those new sales? Do you have a feeling for what the mix might be or what you hope it to be over the course of '05?

Oh, let's see. Who should handle that, Martin or I? I think, certainly, for the next couple of quarters, we should see quite a bit of upgrade interest. And it's part of the marketing program, is to go back to all the old customers who owned a IIc or even a PBS II and demonstrate to them, you know, now's the time to upgrade to a Series 4000. You know, last quarter maybe 20 to 25 percent, in numbers, came from upgrades and trade-ins. And that probably continues for the next several quarters.

And actually, it's an indication, I believe, of how the market views the 4000 system because in -- and I'm going to give you some typical examples is, well, we have users that come to one of our entities, they see the new system, that perks their interest, they run samples side by side and they see the advantages of the new system. And it's been, actually, remarkably easy to propose trade-ins to these existing customers. And I think it's just an indication of how favorable they view the 4000 system. So I agree with Matt, I do foresee that we will continue to see a good level of upgrades happening in quarters -- in the upcoming quarters.

Any bundling or discounting in the quarter?

I don't know about bundling, but, you know -- you know, we always wind up having a discussion with a customer there on, you know, paying list price or making them feel good by throwing something in or giving a discount. So there's some discounting. And when you're going back to these old users and doing an upgrade, every one of those is kind of a customized sale because it depends on how old is their system, how much of a credit do we want to give them for a 1-year-old system versus a 4-year-old system?

I guess, Matt, what I'm getting at is in Q3 there was, you know, kind of a concerted effort, I think, bundle some arrays with --

Oh, yes. Compared to that -- compared to the Q3, sort of, launch quarter we've seen a nice improvement. And we're no longer, you know, including a large number of chips for free with the initial sale. We actually raised pricing, list price, in the fourth quarter versus the third. So I think we're beyond what we did in the third quarter. I don't know if you want to add to that.

Well, the third quarter we had an introduction pricing on the system, introduction bundling, I should call it. And they've operated, you know, in the fourth quarter. So with that respect, our discounting has actually been reduced quite significantly.

Okay. And then, 1 last question. Gross margin improvement over '05, you've got the new chip manufacturing infrastructure up. How should we be thinking about that over the course of '05 and '06 as volume ramps? You know, what kind of improvement might we be able to realize there?

Well, just glancing here for a second. I think the guidance that we want to give is gross margin of 65 to 68. And I think it's going to improve as each quarter goes on, partly due to the impact of the chip manufacturing system, partly due to volume, and, so you know, you absorb more overhead. And also as time goes on, on the system, I think we're going to improve our gross margin. Whenever you launch a new product, you've got some inefficiencies as you're making the product and you're doing design changes and stuff like that. And as the quarters go by those disappear. We've also got something going on in the system where we're going to hopefully replace 1 of the high-cost components with a cheaper alternative, probably early Q2, and that will help gross margin. So I'd say, you know, trending up as the quarters go on. But for the year, in that range of 65 to 68 feels like the right kind of number right now.

Okay. Great. I'll jump back in the queue. Thank you.

Your next question comes from Craig Layton of J.P. Morgan.

Hi, guys. Just a couple of follow-ups here. I'm just wondering how many of your 37 systems were to new customers? And then also, just following up on Adam's question, I assume that, while you mentioned that you give a credit for the trade-in, can you just give me some idea as to how much that is?

Okay. So the -- if you go back to the numbers, as Matt discussed, we have 37 systems, 8 of those represented trade-ins, so that brings 29 new placements there. And again, with regards to the discount with trade-ins it's a bit of a customized situation where, if you would do trade in of a very old system, the discounts are marginal. If it's a very recent system, then the discounts actually do go up, so there is no single rule that we can give you there.

Okay. I was actually looking for systems to new customers as opposed to new systems.

It's probably not profoundly different than the 29. I mean, so what's a new customer? If Pfizer buys a system, is that a new customer? Well, no, Pfizer owns a bunch of systems, but we're probably selling it to a new user within Pfizer, like a different lab. So if we had to approximate -- out of the -- discount the 29, and I don't know, what's your guess, Martin? 27 represent some new -- really new user. It may be within an umbrella organization that we already have a relationship with.

I was going for new accounts, just to simplify it here.

Okay. 27. That's a guess.

Okay. And then, is it too early to really start talking about Series 4000 array use, as in what proportion of arrays came from your Series 4000 placement?

Too early. It's too early to talk about it because we started placements over the last 6 months here. And most of these people, they are -- still have to ramp up to their maximum or their, sort of, stable use. So it's too early for that.

Okay. Thanks.

Again, if you would like to ask a question, press star, then the number 1 on your telephone keypad. Your next question comes from Adam Chazan with Pacific Growth.

One last question. How are you guys feeling about the cash balance? And how might that influence the kind of deal you structure on the diagnostics front?

Well, I think that 1 of the things we did, we mentioned, was in January, we took steps to reduce our cash burn. But beyond that, I think, you know, we will probably during the course of the year need to supplement our cash position. 1 of the main ways that we're looking to do that, and our preferred way to do it is by -- through corporate partnering, either on the diagnostics front, or, you know, we may be able to do something out of the tools division as well. But on the diagnostics fronts, I don't think we're going to let it profoundly influence the kind of deal we do. There's, you know, a lot that goes into what's the right kind of a deal. And I don't think we're going to do anything that just gives us short-term benefit but is a bad move in the long term. I don't know if anybody wants to add something on the diagnostic influence? No?

Thanks.

Again, to ask a question, press star, then the number 1 on your telephone keypad.

Do you have any other questions, Operator?

There are no further questions. This concludes today's conference call. You may now disconnect.

Thank you.