Leerink Report: INVESTMENT THESIS We rate Illumina (ILMN) Outperform for the following seven key reasons: High-density genotyping offers a market opportunity with attractive growth. Genotyping refers to the process of testing an individual’s DNA for the presence of individual genetic mutations that can improve understanding of the basis of, or predisposition to, disease. High-density genotyping seeks to test entire genomes and populations for mutations known as single nucleotide polymorphisms (SNPs) of statistical significance. The recent release into the public domain of over a million genotypes by the International HapMap Project, as well as additional information supplied by a range of other sources, is spurring scientific and drug discovery researchers worldwide to increase focus on genotyping. This information’s availability, which some liken to the release of the Human Genome Project’s sequencing data of several years ago, is driving increasing demand for high-density genotyping products. The high-density genotyping products market is difficult to size currently, but certainly stood under $100 million in 2004 revenues. We expect that demand from drug and biotech companies seeking to use this gene mutation information in drug discovery and development and from disease foundations and governments seeking to characterize the genetic risks among groups or individuals, the size of the market can grow by over 30% annually over the next few years. We believe the competitive landscape in the genotyping products market has yet to be set, and share is available for additive new technologies like ILMN’s. As opposed to several other areas of genomic research (notably high-density gene expression), we think the emerging high-density genotyping products market has yet to declare winners and losers. ILMN has two competitive challenges in high-density genotyping: there is an installed base of a competitor’s (Affymetrix) capital instruments already in the field for gene expression analysis that can be used for genotyping, and AFFX is about a year ahead of ILMN in bringing high-density consumable SNP arrays to market. As long as ILMN can bring its 100K SNP array to market by the end of 2Q-05 as planned and 250K and 500K chips to market by year-end or so, we think it has the ability to compete well for high-density genotyping product business. ILMN’s proven high-density genotyping technology is already vetted and well positioned in this emerging market. ILMN’s BeadArray technology is already proven through its use in the HapMap project since 2002. ILMN was the largest participant in the US phase of HapMap, and its BeadArray technology was used in the creation of over half of the data generated in the project. ILMN’s genotyping assays were found to be sensitive and reproducible. As a result, ILMN stands in a strong competitive position in the high-density genotyping market. A new region-specific ILMN genotyping product is unique, and additive to drug researchers. Earlier this month, ILMN released a testing product that allows researchers to study in great detail the genetic variation in the major histocompatability complex (MHC) region of the human genome.
This region of the genome is thought to be a source of much information about the origins of autoimmune and inflammatory disease. ILMN’s new MHC panels will allow researchers to conduct both high-throughput and high-resolution analysis of a region that has been a challenge for other techniques. We expect this consumable product to both be successful in its own right but also help drive placements of the company’s BeadArray instruments. Away from genotyping, ILMN possesses a powerful cost advantage in oligos, now leveraged through a key partnership. We believe that ILMN’s oligo (short stretches of synthetic DNA for research use) business can grow to $100 million in sales by 2008 from sales we estimate stood in the $10 million – $15 million area in 2004. ILMN has long had a manufacturing cost advantage in this competitive business, but its distribution channel has not been able to fully penetrate customers. With the late 2004 agreement between ILMN and Invitrogen, ILMN enters the biggest part of the oligo market and its products are now being distributed by one of the largest sales forces in the molecular biology business.
ILMN offers a compelling approach to high-density gene expression analysis. While we think that the market leader in high-density gene expression is highly unlikely to be unseated, we expect that ILMN can be one of the reasons that researchers migrate from lower-density or homebrew approaches to high-density array methods for gene expression. ILMN recently introduced whole-genome gene expression analysis consumable products in their 6-sample and 8-sample BeadChips. These consumable products allow for variable per-sample processing costs well below those of the market leader, setting aside the cost of the processing instrument. ILMN’s flexibility may allow it to compete effectively in the custom gene expression microarray area. We expect the high-density gene expression market to grow from 2004’s $450 million to over $650 million in 2008, with ILMN’s share in the 5% to 10% area. Valuation: attractive versus a relevant historical example; price/sales below an index of growth life science companies; DFCF drives 12-month forward valuation range of $14 - $17 and our Outperform rating. We examined where a company that we consider to be the previous most direct comparison to ILMN traded at a similar point in its growth cycle. In the second half of 1998, before the genomics bubble was in full roar, high-density gene expression company Affymetrix (AFFX) was posting quarterly product revenues that were similar to the $12 million that ILMN posted in 1Q-05. At that time, AFFX was trading with a market capitalization of around $600 million, well above the $460 million or so seen today at ILMN. We view the long-term opportunity in high-density genotyping as at least similar in size and scope to today’s high-density gene expression market. While we believe that the competitive landscape may be more difficult for ILMN today than it was for AFFX in the late ‘90’s, we believe researchers should be more comfortable today with adopting high technology solutions to solving biology problems. Comparing ILMN to a group of higher-growth life science consumable product companies shows that ILMN is today trading at 5.9X our 2005 sales forecast of $78 million, below the market capitalization-weighted average of 6.5X. From a discounted free cash flow standpoint and considering historical precedent, we believe that ILMN should trade in a year at $14 to $17 per share, or 16% to 40% above current levels. We rate ILMN shares Outperform.
CATALYSTS We think potential ILMN investors should monitor the following areas for insight regarding the company’s fundamentals and stock performance: The number placed of the company’s BeadStation instruments. To date, 55 of the company’s $200,000 or so BeadStation experiment processing instruments have been placed (along with 10 higher-productivity BeadLabs). We assume that BeadStations will represent 90% of the company’s future instrument placements. BeadStation placements are critical to the sale of follow-on consumable products. By contrast, over 1,250 AFFX GeneChip Systems have been placed for use with that company’s consumable. We think BeadStation adoption is key to ILMN’s success. Consumable product sales growth to customers who already own an ILMN BeadStation instrument. While the company needs to place more instruments, the long-term goal is to use the installed base of tools to sell more consumable products. ILMN has introduced several consumable products since the beginning of 2005, including whole-genome BeadArrays, the DASL gene expression assay, and the MHC fine mapping products. It is important to see traction among existing BeadStation users for new products. Anecdotal evidence of increasing interest in genotyping by commercial researchers. To date, we believe that drug and biotechnology company researchers have been constrained from exhibiting greater interest in large-scale genotyping for two reasons: the cost per data point of genotyping and the lack of a reference set of information against which to easily compare new data gathered. We believe that the HapMap project is providing that reference data and the cost of genotyping continues to decline, thanks largely to advances by ILMN. Positive regulatory agency comments regarding the value of genotyping data. We expect that with improved understanding of the value of understanding genetic variability in the drug development process, regulatory agencies such as the FDA will seek more such information to be included in new drug approval packages. We have seen this process begun already in differential gene expression data and expect that at some point genotyping data will also be sought. Adoption of the company’s DASL assay for gene expression of degraded nucleic acids. In addition to the direct revenue benefits of selling more products, we believe that the customer base for the company’s new DASL assay includes researchers for whom the product will be the first introduction to ILMN’s capabilities. We think that commercial success for the company’s DASL assay will create new customers for a broader range of the company’s products. Emergence of new applications for the company’s products. To date, we believe virtually all of the company’s products are used in the research setting. Placements into clinical trial population monitoring or into clinical diagnostics would represent an expanded opportunity for ILMN. Growth in oligo sales as a result of the Invitrogen (IVGN) partnership. We expect that the company’s oligo sales will begin to accelerate in the second half of 2005 as a result of the December 2004 marketing and product development partnership with IVGN.
While neither ILMN management nor most investors see oligos as the key to long-term value creation at the company, cash flow from oligos improves the company’s financial flexibility and allows overhead to be spread over a larger base of business. Sluggish growth in revenues from this partnership might also be an indication of a diminishing pricing advantage inherent in the company’s Oligator technology, a situation that we clearly do not foresee over the next few years. Progress toward product introductions at development-stage CyVera. ILMN bought privately held CyVera in February 2005 to augment its high-density (over 10,000 data points in a single experiment) solution for SNP detection and gene quantitation with lower-density (10 – 100 or so data points) capability. Management expects products in 2006. Signs of progress toward products from CyVera can be important to the price of ILMN shares.
RISKS Among the risks of investing in mid-capitalization growth stocks, we encourage investors to consider the risks that we perceive as specific to ILMN: ILMN’s success is dependent on sales of capital equipment instruments. ILMN’s long-term success, we believe, is a direct function of the company’s ability to place its $200,000 BeadStation instruments. Should its customers have difficulties purchasing such capital equipment, the company’s ability to sell higher margin consumable products is compromised. The size and growth rate of ILMN’s key opportunity, the large-scale genotyping market, is unclear. We believe that the large-scale genotyping market represents a significant opportunity for ILMN. However, it is important for investors to see that the market is still in formation. Its ultimate size, as well as the extent of ILMN’s participation in this market, is unclear. The leader in high-density gene expression is also focusing on the high-density genotyping market for future growth. Gene expression powerhouse AFFX is building microarray-based products for the high-density genotyping market. Its most recent product, the 100K SNPChip, was introduced in the middle of 2004 and met with strong commercial success. ILMN intends to launch a product that is directly competitive to this AFFX product by the end of the month. AFFX intends to launch a 500K SNPChip by the end of 3Q and we expect ILMN’s 500K product will follow by year-end or so. ILMN should expect AFFX to remain a formidable competitor in high-density genotyping. ILMN’s future revenues from the International HapMap Project are unclear. ILMN has been a large contributor to the HapMap project to define genetic diversity through the discovery of over a million different single nucleotide polymorphisms (SNPs). ILMN has to date received about $9.1 million for the development of SNP assays and for genotyping services using its technologies. In 2004, roughly 30% of ILMN's sales were linked to HapMap. In 1Q-05, about 16% were from HapMap-funded transactions. We believe that ILMN is currently owed no revenues under HapMap contracts. Whether new contracts may be signed is unclear.
Litigation risk. In July 2004, AFFX filed a lawsuit alleging that certain ILMN products infringe six AFFX patents. ILMN filed an answer in September 2004 seeking declaratory judgments that it does not infringe the AFFX patents and filed a counterclaim against AFFX for unfair competition. Any unfavorable outcome to these proceedings could have a negative impact on ILMN’s cash position and ability to reach profitability. No trial date has yet been set. The oligo market is competitive and has been characterized by per-unit price declines. ILMN expects to gain share and profitability in the oligonucleotide market. This market is competitive with several well-funded and entrenched entrants. Per-unit price declines have been seen over the last several years – we estimate that per-base pricing in tube-based oligos have declined by 40% over the last three years. Such a harsh pricing environment is a challenge to management and profitability. Dependence on government and foundation-funded customers. The majority of ILMN’s genotyping revenues have come not from drug or biotech companies but from government and academic labs as well as non-profit disease state foundations. We generally view customers who are not dependent on outside funding as more durable. Recent change to the management team. ILMN has recently hired a new CFO. We expect that this officer will require a process of training and orienting to ILMN’s business and processes.
VALUATION: DFCF, Price/Sales, and A Relevant Look Back We believe that in the near term, ILMN stock will trade as a function of its news flow regarding new products and investor expectations of how news will effect future revenues. We project a revenue growth rate for ILMN of about 43% (2004-07E CAGR). Discounted free cash flow analysis yields 12-month forward valuation range of $14 - $17. Without current earnings, P/E based valuation analysis is not possible. For that reason, we use discounted free cash flow as a valuation metric. We have applied our discounted free cash flow model to the ILMN financial results that we project for the years 2005-2011. Using our terminal EBITDA multiple of 10X and a 20% to 25% discount rate on future cash flows, we arrive at a 12-month forward valuation range of $14 to $17 per ILMN share.
More clarity on the nature of the large-scale genotyping opportunity and funding availability for capital instruments would likely allow us to consider lowering our model’s discount rate. Our model’s sensitivity to changes in terminal EBITDA multiple and discount rate is shown here:
Price/sales and PSG ratio shows ILMN to be valued below a selection of higher growth life science consumable product companies. ILMN today trades at about 5.9 times our expectation of $78 million in 2005 sales, below the 6.5X market capitalization-weighted average of several life science companies with attractive top-line growth profiles. ILMN’s price to sales ratio compares to other growth oriented names in the life science space as follows:
Figure 3: ILMN Relative Valuation vs Peer Life Science Consumable Companies Sales 2005E ($MM) Market Cap ($MM) 2005E Price/Sales Leerink Sales CAGR (2004A – 07E) PSG Ratio AFFX 405 3,500 8.6X 19.4% 44% ILMN 78 461 5.9X 42.7% 14% IVGN 1,195 4,800 4.1X 12.4% 33% LMNX 40 280 7.0X 23.6% 30% SERO 281 862 3.1X 14.5%* 21% TECH 177 1,800 10.2X 9.5% 107% Cap-wtd avg: 6.5X 46% Leerink Swann & Co. estimates; * 2005E – 2007E
Comparing ILMN to another life sciences company standing before an emerging opportunity offers insight. In 1998, high-density gene expression microarray company AFFX was posting similar quarterly product revenues to what ILMN posted in 1Q-05. Like ILMN today, a body of biological data was becoming available for use by researchers, and it seemed that AFFX was well positioned to benefit. The Human Genome Project was an important inflection point in researchers’ desire to conduct large-scale differential gene expression experiments. We believe that the information made available as a result of the completion of the first phase of the HapMap project has provided a like inflection point for researchers. In late 1998, AFFX’s market capitalization stood in the $600 million area with product sales similar to ILMN’s today. By contrast, ILMN today trades in the $460 million area. The comparison is far from perfect. The competitive landscape was then probably more favorable for AFFX than for ILMN today, the US government’s ability and desire to increase the amount of funding for basic research was then certainly more positive than today, and gene expression data is more dynamic than SNP information. Nonetheless, we expect that high-density genotyping will drive new grant applications and experiments just as the Human Genome Project did.
COMPANY DESCRIPTION Founded in 1998, San Diego-based Illumina is a provider of life science tools targeted at the large-scale analysis of genetic variation and function. ILMN’s BeadArray technology uses microscopic beads randomly deposited in wells to form a microarray that permits many assays to be performed in parallel, a method known as multiplexing. We think of ILMN as competing in three areas: oligos, gene expression, and genotyping. The Oligo Market and ILMN’s Oligo-Related Business With its patented Oligator technology, ILMN management describes itself as the low cost manufacturer of synthetic oligonucleotides, known as oligos. Oligos are short stretches of synthetic DNA or RNA used as probes in gene expression and other biology research. Oligo buyers are scientific researchers conducting basic biology or drug discovery studies. Oligos are provided to customers in two primary ways, either by the tens or hundreds in microwell plates, or singularly in tubes. Until recently, ILMN has competed only in the plate-based segment of the market. The oligos ILMN makes run from 15 to 75 chemical bases long. In the fragmented $200–300 million oligo market, major ILMN competitors include publicly traded Sigma-Aldrich (through both Sigma-Genosys and Sigma Proligo), MWG Biotech, and Invitrogen. Privately held competitors include Operon, Synthegen and Bionexus. We believe that in 2004, ILMN’s share of the oligo market was less than 10%. Pricing is aggressive in the oligo business, as the products are not protected by intellectual property and are similar across vendors.
Consultant-conducted market research has estimated that around 10% of oligo buyers switch vendors annually. A single oligo today can cost as little as $0.12 to $0.14 per chemical base within a plate-based bulk purchase or as little as $0.22 to $0.24 per base in lower-density tubes. We estimate that per-base pricing in the oligo market has dropped by around 40% in three years. We believe the plate-based oligo market is roughly $50 million in size, the smallest part of an overall market that is about $200 - $300 million. The tube-based oligo market, which facilitates the vast majority of researcher demand for small numbers of oligos for individual experiments, is $150 million to $250 million in annual revenues by our estimates. ILMN’s Invitrogen Oligo Partnership: A Transforming Event In December 2004, ILMN announced a partnership with IVGN for the production and distribution of ILMN’s oligo products. Under the terms of the deal, IVGN will be the sole distributor of ILMN’s oligos, with the partners splitting the profits 50/50. ILMN will enter the tube-based oligo market with the help of a $3.4 million payment from IVGN to develop next-generation oligo manufacturing technologies that facilitate not only plate-based but also tube-based products. ILMN management has stated that the two companies together generated around $35 million - $40 million in oligo business in 2004. The partnership also stipulates that all sales, marketing, and technical support responsibilities related to these oligos transferred to IVGN. Further, IVGN will continue to manufacture certain oligos that are more difficult to produce and of lesser interest to ILMN due to their specificity to IVGN kits. We think that this partnership is an ideal combination of ILMN’s Oligator technology and IVGN’s industry-leading sales force and distribution channel that should drive higher sales and better market penetration. We believe that this partnership is exactly what ILMN needed in order to leverage its industry-leading oligo production technology. As a result of the agreement, the sales capability of ILMN in oligos goes from 15-20 or so to IVGN’s 350-person sales force. We expect that the ILMN/IVGN partnership can build its share of the oligo business from today’s 10% - 15% or so to 35% over the next three years. We expect that the oligo business covered by this arrangement will reach $100 million in 2008, with operating profitability before the 50/50 partners’ split in the 25% area.
Instruments for Gene Expression and Genotyping ILMN is in the market today with two products. The BeadStation 500 costs $200,000 to $250,000 and includes robotics and software. The company’s legacy BeadLab platform adds automation, LIMS, and sample prep capability and has a price tag of $1 million to $1.5 million. The same instrument platform, with minimal adjustments and software upgrades, can be used for both gene expression and genotyping. As of the end of 1Q-05, ILMN had sold 55 of the $200k area BeadStation, including 10 in the first quarter of 2005 alone; it had also sold 10 of the $1 million BeadLab, including 1 in the first quarter of 2005.
The Bead Station had been introduced at the end of 1Q-04. The International HapMap project had purchased four BeadLab systems for use in large-scale genotyping and assay creation; with the first phase of the project complete, we are awaiting more information regarding how Project managers may seek to deploy its BeadLab assets. ILMN would like to see the instruments in use so that it can sell consumable products such as BeadChips, assays for genotyping and gene expression, and reagents. Genotyping In 2004, custom genotyping products comprised all ILMN product revenues besides Oligator-related oligo sales. Using the company’s Golden Gate genotyping assay, researchers have been able to genotype efficiently and price-competitively. A high-density ILMN fixed content array, the 100K DNA SNP genotyping array, is currently in beta testing and is expected to be on the market by the end of the month. ILMN’s product differs from the already-launched AFFX 100K SNPChip in several ways. We think a key difference is the universe of SNPs that will be offered by ILMN. AFFX’s approach is limited in that it can only detect SNPs that reside within a certain distance of a DNA restriction site. This can leave AFFX’s product poorly equipped to query some SNP-rich areas of the genome in either linkage or association studies. ILMN management observes that its 100K BeadArray includes 70,000 SNPs that are within 10kb of an exon (the protein-coding region of a gene). By contrast, the AFFX 100K SNPChip includes only 27,000 that reside within genes. The implication is that SNPs within or near genes are more medically relevant than those further away. It is important to note that this assertion is intuitive but has yet to be proven. While ILMN readies its 100K SNP product for market, AFFX’s next high-density genotyping product, the 500K SNP chip, is due on the market by the end of 3Q-05. As it uses different reagents and enzymes from the 100K product, AFFX believes that the SNPs that it discovers on the 500K product will not be redundant to the content on the 100K chip. ILMN believes that its 100K BeadChip launch will be followed in short order by a 250K SNP detection product and then a 500K product. Ultimately, management has stated that it intends to have a 1 million SNP detection array on the market by the middle of 2006. Gene Expression – Differential Microarrays We think the differential gene expression microarray business is an $800 million segment, which includes both microarrays and lower-density technologies (spotted arrays, homebrews, etc.) for differential gene expression. ILMN recently launched a whole genome expression human chip called the Sentrix BeadChip. The product uses a single 50-mer probe per gene, attached to a 23-mer unique address that is attached to a bead. Management states that the company plans to launch mouse and rat genome chips by year-end 2005. ILMN’s core Sentrix BeadChip costs about $960 and can process 6 samples on a chip. With full utilization by the researcher, the cost is $160 per human genome, well below the $400 area per sample from the competition. ILMN asserts that the sensitivity of its product is better than AFFX’s chip.
To be specific, ILMN asserts that its product carries linear range and data reproducibility that represents an improvement over legacy technologies in the market. Management states that academic and internal research will be published in the next few months that will point to the positive differences between ILMN and competitors. It cited a single study on brain/liver tissue where ILMN’s self-concordance was above that of other microarray technologies and ILMN discovered more genes than the competition. Until a body of research comparing microarray technologies is made available and vetted, we remind investors that variability can be great among companies seeking to conduct technology comparison studies. Operator bias and error is frequent, and study design is sometimes not in line with what is useful to product users. As well, investors should recall that several such studies comparing microarray technologies have been offered over the years, each comparing its technology favorably against industry leader AFFX. After it all, AFFX remains by far the commercial leader in high-density DNA microarrays with market share that we estimate at 75% to 80%.
We believe the installed base of AFFX tools and the mindshare that AFFX holds (including the thousands of scientific papers that have been published which cite high-density DNA microarrays and frequently AFFX’s technology) continues to bias gene expression researchers heavily toward the AFFX GeneChip product. Other ILMN Gene Expression Products Focused expression arrays. This is a custom product, often for researchers who have completed whole-genome analysis and are now bearing down on a small number of genes that are proving to merit further study. With this product, a research stipulates up to 700 genes of interest that can be interrogated and quantified in a gene expression experiment.
ILMN prepares beadsets that each contain a capture molecule for one of the 700 genes. The beads with oligos attached are then placed in a single vessel to await the addition of a prepared sample. As these are plate-based, researchers may process 96 samples in a single experiment. A researcher can designate up to 700 genes of interest for which ILMN makes probes for use with a 1,500 head fiber bundle on which the beads self-assemble. The redundancy (700 genes, 1500 spaces) is to assure that all of the genes of interest assemble on the array for analysis. The cost is $100 per sample, so the 96-sample plate costs $9,600. ILMN does not sell partial plates. DASL assay. ILMN’s new DASL assay can help researchers derive gene expression information from DNA contained in degraded settings such as formalin-fixed, paraffin-embedded tissue samples. These samples, often tumor biopsies, have value to researchers as they may be several years old and carry years’ worth of relevant clinical history with them. Researchers can go back and look at the gene expression patterns and see how they relate to patient outcomes. Similar to the custom assay product, DASL is used with ILMN’s Universal Array. Cost is about $100 per patient sample, and the researcher must identify their genes of interest. ILMN creates an oligo pool for interrogation of the sample. Importantly, ILMN needs 50 intact bases of DNA as a starting point to identify a gene and discern an expression pattern. We view the DASL opportunity as finite in nature.
Today, clinicians and pathologists are doing a better job of archiving tissue samples in ways that are not as limiting in terms of future use. We suspect that in several years, a product like the DASL assay will not be as important. Although there are other homebrew methods for extracting gene expression information from these samples, they are generally qPCR-based and low-throughput in nature and therefore not amenable to large samples. However, we note that DASL is not by any means the only method for examining archival DNA. For example, the Paradise Reagent System, developed by Arcturus Bioscience, www.arctur.com, provides an integrated system to isolate and amplify mRNA for analyzing global gene expression in archival specimens.
FINANCIAL STATEMENT ANALYSIS A review of ILMN’s financial statements leads us to conclude that the company has ample financial flexibility to implement its strategy and reach sustainable profitability.
Income Statement Sales: ILMN has three reporting units that contribute to total sales: products, services, and research. In 2005, we expect the company to generate the vast majority (about 84%) of sales from products, with the remaining 14% and 2% coming from services and research, respectively. We do not expect this mix to change dramatically as the ILMN’s installed base of instruments grows. In 2007, we expect instrument sales to constitute about 91% of total sales, with services contributing 7.5% and research totaling just 1%.
Margins and Expenses: Gross margins, which stood at 74% in 2004, should be in the 69% to 71% area through 2007 thanks to continuing new product introductions and stable pricing for instruments. Although we acknowledge that oligo sales have traditionally experienced poor pricing power, we believe that ILMN’s low-cost leadership and increasing market share will offset further declines. Further, we expect oligos as a percentage of total revenues to decline through 2007 as ILMN’s BeadStations gain traction. We believe that ILMN will continue to spend on R&D and SG&A as the business ramps into 2H:05. Management has guided to approximately $28-$30 million in R&D spending in 2005; we are modeling approximately $29 million. Regarding SG&A, management guidance is for $31-33 million; we anticipate about $31 million in 2005. Earnings per share: We project that ILMN will post EPS results in 2005 similar to those seen in 2004, albeit for different reasons.
We expect ILMN to narrow its 2005 EPS loss slightly over 2004’s performance, from ($0.17) to ($0.10). We expect the first full-year profitability at the company in 2006, where we model EPS of $0.21. Off a base of negative EPS in 2004A and 2005E, EPS CAGR analysis is not likely to be useful for investors. We think revenue growth analysis will be a more useful metric in the near term.
If stock-based employee compensation expense determined under a fair value based method was incorporated into the company's income statement and not simply footnoted as allowed under SFAS 123, the company's ($0.17) GAAP EPS loss in 2004 would have been ($0.41). Cash flow statement and balance sheet: Net of adjusting for cash moved to marketable securities in 2004, ILMN’s cash flow performance is roughly in line with its net income losses. As of the end of 2004, ILMN had cash and marketable securities of about $55 million. We expect a loss of $4.5 million or so in 2005 to be seen at the operating cash flow level, which the company can handle well from its existing cash position. Inventories and accounts receivable seem to be in line with the changes in the rest of the business. We believe that capital expenditures at the company will run in the $3.5-4.0 million area annually over the next few years.
RECENT EVENTS 1Q-05 Results: Street-beating Sales, Positive Outlook In late April 2005, ILMN reported 1Q-05 sales of $15.1 million, above the company’s guidance range of $13 to $15 million. This performance marked the 15th consecutive quarter of sequential revenue growth. Product and services revenue of $14.9 million rose 50% over the previous year’s quarter. GAAP EPS of ($0.03) compares to ($0.12) in the previous years’ quarter. FY-05 revenue guidance of $75 to $82 million was reaffirmed, as was EPS guidance of ($0.14) to ($0.01). Management noted that it expected a more back-end loaded 2005 than seen in some previous years, due to the expectation that sales of whole-genome BeadChips for gene expression and genotyping would be concentrated in the second half of 2005. Whole genome BeadChips for gene expression were only introduced during 1Q-05, while 100K SNP genotyping BeadChips were expected to be introduced by the end of 2Q-05.
Guidance was offered for 2Q-05 of a range of ($0.08) to ($0.04) in EPS. CyVera Acquisition In April 2005, ILMN completed the acquisition of CyVera Corporation, a development-stage company that allows researchers to detect molecular binding events, for $17.5 million. The deal consisted of 1.6 million shares of Illumina common stock and $2.5 million cash. ILMN’s rationales for the transaction seemed to be two: that CyVera’s intellectual property and expertise in molecular binding detection and labeling beads were novel and compelling, and a technology for cost efficient lower density multiplexing of molecular binding events would offer improved leverage of ILMN’s franchise and fixed costs. CyVera owns its solid phase addressing technology, which is centered on a unique hologram printed onto a bead; at the detection phase of an experiment, a laser would detect the hologram and infer the capture molecule on the bead. At the price paid for this development-stage company, the decision to make the purchase seems to pass muster strategically even if only the first rationale is justified in the long run. ILMN’s self-assembling array technology identifies individual molecules through quantitative algorithms, and augmenting that method seems to us like it could increase the flexibility of ILMN’s approach and results. ILMN management expects that CyVera will have products to market in 2006. The CyVera instrument system is expected to cost $50k to $75k.
MANAGEMENT Jay Flatley has served as ILMN’s President and CEO since 1999. Previously, he was President and CEO of Molecular Dynamics where he led the company’s IPO in 1993 and eventual sale to Amersham Pharmacia Biotech in 1998. Flatley serves on the Board of Directors of Bruker AXS and received degrees from Claremont McKenna College and Stanford. Christian Henry has served as Vice President and Chief Financial Officer since 2005. Most recently, he served as CFO of Tickets.com, a publicly traded online ticket provider. Previously, he was Vice President, Finance and Corporate Controller at Affymetrix and held a similar position at Nektar Therapeutics. He holds degrees from the University of California, San Diego and the University of California, Irvine.
David L. Barker, Ph.D., has served as Vice President and Chief Scientific Officer since 2000. He was previously Vice President and Chief Science Advisor at Amersham Pharmacia Biotech and held senior positions at Molecular Dynamics, Inc. He holds degrees from the California Institute of Technology and Brandeis University.
John Stuelpnagel, DVM, is ILMN’s Founder and currently serves as Chief Operating Officer and a Member of the Board of Directors. He founded Illumina as a Venture Capital Associate with CW Group and previously worked at Catalyst Partners. He received degrees from the University of California, Davis and The Anderson School at UCLA. Scott Kahn, Ph. D, joined ILMN in 2005 and serves as Chief Information Officer. He previously served as Chief Scientific Officer at Accelrys Inc. He holds degrees from the University of California, Irvine and did post-doctoral work at Cambridge University, UK.
As of March 8, 2005, all officers and directors of Illumina as a group owned 10.4% of the Company’s common stock.
CONCLUSION The macro opportunity that stands before ILMN can be as attractive as anything we’ve seen in life sciences since DNA sequencing and gene expression volumes expanded dramatically in the middle and late 1990s. ILMN’s genotyping products and services have already been vetted through the HapMap project. We expect a key new ILMN product for genotyping, a 100K SNP BeadArray, to be introduced for commercial sale by the end of 2Q-05. The competitive landscape for ILMN is a challenge, but we believe the market is likely to grow quickly enough to reward several suppliers.
We rate ILMN shares Outperform. Illumina, Inc.—
June 10, 2005 P 15
Leerink Swann & Company |
