Market patterns emerge for biomarker companies.
Identifying patterns of proteins sloughed off by diseased tissues into the bloodstream and other body fluids offers promise for pharmaceutical development. Finding these protein biomarkers can help researchers predict drug efficacy and toxicity in pre-clinical research, screen and classify patients during clinical trials, find optimal drug doses and even identify discovery targets.
Instrument companies also see wide-reaching clinical diagnostic opportunities, including the potential to detect illnesses such as Alzheimer's disease, Parkinson's disease, HIV and cancer, to name a few. Although the idea of placing mass spectrometers in doctors' offices may not seem feasible, scientists are seeing steady progress toward the day when physicians will be able to accurately diagnose diseases with a simple blood test and precisely tailor treatments to individual patients.
Dr. Mark Flocco, business development manager for Clinical Proteomics and Biomarker Discovery at Bruker Daltonics, says that one effect of the growing biomarker trend has been interest from a new crop of researchers using mass spectroscopy (MS) to analyze proteins. "The market before was mostly technical or analytical groups. Not to say all, but many who used the instrumentation came from a different educational or technical background. We are now approaching a new group of clinical researchers, biologists and therapeutic groups that we have had little contact with before this. These are people asking more biological questions than identification or structural questions," he says.
To make the jump to MS as seamless as possible, these new users require more robust and high-through-put instrumentation that automates much of the tedious sample preparation process as well as offers user-friendly back-end informatics. In addition, says Dr. Flocco, instrument companies will need to be increasingly sensitive to the training and support needs of these scientists.
In response to this changing demographic, many instrument companies have created specialized proteomics solutions that address such crossover concerns. Last year, for instance, Bruker introduced its ClinProt system (see IBO 9/15/03), which brings together the company's magnetic bead-based sample preparation and liquid-handling robotics with its AnchorChip targets and matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) analysis. More recently, at the American Society for Mass Spectrometry Conference in May, Applied Biosystems unveiled two products geared toward the biomarker researcher--its iTRAQ Reagents, which allow proteomics researchers to draw from multiple samples, and its Pro QUANT Software for automatic data processing (see IBO 6/15/04).
The relative newness of biomarkers discovery, and the lack of a single standard platform, has created additional opportunities for instrument companies angling for a piece of the market. One disagreement between researchers is whether MALDI-MS or electrospray LC/MS (liquid chromatography/MS) is more appropriate for analyzing proteins of interest. "It's like religious factions," says Dr. John Michnowicz, LC/MS marketing manager at Agilent Technologies. "You have people who absolutely believe in MALDI and people who absolutely believe in electrospray LC separations." Ionization techniques is a matter of sample preparation preference--either 2D gel electrophoresis or LC. Regionally, this divide is more pronounced, with many European researchers leaning toward 2D gels and MALDI, and US labs tending to favor liquid separations and electrospray. Bruker and PerkinElmer tell IBO that biomarker applications have stirred interest in their respective MALDI-TOF instruments. PerkinElmer claims that the orthogonal design of its prOTOF 2000 MALDI OTOF, which was developed with MDS Sciex, helps bypass consistency problems sometimes caused by sample ionization. Similarly, the MDLC LTQ--a collaboration between Thermo Electron and GE Bio-Sciences--complements 2D electrophoresis (see IBO 6/15/04). The proteomics LC/MS system released by Agilent in May is based around Agilent's 1100 Series LC platform. The technology also offers the Multiple Affinity Removal System, which comprises a special column that allows low-abundance proteins to pass through while retaining high-abundance ones. Waters's offerings lean toward electrospray with its Protein Expression Profiling System, which employs the company's nanoACQUITY UPLC and Q-Tof Premier MS (see IBO 6/15/04), but also include the Waters Micromass MALDI micro TOF (see IBO 6/ 15/04). Shimadzu Biotech sells both electrospray and MALDI systems, as does Applied Biosystems.
While these competing techniques may contribute to market polarization, Dr. Michnowicz says that he doesn't see this technological rift closing anytime soon. "I think the tide has slowly changed to primarily using LC/MS techniques for proteomics, but for biomarkers there are still a significant number of people using MALDI because of the influence of the Ciphergen [ProteinChip] System."
The ProteinChip System, which uses Ciphergen's proprietary surface enhanced laser desorption/ionization technology, was used in a high-profile ovarian cancer biomarker study conducted by biotechnology company Correlogic Systems along with the National Cancer Institute (NCI) and the Food and Drug Administration (FDA), and published in 2002. Although the study boasts sensitivity of 100%, specificity of 98% and positive predictive value of 94%, some have raised questions about its reproducibility. In an NCI interview released at the same time, FDA scientist Dr. Emanuel Petricoin, who co-authored the study, said that to be suitable for women in a general population (as opposed to those in high-risk clinics) the false positive rate would need to be much smaller.
The rigorous attention focused on validating biomarkers points to the peculiar nature of such tests, where the pattern itself serves as the diagnostic. Although pattern recognition as an application is not a new concept, it places an additional burden of proof on the companies producing the tests. Low-resolution MALDI-TOF analysis can generate more than 15,000 data points, and high-resolution MS can create up to 400,000. Since individual proteins markers often do not correlate with a specific disease, the difficult work of discovering patterns is left to software that can perform sophisticated algorithms--an area to which instrument companies are paying close attention.
Last month, for example, PerkinElmer signed an agreement with Predictive Diagnostics to access the company's Biomarker Amplification Filter (BAMF) informatics tools for biomarker profiling and analysis. According to Dr. Mary Lopez, director of biochemistry at PerkinElmer, the deal enables PerkinElmer to offer a complete platform to researchers developing diagnostics or screening populations for clinical trials.
Agilent, which licensed Millennium Pharmaceuticals' SpectrumMill software for biomarker discovery in 2002 (see IBO 12/31/02), is also actively developing new computer tools for biomarker researchers. "There are several different ways of doing biomarker discovery in software," explains Dr. Michnowicz. "One of them is to do it at the protein level, which is what SpectrumMill does, and the other way is to do it at the peptide level. We're developing software and exploring bioinformatics concepts which will make the discovery at the peptide level much more automated and relatively easy for mass spec people to do." Dr. Michnowicz says he cannot divulge additional details about the project until Agilent nears an introduction date.
"Multimarker, or pattern recognition, analysis has become a key component in going from clinical research [to diagnostics]. However, one must be careful not to over-train the model-generating software to obscure any valuable information," says Dr. Flocco, who cautions that it is essential for results not only to be statistically sound, but also to be biologically relevant. "A diagnostic will be closely monitored for statistical and biological credibility before being accepted and used in the diagnostic market," he says.
In November 2003, biomarkers moved closer to becoming a diagnostic reality when the FDA released its "Pharmacogenomic Data Submissions" draft guidance, providing a system for classifying the proteins (see IBO 11/15/ 03). According to Dr. Lopez, the momentum behind biomarker research is pointing more and more toward government approval. "I think everyone is waiting with bated breath for the FDA to approve one of these types of pattern-recognition-based tests as a diagnostic," she says, adding that she believes that Correlogic's OvaCheck is probably closest to achieving this milestone. In November 2002, Correlogic licensed its ovarian cancer detection technology to Quest Diagnostics and LabCorp (see IBO 3/15/04). Also, the company has continued to work with the NCI and FDA. "I think that that is inevitable. I think that this approach is an approach that will be adopted by the FDA. At the moment, it is about a $100 million market, which is nothing to sneeze at. It's growing at a rate of around 30%-33%. So, it's a pretty respectable market. It's volatile, it's rapidly emerging, and I think the moment yon see the FDA approve one of these tests, it's really going to take off."
Still, despite the enormous growth potential that biomarkers offer and the absence of any dear leader in the field, Dr. Lopez notes that there are surprisingly few companies jockeying for position. "Once the first test is approved, I think you will see a domino effect, because this concept will pretty much take over, and that's when you'll see the companies and the providers that can give high-quality execution come to the forefront," she predicts. In the meantime, she says, instrument manufacturers are eyeing the market with a great deal of interest, but also wariness. "You see a lot of companies beginning to make their move, but the field is not very crowded in this area."
|Printer friendly Cite/link Email Feedback|
|Publication:||Instrument Business Outlook|
|Date:||Oct 15, 2004|
|Previous Article:||The bottom line.|
|Next Article:||Varian to acquire UK magnet maker.|