By Robert M. Frederickson
October 10, 2003 | Target validation remains a pivotal factor that determines whether genomics-based drug discovery can turn findings into new products to treat disease. However, most preclinical work is done in animal models and cultured cells, which are not necessarily good predictors of how a drug will act in the human body.
"The use of animal models in biomedical research is very important, but ultimately the full understanding of human disease requires the study of humans," said bioinformatics expert Mark Boguski of Vulcan Inc. in his keynote address at the 1st Clinical Genomics Symposium in June 2001. Moreover, workers do not always have on hand appropriate models of a disease's stage or its genetic origins. Clinical genomics is beginning to bridge the gap between preclinical hopes and clinical realities.
By examining well-characterized and appropriately preserved patient samples, researchers can tweeze out which of hundreds or thousands of genes are key to the origin, progression, and perhaps even treatment of human disease. Indeed, it has been suggested that we might categorize or "recategorize" a disease based on its genetic profile.
The pharmaceutical and biotech industries have recognized the potential research value of clinical samples. We are seeing "a change in mindset, from that of traditional drug discovery research to that of population science," Boguski continued in his address. "What we are witnessing today is the shifting upstream toward discovery research of the clinical trial discipline: namely, the power of large numbers of patient cases and the systematic application of relevant clinical data to correlate with molecular changes."
While collaborative studies have been set up between particular research institutions and individual companies, most clinical samples are discarded by the academic centers and hospitals that collect them. Indeed, until recently, no group has attempted to collect and organize a national repository of high-quality clinical samples and related patient information.
Ardais was founded in 1999 with the aim of creating such a repository. A privately held biotechnology company, Ardais announced the launch of its nationwide "clinical genomics" initiative with leading academic medical centers in September 2000.
Mixing Biological and Medical Data
Ardais' plan was to partner with these centers to collect, process, and use high-quality clinical materials and the associated information — all under the auspices of the necessary scientific and medical expertise, bioinformatics tools, and appropriate bioethical considerations. Said Martin Ferguson, senior vice president of bioinformatics at Ardais, in a 2001 press release: "The clinical data that accompanies tissue and molecular derivatives must be recorded in a comparable way from patient case to patient case, from center to center, expressed in standardized vocabularies, and accessible across large numbers of diverse patient populations."
| Tissue type: Ardais has gathered a vast collection of clinical tissue samples and data.
"[Our plan is to create a platform that integrates] the huge volumes of biological data from the human genome and molecular biology studies with the vast repository of medical knowledge gathered over decades of clinical experience," Ferguson added.
To date, 180,000 samples have been logged, representing nearly 11,000 patient cases. Forty-five companies had signed on by August, including six of Big Pharma's top 10. Ardais averages about eight new subscribers per quarter.
What these subscribers get is access to the vast sample collection and its annotated database of rich clinical information. Samples include mRNA, protein, or any biochemical derivative of interest to the client. Standardized disease or custom tissue microarrays are also available. Ardais takes care of the regulatory, ethical, and quality control issues and indemnifies the researcher from any claims. Key to the success of its system has been a $13-million investment in IT to create an ontology to accurately quantify clinical data.
Last year, the company was awarded a two-year, $2-million Advanced Technology Program grant from the National Institute of Standards and Technology. In October 2002, Ferguson said that the funds would be used "to develop and test prototype software tools, data structures, and applications for linking quantitative, structured clinical information derived from the medical records of patients to the molecular profiles derived from their donated tissue samples."
Earlier this year, Ardais' investors granted the company $13.4 million in new financing. "Ardais has successfully validated its business model by capturing the leadership position in the emerging field of clinical genomics, and by forming commercial relationships with the leading firms in the biotech and pharmaceutical industry," said William Mills, partner at Advent International, in an official statement.
Ardais plans to leverage its technology to offer "tissue collection services" for gathering and analyzing samples during a clinical trial. The idea is to deploy the system for use by academic centers that do not want to re-create the wheel Ardais has already fashioned. Ardais also recently launched immunohistochemistry services and is licensing software and protocols for full biorepository management services.
While others, such as Asterand, Clinomics Biosciences, and LifeSpan BioSciences, vie for space in the clinical genomics arena, Ardais seems to be leading the pack at the moment.
Robert Frederickson is a biotech writer based in Seattle. He can be reached at email@example.com.