Appendix Q For-Profit, Minimal Government Funding Example: Ardais Corporation Ardais Corporation is a privately held clinical genomics company whose goal is to accelerate biomedical research by applying actual human disease, in the form of human tissue samples, as the discovery model in pharmaceutical research. The Ardais Biomaterials and Information for Genomic Research (BIGR™) System encompasses a unique repository, called the BIGR™ Library, with more than 120,000 research-quality tissue samples representing a broad diversity of disease. The samples are collected through the National Clinical Genomics Initiative, a strategic collaboration between Ardais and four leading U.S. medical centers. The formation of the National Clinical Genomics Initiative was based on the identification of the broad need for tissues and data to support genomics-based biomedical research by researchers in academe and industry. The goals of the Initiative are to develop systematic, large-scale procedures to comprehensively collect, process, and store research-quality clinical materials and associated information; to provide these critical resources in highly optimized formats for efficient and robust design of biomedical research studies; and to support the research and clinical programs at each participating medical institution. Ardais began the Initiative in January 2000 with Duke University Medical Center (DUMC) and Beth Israel Deaconess Medical Center, a Harvard University Medical School affiliate. The initiative was launched with these two participants in September 2000 and since has expanded to include Maine Medical Center and the University of Chicago. To date, the Initiative has collected, processed, and stored over 160,000 biospecimens from approximately 10,000 patients at its partner institutions. Approximately 50,000 samples per year are collected from a total of about 4,000 patients seen at the four sites. The Initiative presented many challenges in tissue and data collection. Broad-based tissue collection by a for-profit company would require iron-clad patient protection. To address this issue, Ardais designed a comprehensive bioethics strategy that considered legal requirements, the recommendations of the National Bioethics Advisory Commission (NBAC), and suggestions from experts in government, industry, and academe to protect the privacy of patient donors. Maintaining the quality of tissue samples and clinical information was a primary focus of the Initiative. Ardais developed standard operating protocols covering tissue sample access and archiving patient data collection and patient identification protection and dissemination to qualified researchers. These protocols were designed to maximize the molecular and histological quality of the samples. Ardais emphasizes the importance of high standards to disease research because: Only if one complies fully with all existing Federal, State, and institutional requirements regarding patient privacy can clinical materials be appropriately licensed to a researcher. Only if samples and data are collected under a unified set of standards can a researcher focus on biological variations between diseased samples rather than on variations due to poor sample handling and processing. Only if the relevant clinical and pathological information has been captured consistently in structured formats can a researcher select the appropriate samples easily and cost effectively. Only if a rigorous quality assurance/quality control (QC) process of sample validation and pathology verification has been applied can a researcher be sure that the sample is consistent with the hospital diagnosis and is of sufficient molecular integrity. The need for a comprehensive information technology (IT)/bioinformatics system was also identified. Specific needs included: Sample tracking Case linking Web-based access Experimental design Browsing Remote deployment Multisite coordination To satisfy IT and bioinformatics requirements, Ardais created the BIGR™ System as a discovery platform for application to drug discovery and development. The system includes a centralized, shared clinical genomics repository that encompasses tissue samples, molecular derivatives, and associated clinical information accessed by an array of bioinformatics tools. Finally, Ardais developed standardized protocols to address the complicated logistics requirements (training, supplies, barcoding, shipping, storage, inventory control, infrastructure, and equipment) of a multisite repository system. Program Structure The Genomics Initiative pays rigorous attention to legal and bioethical issues. These steps include: Active dialogue with participants Formalized policies Formation of a Bioethics Advisory Board Required institutional review board (IRB) approval of all protocols (linked and unlinked) IRB oversight of distribution of samples to commercial entities A fully informed consent process Deidentified data Title for research use A stated policy of no interference with necessary medical procedures Procedures to obtain longitudinal clinical data followup Fully Health Insurance Portability and Accountability Act-compliant patient identity coding processes In line with NBAC report and the Code of Federal Regulations. Ardais developed stringent collaboration operating procedures to minimize conflict with the participating institution. There is no interference with independent, ongoing tissue collection procedures. There is a stated desire to efficiently address the tissue collection needs at each site. Surgeon approval is required before patients can be approached. Consenting activities are coordinated during preoperative screening procedures to ensure that patients are not unnecessarily burdened with multiple consent requests. The program structure at each site includes a dedicated staff and infrastructure, to ensure full focus on appropriate data and sample collection. The institution hires and pays the staff, with salary and benefit support provided by Ardais. Uniform protocols are implemented across all sites. All supplies and equipment for each site are obtained from the same vendors and are identical. Specific technical training procedures are utilized across sites. There is ongoing inservice training and QC monitoring throughout the Initiative. The program management team at each site consists of representatives from Ardais Executive Management and the home medical center. Day-to-day management is the responsibility of the Banking Staff Management team (pathologists, bank manager, research nurse, project leaders) and the Ardais Alliance Management team. Joint steering/operating committee meetings occur at prescribed intervals (monthly via telephone or onsite). Open communication has been found to be critical to management success. A formalized structure has been developed for identifying and deploying best practices. Consent Ardais consent and confidentiality procedures are standardized across the four participating institutions. An informed consent interview (10 to 30 minutes in length) takes place with each patient. Appropriate patients are selected by research nurses, who have a success rate of over 95 percent in obtaining consent from those patients who are approached. A unique Ardais identifier and case identification are assigned to each consented donor for use in the BIGR™ Library. The link to the patient is maintained at the home medical center in a secure database (only banking staff have access). A time-stamped consent tracking form has been developed, which includes: Verification of every consent after sample banking (required prior to case release) Validation of time stamp on initial informed consent Indication if specimen was successfully banked Sample Processing Sample collection and banking procedures follow set guidelines at all institutions. After samples are retrieved from the patients, they are evaluated at a grossing station. A pathologist determines which surplus tissues can be banked, and both normal and diseased tissues are dissected. At a processing station staffed by trained, project-dedicated personnel, samples are cut and processed into modules. Samples are weighed, the time of each procedure is recorded, and all information is entered in the informatics system. Samples are linked to barcodes and stored in Ardaisdesigned protective containers (cryosettes). Samples are: Optimally frozen to enhance preservation of histologic quality and enable molecular analysis with histological specificity Matched to relevant, structured, and in-depth clinical data Available in frozen/optimal cutting temperature-embedded, and formalin-fixed/paraffinembedded formats Of uniform size and shape Pathology-verified by board-certified pathologists and accompanied by verification data Preserved within 1 hour, postresection—actual time is tracked QC and verification data associated with each sample are available through the BIGR™ Library. Medical center IRB-approved protocols are in place, with linked samples accompanied by patient informed consent that permits broad and ongoing data collection. All samples are held at the home institution for an embargo period—usually 30 days. This permits the attending pathologist to retrieve additional tissue specimens for patient care, if needed. Once the pathologist has signed off on the samples, the embargo period ends and the samples are shipped to the tissue repository in Lexington, Massachusetts. Shipping procedures follow similar uniform procedures and are tracked by the informatics system. Shipping containers consist of unified kits designed specifically for these purposes. There is strict adherence to Government regulations (Occupational Safety and Health Administration, U.S. Department of Transportation, and International Air Transport Association) for shipment of biomaterials. Personnel involved in shipping operations are highly trained. The informatics inventory management system can track where all samples are at any time and at any site—a “virtual” repository. The specific location of samples in a freezer can be ascertained. It is necessary to maintain the link to the donor for ongoing data collection efforts, while adhering to all donor confidentiality precepts. At DUMC, the Medical Assistant on the World Wide Web (MAW3™) system was created to collect longitudinal data. Ardais is linked to this system as a central hub, with MAW3™ as a donor site. Each system collects and stores structured, comparable data. Systemized Nomenclature of Medicine—Clinical Terms is used to encode data elements in a common dictionary for communication between the systems. At DUMC, data extraction from the database is needed. At Ardais, the necessary data are abstracted and put into an electronic, structured, deidentified form that does not allow the capture of patient-identifying information. Case report forms include 250 to 400 disease-specific fields for structured data as well as fields for demographics, history of the illness, surgical pathology findings, postoperative notes, discharge summary, and follow-up information. Image capture capability allows for display of representative images and translational detail of samples. Clinical Genomics Ardais provides quality samples for research use, with tissue collection based on identified research needs. All major cancers are represented in the repository as well as some of the rarer cancers. Ardais has board-certified pathologists on staff or serving as consultants and can provide sample qualification through pathology verification. All samples are pathology-verified prior to distribution to researchers. In addition, the company offers consultation for product development and human molecular pathology services. Company histotechnology capabilities include sample QC, slide production, and immunohistochemistry. Pathology data are disconnected from the repository samples because pathological evaluation is conducted on the tissues retained for diagnosis, not on the samples donated to the repository. Therefore, pathology data are verified for the samples to make sure that the findings are consistent with the diagnosis and tissue composition of the original samples. A mechanism exists for reporting back major discordants in diagnosis. The event chain of exactly what happens to the samples after their removal from the patient is meticulously tracked. It is possible to tell who requests a sample, which personnel performed various analyses, and what specific processes were used. Samples are either frozen or formalin-fixed (for 8 to 15 hours before processing)/paraffinembedded. Ardais offers a variety of clinical genomics applications for sample processing: DNA extraction RNA extraction (90 percent of samples) Molecular derivatives Laser-capture microdissection Protein extraction Cell cultures Qualification data Microarrays Macroarrays (for angiogenesis studies) Tissue sections Available frozen tissue arrays have 2-mm core dimensions and up to 40 samples per array, and are qualified for immunohistochemistry analysis. RNA is qualified for research use in: Gel-based analysis Bioanalyzer analysis Polymerase chain reaction-based analysis Array-based analysis Quantitative measurement of gene expression QC In order to prioritize which samples will be collected at each site, target scores were created to determine what types of samples are of interest to the institution’s researchers. Monthly reports are generated to identify how well the targets are met, if demands are satisfied, and if corrective measures are needed. There is structured quality feedback to the medical institutions. Bioinformatics Library The BIGR™ Library is a scalable, Web-deployable, Java-based system architecture that is managed from a central database. The Library is multi-institutional and will allow Web-based access to the repository on a researcher’s desktop computer. Each medical institution establishes a “Research Strategy Committee” to review requests for samples and ensure IRB compliance. It is possible to designate samples for restricted allocation to the home institution. Ardais has instituted an education protocol (involving beta system utilization) to inform researchers about the benefits of using BIGR™. To date, DUMC has trained 8 investigators, Beth Israel has trained 6 investigators, Maine has trained 4 investigators, and Chicago has trained 12 investigators to use the BIGR™ Library. The idea is to train the investigators so that they can/will apply to the committee for access to the live system. Once the distribution program is operational, there will be an open enrollment period for researchers. It is possible to search for samples using: Diagnosis Identification (patient identifier) Attributes Tissue Appearance All of the above It is also possible to click on a microscope icon and see a digitalized pathology image of the samples on the screen. When an RNA derivative is generated, all of the images and data are available and immediately viewable.

Return to Table of Contents Next Appendix: Appendix R: For-Profit, Primarily Private Funding Example: First Genetic Trust Top of Page