January 15, 2014 (Vol. 34, No. 2)
Biorepositories provide access to a vast array of biospecimens, which are rapidly becoming crucial resources in drug discovery and development.
These biospecimens must be of high quality to ensure the results attained are reliable and reproducible, and to avoid wasting time and money by reliance on inadequate specimens.
The challenges in collecting, processing, storing, and tracking biospecimens were addressed at CHI’s recent conference entitled “Leaders in Biobanking: Maximizing Your Investment in Biospecimens”.
Building quality biological collections begins with the incorporation of best practices, guidelines that provide high-level strategies for creating, managing, and even closing a biorepository. Two sets of best-practice guidelines are currently published—one by the International Society of Biological and Environmental Repositories (ISBER) and the other by the National Cancer Institute (NCI).
“It can be challenging to take those guidelines and effectively and efficiently apply them to everyday practices,” advised Katheryn Shea, vp of Precision Bioservices. According to Shea, one must begin with basic design considerations: “The scope needs to be defined, including the short- and long-term objectives of the biorepository collection. Everything from the facility infrastructure to the legal and ethical policies, standard operating procedures, equipment platforms, and information technology requirements must be well thought out.”
Overall, the objective of every biorepository is to collect and house high-quality biospecimens. Multiple groups, including the Biorepositories and Biospecimen Research Branch (BBRB) of the NCI, have reported that only a small percentage of collected samples are of usable quality.
Samples need to be collected consistently across centers; variability can introduce bias into testing. Depending on the research study, variability can be more impactful. For example, it matters whether biospecimens will be used to evaluate a stable or a labile biomarker. Differences in results need to be interpreted, and if data on sample collection and preservation is not available, that information cannot be considered in the results analysis.
To control preanalytical variability, collection kits are critical. Complete all-in-one collection kits, preferably produced under GMP conditions, facilitate the process for medical professionals and raise compliance with collection protocols.
“People come to us for our scientific and logistics expertise. Being able to help them design the collection, handling, and preservation conditions helps achieve first-time-right quality, leading to consistency in results and acceleration of studies,” continued Shea.
“As a community we need better control of how biospecimens are collected and annotated. New reporting requirements are needed for the preanalytical variables associated with the biospecimen and we are happy to see the development of new tools for this such as the Sample PREanalytic Code (SPREC) developed by ISBER.”
The College of American Pathologists (CAP) has worked closely with biorepository professionals, including key members of ISBER, to develop an accreditation program specifically for biorepositories to ensure standards are in place, and help raise the quality and consistency of collected biospecimens.
Researchers are beginning to understand the critical steps—and the range of acceptable values—in cryopreservation that result in minimal damage. As that wisdom accumulates, practices in terms of protocols and therefore the preservation process will improve.
In the case of freezing cells obtained from the body, there are five core steps that can adversely influence the viability of a cell: addition of cryopreservation agents, the cooling rate, storage conditions, the thawing rate, and removal of the solution the cells were stored in prior to downstream use.
“For cells, we understand the principles of each of the steps and how to minimize damage to get overall quality from pre-freeze to post-thaw. Improvement of biofluid specimens will take place in the near future using some of the same principles that are used for stabilizing proteins,” offered Allison Hubel, Ph.D., a professor of mechanical engineering and director of the Biopreservation Core Resource at the University of Minnesota.
“It is going to be difficult and time-consuming to improve the preservation process for organized tissues so that critical biomarkers are stabilized. But that will come down the road as well,” Dr. Hubel added. “We need to continue to do research so we can create an overall high-quality preservation protocol.”
“We cannot use a biobank as a low-temperature garbage can,” Dr. Hubel emphasized. “We want to dispel the cold black-box myth. There are scientific principles behind each of the steps of the freezing process that can be controlled to achieve overall improved quality.”
Several layers of cost go into the overall preservation process. A large cost is affiliated with biospecimens that may have no downstream use; long-term storage should only be a small fraction of the overall collection and be carefully planned. An example is biospecimens for longitudinal studies. Other costs such as collection, processing, and placement in a biorepository are being increasingly mitigated by automation.
The Biopreservation Core Resource provides education and training on cryopreservation by teaching principles that can be translated into practice to improve biospecimen quality. The center helps individuals break their cryopreservation process into segments and to validate each segment to identify problem areas.
Research findings need to be attributable to a biological phenomenon and not a technical issue. Approximately 2% of samples stored within biorepositories are incorrectly labeled.
In addition, although biospecimens are externally labeled and STR (short tandem repeat) analysis may be used to obtain a genetic ID of the sample, neither of these methods assesses quality. In contrast, a comprehensive genetic fingerprint of the sample shows it is unique and correctly identified.
The research-use-only MassARRAY system provided by Sequenom combines the benefits of simple, reproducible primer extension reaction chemistry (PCR) with MALDI-TOF mass spectrometry to quickly and cost-effectively characterize genotypes with high levels of accuracy.
“The system allows multiplexing of a large number of SNPs in a single well, which helps with the power of discrimination during sample identification. When you are comparing a tumor sample to a nontumor (normal) sample, invariably there is goingto be loss of heterozygosity, and that automatically drops the number of SNPs that can be analyzed by 50%,” explained Marisa Pearce, director of product management at Sequenom.
“Other methods, which may typically start with 10–15 SNPs, end up with 5–7. Then it becomes a lot harder to predict with certainty the identity of that sample,” commented Pearce. “We start with 44 and end with 22.” The Sequenom system can be used for a variety of DNA sample types including cell lines and FFPE tissue. It is designed to perform a number of off-the-shelf or customized research-use-only genetic assays.
The MALDI-TOF mass spectrometer analyzes the data and software compiles the 44 SNPs into a single sample report that is collected into a database, providing a reference point for all samples curated in and added to the database. The historical comparison determines sample relationships via global search comparisons within the database and accommodates a growing number of samples for medium-to-large-scale biobanks. The number of amplifiable copies of DNA can also be quantified to assess whether or not the sample can be used for other genotyping or sequencing assays.
“Capital equipment represents the largest cost. It is offset by the low cost of performing each panel, which is about $5/sample. Sample identification is evolving. In the future, more rigor to ensure the identification of the samples will help guide research better and avoid sample mix-up. Better quality of sample will help in figuring out unexpected results,” concluded Pearce.
Sample Screening and Data Analysis
Biobanks not only need to assess sample quality, they also need to meaningfully combine biological information with phenotypic information. By meeting both of these challenges, the biobank adds value to the sample collection for discoveries of new disease associations and biomarkers. As sample sizes grow, biobanks are developing more specialized and centralized infrastructures to aid in sample management and data analysis.
Genetic-based information can support accurate sample stratification, fingerprinting, and sample screening for known genetic variants that are relevant for specific disease areas. Tools include sophisticated databases or laboratory information management systems (LIMS) for sample tracking and capturing phenotypic data.
Still, many biobanks face challenges. These include genetic analysis, fingerprinting analysis, throughput, workflow automation, sample tracking, database integration, data analysis, and bioinformatics and database implementation for sample stratification from genotypic results.
According to Alem Taye, a market development specialist at Illumina, the company offers high-throughput genetic screening platforms ranging from SNP genotyping to targeted, exome, and whole-genome sequencing of biospecimens, in addition to a modular automation solution for various throughput needs and downstream data processing.
These platforms enable biobanks to fingerprint samples to determine specific traits, ethnicities, gender, as well as typing of disease markers like copy number variations and known disease-associated variants to produce a database for sample stratification and characterization. This genetic-based information can add value to sample collections.
Illumina’s range of assays can lower processing costs while increasing throughput up to 2,000 samples per week. The platforms leverage the new 24-sample BeadChip, which can support up to 750,000 custom and predesigned markers per sample. These capabilities, along with sequencing, can enable biobanks to offer genomic services currently only found in large genome centers.
Illumina’s array LIMS is designed to provide biobanks an integrated system that is automated to achieve increased efficiency and accuracy and decreased variability. The system can track samples, reagents, and quality control data through every step of the workflow.
U.S. Veterans to Be Genotyped on Affy Platform
The U.S. Department of Veterans Affairs recently awarded a five-year contract for genotyping in support of its Million Veteran Program to the BioProcessing Solutions Alliance, a partnership between a research center at Rutgers, The State University of New Jersey, and BioStorage Technologies, which provides sample management solutions for the bioscience industry.
RUCDR Infinite Biologics (Rutgers) and BioStorage will use Affymetrix’ Axiom genotyping solution to build a custom microarray tailored to specific genetic and disease-related targets identified by VA scientists and physicians.
According to Marijo Gallina, strategic marketing manager for genotyping at Affymetrix, the Axiom is the company’s newest genotyping platform and “increases marker efficiency, delivering more genomic coverage per variant on the array.” As opposed to bead technology, “Axiom arrays are manufactured using a process that is completely reproducible and doesn’t suffer from marker drop outs,” added Gallina.