Biobanking Embraces Specialization and Interdependence

In the biobanking community, relationships deepen among researchers, specimen providers, technology developers, and standards bodies


Although the word “biobanking” emerged in the 1990s, it has yet to acquire a set meaning. For some, the word encompasses the long-term storage functions fulfilled by biorepositories; for others, it refers to the administrative work and scientific projects that biorepositories support. Perhaps the word is in flux because biobanking itself is in flux.

Initially, biobanking was a DIY pursuit. It was something that small, individual laboratories would do using ordinary freezers and handwritten notes. More recently, biobanking began projecting a “big science” manner. For example, ambitious biobank projects such as the UK Biobank and the All of Us Research Program were launched with the assistance of government agencies. (Supporters of the UK Biobank include the UK Department of Health, the Medical Research Council, and the Scottish Executive. The All of Us Research Program is led by the National Institutes of Health.)

Although biobanking is most readily imagined as a cottage industry or as a massive governmental initiative, it is, for the most part, something in between. It is a maturing industry. And like any such industry, it has a division of labor. As this article illustrates, biobanking’s division of labor encompasses the curation of data registries to support precision medicine; the development of biomarkers to simplify assessments of biospecimen quality; the building of online communities to help researchers connect with biospecimen providers; the development of technology to further automation; and the promulgation of standards for validation and accreditation.

Biobank project informs the selection of more precise therapies

“Our mission is to bring precision medicine to the field of obesity,” says Mark Bagnall, CEO of Phenomix Sciences. The individual variability in response to obesity treatment has been known for decades and represents a major therapeutic challenge, but it is poorly understood.

Phenomix’s technology is based on research from its physician founders, Andres Acosta, MD, PhD, and Michael Camilleri, MD, of the Mayo Clinic. They studied almost 1,000 patients in multiple clinical studies and revealed that obesity comprises multiple conditions. “At least four of those conditions account for about 90% of obesity patients,” Bagnall notes. “We built Phenomix based on that insight.”

After Acosta and Camilleri founded Phenomix 2017, the company leveraged the 1,000-patient study to establish the Phenomix Sciences Obesity Platform. “This database contains about 20 billion discrete data points,” Bagnall points out. Based on these data, Phenomix scientists are developing blood tests to identify the subtype, or phenotype, of obesity in an individual and guide the choice for the best therapeutic. Previous work revealed that knowing a person’s obesity phenotype can double the amount of weight loss if the treatment is tailored to that specific phenotype.

MyPhenome diagram
Phenomix Sciences is developing MyPhenome, a blood test designed to help doctors prescribe more precise obesity treatments. It uses artificial intelligence to process multiomics information and recognize obesity phenotypes such as “hungry brain,” “hungry gut,” “emotional hunger,” and “slow burn.” To advance development, Phenomix launched a biobanking registry and outcomes study

“This is the foundation for precision medicine for obesity,” Bagnall declares. “This is the rationale for our biobanking study.”

Scientists at Phenomix are supplementing this dataset with outcomes from 2,000 participants that will undergo obesity treatment over five years and will provide samples at baseline and at regular intervals to build a longitudinal database. Biological data that originate from DNA, hormones, metabolites, and stool samples that are informative about the microbiome, together with behavioral data such as exercise and eating behaviors, are integrated into a specialty app that participants can access and from where they will receive individual feedback.

“This will build our understanding of the disease,” Bagnall relates. “We will look through the lens of obesity phenotyping in a way that has not been done before.” In addition to the initial study in collaboration with the Mayo Clinic, Phenomix is in discussion with other sites to conduct further biobanking around the United States. The first test developed by Phenomix will identify people who respond to glucagon-like peptide-1 receptor antagonists, such as semaglutide.

“If someone is overweight or has obesity, their primary care physician could recommend a test to determine the phenotype,” Bagnall says. “The test would indicate which intervention would be most likely to produce the desired response.”

Approximately 60% of Americans currently are overweight or obese. “The associated morbidity is stunning, and the obesity epidemic consumes a gigantic portion of healthcare costs,” Bagnall observes. “We will never get healthcare costs under control unless we control obesity.”

Biomarker delivers freeze-thaw insights

“My interest in biobanking is to help provide quality control and assessment tools,” says Chad Borges, PhD, a researcher at Arizona State University with joint appointments in the Biodesign Institute and the School of Molecular Sciences. “We need better tools to assess the integrity of samples used in research, especially if they have been banked for a while and/or have changed owners.”

Tools of interest to Borges include assays that could estimate how much time, in aggregate, biological samples have spent above their storage temperatures. Such assays are necessary because freeze-thaw histories are seldom well documented.

To help researchers quantify the cumulative exposure of plasma and serum to thawed conditions, Borges and colleagues developed the ΔS-Cys-Albumin test. ΔS-Cys-Albumin indicates the relative abundance of an albumin proteoform called S-cysteinylated albumin. That is, ΔS-Cys-Albumin shows how much albumin in a sample has been oxidized via S-cysteinylation, a process that accompanies the thawed state.

In blind challenge studies, Borges and colleagues showed that using the ΔS-Cys-Albumin biomarker, they can predict with a 98% accuracy whether a sample is pristine or has been exposed to thawed conditions. The researchers also demonstrated that the instability of biomolecules with known degradation profiles, such as matrix metalloproteinase-9 and a2-macroglobulin, can be linked to changes in ΔS-Cys-Albumin, sample by sample.

Unfortunately, stability information is unavailable for many protein molecules or analytes of clinical interest. To investigate how other proteins besides albumin change over time as plasma and serum samples are exposed to the thawed state, Borges and colleagues used a liquid chromatography–mass spectrometry approach to simultaneously measure the stability of S-cysteinylated albumin and 21 clinically relevant proteins at room temperature, in refrigerated conditions, and in freezing conditions for different time spans.

The researchers found that there was a linear inverse relationship between the percentage of proteins destabilized and ΔS-Cys-Albumin, irrespective of the thawing temperature or the time. “Multiple instabilities existed at every single temperature,” Borges remarked. “And several of them occurred at fairly early time points.”

According to the researchers, their studies showed that ΔS-Cys-Albumin measurements can be used to approximate the percentage of other proteins that have been destabilized, even if information about an archival sample’s freeze-thaw history is unavailable. “Over the next five years,” Borges adds, “I would like to see the biobanking field at a place where every biobank has the tools that it needs to assess the integrity of its samples.”

Online platform links researchers and specimen contributors

“Instead of building another commercial biobank, we built a platform to connect two networks of players,” says Christopher Ianelli, MD, PhD, founder and CEO of iSpecimen. On one side of the platform, there is a network of providers across the healthcare landscape, such as hospitals and physician practice groups, which have access to patients and clinical samples. On the other side of the platform, there is a network of life sciences researchers.

The platform is called the iSpecimen Marketplace. “It allows researchers to search the exact patient samples and datasets that they need by asking healthcare providers to send the data from their systems,” Ianelli explains. In other words, it helps establish a connection between the life sciences researchers that need a specimen and the providers that have that specimen.

iSpecimen says that procedures for biospecimen procurement comply with ethical guidelines. Moreover, data exchanges are secure. “All the data we receive is stripped of patient identifiers,” Ianelli asserts. In addition, the data sets are harmonized and cleaned so that everybody can be mapped to a standard uniform database.

Over 200 healthcare providers and 400 customer relationships have moved through the iSpecimen Marketplace so far. “We have satisfied well over 2,000 research projects by getting the right specimens to research programs,” Ianelli reports.

iSpecimen does not charge hospitals or healthcare providers, and its revenue originates from helping commercial researchers. This helps accelerate research on the life sciences side, where many subdisciplines have had to deal with the same bottleneck, namely, accessing biospecimens from a very particular group of patients. “We open up that bottleneck by giving researchers access to our prebuilt network,” Ianelli remarks.

A unique challenge that iSpecimen scientists are currently addressing stems from the digitization of healthcare. “Digitization has existed in finance for years,” Ianelli notes, “but it is relatively new for hospitals and healthcare providers.” Digitization depends on interoperability, which is the ability of two or more hospitals to exchange information. “There have been many problems with interoperability, and that slowed our progress because we had to invest to clean our datasets,” Ianelli admits, “but a lot of work has been made more recently to improve it.”

Barcode system defies frost, aids automation

Ziath's DataPaq Mirage
Ziath develops instrumentation control and information management products to simplify the automation of sample-tracking applications. For example, the company offers the DataPaq Mirage, a rack scanner that incorporates a camera for reading 2D barcodes. (In this image, the DataPaq Mirage is an integral part of an automated workflow.) All of Ziath’s Datapaq camera-based scanners may be run with the company’s DP5 2D barcode scanner software. Another Ziath software, Samples, is designed to organize sample storage inventories.

“Biobanks need to appreciate the value of automation over its cost, because this would provide protection during sample handling and storage,” says Neil Benn, co-founder and managing director of Ziath. The company is working to bring sample tracking and management solutions to the market. It has four main product groups: 2D barcoded tubes, devices for handling tubes, 2D barcode scanners, and software.

Ziath appreciates that many biobanks are struggling to upgrade their practices, particularly where automation is taking hold. “We regularly get contacted by established biobanks that are still using Excel to track their samples and putting sticky labels on their tubes,” Benn relates. Such rough-and-ready practices often lead to misidentified, lost, or corrupted samples and invalid data. “In our experience,” Benn says, “the United States is currently lagging well behind Europe and Asia in this respect.”

Ziath's CryzoTraq tubes
Ziath provides cryogenic vials that can store biological specimens at temperatures as low as −196°C. They are designed to resist leaks, they are sized to contain 2- or 5-mL volumes, and they have internally or externally threaded caps. Ziath calls the vials CryzoTraq tubes. Each tube has a 2D datamatrix barcode inserted in its base.

Ziath considers itself a global leader in the use of 2D barcoded sample tubes to maintain large sample libraries. “The only safe, reliable, and efficient way to track large numbers of stored samples is by using tubes that have had a 2D data matrix barcode laser-etched onto them,” Benn insists.

Although 2D datamatrix barcodes provide an excellent solution for tracking, once they are covered in ice, they may be difficult to read. The latest camera-based tube rack readers developed by Ziath use machine learning and artificial intelligence to help users find and decode barcodes more quickly, even when they are iced over. “We expect new digital technologies will enter this field,” Benn declares, “and they will help scientists find the correct rack and the right sample tube more quickly and efficiently.”

Biobanking accreditation and validation

“I would like to see more methods that organizations could use to process or store their materials,” says Cory Arant, program manager at the American Association for Laboratory Accreditation (A2LA). He believes there should be “some format that can be shared” so that organizations familiar with validation methods could be a resource for organizations are not.

The A2LA, one of the largest accreditation bodies in the world, uses the ISO 20387 standards to assess accreditation for biobanking. When representing the A2LA at the 2022 meeting of the International Society for Biological and Environmental Repositories, Arant talked about best practices and standards in biobanking and about nonconformities and corrective actions.

In the testing field, many methods are available to the public, and various organizations are working toward robust and standardized approaches. Nonetheless, accredited organizations generally use their own internal procedures. Typically, these are validated procedures. “Even though they may not be well recognized, generally they are probably very similar,” Arant observes.

If accreditation and validation are lacking, researchers are more likely to waste precious resources on generating irreproducible results. As Arant notes, researchers may fail to keep (or discard) information in accordance with widely used standards. He adds that researchers “can make preanalytical errors when they use the material that they receive.”

Arant is also interested in addressing the need to distribute materials across international borders. “If someone tries to get something imported, it may stay in customs for an extended amount of time,” Arant notes. “The material may be compromised.”

Previous articleBringing Big SynBio to the Big City
Next articleThe State of Phage Therapy Research