Contributed Commentary by Cathie Miller, PhD, BioIVT

March 18, 2022 | Following a standard protocol when collecting biospecimens and conducting a quality review to confirm that you are working on appropriate samples are critically important processes for downstream analysis in medical research. If biospecimens are not collected and processed in the same way, subject to the same quality controls, diagnosis verification, and clinical data collection, researchers will not be able to determine whether data outliers are due to disease progression or process inconsistencies.

If biospecimen collection protocols and quality control processes are followed, any samples, regardless of their age, can be run through an assay and staining system, or other downstream application, and they will act in a similar manner. Thus, helping researchers understand if variables that occur arise from sample processing (occurring in most samples) or need to be investigated. Consistency is particularly important when studying rare cohorts.

Impetus to Develop Standards

I was a researcher at the Hermelin Brain Tumor Center in the Henry Ford Health System in Detroit when it began working with the NIH Human Genome Project. I later joined the National Disease Research Interchange, which was part of the Genotype-Tissue Expression (GTEx) project. Both working groups were funded by the federal government.

GTex was established in response to the Human Genome Project and some of the issues that arose. GTex was specifically designed to create a cohort of normal tissue signatures (both genomic and proteomic) to be used as normal controls for other projects. One of the biggest challenges those programs faced, and chose to address, was sample variability. The original samples had been collected over many years and there were no set standards regarding how the tissue or data were collected or processed. Many samples proved useless because either they were not processed correctly or there were insufficient data associated with them to be used in downstream analysis. Researchers needed to know whether the sample came from someone with a disease, and if so, what disease? Also, what was the donor’s gender, race, age, and what treatment had they received?

On the tissue side, if the sample sat on a bench for three hours before it was snap-frozen and put in the freezer, it had degraded. If it had been fixed during the formalin-fixed paraffin-embedded (FFPE) process for too long, it could not be stained.

There was an urgent need to develop standards. In response, the National Cancer Institute (NCI) published its Best Practices for Biospecimen Resources, and the International Society for Biological and Environmental Repositories (ISBER) introduced its Best Practices: Recommendations for Repositories. Both organizations issued protocols for the industry to follow. Reputable biobanks have been using them as a reference, and default standards, ever since.

Expert guidance is also provided by the Best Practices, Biobanking and Biomolecular Resources Research Infrastructure European Research Infrastructure Consortium (BBMRI ERIC), and the European, Middle Eastern and African Society for Biopreservation and Biobanking (ESBB).

Tissue Biospecimen Protocol

When a standard protocol is followed, the following questions will be answered about each tissue sample: When was the sample removed from the donor? How long was the tissue clamped off from the blood supply? How long did the tissue sit in the formalin fixation solution? How long was the paraffin embedding process? If it is a frozen sample, how quickly was it frozen? These are all crucial factors when studying biomarkers because they impact antibody staining and intensity.

The tissue specimen also needs to be an appropriate size (less than 2.5 x 2.5 x 0.3 cm). If it is too large, the fixative and paraffin will not diffuse through the tissue well enough. If the tissue block is not fixed correctly, and the paraffin is soft or brittle, it cannot be sectioned or stained properly.

The biospecimen must also include both diseased and normal tissue, so that a board-certified pathologist can confirm the diagnosis and origin.

Important QC Processes

Some analytes and assays are sensitive to freeze-thaw stability issues. For these, it is important to track how many times biofluids—serum, plasma, urine, or ascites—have been freeze-thawed because that process can affect measurement and activity of some of the analytes that researchers are looking for and start to degrade the specimen.

For biomarker development and analysis studies, researchers want serum, plasma, or urine specimens that have been aliquoted at the original collection time. If a larger volume is required, it is better to take out three aliquots than freeze-thaw and mix the biofluid. The former approach ensures the right representation of the chosen analytes in all the matrices being studied.

For cells, flow cytometry is used to both classify the cell population in a vial as well as determine cell viability. With primary cell isolations, which can be immune cells, dissociated tumor cells, or cells isolated and grown up from tissue, that is the only way to really characterize the cell population. For example, it will determine whether the peripheral blood mononuclear cell (PBMC) population is predominantly T cells (CD4, CD8 or CD19), B cells or monocytes. That can vary greatly depending upon the donor, the disease, and how far the disease has progressed.

Reducing Drug Development Failures

Starting with a well-defined, accurately represented cohort of biospecimens is extremely important at every stage of the drug development process. All samples need to go through a clinical data and pathology review to ensure, for example, that the entire cohort of donors had adenocarcinoma of the lung and that none of them had squamous cell carcinoma of the lung.

What gender are the donors? If they are all male, that is not representative of a cohort. If they are all one race, that is not representative of the disease. Having a well-defined cohort that has been validated, ensures that you are seeing results that matter.

It is also important to have a board-certified pathologist review every tissue block to ensure that each one contains a representative sample of the donor’s tumor and not just inflammation. If a block is mischaracterized, it will throw off the results.

Mutation data analysis—enabled by next-generation sequencing, proteomics, and RNA-seq analysis—is also being used to further segment cohorts. Instead of ordering a cohort of 300 lung cancer specimens, researchers can now request 50 adenocarcinomas of the lung with a G12C mutation in KRAS, or an EGFR 19-delta variant. This allows a drug or diagnostic to be created for a specific mutation profile.

Harnessing Artificial Intelligence (AI)

Another benefit of creating a consistent collection of biospecimens with validated diagnoses, associated clinical data, and stained images from the original tissue block is that an AI system can review those massive amounts of data and look for patterns. It can identify consistencies and similarities in the biospecimen histology and associated demographic, clinical and outcomes data, and make cohorts from those. This approach could reveal pertinent biomarkers that no one has looked for before. It holds enormous potential as an innovative approach to developing therapies for complex conditions such as cancer and Alzheimer’s disease.

Dr. Cathie Miller is Senior Director of Product Marketing at BioIVT. Dr. Miller has more than 20 years’ experience as a bench scientist, director, and global product manager, using biospecimens to advance cancer, immunology, and virology research. She received her PhD from the University of Louisville School of Medicine and completed her post-doctoral fellowship at the Wistar Institute and University of Pennsylvania. She can be reached at CMiller@BioIVT.com. BioIVT’s Asterand Human Tissue Repository has been following the NCI and ISBER protocols across its sites for more than 15 years.