We spill a good bit of virtual ink in our blogs and eBooks discussing biospecimen integrity, but the emerging term that is being used more and more often, is “fit–for-purpose.” Identifying a sample as fit-for-purpose implies molecular integrity, but goes beyond this to encompass all the variables that make a biological sample useful for research. Sample integrity focuses on the quality of individual samples, while the term fit-for-purpose can be interpreted as addressing entire collections.
We are familiar with the basic issue: too much research is conducted using biological samples with unknown pre-analytical variability and quality, and true to the principle of “garbage in, garbage out,” the result is research data that cannot be replicated. Given the estimated 95 percent failure rate for new drug development, we cannot ignore this issue. But what can we do, and where do we start?
The attributes that make a biospecimen collection fit for its intended purpose depend on the questions that the specific research hopes to answer. This is the case whether the specimen is a fluid or a tissue, although these two sample types must be processed differently to stop their biologic activity and stabilize molecular composition. For instance, with respect to cohort studies and searches for biomarkers, fit–for-purpose ideally refers to a biospecimen set that includes pre-disease, pre-treatment, disease/normal tissue, post-treatment, and follow-up samples—blood, saliva, urine, tissue, etc .
Overall, we can create three categories in the fit-for-purpose bucket for discussion:
- Donor consent
- Data quality and integrity, and
- Specimen quality and integrity.
The issue of proper biospecimen consent requirements does not vary with the nature of the research. The end user of the sample must know that the donor gave proper consent and that the consent met the criteria of the associated institutional review board.
Determining and appending the correct data—the clinical and other patient data that will yield the needed insight into the biology of the disorder under investigation—delves deep into IT issues as well as data collection. How do scientists doing biospecimen-based research overcome the barriers and challenges of access to and availability of clinical data? Was the diagnosis correct? Is the specimen representative of the disorder? With the emerging findings of microbiome involvement how does this play out with respect to root cause analysis of disorders?
Our third subdivision may be the biggest current unknown, and that is ensuring the specimen you are working with meets the requirements of the analytic technology. The more sensitive and powerful the technology, the greater the need for samples that were handled and preserved correctly, or the analytic platform is merely measuring noise.
However, there are additional issues that span these three overly neat categories. For instance, with regard to matching the sample quality to the technology, researchers and scientists in laboratories often know little about clinical medicine, how tissue is obtained during surgery and handled after collection, the effects of the bore of a needle, and other pre-analytical variables, and thus cannot account for the impact of these variables on the biospecimens they use. Building a fit-for-purpose specimen collection calls for a multi-disciplinary perspective.
Another issue that is mostly overlooked is that most patients (including 80 percent of cancer patients) are treated at the community level, and only a small proportion at academic or institutional research centers. Few of these research collections are representative of the population with regard to ethnic origin.
Another overlooked issue is standardization and uniform quality between clinical sites and also between laboratories. Fit-for-purpose specimens require standardized collection and handling practices and processes across sites, and high quality laboratory processing. [An excellent step in this direction is available by participating in the Integrated BioBank of Luxembourg (IBBL) Biospecimen Proficiency Testing Program, endorsed by the International Society of Biological and Environmental Repositories (ISBER).]
It has been about five years since Begley and Ellis’ commentary on non-reproducible results was published in Nature1. They state, and I quote, “Perhaps the most crucial element for change is to acknowledge that the bar for reproducibility in performing and presenting preclinical studies must be raised.”
Is your inventory automation-friendly? Chances are if you didn’t launch your biobank with automation in mind, or if a large percentage of your inventory is legacy collections, you’ve got some serious challenges ahead of you. Download our eBook Automating Your Sample Collection for Biobanking: 10 Things to Consider to learn more!
Begley, C.G. & Ellis, L.M. (2012). Raise standards for preclinical cancer research. Nature, vol. 483, pp. 531–532. 483531a.pdf