As Director of the Office of Biorepositories and Biospecimen Research (OBBR) at the National Cancer Institute, Carolyn Compton, MD, PhD, oversees NCI's effort to improve the quality and availability of biospecimens. Dr. Compton has stressed the need for high-quality biospecimens in the era of personalized medicine (OT, 10/10/10 issue), and here she talks about how the community can achieve those high standards.
Why is the issue of biospecimen quality coming to the fore now?
Carolyn Compton, MD, PhD: I think there are two main reasons. One is technology development. We now have tools for molecular analysis of biospecimens that are orders of magnitude more powerful than ones we've had in the past. And technology development is so very rapid that the technologies get better and cheaper with every passing day, which means they become not only research tools with great transformative power, but also part of the clinical armamentarium to interrogate patient samples for data that is relevant to patient management in real time.
When the technology was insensitive or nonspecific by comparison, we could tolerate a lot more variation in quality in human samples and still get an answer that was usable. But now the ability to interrogate thousands of molecules in a single class all at once, to be able to multiplex tests to look at multiple classes of different molecules all at once, all of these things start to refocus attention on the quality of the analyte, along with the quality of the analysis.
The simple way to think about this is that you can have the perfect analytical tool and still get the wrong answer if you compromise the specimen you are analyzing.
Number two is the inflection point at which we sit right now in transitioning from traditional medicine to personalized medicine. In personalized medicine, the biospecimen from the patient is the receptacle of all of the molecular information that, when mined, will determine downstream patient management. That means the biospecimen takes on a new significance.
In the past, you've mentioned the importance of understanding that biospecimens are living until fixed. Can you elaborate on that, and how your team aims to cope with it?
Cells and tissues don't immediately die because you remove them from the body. If we want to think about this in very simplistic terms, they are basically holding their breath. They have been disconnected from their blood supply. They have been removed from the warm body temperature. But they are still alive.
In fact, the disconnection from their blood supply and the change in their environment from body temperature to room temperature—and those are only two variables I might mention—represent extreme biological stress on the tissue.
Eventually the cells will die, but that process is one that takes place over a course of time. And during the time frames that we handle specimens in the clinical setting, the biospecimens are very much alive and capable of reacting to environmental stresses and do react to them.
These reactions to environmental stresses that we, the physicians, cause in removing them from the body or handling them in the pathology suite, are not a reflection of the biology of the disease the patient has. They are completely artifactual, and we have to be able to separate that artifact from their biologic behavior and activity when they were back in the patient happily existing as part of the patient's anatomy.
We have to be able to separate those two things and not confuse them—not misinterpret artifact as medical reality.
How do you do that when there are so many variables to consider?
We have to go about this in a scientific fashion. We have to study the biospecimen as a focus of research to understand how it, the biospecimen, reacts to these different variables. We, at the NCI through OBBR, have sponsored the very first research program of this type ever funded through the National Institutes of Health—the Biospecimen Research Network, which is bringing investigators together to focus on these issues, one biospecimen type at a time, one variable at a time, one class of biomolecules at a time.
It is complex but it is not infinitely complex, and it can be understood. And with the right experimental designs, we can map out the so-called lifecycle of a biospecimen and understand the biologic stresses and their impact on different types of biospecimens as reflected in the different classes of biomolecules: DNA, RNA, protein, other subclasses of biomolecules, lipids, sugars, carbohydrates.
Is the goal to develop standard operating procedures for handling biospecimens?
The goal is to take the data coming out of a research network like this with a new understanding about what is artifact and what is not, and create—actually first of class—evidence-based, data-driven standard operating procedures for collecting specimens of different types to make certain that the analysis gives you reliable data.
Until then, how should surgeons, pathologists, researchers maximize the quality of the samples they are collecting now?
We can't let the lack of knowledge in biospecimen biology that exists right now stop us from going forward. There are existing best practices that represent the state of the science—albeit the state of the science being as weak as it is, there is some data out there.
NCI/OBBR is putting together a web-searchable tool to bring the existing published data on biospecimen science to the community. So you're be able to go onto a website—the Biospecimen Research Database [https://brd.nci.nih.gov/BRN/brnHome.seam]—and if there is existing objective scientific data on how a variable affects a biospecimen of a given type as far as a molecular analysis goes, you will be able to find that data and use that data for the basis of creating your own optimized approach.
If there is no data, the best possible advice is to create a standardized approach that is adhered to, and is documented, so you at least know what you did with the specimen—Right now there is no requirement in the medical world, and certainly none in the scientific world, to even record these variables to know what happened to the specimen on its way to the molecular analysis; it could have sat at room temperature for 20 minutes or two days. Then you have a record of the history of this biospecimen and it gives you much greater ability to judge its quality at the end of the day—instead of it being a complete mystery.
If we do that, and if we continually improve the scientific basis for the standard operating procedures, we can compare back to an earlier standard operating procedure, and document the increase in quality that we are able to achieve. But if we don't have any record, then we won't know what we are comparing to. We have to have two solid data points to make the comparison, so we should start with a standard and stick to it.
Anything you would like to add?
This is really the key to personalized medicine. And if we don't get this right, we won't have any personalized medicine. This is critical.