Are We Insane? How Can We Expect Great Biomarkers and Therapies when There Is Poor Sampling?


Author: 
Theral Timpson

When the former deputy director of the NCI, Anna Barker, suggested we start this year’s Biomarker Development Series with a show on biospecimen quality, I admit: I wondered how I was going to make it interesting.  

Biosampling?  What is there to talk about?

Then I met Carolyn Compton, a former pathologist and now the Chief Medical and Science Officer for the Naitonal Biomarker Development Alliance. Carolyn has a provocative message:  sampling issues are becoming one of the core problems in biomedical research. 

Today, I’m happy to announce the upcoming series, Back to Basics: Improving Biospecimens.  This series of five interviews will give us a chance to explore how big of a problem we have with sampling issues and what can be done about it. Quintiles is one of the largest CROs in the world who deals with sampling issues on a massive scale.  They’ve agreed to underwrite the series to bring more awareness to a topic that is too often overlooked. 

And just why should we talk about sampling?

Let’s start with the issue of non-reproducible research.  According to a study by Amgen in 2012 (Nature 483, 531-533, 2012), only 11% of 53 seminal publications on drug targets or toxicity could be reproduced by their team.  Let's add to that a fact that ASU’s George Poste points out in a Nature article of 2011 (Poste G. Nature-469, 156-157, Jan 2011) that out of over 150,000 biomarkers that have been written about in scientific and medical literature, only about 100 have been commercialized.  Why the stunning lack of success?  

“Garbage in, garbage out,” Carolyn says, borrowing a phrase from the big data folks.

The samples are not procured correctly, they’re mishandled, and they degrade.  It's critical that the industry works to improve standards.

I pulled the studies mentioned above from a talk Carolyn gave on these issues at a recent skin cancer conference in Scotland.  It’s a terrific introduction to the topic and is embedded at the end of this post.  

We’re going to unroll the series in the following format:

Our first show will be an overview.  We talk with Jim Vaught, President-elect of the International Society for Biological and Environmental Repositories (ISBER).  In his interview, Jim gives a shocking example of where poor sampling is causing an issue.  He says there is such variability between various labs and their assays for doing the Her2 testing for breast cancer.  Jim says that about 20% of the Her2 testing generates a false positive and about 20% generates a false negative.  This means that too many women with breast cancer are treated with the companion therapy, Herceptin, that is not making any difference.  And too many are not treated with Herceptin when they could be.  Jim ties this back to lack of standards in handling the biospecimens.

In our second show we’ll be looking at just what is involved in procuring, handling, and storing a sample.

With the third show, we talk to the Medical Director of Quintiles, Diane Fahri, about the issues she sees in sampling for clinical trials.  Diane has made it a priority to ensure that there is “harmonization” in procuring and shipping samples.  

One country may have a holiday and their shipping lines close.  This can affect the samples, Diane says.   There’s a myriad of issues when it comes to working with samples from such diverse labs around the world.

Fortunately, a new science of sampling has been emerging.  There are new programs, particularly in Europe which are offering training, even a Ph D in handling biospecimens  Our fourth show will be devoted to understanding better this new science.

To finish up, we’ll again have Carolyn Compton to the program to talk about the work she’s doing to improve standards.  So far SOPs have varied largely from institution to institution.  Carolyn is hosting a special conference in early December with many of the key opinion leaders to come up with some basic standards which can then be adopted by the professional organizations, such as the College of American Pathologists or CAP.

Nonreproducible science has become one of the worst plagues of our industry.  Better standards in sample collection, handling, and storage will go a long ways toward improving research and clinical efforts to bring great diagnostics and therapies to patients.



New to Mendelspod?

We advance life science research, connecting people and ideas.
Register here to receive our newsletter.

or skip signup