A Call for Antibody Quality Control

February 17, 2015
Randi Hernandez

Randi Hernandez was science editor at BioPharm International from September 2014 to May 2017.

Antibodies in research should be standardized and categorized using a barcode-like classification system, according to research published in Nature.

To ensure reproducibility of experiments with biologics, scientists, researchers, and pharmaceutical companies should define antibodies by the sequences encoded within their subunits, according to researchers from Los Alamos National Laboratory in an article published in the journal Nature. Their research asserts that protein-binding reagents used outside of most clinical trials are “wildly variable,” and researchers should be able to have access to identical reagents.

Unlike genes, oligos, plasmids, and recombinant proteins, “Antibodies are the only widely used reagents in biological research that are not defined at the sequence level,” researcher Andrew Bradbury said in a release accompanying the research. Bradbury insists that the current “scattershot harvesting” of antibodies is a wasteful process producing antibodies with less than ideal specificities in varying concentrations; only 0.5-5% of the antibodies in a polyclonal reagent bind to their intended target. The ultimate goal, he says, is for everyone to use the same binding reagent to remove the variability in the quality of antibodies across manufacturers.

Andrew Bradbury and Andreas Plückthun collected more than 100 signatures from researchers in support of their antibody project, and in the Nature piece, call upon agencies such as the National Institutes for Health and Affinomics by the European Union to help fund the collaborative enterprise, which was likened to the Human Genome Project. “Publishers and funding agencies should mandate that in, say, five to ten years’ time … all binding reagents in published papers are recombinant and defined at the sequence level,” the authors wrote.

“In many cases, the sequences that determine an antibody’s specificity have either not been defined or have not been made publicly available,” said Daniel Christ, associate professor of Sydney’s Garvan Institute of Medical Research and a supporter of the proposal. “I think it would be desirable if the sequences of antibodies were deposited online, and made accessible to the wider community.” Another signatory of the article, C. Glenn Begley, was the scientist who showed that results from only six of 53 landmark preclinical studies could be reproduced.

Bradbury estimates the development of a database would require an investment of at least $1 billion, although he says it would ultimately save researchers in the long run on money poured into antibody characterization efforts when antibodies do not bind as expected.

Intellectual property issues
This change in the way antibodies are supplied, however, may not be embraced by entities that make their money off selling their reagents to laboratories. Indeed, the authors admit that the “absence of these reagents in the research-antibody marketplace stems mostly from economic considerations, rather than technical challenges,” so it may not necessarily be in these companies’ best interest to alter their production methods. In addition, if the method of making a reagent (be it upstream or downstream processing) is considered a trade secret, life-science manufacturers may not want to share information about their antibody sequences, Stacie L. Ropka, PhD, JD of Axinn, Veltrop, & Harkrider LLP told BioPharm International.

Specifically, at least one of the suggestions made by the researchers-that all antibodies made with hybridoma technologies be switched to recombinant cell technologies-may be blocked until at least December 2018 by a couple of Cabilly patents, says Ropka. These patents, which are owned by Genentech and City of Hope, protect technology required for the artificial synthesis of antibody molecules. “In short, there are patents that claim specific antibodies; there are patents that claim methods of making antibodies (some, like Cabilly, are fairly broad and not limited to a specific antibody); and there are patents directed to how antibodies are isolated and purified,” says Ropka. “If the research community needs to employ such methods, they run the risk of infringing these types of patents. Changing methods by which antibodies are produced may require an understanding of the patent landscape and a review of patent claims to determine if there are claims that have the potential to be asserted by a patentee against the manufacturer, user, or seller of antibodies (or other proteins mentioned in the article) that are patent protected or [if there are] antibodies/other proteins made with methods that are patent protected.”