Three steps to fix the antibody market

Buyer beware may be a common adage, but purchasing antibodies has become a gamble, says Dr Jason Li. Here’s how he believes the marketplace can shift the odds back to the researcher’s favour

Suppose you are shopping for a book online, but the website doesn’t show the author. You email the company and ask how you’re supposed to ensure you’ve found the right book. In response, the company tells you that most customers buy half a dozen or so different books with the same title and decide for themselves which one they’ll read. Be warned, they caution, many of the books have multiple listings with different covers but, in fact, have the same contents.

This marketplace would certainly collapse. However, the antibody market has operated just like our fictional bookseller for the past twenty years. If we want to accelerate scientific progress and attack the reproducibility crisis, we must first fix how we buy and sell antibodies.

According to a 2015 survey by The Scientist, over 75 percent of researchers are replacing failed antibodies on a routine basis. This inefficiency delays research and wastes precious grant money. Unlike the analogy of buying a book, faulty or duplicate antibodies might not be immediately evident. Cross-reactive antibodies can bind to the wrong target but produce data just the same, leading to misinformed conclusions. Variable antibodies from inconsistent manufacturing can generate irreproducible data. When scientists don’t have the right information to buy the proper antibody, science suffers.

In our current system, distributors do provide a great service to the scientific community by providing antibodies to labs in a tremendous variety of fields throughout the world with tens of thousands of targets in their catalogues. The emergence of such companies was a salve to the painstaking research in antibody production that preceded it. Following the lead of Georges Köhler and César Milstein’s initial success in growing monoclonal antibodies in 1975, most researchers initially grew antibodies for their research in their own labs. Naturally, researchers began to communicate and exchange antibodies with each other when they were studying the same protein. Quality was high both because a scientist’s research depended on it but also because his or her reputation rose or fell with the antibodies they were sharing. In time, third parties began collecting extra antibodies from labs and splitting the profits of distributing them to scientists who needed them. By the 1990s, private manufacturers emerged and the age of commercial antibodies was in full swing.

While this was a boon for researchers who needed easy access to antibodies, this resulted in declining quality in some cases. No longer was a scientist’s reputation on the line if he or she produced or distributed a faulty antibody. Distributors also shielded their suppliers from the consequences of putting lousy reagents on the market by relabeling antibodies as their own, obscuring the origins of their products.

As a result, alongside the many high-quality antibodies on the market, there are hidden ones that disrupt and delay science. Finding the high-quality antibodies requires considerable effort, and researchers commonly rely on their own validation studies after purchasing several antibodies for the same protein target.

At times, faulty antibodies have been exposed only when a paper was retracted and an earnest scientist’s reputation ruined. Three fundamental changes must occur in the antibody market to fix these problems.

1. Identify high-quality antibodies Many good antibodies fill the market because scientific progress continues across vastly different fields. However, as a scientific community, we must separate the good antibodies from the bad to protect ourselves from cross-reactive, variable antibodies that delay progress, waste money, and frustrate researchers.

One way to accomplish this would be to develop a universal serial number system that would trace all research antibodies back to the original manufacturer and lot number. This would allow manufacturers to reach out to consumers should an issue arise (like a product safety recall in other industries) regardless of which distributor researchers bought the reagent from. Doing so would also increase reproducibility in peer-reviewed publications because we would recognise antibodies despite their many labels. This information would reduce the risk of switching to a different antibody in the course of research.

2. Improve manufacturing validation methods The gold standard in antibody validation is knockdown, knock-out or mass spectrometry, depending on the application. The industry is slowly embracing the importance of these methods. Proteintech began knock-down validation in 2014 and Abcam recently made an announcement announcing a knock-out program of their own.

• Knockdown. siRNA disrupts a cell’s ability to produce a particular protein by preventing translation of messenger RNAs. If the antibody binds correctly, signal should decrease when a target is knocked down on a Western Blot. • Knock-out. CRISPR and other gene editing methods can eliminate a gene from DNA, preventing the associated protein from ever being produced. By default, an antibody binding to any protein in this environment is binding to the wrong protein.  • Mass spectrometry. This method is useful if the protein can be immunoprecipated. Large and complex proteins bound with antibodies potentially yield unique spectra that easily differentiate them from any other bound molecule.

3. Reduce production variability The biological processes that produce antibodies are complex, and many characteristics of the molecules depend entirely on the animal producing them. However, laboratories can ensure significant lot-to-lot consistency when they control the experimental conditions. Consistently using the same species, the same sex and the same basic environment (such as temperature and time of day) goes a long way to reproducing identically performing antibodies. Statistically, any two animals expressing the same MHC class II molecule on the surface of their immune cells have a very high probability of creating the same antibodies when presented with the same antigen. Manufacturers can improve the quality of their antibodies by adopting more rigorous reproducibility standards.

While the above steps would improve the overall climate in the antibody market, we must ultimately reconcile that no consumer is exempt from due diligence in making informed purchases.

Finding the correct antibody in many ways is its own scientific research. Case in point, some best practices include: • Review the literature. Both the quantity and quality of published papers using a particular reagent count when identifying antibodies trusted in the scientific community. This step can narrow down the hundreds of thousands of antibodies available to a manageable number to investigate further, although it can prove difficult if you are studying a new protein or one that is not well researched. • Check validation data. Knockdown, knock-out or mass spectrometry may be the gold standard validation methods, but they’re still relatively uncommon. Alternatively, Western Blot tests on primary tissues or unmodified cell lines (not only on cells over-expressing the target antigen) can provide useful data when shown that the tests produce reproducible data. Make sure there is enough data to show that it’s reproducible. • Compare applications. Check whether an antibody has been validated for your application. No matter how well it binds on the molecular level, your data are what furthers your research, and if your protocol relies on Western Blot, an antibody that does not bind correctly in that application can derail your research just as much as a poor quality one.

The current crisis of the antibody market prevents scientists from forwarding their research with high quality antibodies. Only when we improve transparency in the marketplace can we hope to resolve other problems such as irreproducible science. You would not buy a book without confirming the author. Why would you buy antibodies without knowing their manufacturer?

The author: Dr Jason Li is CEO of Proteintech Group.