This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American
An article in Nature in 2015 estimated that $350 million is wasted annually on biomedical research simply because poor quality materials lead to false results.
Not only is this expensive, but waste in the biomedical sector is dangerous, slowing the development of life-saving drugs and procedures. Scientists build off each other’s published work, and there is a narrow margin for error when compounds and procedures developed in one lab are used to create life-saving drugs in another. Key results must be repeatable.
According to a survey by Nature, more than 50 percent of researchers agree that science is facing a significant reproducibility crisis. A number of factors seem to be at play, from bias in what journals choose to publish to the manipulation of statistics to overstate a finding. But in the biomedical sector, an added layer of reproducibility problems has turned out to be purely economic.
On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Antibodies are one of the most frequent culprits. The current state of the market prevents scientists from identifying low-quality manufacturers. Increasing transparency in the antibody market can punish manufacturers that cut corners and promote high-quality products that lead to reproducible results.
Does antibody know what’s happening?
Antibodies are small proteins produced by your immune system to help fight off bacteria and viruses. They work by “tagging” specific invaders, attaching to them and signaling the target for the next strike by the immune system’s heavy hitters.
Their ability to tag specific kinds of cells is extremely valuable in research, too. When the Human Genome Project revealed our complete genetic map, a surge of research flowed into understanding the great variety of proteins our DNA is coded to produce, and this created a high demand for antibodies which could tag and isolate these specific proteins. Commercial manufacturers popped up to fill the need, as did commercial distributors.
Unfortunately, only 50 percent of the antibodies produced commercially actually do what they’re supposed to, meaning they tag the wrong molecules or tag the right ones inconsistently from batch to batch. Scientists who assume that what they bought is working correctly may publish skewed and irreproducible results.
For most consumer goods, this wouldn’t be a problem. Retailers like Amazon show the manufacturer’s brand to the customer, so low-quality products are easily discovered and avoided. In the antibody industry, however, distributors re-brand products. There’s no way to discover the original manufacturer, and scientists have no way of knowing what quality of antibodies they receive.
Without transparency, scientists can’t make the decision to avoid suspect manufacturers. Scientists are put in a no-win situation—they are forced to buy antibodies but have no way of knowing whether the products they get will actually work. The effects are serious and far reaching, as this problem leads to failed biomedical research, delayed treatments, and wasted hard-to-come-by research dollars.
Why should antibody care?
Even though antibodies are critical to research, producing them takes a substantial amount of time. Before the commercial antibody industry developed in the late 1990s, experiments could be set back 6 months just because of antibodies. The advancements in practice to enable the ability to buy antibodies of any quality off the shelf is a huge boon for scientists and science.
However, the long production time of antibodies also presents challenges to manufacturers. Antibodies can be made in thousands of varieties, and manufacturers have no way of knowing which antibodies will be in demand at any given moment. If they wait to find out before creating products, they quickly go out of business. As such, companies produce a wide range of antibodies, so that the profits of the few that end up in high demand pay for the development of the rest.
In the early days of the antibody market, manufacturers were responsible for production, marketing and sales. However, sky-high production costs made it difficult for manufacturers to invest resources into getting customers. The system was bad for scientists too, as they needed to run from manufacturer to manufacturer looking for the antibodies they wanted. The process was extremely messy—and then the distributors stepped in.
Distributors, as in most marketplaces, collect a range of goods from a range of manufacturers, allowing scientists to visit a single vendor and fulfill their antibody needs. Manufacturers have virtually guaranteed sales to distributors. In theory, everyone wins in this situation.
Unfortunately, that isn’t the case for antibodies. The large number of investor-backed manufacturers used to mean that distributors could acquire antibodies extremely cheaply. By rebranding products and obscuring the original manufacturers, distributors could make a tidy profit.
But when investor funds began to dry up, manufacturers needed to drive more revenue to keep up with their lower margins. Quality control is expensive and accounts for two-thirds of manufacturing costs. It’s the obvious first place to cut corners.
Some transparency of my own
In every aspect of the reproducibility crisis, it’s tempting to seek out one specific culprit. The crisis seems much less threatening if “just a few” bad-egg scientists or manufacturers can be blamed, but the true causes of reproducibility problems are systemic, not scientific or technological. There is ultimately no convenient single party to shoulder the blame.
Fixing the market problem will not outright solve problems with replication across scientific fields, either. Some fields, such as the under-fire field of social psychology, are unlikely to be affected by antibodies at all.
However, insisting on transparency in antibody businesses is a critical first step in mending the biological and biomedical sciences. Until the market works correctly, and scientists have trust in their tools and their results, correcting other reproducibility issues is likely to be frustratingly slow.