Activities Discovered for Some Inactive Drug Ingredients
Activities Discovered for Some Inactive Drug Ingredients

Activities Discovered for Some Inactive Drug Ingredients

Screens of hundreds of drug excipients reveal that some can interact with biological targets, contradicting their FDA categorization as inert.

Ruth Williams
Ruth Williams
Jul 23, 2020


A series of tests including computational predictions, biological assays, and live-animal studies has revealed that while the majority of inactive drug ingredients are indeed inert, many can interact with biological targets with the potential to cause physiologically relevant effects. The findings are published today (July 23) in Science.

“It seems to me to be quite a tour de force,” says epidemiologist Mady Hornig of Columbia University who was not involved in the study. “We’ve been quite data-starved in this area in general,” she continues, “and this represents an important leap forward in terms of a strategy for identifying candidate [excipients] of concern.” She adds, however, that “it is so important to emphasize . . . these are candidates for concern, not definitive agents of concern.”

The success of a drug often depends not simply on the active ingredients it contains, but on how it is formulated, explains Thomas Hartung, a pharmacologist at Johns Hopkins University who also did not participate in the research. The inactive components may stabilize the drug, prevent contamination, control the drug’s metabolism, or improve its taste or identification. But, Hartung continues, there is a sort of “toxicological ignorance” about these substances in part because they are largely considered safe and because to screen them is an “enormous burden” and costly.

The new study, which was an academia-industry collaboration funded in part by the US Food and Drug Administration (FDA), has made great inroads to addressing this “silent area,” Hartung says, and has shown “that among very many innocent substances there can be a black sheep” that we should “have an eye on.”

While all excipients in pill-based or injectable drugs have been deemed safe either because they are food-stuffs, such as sugars and starches, or because they have been tested to rule out acute toxicity, such tests “can hide a lot of sins,” says pharmaceutical chemist Brian Shoichet of the University of California, San Francisco, who coauthored the study.

For example, an excipient may be harmless at the recommended dose, but if a patient is taking multiple drugs at once that contain the same substance, or is taking a drug daily for many years, doses of the excipient may reach levels that were not anticipated, explains coauthor Lazlo Urban of pharmaceutical company Novartis.

To study the potential biological activities of excipients, the team performed computational predictions of interactions between 639 excipients—chosen from the FDA’s list of inactive ingredients—and 3,117 well-characterized target proteins that have known natural ligands and, in some cases, drug interactions.

Of the 20,000 or so plausible interactions that were reported, the team prioritized 69 for in vitro biological interaction assays because of the excipients’ similarities to the natural ligands for the proteins. These assays resulted in the identification of 19 excipients that interacted with at least one target. In a parallel approach, the team screened 73 of the most commonly used excipients against 28 targets that are commonly used to assess drug safety and found 32 that were active against at least one target. Many of the excipients identified by the two approaches were the same, creating a total of 38 ingredients with in vitro biological activities.

The team then selected seven of these to investigate further, using both animal experiments and data from existing publications to determine the pharmacokinetics of each excipient. Five of the substances, when given at clinically relevant doses, did not reach concentrations in the blood that would be physiologically relevant—that is, high enough for target binding and activation to be significant. “[They] were either trapped in the gut or metabolized very quickly,” explains Shoichet. But the other two potentially could.

These were thimerosal—a mercury-containing compound that acts as an antimicrobial to prevent bacterial contamination of multidose vaccines—and cetylpyridinium chloride—an antiseptic found in certain mouthwashes, toothpastes, and breath fresheners. Being a mercurial compound, thimerosal has been controversial for decades and, in the US, has already been removed from all childhood vaccines as a precaution. It is still present in multidose, but not single-dose, adult flu vaccines.

While the science supports further investigations of the substances identified in the study, what further action the FDA might take regarding their categorization or use, remains unclear. A representative of the FDA responded to The Scientist’s request for comment with the following emailed statement: “It would be premature for FDA to comment on this study. . . . Generally, the FDA does not comment on specific studies, but evaluates them as part of the body of evidence to further our understanding about a particular issue and assist in our mission to protect public health.”

It is reassuring to know “many of the excipients we tested proved to be negative in our assays,” says Urban, and that for those that did show concerning activities, there may be “options for replacement” or “options not to use as many.” Norvartis is pursuing further studies to that end, he says. “We take it very seriously for the safety of our patients and we hope that . . . other [drug manufacturers] follow us.”

As well as “laying a groundwork” for such studies, says Hornig, the paper provides a strategy for identifying such substances not just in drugs but also foods and cosmetics. “It underscores the call that so many of us have been making for such a long time to look at all the various agents that we put in our bodies,” she says.

J. Pottel et al., “The activities of drug inactive ingredients on biological targets.” Science, 369:403–13, 2020.