Survey Identifies Drugs Most Likely to be Found in the Environment

Contact: Charmayne Marsh 202-872-4445 in Washington

April 4--11, 2002, in Orlando407-685-8070

EMBARGOED FOR RELEASE: Wednesday, April 10, 6:00 p.m., Eastern Time

Survey identifies drugs most likely to be found in the environment

ORLANDO, Fla., April 10 -- A team of Johns Hopkins researchers says antidepressants, anticonvulsants, anticancer drugs and antimicrobials are the types of pharmaceuticals most likely to be found at "toxicologically significant" levels in the environment. Their conclusions are based on a survey of the 200 most sold and prescribed drugs in the United States.

Padma Venkatraman, Ph.D., one of the study's lead authors, presented estimates of "probable environmental concentrations" (PECs) of the top 200 drugs to scientists gathered here for the 223rd national meeting of the American Chemical Society, the world's largest scientific society.

"We're trying to make an intelligent guess as to what's out there in the environment and what's probably toxic," Venkatraman said during an interview. As for any danger to people, "We certainly don't have any evidence that most pharmaceuticals pose a human health risk, although the presence of carcinogens or teratogens even at low concentrations is of potential human health concern," she emphasized.

Any danger from the drug concentrations is more likely to be to aquatic organisms than humans, according to Venkatraman. "Our research suggests that certain existing drugs may be present at levels at which they have the potential to exert sublethal effects."

The survey is based on calculated total drug sales and prescriptions and a search of the medical literature, said Venkatraman. "Based on the existing data about the drug's biochemistry and existing data on its metabolism, we came up with the estimates."

Because some drugs are not metabolized before elimination, the researchers calculated PECs with and without metabolism, Venkatraman noted. "Also, there is the fact that expired medications are often flushed down the toilet," she added.

Although the survey focused primarily on prescription drugs, some over-the-counter drugs, such as ibuprofen, were included, Venkatraman said.

Although Venkatraman would not specifically name the top individual drugs that are likely to be found at significant concentrations, she did highlight "the classes of compounds that we think might be important."

Antimicrobials, anticonvulsants and anticancer drugs are among the classes of compounds that the researchers suspect may be found at the highest concentrations, according to Venkatraman.

Measuring environmental contaminants is target-specific, Venkatraman noted. The aim of the research is to help other scientists better target their searches for pharmaceutical compounds of environmental importance, she said. "We're hoping we'll be able to give them a basis to guess what's likely to be persistent and what's likely to be used in great quantities and therefore likely to be out there in the environment also."

Research into the fate of pharmaceuticals in the environment got a kick-start in this country from similar studies done in Europe several years ago. But there are differences between those studies and the Johns Hopkins study, said Venkatraman.

"For instance, there was a Danish study that looked at the top 20 [drugs sold and prescribed in Denmark]. If you compare our study to theirs, only about one-third, or less, of the drugs that show up in their top 20 even show up in our top 200. There's not much of an overlap."

Differences in prescription practices, climate, and drinking water and sewage treatment account for the disparity between the studies, according to Venkatraman. As an example, she noted: "Many U.S. cities employ sewage treatment that's more rudimentary than what you'd see in most European cities. Ozone, which is more reactive than chlorine, is widely used as a disinfectant in Europe, while we use chlorine. In Europe, the use of activated charcoal to remove trace organics in drinking water treatment is also much more prevalent than in our country."

The Johns Hopkins study "complements the results of the first nationwide study of drugs in the nation's streams," according to the researchers. That study, done by the U.S. Geological Survey was published last month in Environmental Science and Technology, one of the American Chemical Society's leading peer-reviewed journals.

One area of the research that Venkatraman says she is very excited about is the use of "innovative analytical techniques," to measure drug concentrations in the environment. As an example, she noted, "We are modifying existing derivatization techniques to adapt them to our analytes of interest." Derivatization basically involves chemical modification of compounds in order to make them more amenable to analyses by gas chromatograph.

The other key Hopkins researchers for the EPA-funded study are principal investigator Lynn Roberts, Ph.D., co-principal investigator Edward Bouwer, Ph.D., both of the university's geography and environmental engineering department, and Kelvin Chan, a student at the university.

# # #

The poster on ENVR 167, will be presented at 6:00 p.m., Wednesday, April 10, at the Convention Center, Hall C, during a general poster session.

Padma Venkatraman, Ph.D., is a postdoctoral fellow in the Department of Geography and Environmental Engineering at Johns Hopkins University in Baltimore, Md.

A. Lynn Roberts, Ph.D., is an associate professor in the Department of Geography and Environmental Engineering at Johns Hopkins University in Baltimore, Md.

Edward J. Bouwer, Ph.D., is a professor in the Department of Geography and Environmental Engineering at Johns Hopkins University in Baltimore, Md.

-- Marvin Coyner

#13198 Released 4/10/2002

Embargoed: Wednesday, April 10, 6:00 p.m., Eastern TimeENVR 167 [520603] Environmentally significant human pharmaceuticals: A preliminary risk assessment

A. Lynn Roberts, and T. V. Padma, Department of Geography and Environmental Engineering, Johns Hopkins University, 3400 N. Charles Street, 313 Ames Hall, Baltimore, MD 21218, Fax: 410 516 8996, [email protected], Phone: 410 516 4255

Analytical tools and techniques used in attempts to measure environmental contaminants are highly target-specific. It is necessary to prioritize such efforts so that time and energy are not wasted in trying to measure contaminants that are unlikely to occur at detectable levels or at levels capable of producing ecotoxic effects. To begin to assess potential human health and ecotoxic consequences associated with pharmaceuticals in the environment and to focus attention on existing drugs most likely to be encountered at environmentally significant concentrations in the United States, we believe that estimates of probable environmental concentrations (PECs) are of merit. We have computed PECs for the "top 200" drugs sold and prescribed in the United States. These data should provide much needed information on "large-volume" drugs most likely to occur at quantifiable and toxicologically significant environmental concentrations. Our research suggests that certain existing drugs may be present at levels at which they have the potential to exert sub-toxic effects.

Non-technical summary of paper # 520603Environmentally significant human pharmaceuticals: A preliminary risk assessment. A. Lynn Roberts and T. V. Padma*

"All substances are poisons; there is none which is not a poison. The right dose differentiates a poison and a remedy" (Paracelsus). Although human pharmaceuticals are designed to elicit biological responses in organisms and thus may exhibit toxic effects, the notion that excretion of unmetabolized drugs could represent an important route for their introduction to the environment was largely ignored. Instead it was tacitly assumed that they would be degraded in sewage treatment plants to levels below those capable of harming the health of aquatic organisms or humans. Recent evidence (mostly based on European studies) indicates, however, that pharmaceuticals may be more persistent and widespread in the environment than previously thought. But which pharmaceuticals pose the greatest risk if present as environmental contaminants? In other words, how do environmental concentrations correspond to different thresholds for toxic effects?

Answering this question is not simply a matter of going out and analyzing environmental samples to see what's present, and at what concentrations. As any analytical chemist knows, "what you see depends on what you look for". Environmental scientists are constantly faced with the problem that non-target contaminants that might be present could elude detection. This problem is particularly acute for human pharmaceuticals, because their generally polar nature and often low volatility means that few are amenable to analysis via the solvent extraction and GC/MS protocols that have been the mainstay of environmental analysis. In the past, environmental chemical analysis largely emphasized non-pharmaceutical contaminants, such as the organic solvents, organochlorine pesticides, polychlorinated biphenyls and polycyclic aromatic hydrocarbons that figure so prominently in EPA's "Priority Pollution List". An important criterion for inclusion on this list was that analytical methods for analyte determination in environmental samples had to exist. There is now a growing recognition that this criterion only provides "the tip of the iceberg" in terms of contaminants of potential concern. A need exists for reliable and robust means that could be used to quantify "unconventional pollutants", such as human pharmaceuticals, in environmental samples. This represents an enormous analytical challenge, as some 3000 pharmaceuticals are in current use, comprising a wide array of individual species from many structurally diverse and dissimilar classes.

In order to focus efforts on those pharmaceuticals most likely to occur at levels potentially capable of producing human health and/or ecotoxic effects, we have been compiling data pertaining to amounts of different pharmaceuticals sold each year in the US. Data pertaining to production values for individual drugs are confidential and are difficult to acquire. Such statistics can however be estimated from annual sales or prescription data, coupled with information pertaining to average wholesale price or typical dosage conventions. The results are being used to estimate probable environmental concentrations (PECs), which factor in dilution and metabolism or other sources of attenuation. These PEC values can in turn be compared to "probable no-effect concentrations" or PNEC values (obtained from toxicological studies). A high ratio of PEC to PNEC indicates a pharmaceutical with a high environmental risk.

Our results indicate that a number of different drugs may be present at levels that can be detected by appropriate analytical instrumentation (sometimes following appropriate chemical derivatization to volatile forms in the case of GC/MS analysis), and at levels potentially capable of exerting subtle effects on aquatic organisms. These compounds therefore represent a high priority for analytical method development and future monitoring activities. Our results also suggest that patterns of pharmaceutical usage between the US and Europe may be sufficiently different as to complicate direct extrapolation of monitoring results across geographic boundaries.

###