Pharmaceutical data eludes environmental researchers
The discovery that European and North American rivers and streams contain a witch’s brew of pharmaceuticals at low concentrations has raised obvious questions: What are the environmental effects, and should these drugs be regulated? However, scientists are stymied by their failure to collect some of the most fundamental data on these drugs. Nothing illustrates that better than carbamazepine, a popular anti-epileptic drug. In the late 1990s, scientists began detecting carbamazepine in our waterways, and they later found that the drug resists degradation in drinking-water treatment plants (Environ. Sci. Technol. 2004, 38, 392A–399A).
Unpublished research by the U.S. Geological Survey (USGS) found that carbamazepine pollution is widespread. In a sampling of 44 river sites across the United States, USGS scientists found an average carbamazepine concentration of 0.06 parts per billion (ppb) in water and 4.16 ppb in the sediment.
“The difference between the water and the soils may reflect sorption to the solids and associated organic matter,” says Ed Furlong, a USGS research chemist and one of the study’s authors. The study also found average carbamazepine levels of 20.9 ppb in the solids of sewage treatment plants, which discharge into some of the rivers tested. Other researchers have found that carbamazepine fails to degrade when passing through water treatment plants and is very persistent once it enters the environment.
As part of its regulatory process, the U.S. Food and Drug Administration (FDA) considers the environmental impact of new pharmaceuticals. “In the 1970s, the FDA considered itself a medical agency and was not concerned with effects on the environment,” recalls Florian Zielinski, an FDA chemist who works on the environmental assessment of drugs. Zielinski says that change began in 1969 after passage of the National Environmental Protection Act. However, it was not until the late 1980s that the FDA began requiring companies to submit environmental risk data, and these regulations were not finalized until 1998. Zielinski says this means that drugs approved before the late 1980s lack environmental risk assessments. In addition, the FDA has no definitive estimate of how many currently marketed drugs have been evaluated for their environmental impact.
It can be difficult for scientists outside the FDA to perform environmental risk analyses because much of the information on pharmaceuticals is proprietary. Companies submit anticipated production and sales figures to the FDA during a new drug application (NDA), but this information is then withheld from the public to protect the company from competitors. This makes it impossible to determine how much of any pharmaceutical is pouring into waterways.
Further, under current regulations, a company can obtain a “categorical exclusion” and not have to perform an environmental assessment if they manufacture less than 40,000 kilograms (kg) per year. “Forty thousand kilos correlates to about 1 ppb in the aquatic environment,” says Zielinski. He adds that this figure assumes that the drug is spread uniformly across the United States.
A categorical exclusion does not take into account the input from multiple companies that might all be making the same drug. For instance, if 10 companies are manufacturing a drug at 30,000 kg each, for a total of 300,000 kg, there is no trigger to perform an environmental assessment. And if one company surpasses 40,000 kg, that company’s environmental assessment would not account for production from other companies.
The issue for environmental researchers is complicated further because a generic drug can be sold under various names by many different companies. Geigy, a manufacturer in what was then West Germany, first put carbamazepine on the European market in 1964 as Tegretal. It was then brought to the United States as Tegretol in 1968. It has been sold under a number of brands, including Mazepine (Canada), Timonil (West Germany), and Stazepine (Poland). According to the FDA’s Orange Book of approved drug products, 12 companies now manufacture carbamazepine in the United States under the proprietary names Carbatrol, Tegretol, Teril, and Epitol. Tegretol-XR, an extended-release formula from Novartis, came on the market in the 1990s.
The FDA has approved carbamazepine for treating epileptic seizures and trigeminal neuralgia, a condition of one of the nerves feeding into the head that causes stabbing pain to the jaw or cheek. However, carbamazepine is also used “off-label” to treat bipolar depression, excited psychosis, and mania. In addition, Timothy Berigan, chief behavioral psychiatrist at Fort Huachuca, Ariz., says that Scandinavians prescribe the drug to aid alcohol detoxification. Novartis lists the standard daily dosage for adults as 800–1200 mg, although dosage may reach as high as 2000 mg per day.
Lynn Roberts, a professor of geography and environmental engineering at Johns Hopkins University, says that figuring out the annual consumption of carbamazepine is almost impossible. On the basis of U.S. sales of generic carbamazepine, she and Ph.D. student Kevin Bisceglia calculated that U.S. production ranged from 43,000 kg in 2000 to 35,000 kg in 2003. However, Roberts has been unable to get data on brand-name carbamazepine since 1999, when sales were $93 million. In that year, $67 million of generic carbamazepine was sold. She adds that these numbers do not include possible imports from Canada.
The NDA for Tegretol-XR contains metabolic studies that shed light on potential environmental risks. Novartis reports that for patients taking the drug orally, about 3% of the pharmaceutical is not metabolized. At a daily dosage of 1200 mg, this means that a single patient excretes at least 36 mg of carbamazepine into the environment every day. The rest comprises hydroxylated and conjugated metabolites; the main one is carbamazepine epoxide, which also has strong anticonvulsant properties. Carbemazepine epoxide is less cytotoxic than the parent drug and is negative for the Ames test. No environmental research has been done on this metabolite.
Carbamazepine is considered carcinogenic in rats but does not cause mutations in mammalian cells. Sublethal effects were observed in Daphnia at 92 ppb, and the lethal concentration in zebra fish is 43 ppb after 4 days.
On the basis of its annual production of Tegretol, Novartis calculated an expected concentration in the environment. From a comparison of this number to the lethal concentration in zebra fish, the company cites a safety factor of at least 100. However, the production values used to formulate the safety factor are confidential.
Researchers are just beginning to study the ecotoxicology of carbamazepine and other drugs. Eve Dussault, a graduate student at Guelph University (Canada), presented data on carbamazepine at a Society of Environmental Toxicology and Chemistry meeting in November. She found that Daphnia were killed when carbamazepine was present at 17.2 ppm and that midges died at 34.4 ppm. However, she also found that carbamazepine inhibited growth at 12.7 ppm for Daphnia and at 9.2 ppm for midges. Dussault says that once carbamazepine enters the environment, its half-life is about 80 days. However, she adds that the environmental levels are much lower than what she found was lethal to benthic organisms.
“But this is only acute toxicity, a 10-day test,” she says. “We don’t have any clue about chronic toxicity in the environment.”
