A Resolution of Sorts on TCE

TCE is a prime example among many of the difficulties we face trying to assess and manage the risks from low levels of chemicals in the environment. EPA has been undergoing a torturous risk assessment process since 2001 in order to set regulatory limits. The original TCE risk assessment received abundant criticisms from stakeholders, principally DOD and industries undergoing cleanup of TCE-contaminated sites. Some of those comments could be considered exercises in “manufactured uncertainty”; however, others were on the mark in identifying flaws in EPA’s analysis. Eventually the TCE risk assessment ended up with the National Academy of Sciences, which in 2006 issued a report that partially vindicated EPA, and partially confirmed what EPA’s critics were saying, but in the end, recommended that EPA revise its risk assessment, and quickly. Last year, Congress started getting into the act, sponsoring legislation to set a timeline for EPA to complete its scientific review and propose numbers that could be used for setting cleanup and drinking water standards. This only heightens the unreality of the situation; one would think with war, global climate change, economic meltdown and a crisis in healthcare, Congress would have better things to do with its time than debate the cleanup standards of ONE Superfund contaminant, no matter how prominent it is.

Earlier this month, there was a bit of a resolution on this issue, as reported in the March 18, 2008 Risk Policy Report, published by Inside EPA. At the meeting of federal facilities managers sponsored by the Association of State and Territorial Solid Waste Management Officials (ASTSWMO), EPA announced they it had reached an agreement in principle with the Department of Defense (DOD) on cleanup levels to address inhalation exposures to TCE associated with vapor intrusion. According to the Risk Policy Report:

The move will provide DOD and industry a long-sought consistent national approach to the issue, a DOD source says, as well as allow regulators across the country to make long-delayed cleanup decisions to address vapor intrusion, a previously unrelated pathway that results from toxic vapors from underground solvents that enter homes, offices and other structures

EPA has developed provisional guidance that will adopt the risk assessment prepared by the California Environmental Protection Agency for purposes of setting cleanup standards for vapor intrusion pathways. This will provide some badly-needed consistency in cleanup policy, which will allow remedial actions to move forward (I’ve personally observed sites where action has been stalled over differences in opinion about appropriate risk-based standards for TCE). Currently, state and federal agencies are using a variety of risk assessments which provide up to a 65-fold range in cleanup levels, for the same target cancer risk. EPA states that cleanups should achieve indoor air concentrations ranging from 1 to 10 ug/m3; the ability to protect public health with these levels in air is a matter for further discussion (I’m working on a longer post discussing that issue), but it’s possible that 1 to 10 ug/m3 in air represents levels that can be reasonably achieved with Superfund remedial technologies.

I’ll be back with more on this topic.

What’s New with TCE?

Not a lot, it seems. TCE has been in the center of a regulatory tangle going back over seven years, which has delayed the development of health criteria based on the most current health effects data. Over 15 years ago, EPA withdrew the cancer slope factors used to assess human cancer risks from TCE, because it was questioned that those values, derived from studies in rats and mice, adequately reflected the cancer hazard in humans. About seven years ago, the EPA began reassessing the science associated with the health risks of TCE, and issued in 2001 a draft report of that reassessment. That report concluded that evidence for TCE being a human carcinogen was stronger than previously believed, and that TCE was a more potent carcinogen than previously believed. After a contentious public review period with the science in EPA’s draft reassessment being closely questioned, the National Academy of Sciences took up a review of the EPA reassessment in 2004. Over a year ago, the National Academy of Sciences published its findings, validating some of EPA’s conclusions, and sending others back for further study; for example, EPA’s conclusion that TCE was a more potent carcinogen was judged by the NAS to be based on weak evidence.

EPA’s TCE web page doesn’t provide a schedule for when its risk assessment is going to be updated. There doesn’t appear to have been a lot of new data generated in the past year, based on what’s available in PubMed. One meta-analysis of occupational epidemiology of TCE exposure and liver cancer attempts to rebut the finding from the NAS study that high-dose TCE exposure, such as found in some workplaces, may be relevant to human cancer risk. This is relevant with regard to liabilities for past exposures (and is probably the origin for the study), but may eventually become a moot point. TCE releases are decreasing with time according to the TRI, which suggests that between pollution prevention initiatives and lingering concerns about groundwater contamination and exposure liabilities, businesses using metal cleaners or solvents are finding alternatives for TCE.

Another study investigated the association between TCE exposure and mutations in the von Hippel-Lindau (VHL) gene, which has been identified as a mechanism for kidney cancer. They examined several kidney cancer patients who may have had high TCE exposures in the past but didn’t observe a relationship between VHL mutations and TCE exposure. This doesn’t necessarily mean that TCE isn’t a kidney carcinogen in humans (the NAS concluded it’s likely there’s a relationship between TCE exposure and kidney cancer), but that further investigation may be needed to understand the mode of action for kidney cancer.

EPA scientists published a paper earlier this year discussing the difficulties in understanding the mode of action for TCE (in this case, with regard to liver cancer), which I suppose explains in part why a finalized risk assessment isn’t available yet. Congress is getting into the act with TCE, a development that absolutely floored me when I first heard about it back in August of this year. Senate Bill 1911 (S.1911), sponsored by Sen. Hillary Clinton and with bi-partisan support, would “amend the Safe Drinking Water Act to protect the health of susceptible populations, including pregnant women, infants, and children, by requiring a health advisory, drinking water standard, and reference concentration for trichloroethylene vapor intrusion (note: someone should remind Congress that the RfC doesn’t address cancer effects), and for other purposes”. It’s currently in the Senate Environment and Public Work Committee. Nothing has happened yet, but it’s still surprising that Congress is getting involved. Three years ago, I asked the question:

What’s next if the NAS can’t provide some good recommendations? Take it to Congress? I can’t wait to see that: “quick, we need to put aside the war on terror so we can debate this drinking water standard for TCE”.

This is one case where I hate being right.

A Little Light Weekend Reading – The National Academy of Science’s TCE Report

In its introduction, the NAS committee strikes a note of urgency with regard to assessing TCE risks:

The committee found that the evidence on carcinogenic risk and other health hazards from exposure to trichloroethylene has strengthened since 2001. Hundreds of waste sites in the United States are contaminated with trichloroethylene, and it is well documented that individuals in many communities are exposed to the chemical, with associated health risks. Thus, the committee recommends that federal agencies finalize their risk assessment with currently available data so that risk management decisions can be made expeditiously.

One of the controversial elements of EPA's 2001 risk assessment was the hazard identification, or the compilation of evidence used to determine if TCE posed a threat to human health. EPA relied on a meta-analysis of the available epidemiological studies, prepared by Daniel Wartenberg at Rutgers. Another meta-analysis had been presented to the NAS in 2005 (which looks to have been industry-funded based on the authorship). The NAS concluded that both had limitations, that neither should be used in the TCE risk assessment, and recommended preparing a new meta-analysis of the epidemiological data. You'll probably hear opinions to the contrary, but this may not be terribly significant for the risk assessment. The NAS seemed persuaded by the evidence that TCE poses a kidney cancer risk (see below).

The NAS committee noted that TCE has been shown to be toxic to the kidneys, both in laboratory animal studies and in humans. It concluded that TCE was a potential kidney carcinogen, based on laboratory animal studies, mechanistic studies of the metabolism of TCE, and epidemiological studies. It observed that kidney cancer in animals was preceded by kidney toxicity and more pronounced in male rats than female rats, and absent in mice. TCE requires metabolic activation to produce kidney cancer both in animals and humans (in other words, a metabolite of TCE is the carcinogenic agent). You may hear arguments that the studies in animals are not compelling, because the cancer in rats is an artifact of aged male rats, and that the metabolic pathways in rats are not significant in humans - suggesting TCE is not a human kidney cancer risk. Again, it sounds as if the NAS seemed persuaded by the evidence that TCE poses a kidney cancer risk. The details may be in the body of the report (this commentary is drawn from the executive summary).

The NAS committee noted that susceptibility to kidney cancer risk may be related to variability in levels of a key enzyme that metabolizes TCE, and with the occurrence of mutations in the von Hippel-Landau (VHL) tumor suppressor gene. Could these factors point to a population that is potentially sensitive to kidney cancer risks from TCE exposure? The NAS doesn't say that in so many words, but other investigators (here and here) suggest that might be the case.

The NAS committee concluded that humans were much less susceptible to liver and lung cancer from TCE exposure, compared with laboratory animals, due to the differences in metabolism between the species. It observed that animal and epidemiological studies suggested that TCE may be associated with reproductive and developmental toxicity, including male and female infertility and cardiac malformations, though the relevance of some of animal study results to humans was unclear. The NAS committee also noted mixed results (in other words, some positive, some negative) in epidemiological studies of congenital defects in heart valves in communities exposed to TCE, but that the heart valve defects observed were the same in laboratory animal and the epidemiological studies.

Susceptibility to TCE toxicity was a major issue in EPA's TCE risk assessment. For example, the NAS committee observed that several factors can contribute to an individual’s susceptibility to the toxic effects of TCE, including disease states and differences in the expression of enzymes involved in TCE metabolism. These include conditions such as alcoholism, obesity, and diabetes, which are known to induce the expression of the CYP2E1 enzyme; that's the oxidative enzyme primarily responsible for converting TCE to toxic metabolites. While there are methods for addressing variability in populations in a risk assessment, it's clear we still don't understand susceptibility very well yet. Regardless of the method selected to quantify the risk from TCE, and develop cleanup levels for it, there will always need to be the awareness that these levels may not be protective for a certain portion of an exposed population. However, it's an open question of how sizeable that portion might be. For example, CDC reports that 33 states show rates of obesity in the population ranging from 20-24 percent, with 9 states with rates over 25 percent. In addition, CDC reports that the nationwide incidence of diabetes is 7 percent in individuals of all ages.

One last thing: The NAS committee stated that none of the existing epidemiologic data is suitable as a primary means of quantifying cancer risks, which should put to rest the "TCE is 40-times more carcinogenic than previously believed" meme that was running around the internets earlier this year.

My sense of the overall story so far:

As always there are uncertainties in the evidence, however the principal health concern with TCE may be kidney cancer. Other concerns include possible developmental toxic effects, particularly cardiac valve defects. The risk of adverse effects may be more significant for susceptible populations; some of the factors contributing to potential susceptibility, such as obesity and diabetes, are relatively common in the U.S. population. TCE exposures are widespread across the U.S., and the NAS is encouraging agencies to finish up the TCE risk assessment so that risk management decisions can be made expeditiously.

More on TCE later. This was just from the executive summary.

Release of the National Academy of Sciences TCE Report

Today, the National Academy of Sciences has released it's report of EPA's reassessment of TCE health risks. I'll be reviewing the NAS's report in more detail and providing updates, but the significant result appears to be that NAS concurs with EPA that TCE poses a kidney cancer risk.

WASHINGTON -- A new report from the National Academies' National Research Council recommends research to improve understanding of how the environmental contaminant trichloroethylene causes cancer and other adverse health effects, but adds that enough information exists for the U.S. Environmental Protection Agency to complete a credible human health risk assessment now.

In 2001 EPA issued a draft risk assessment on trichloroethylene, a solvent widely used as a degreasing agent that is contaminating air, soil, and water at several military installations and hundreds of waste sites around the country. The release of the draft risk assessment was followed by much debate about the quality of evidence on trichloroethylene and how that evidence should be assessed. This prompted an interagency group to request that a Research Council committee review issues related to assessing the health risks from exposure to trichloroethylene, commonly referred to as TCE. The committee was not asked to conduct a risk assessment of its own.

The evidence on cancer and other health risks from TCE exposure has strengthened since 2001, the committee found. It pointed out that research, including studies of human populations, supports the conclusion that TCE is a potential cause of kidney cancer. Research shows that the chemical may cause other kidney problems as well, but the level of exposure needed to produce kidney damage is not clear. Animal data indicate that relatively high doses of TCE are needed to induce liver toxicity and cancer. Some epidemiology studies indicate a higher incidence of liver cancer among populations exposed to TCE, but the evidence is inconsistent. Studies of people exposed to TCE at work do not show a strong association between exposure and lung tumors, the report notes.

Animal research and human population studies suggest that TCE exposure may also be associated with other health effects, such as reproductive and developmental problems, impaired neurological function, and autoimmune disease. The committee recommended studies to advance understanding of the mechanisms by which TCE causes cancer and other health problems; which populations are most sensitive to TCE's effects; and how exposure to a mixture of TCE and other chemicals affects human health.

A large body of epidemiological data on TCE and cancer is available, but a new analysis of that data is needed to better characterize the hazard that TCE presents to humans, the committee said. It found several weaknesses in the analysis that EPA used in its draft risk assessment, as well as in an analysis developed by researchers since the draft was issued. To overcome these weaknesses, the new analysis should establish clear criteria for including epidemiological studies based on objective characteristics, the committee said. It added that it would be appropriate for EPA to use a model jointly developed with the U.S. Air Force to simulate how the body metabolizes TCE, although the model does not resolve uncertainty about the mechanisms by which the chemical causes cancer.

A model is being used to extrapolate from animal studies an estimate of the cancer risk posed by TCE at low doses. The risk is extrapolated below a "point of departure," which is associated with an incremental effect, such as 5 percent more cancers. EPA should consider a range of points of departure in its risk assessment, the committee recommended. Because there is not enough evidence on how TCE triggers cancer to choose the best model for relating the body's response to different dose levels -- a so-called dose-response model -- it is appropriate under EPA's cancer guidelines to extrapolate the risk using a linear model, in which cancer risk rises in proportion to dose.

The committee's report was funded by the U.S. Environmental Protection Agency, U.S. Department of Defense, U.S. Department of Energy, and NASA. The National Research Council is the principal operating arm of the National Academy of Sciences and the National Academy of Engineering. It is a private, nonprofit institution that provides science and technology advice under a congressional charter.

TCE and PCE Stories

The persistence in the marketplace of chlorinated solvents such as TCE (trichloroethylene) and PCE (perchloroethylene) is astonishing. One would have thought that the liabilities of soil and groundwater contamination, known health risks to workers, questions about health risks to the general public, burdensome regulations related to hazardous waste, air toxics and ozone depletion and the availability of alternatives, would have been more effective in phasing these chemicals out of the market.

The notable point about TCE and PCE from the viewpoint of environmental health is that their use patterns lead to widespread human exposure. A primary use for TCE and PCE are for solvent cleaning. Half of all PCE use is for dry-cleaning clothes. This means that TCE and PCE uses cut across a large number of industry and workplaces. NIOSH has estimated that approximately 390,000 workers are potentially exposed to TCE nationwide, and that approximately 690,000 workers are potentially exposed to PCE, based on the National Occupational Exposure Survey (NOES). While the NOES was completed over 20 years ago, and there hasn’t been an update, these numbers are likely to be reasonably consistent today. Chlorinated solvent use appears to have declined over time (trends for the U.S. have been hard for me to find – if I uncover more information, I’ll post it), but it is likely that a lot of these solvents are still sold. Occupational exposure is the most significant human health concern with TCE and PCE exposure (a topic for another day – I’ll be writing a number of posts on these chemicals).

Every day, millions of people bring home clothes from the dry cleaners that are off-gassing PCE. Dry-cleaned clothes represent one of the largest sources of exposure to PCE by the general public (see “Everyday Exposure to Toxic Pollutants”, by Wayne Ott and John Roberts, February 1998 Scientific American, downloaded for a price from sciam.com). This is consistent with a wealth of studies (most recently here) indicating that a preponderance of volatile organic compound exposure occurs indoors.

According to the ATSDR, over 1,309 Superfund sites in the U.S. have soil or groundwater that is contaminated with PCE; 1,460 Superfund sites are contaminated with TCE. This does not include thousands of other sites under the jurisdiction of state agencies and undergoing cleanup as part of RCRA Corrective action, which are not listed on the National Priority List. TCE exposure at contaminated sites is of sufficient concern that ATSDR has established a TCE subregistry within its National Exposure Registry.

In many cases, people living near these sites have been exposed to these chemicals in domestic-use water, from ingestion, skin contact or inhaling chemicals that volatilize from water. However, what has become a greater concern is vapor intrusion, a pathway where these chemicals volatilize from soil or groundwater, migrate through soil near building foundations and are drawn into indoor air through egresses in foundations. This issue took regulatory officials a bit by surprise, and has been a growing issue for hazardous waste site cleanups across the U.S. In New York, state officials are going to revisit 400 sites that had cleanup decisions made before 2003 to investigate the potential for indoor exposures to volatile compounds from vapor intrusion (this comes from the TCE Blog, your source for all things related to chlorinated solvents in the environment).

As you can imagine, there’s much more to this story, beyond the fact that TCE and PCE exposures are widespread. Who needs to be concerned about potential exposures? Under what conditions could people be at significant health risks? What can they be doing to reduce their risks? What can other stakeholders, such as governments and industries, be doing to reduce health risks from these chemicals? How do the potential risks from TCE and PCE exposure compare with other chemicals in a person’s environment?

Stay tuned for more posts on this topic. Until then, check out TCE Blog.