Advisory Panel Poised To Back EPA's Landmark Cancer Finding On TCE

Salem-News.com has been reporting the massive TCE contamination of MCAS El Toro and Camp Lejeune for three years.

(WASHINGTON D.C.) - A panel of experts reviewing EPA's draft risk assessment of the solvent trichloroethylene (TCE) appears to be poised to back the agency's proposed listing of the ubiquitous groundwater contaminant as a human carcinogen, a decision that exceeds classifications by other health agencies and one that is hotly contested by industry, due in part to the costly cleanups that it will drive.

At the panel's May 10-12 meeting, several panelists argued that human epidemiological data, particularly studies indicating TCE exposure leads to kidney cancer, is particularly strong, and can be the basis for EPA's determination.

But industry representatives and some members of the review panel questioned whether the epidemiology data and the animal toxicology data EPA presented in the draft assessment are sufficient for the agency to make the determination that TCE causes cancer in humans by all routes of exposure. Caffey Norman, an attorney representing the Halogenated Solvents Industry Alliance (HSIA) argued during public comments at the panel's meeting in Washington, DC, that the existing evidence supported a lesser classification. "The proposed classification of TCE as a known human carcinogen cannot be supported by the evidence," Norman said, pointing to a 2009 report from the National Academy of Sciences (NAS) regarding historical contamination of drinking water at Camp Lejeune, NC, base housing with TCE, perchloroethylene, and other chemicals.

The NAS report, Contaminated Water Supplies at Camp Lejeune: Assessing Potential Health Effects, shows "no sufficient evidence of causation," Norman argued.

But EPA's Peter Preuss, director of the program that writes the IRIS assessments, argued that the NAS' Camp Lejeune report findings had little bearing on the agency's assessment of TCE's cancer risks. "There's also been a lot of talk about the [NAS] report," Preuss told the panel. "These reports are [conducted] under a specific act of Congress for renumeration. That's very different than what are the right uses for epidemiology studies." Norman was followed by HSIA toxicologist Paul Dugard, who urged the panel to reject EPA's finding because "TCE is ubiquitous [and] unduly low targets would result in billions of dollars [in risk mitigation] that could be better spent." But the panel members seemed to agree with the conclusions of a subgroup of epidemiologists presented May 12, the last day of the meeting, concurring with EPA's conclusions. "We support the conclusion that TCE is carcinogenic to humans by all routes of exposure as outlined in the EPA guidelines," said John Vena, head of the epidemiology department at the University of Georgia.

"The consistency of the findings are remarkable. The pooled risk estimates [in EPA's meta analysis of human epidemiology studies] though modest, are robust, with no indication of publication bias." EPA makes determinations of carcinogenicity based on its 2005 cancer risk assessment guidelines, which indicate what level and types of evidence is required for each classification. In this case, EPA had to determine that either the epidemiology data was sufficient to make the cancer finding alone, or that the data had to be sufficiently supported by animal data that met additional criteria. Listing 'Known' Carcinogens Ivan Rusyn, a panelist from the University of North Carolina at Chapel Hill urged the other panelists to "think very carefully" about listing a chemical as being "known" to be carcinogenic to humans and ensure the panelists are "all comfortable with that level of evidence." Rusyn noted that other agencies considering whether TCE is carcinogenic, including the National Toxicology Program (NTP) and the International Agency for Research on Cancer (IARC), deemed it the lower classification of likely to be carcinogenic to humans.

"It seems to me that strong, convincing and causal association [of carcinogenicity] for these other agencies is usually stronger literature -- dozens of studies and multiple positive meta analyses. This clearly doesn't rise to the same level," Rusyn said.

But Weihsueh Chiu, EPA's TCE chemical manager, clarified that the agency is "relying on the epidemiology" data to make its determination. "And really the kidney cancer evidence," Chiu added. "Without that, we would've made the likely [carcinogenic to humans] determination."

Epidemiology experts on the panel quickly came to EPA's defense, arguing that the epidemiology data presented in the agency's draft report strongly supported the finding that TCE is a human carcinogen. "The kidney evidence, there are a number of studies with positive results. Some show a dose-response relationship," said Anneclair De Roos of the Fred Hutchinson Cancer Research Center. "I fully support EPA's conclusion that the carcinogenic descriptor is appropriate."

Another epidemiologist on the panel agreed, saying that EPA "makes a good case for carcinogenicity," and that the evidence for a likely classification "is a much lower bar than this. We're definitely not at that level of evidence to me. To me, there is very convincing epidemiological human evidence for these three [cancer] sites."

And Aaron Blair, a scientist emeritus at the National Cancer Institute, said that his conclusion that TCE is a carcinogen "isn't solely based on the epidemiological data. I wouldn't vote on that. The data really hangs together with the toxicological evidence I hear from you." Rusyn thanked the epidemiologists for their opinions, saying that he withdrew any objections to the panel supporting EPA's classification. "I get the sense that the epidemiologists feel this is convincing," he said. "I think I am convinced as a toxicologist."

EPA based its quantitative estimates of TCE's potency by oral exposure and by inhalation exposure on an epidemiology study of French screw cutting workers, by Barbara Charbotel and others and published last year. Another NAS report, Assessing the Human Health Risks of Trichloroethylene: Key Scientific Issues, published in 2006, "didn't consider epidemiology data strong enough" to be the basis for these quantitative calculations, said Dugard, the HSIA toxicologist. "This study is new [since NAS' 2006 report] but you'll find it's very small."

"The Charbotel study has significant limitations," argued Michael Kelsh, an epidemiologist at the consulting firm Exponent told the panel. "Ultimately, it depends on self-reported work history. I'm always concerned when people try to take epidemiology data to quantitative risk analysis." -- Maria Hegstad