|
Items in cart [1] |
Questions? Call 888-624-8373 |
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 90
4
Review of Toxicologic Studies
This chapter summarizes findings of animal studies of trichloroethylene (TCE) and tetrachloro-
ethyle (perchloroethylene, PCE) toxicity and relevant end points. The review was based in part on previ-
ously published comprehensive reviews on the two chemicals of interest, but numerous published studies
were reviewed individually in greater detail. Studies were examined according to criteria that reflected
robustness of study design related to the hypothesis being tested and that included such characteristics as
number of animals tested, measurement methods used, appropriateness of statistical methods, and con-
cordance of conclusions with data presented. Studies substantially lacking in some of or all those and
other measures of study quality and studies whose outcomes were not able to be repeated in later studies
or in other laboratories were given less weight in the evaluation. Salient findings on principal health end
points are summarized by chemical and organ system. The administered doses or the doses associated
with the no-observed-adverse-effect levels (NOAELs) or the lowest-observed-adverse-effect levels
(LOAELs) are reported when possible. At the conclusion of this toxicologic review, a hazard evaluation
of TCE and PCE exposure at Camp Lejeune was conducted for selected health end points. A hazard
evaluation is conducted to provide information on the intrinsic toxic potential of an exposure and is not
meant to provide a quantitative risk assessment.
As noted in Chapter 2, the committee identified nine volatile organic compounds (VOCs) of con-
cern. To manage the vast amount of information on each, we provide different degrees of review accord-
ing to the findings from the exposure assessment regarding the frequency and concentrations of the con-
taminants in the affected drinking-water systems. This chapter presents detailed toxicologic evaluations of
the two chemicals of greatest concern, TCE and PCE. Information on the metabolism of TCE and PCE
and factors that influence their toxicity was presented in Chapter 3 and is drawn upon in this chapter.
Chapter 7 provides an integrated discussion of the toxicologic evidence in context with the epidemiologic
evidence on TCE and PCE. For completeness of the literature review, Appendix D provides brief reviews
of the toxicologic data on the seven other chemicals.
TRICHLOROETHYLENE
Data on the toxicity of TCE were summarized in a report by the National Research Council (NRC
2006). In some cases, more recent literature reviews on particular subjects were available (e.g., Lamb and
Hentz 2006; Watson et al. 2006), and they were relied on for defining the body of literature available up
to the time of publication. In addition, a literature search of Medline was done to determine whether any
relevant new publications were available. Conclusions drawn for the present report were based on a re-
view of the body of available peer-reviewed literature. Because TCE and PCE have some of the same me-
tabolites and effects, salient finding of studies of PCE are discussed in relevant sections of the TCE re-
view. More detailed review of the PCE literature is provided later in the chapter. To facilitate a
comparison of the toxicologic data with the epidemiologic data in Chapter 7, the toxicologic data are pre-
90
OCR for page 91
Review of Toxicologic Studies 91
sented below according to organ system and in some sections divided to consider toxic effects separately
from carcinogenic effects.
Hepatic Effects
Toxicity
TCE, even in high doses, produces only a modest degree of injury of hepatocytes in laboratory
animals. Klaassen and Plaa (1966) compared the acute hepatotoxicity of TCE with that of other common
halogenated aliphatic hydrocarbons (halocarbons) in male mice dosed by intraperitoneal injection. The
dose of TCE required to produce an increase in serum alanine-aminotransferase activity, 1.6 mL/kg, was
almost as high as the dose that was lethal in 50% of test animals, 2.2 mL/kg. Oxidative stress was as-
sessed by measuring thiobarbituric-acid-reactive substances in the livers of male Fischer rats that received
one intraperitoneal injection of TCE at 0, 100, 500, or 1,000 mg/kg (Toraason et al. 1999). Thiobarbituric-
acid-reactive substances were increased in the 500- and 1,000-mg/kg groups. Hepatic concentrations of 8-
hydroxy-2′-deoxyguanosine adducts, induced in DNA by oxygen-based radicals, were also increased at
500 mg/kg and presumably at 1,000 mg/kg. It should be recognized that the 500- and 1,000-mg/kg doses
produced stage II and stage III-IV anesthesia, respectively. Channel et al. (1998) gave male B6C3F1 mice
TCE at 0, 400, 800, or 1,200 mg/kg in corn oil by gavage 5 days/week for 8 weeks. Transient increases in
cell and peroxisome proliferation, centered around day 10, were observed only at the highest dose. There
were no differences from controls in the incidences of hepatocellular apoptosis or necrosis. Thiobarbi-
turic-acid-reactive substances were significantly increased in the groups treated with TCE at 800 and
1,200 mg/kg on days 6-14. 8-Hydroxy-2′-deoxyguanosine adducts in liver DNA were significantly in-
creased throughout much of the study with TCE at 1,200 mg/kg. Buben and O’Flaherty (1985) saw a
modest increase in serum alanine aminotransferase and decrease in hepatic glucose-6-phosphatase activity
in mice given TCE at 500 mg/kg or greater in corn oil by gavage five times a week for 6 weeks. Mice re-
ceiving as little as 100 mg/kg per day had an increase in relative liver weight. It is clear that TCE, even
when given repeatedly to mice and rats at narcotic doses, has little ability to damage hepatocytes.
Adverse effects of TCE on the liver are usually attributed to metabolites of the cytochrome P-
450-mediated oxidative pathway (Bull 2000). Buben and O’Flaherty (1985) reported that plots of their
mouse subchronic-hepatotoxicity data against urinary-metabolite excretion values indicated that TCE’s
effects are directly related to the extent of its metabolism. As described in Chapter 3, TCE is oxidized by
cytochrome P-450s (notably CYP2E1 at low to moderate TCE doses) to chloral, which is converted to
chloral hydrate. That intermediate has a short half-life; it is rapidly oxidized to trichloroacetic acid, which
is reduced to trichloroethanol (Lash et al. 2000a). Relatively small amounts of dichloroacetic acid may
arise from trichloroacetic acid or other metabolites. Induction of CYP2E1 in rats with pyridine increases
the toxicity of TCE to isolated rat hepatocytes (Lash et al. 2007). High concentrations of trichloroacetic
acid and dichloroacetic acid are not toxic to hepatocytes freshly isolated from B6C3F1 mice (Bruschi and
Bull 1993); the researchers proposed that trichloroacetic acid and dichloroacetic acid cause peroxisome
proliferation and the ensuing generation of reactive moieties that deplete glutathione and can cause oxida-
tive injury. Dichloroacetic acid does not induce peroxisome proliferation in male B6C3F1 mice in the
same dose range at which it produces hepatic tumors (DeAngelo et al. 1999). Laughter et al. (2004) found
that high oral doses of TCE increased liver weight, peroxisome proliferation, and hepatocellular prolifera-
tion in male mice. Those effects appeared to be due primarily to trichloroacetic acid’s activating a nuclear
protein known as the peroxisome-proliferator-activated receptor alpha (PPARα). PPARα-dependent
changes seen in gene expression may contribute to the carcinogenicity of TCE in mouse liver.
TCE-induced hepatic injury is not a common finding in humans and was rarely reported in pa-
tients when TCE was used as an anesthetic (Lock and Reed 2006). Clearfield (1970) described hepatocel-
lular degeneration in two men who intentionally inhaled extremely high vapor concentrations of TCE for
their intoxicating effects. In contrast, James (1963) saw only small foci of fatty accumulation in the liver
OCR for page 92
Contaminated Water Supplies at Camp Lejeune—Assessing Potential Health Effects
92
(steatosis) of a man who died after 10 years of TCE abuse. Bruning et al. (1997) found renal injury but no
evidence of hepatotoxicity in a man rendered unconscious for 5 days by drinking about 70 mL of TCE in
a suicide attempt. Pembleton (1974) reported a transient postoperative rise in serum aspartate aminotrans-
ferase activity in four of 100 patients anesthetized with TCE for surgical procedures. A study of 289 Brit-
ish workers who experienced central nervous system (CNS) symptoms from TCE inhalation and dermal
exposure in the workplace revealed no clear diagnoses of hepatotoxicity (McCarthy and Jones 1983).
Such findings over the last 50 years indicate that acute or repeated high-dose exposures to TCE will pro-
duce a modest degree of hepatic injury in some people but not in most people (ATSDR 1997a).
Cancer
The carcinogenic effects of TCE and its metabolites have been assessed in a number of lifetime
studies of several strains of mice and rats (NCI 1976; Fukuda et al. 1983; Henschler et al. 1984; Maltoni
et al. 1986; NTP 1988, 1990a). Results of studies of TCE induction of hepatic tumors in rodents are
summarized here on the basis of the extensive National Research Council review (NRC 2006).
It has been well established that TCE, when administered chronically in very high doses by ga-
vage, can produce an increased incidence of hepatocellular cancer in B6C3F1 mice. In the original bioas-
say (NCI 1976), technical grade TCE (containing epichlorohydrin and 1,2-epoxybutane as stabilizers) had
this effect. Concern that these stabilizers are well-established mutagens and contributed to TCE’s appar-
ent carcinogenicity led scientists to utilize TCE without these stabilizers in future bioassays. Henschler et
al. (1984) saw no increase in liver tumors in either sex of Swiss ICR/HA mice, rats, or Syrian hamsters
that inhaled highly-purified TCE (stabilized with 0.0015% triethanolamine) for 18 months. Exposure of
male and female B6C3F1 mice to epichlorohydrin-free TCE by corn oil gavage at 1,000 mg/kg/day for 2
years caused increases in hepatocellular carcinoma. No such increase in liver tumor incidence was mani-
fest in F344/N rats (NTP 1990a). Another study of four additional strains of rats of both sexes ingesting
epichlorohydrin-free TCE at 125-1,000 mg/kg also showed no increase in liver tumors (NTP 1988). Thus,
it has been demonstrated that TCE itself, when administered chronically in very high oral doses, results in
an increased incidence of liver cancer limited to male and female B6C3F1 mice.
The major oxidative metabolites of TCE—trichloroacetic acid, dichloroacetic acid, and chloral
hydrate—have also been extensively studied in rodents (Herren-Freund et al. 1987; Bull et al. 1990;
DeAngelo et al. 1991, 1996, 1997, 1999; Daniel et al. 1992, 1993; Pereira 1996; George et al. 2000; NTP
2002a,b,c; Leakey et al 2003). Trichloroacetic acid is a species-specific carcinogen that induces perox-
isome proliferation and hepatocellular carcinomas when administered in drinking water to male and fe-
male B6C3F1 mice (B6C3F1 mice are particularly susceptible) (Herren-Fruend et al. 1987; Bull et al.
1990; DeAngelo et al. 1991). The blood concentration of trichloroacetic acid required to induce hepatic
tumors in mice is in the millimolar range. Effects have been observed with drinking-water concentrations
of trichloroacetic acid of 0.05-5 g/L. TCA did not induce hepatic tumors in male F344 rats under similar
treatment conditions (Daniel et al. 1993; DeAngelo et al. 1997). B6C3F1 mice produce a large amount of
trichloroacetic acid after exposure to TCE relative to unresponsive mouse strains (see Chapter 3). Tri-
chloroacetic acid increases the rate of hepatocellular proliferation, production of reactive oxygen species,
hepatocellular hyperplasia, and hepatomegaly (see Chapter 3). Marked species differences in susceptibil-
ity to peroxisome proliferation associated with liver cancer after increased fatty-acid beta oxidation and
modulation of hepatocellular replication related to activation of the PPARα nuclear receptor by TCE and
its metabolites have been investigated and reviewed in detail (Klaunig et al. 2003; Cattley 2004; Laughter
et al. 2004). Rats exhibit saturation of TCE oxidative metabolism that results in amounts of trichloroacetic
acid that are probably insufficient to induce hepatic peroxisome proliferation. It is thought that humans,
like rats, have lower rates of oxidative metabolism and higher rates of conjugation than do mice.
Trichloroacetic acid produces hepatic tumors only in B6C3F1 mice, but dichloroacetic acid in-
duces them in mice and in F344 rats at exposures up to 5 g/L in drinking water for 104 weeks (Herren-
Freund et al. 1987; Bull et al. 1990; Daniel et al. 1992; DeAngelo et al. 1996, 1999; Pereira 1996; NRC
OCR for page 93
Review of Toxicologic Studies 93
2006). Dichloroacetic acid is a major metabolite of TCE in B6C3F1 mice but a minor metabolite in Spra-
gue-Dawley rats (Larson and Bull 1992). Marked liver enlargement and cytomegaly in dichloroacetic
acid-treated mice also indicate that induction of hepatic tumors depends on stimulation of increased cell
division secondary to hepatoxoic damage (Bull et al. 1990). Inhibition of dichloroacetic acid metabolism
by the parent compound at less than 1 to 500 µM (Kato-Weinstein et al. 1998) is thought to contribute to
the variation in mouse hepatic tumors observed at this dose range (Bull et al. 2002).
Choral hydrate induces hepatic tumors in male B6C3F1 mice but not in female mice or F344 male
rats (George et al. 2000; NTP 2002a,b; Leakey et al. 2003). Female B6C3F1 mice given choral hydrate in
water by oral gavage for 104 weeks at up to 100 mg/kg per day had no increase in hepatic tumors (NTP
2002a), whereas exposure at the same doses in two groups of male mice fed ad libitum (NTP 2002a,b) or
fed a calorie-controlled diet (Leakey et al. 2003) had increased incidences of hepatocellular adenoma or
carcinoma (combined). Dietary control of caloric intake in the latter study was thought to improve sur-
vival and to decrease interassay variation. Choral hydrate is metabolically converted to trichloroacetic
acid or dichloroacetic acid, and this contributes to its weak carcinogenicity. Overall, choral hydrate is an
ineffective hepatic carcinogen that induces tumors only in male mice.
An epidemiologic study was conducted of short-term clinical exposure to choral hydrate used as a
hypnosedative and possible cancer risk in humans (Haselkorn et al. 2006). An increasing risk of prostatic
cancer with chloral hydrate was found, but the trend wat not statistically significant. Thus, the authors
concluded that there was no persuasive evidence of a causal relationship between choral hydrate exposure
and cancer in humans, but they were unable to rule out a causal relationship because statistical power was
low. Trichloroacetic acid elicits hepatic tumors in mice with a phenotype typical of peroxisome prolifera-
tors, whereas dichloroacetic acid produces hepatic tumors with a distinctly different phenotype and also
increases tumor growth (Bull 2000; Thai et al. 2003).
The relevance of TCE- and PCE-induced hepatic tumors to humans has been the subject of a
great deal of research. Oral and inhalation carcinogenicity bioassays of TCE in rodents have shown that
adenocarcinomas are strain- and species-specific (that is, are limited to the B6C3F1 mouse). Haseman et
al. (1998) reported a spontaneous hepatic-tumor incidence of 42.2% in male control B6C3F1 mice used in
National Toxicology Program (NTP) studies. The NTP recently held a series of workshops to determine
whether another mouse strain and a rat strain should be adopted. In light of the high background hepatic-
tumor incidence, it was recommended that the NTP explore the use of multiple mouse strains (King-
Herbert and Thayer 2006).
It has been clearly established that the toxicokinetics (target-organ dosimetry) of TCE and PCE of
the mouse and the human are different (see Chapter 3). Mice absorbed substantially more TCE and PCE
because of their greater respiratory and alveolar ventilation rate, cardiac output and pulmonary blood flow
rate, and blood:air partition coefficient. Mice also metabolically activate substantially more of their ab-
sorbed doses to bioactive substances (Lipscomb et al. 1998). On an equivalent inhalation exposure to
PCE, rats exhibited markedly higher blood and urinary concentrations of trichloroacetic acid and di-
chloroacetic acid than humans (Volkel et al. 1998). The rats’ blood also contained much higher concen-
trations of protein adducts (Pahler et al. 1999). Physiologically based toxicokinetic models similarly pre-
dict that mice will produce higher target-organ (liver) doses of trichloroacetic acid than humans after
exposure to PCE (Clewell et al. 2005) and TCE (Clewell and Andersen 2004).
The primary mode of action of trichloroacetic acid, and to a smaller extent dichloroacetic acid, is
activation of PPARα. Stimulation of PPARα can enhance DNA replication, resulting in expansion of
some clones of hepatocytes and suppression of apoptosis, so initiated and precancerous cells will be
spared. Male wild-type mice dosed orally with TCE exhibit hepatocyte proliferation and changes in ex-
pression of genes involved in cell growth (Laughter et al. 2004). PPARα-null mice are refractory to those
effects, which are associated with carcinogenesis. Mice expressing human PPARα fail to show increases
in markers of cell proliferation and are resistant to liver cancer if treated with PPARα agonists (Morimura
et al. 2006; Yang et al. 2008). The concentration of PPARα in human cells is about 10% of that in the liv-
ers of rodents (Palmer et al. 1998; Klaunig et al. 2003; Lai 2004). The interpretation of mouse hepatic-
tumor induction in 2-year bioassays relative to the inducing compound’s mode of action, including induc-
OCR for page 94
Contaminated Water Supplies at Camp Lejeune—Assessing Potential Health Effects
94
tion of peroxisome proliferation, has been assessed in a human-relevance framework (Cohen et al. 2003,
2004; Meek et al. 2003; Holsapple et al. 2006; Meek 2008). The relevance of B6C3F1 mouse hepatic tu-
mors to humans is also weakened by the observations that the background incidence of hepatic tumors in
unexposed B6C3F1 mice is about 60% and that large numbers of chlorinated compounds induce such tu-
mors in mice (Gold and Slone 1995). The human is likely to be much less responsive toxicodynamically
than the mouse to the cellular effects of trichloroacetic acid and dichloroacetic acid.
Many toxicologists have judged that the mode of action for hepatic carcinogenesis observed in
mice after administration of peroxisome-proliferation-inducing drugs and other chemicals (such as TCE
and PCE) makes it unlikely that such chemicals pose a hepatic-cancer risk in humans (Cattley et al. 1998;
NTP 2000; Clewell and Andersen 2004; NRC 2006; Klaunig et al. 2007). It was concluded by the Na-
tional Research Council that the PPARα mode of action for liver cancer in mice is not relevant to humans
(NRC 2006). However, others have raised questions about the interpretation of PPARα actions and
whether it is the only relevant mode of action for such chemicals (Keshava and Caldwell 2006), and this
continues to be a subject of active debate (Peters et al. 2005; Klaunig et al. 2007; NRC 2008).
Toxicodynamic mechanisms of hepatic carcinogenicity other than peroxisome proliferation have
been explored. Both trichloroacetic acid and dichloroacetic acid apparently contribute to hepatic tumori-
genesis in mice (Bull et al. 2002; Caldwell and Keshava 2006). High, repeated doses of those TCE and
PCE metabolites initially stimulate and then depress the growth of normal liver cells (Bull 2000). That
may confer a growth advantage on initiated cells. Dichloroacetic acid at high concentrations also appears
to act by increasing the clonal expansion and decreasing apoptosis of such precancerous cells. Moderate
amounts of dichloroacetic acid are apparently produced from trichloroacetic acid and trichloroethanol in
mice, but only trace amounts were found in one of three studies of TCE-exposed humans (see Chapter 3).
It is important to recognize that stimulation or inhibition of cell growth through PPARα activation ceases
when the metabolites are eliminated (Miller et al. 2000). Thus, such alteration of cell signaling is not a
genotoxic mechanism of action. Very high concentrations of dichloroacetic acid and chloral hydrate have
a weak genotoxic action in vitro. Bull (2000) and Moore and Harrington Brock (2000), however, con-
clude that it is unlikely that those metabolites would cause tumors in any organ through genotoxocity or
mutagenicity at exposure concentrations relevant to humans.
Renal Effects
Toxicity
TCE has limited capacity to produce renal injury in rodents that are subjected to high oral expo-
sures for extended periods. Jonker et al. (1996), for example, gave female Wistar rats TCE at 500 mg/kg
by corn-oil gavage for 32 consecutive days. Urinalyses showed only slight increases in N-acetyl-β-
glucosaminidase and alkaline phosphatase activities. A comparable exposure to PCE produced somewhat
larger increases. Kidney weights were modestly increased by both chemicals. Microscopic examination
revealed multifocal areas of vacuolation and karyomegaly in the animals’ renal tubules. Male Eker rats
received TCE at 50, 100, 250, 500, or 1,000 mg/kg by corn-oil gavage 5 times a week for 13 weeks
(Mally et al. 2006). There were no changes in γ-glutamyltransferase activity or other urinary indexes of
renal cytotoxicity. There was tubular-cell proliferation at 250 mg/kg or greater and karyomegaly at 500
mg/kg or greater. Overt nephropathy was restricted to the 1,000-mg/kg group. Nephropathy has been a
common finding in rats and mice in chronic, high-dose cancer bioassays of TCE (NCI 1976; NTP 1986a,
1988, 1990a). Nephrosis and cytomegaly were more severe in the rats than in the mice, and male rats
were generally affected more severely than females. Cytomegaly was manifested as frank enlargement of
the cytoplasm and the nucleus of scattered tubular cells in the inner cortex and outer stripe of the medulla.
Karyomegaly was later observed in the proximal tubular epithelial cells of the pars recta. The affected
tubules were dilated, and the cells were flattened and elongated and contained enlarged, hyperchromatic
OCR for page 95
Review of Toxicologic Studies 95
nuclei with irregular shapes. A low incidence of renal tumors was seen consistently in several strains of
male rats in the bioassays.
TCE has also been found to have some adverse renal effects when inhaled acutely or repeatedly at
high concentrations for long periods. Proximal tubular damage was reported in male F344 rats exposed
for 6 h to TCE vapor at 1,000 or 2,000 ppm (Chakrabarti and Tuchweber 1988). Mensing et al. (2002)
subjected male F344 rats to TCE at 500 ppm for 6 h/day 5 days/week for 6 months. Glomerulonephritis
was seen on histopathologic examination, but urinary biomarkers of glomerular damage were not found.
Increases in urinary N-acetyl-β-glucosaminidase and low-molecular-weight proteins reflected mild
proximal tubular damage.
Adverse effects of TCE on the kidneys are due largely to metabolites formed via the glutathione
conjugation pathway (Lash et al. 2000b). As described in Chapter 3, conjugation of TCE with glutathione
to form S-(1,2-dichlorovinyl)glutathione (DCVG) occurs primarily in the liver. DCVG is secreted into
bile and blood. That in the bile is converted to S-(1,2-dichlorovinyl)-L-cysteine (DCVC), which is reab-
sorbed into the bloodstream. As noted in Chapter 3, humans have a lower capacity than rats to metabolize
TCE by the glutathione pathway. Lash et al. (1999) were able to detect DCVG in the blood of humans
who had inhaled TCE at 50 or 100 ppm for 4 h, but Bloemen et al. (2001) could not find DCVG or
DCVC in the urine of similarly exposed subjects. DCVG in the blood is taken up by the kidneys and me-
tabolized to DCVC by γ-glutamyltransferase and a dipeptidase. Lash et al. (2001b) observed the follow-
ing decreasing order of toxic potency in freshly isolated rat cortical cells: DCVC > DCVG >> TCE.
DCVC can be detoxified by acetylation and activated further by two pathways: (1) cleavage by renal cy-
tosolic and mitochondrial β-lyases to dichlorothioketene, which in turn can lose a chloride ion to yield
chlorothioketene or tautomerize to form chlorothionacyl chloride (the latter two moieties are very reactive
and acylate proteins and DNA), and (2) oxidation by renal cytochrome P-450s or flavin-containing mono-
xygenases to the epoxide, DCVC sulfoxide (DCVCS). Lash et al. (1994) reported that DCVCS was a
more potent nephrotoxin than DCVC in vitro and in vivo in rats. Apoptosis was observed after as little as
1 h of incubation of cultured human renal proximal tubular cells with DCVC and DCVCS (Lash et al.
2003, 2005). Cellular proliferation accompanied by increased expression of proteins associated with cel-
lular growth, differentiation, stress, and apoptosis was also an early response to low doses. Necrosis,
however, was a late, high-dose phenomenon in this cell system. Exposure of human renal proximal tubu-
lar cells to DCVC at lower concentrations for 10 days also resulted in expression of genes associated with
cell proliferation, apoptosis, and stress (Lash et al. 2005) and repair and DCVC metabolism (Lash et al.
2006).
Proximal tubular-cell damage, as discussed above, appears to be a prerequisite for renal-cell can-
cer. Bruning et al. (1996) observed urinary protein-excretion patterns indicative of tubular damage in all
of a group of 17 workers exposed for years to peak TCE vapor concentrations that caused CNS depres-
sion. They later reported small increases in urinary excretion of glutathione S-transferase α and α1-
microglobulin in a group of 39 cardboard workers without renal-cell cancer who had been heavily ex-
posed to TCE for about 16 years (Bruning et al. 1999). Both indexes are markers of proximal tubular in-
jury. Higher α1-microglobulin excretion was reported in renal-cell cancer patients with TCE exposure
than in renal-cell cancer patients without TCE exposure in an updated study (Bolt et al. 2004). Green et
al. (2004) described similar findings in 70 electronics workers who inhaled TCE at an average concentra-
tion of 32 ppm for about 4 years. A battery of tests for nephrotoxicity was assessed after 4 days of expo-
sure. Urinary albumin and N-acetyl-β-glucosaminidase were higher than in controls, although there was
no correlation with the magnitude or duration of TCE exposure. There was also a suggested increase in
urinary glutathione S-transferase α activity that correlated with the intensity but not with the years of ex-
posure. Finally, Bruning et al. (1998) evaluated renal damage in a man who ingested about 70 mL of TCE
in a suicide attempt. He was rendered unconscious for 5 days. His urinary glucose and protein concentra-
tions were normal, but α1- and β2-microglobulin, N-acetyl-β-glucosaminidase, and several low-molecular-
weight protein concentrations were increased. Such modest, reversible signs of renal injury demonstrate
that TCE, even in extreme exposure conditions, has quite small nephrotoxic potential in humans.
OCR for page 96
Contaminated Water Supplies at Camp Lejeune—Assessing Potential Health Effects
96
Cancer
TCE was given in corn oil to F344/N rats and B6C3F1 mice of both sexes by oral gavage at doses
up to 1,000 mg/kg in rats and 6,000 mg/kg in mice in a 13-week study and up to 1,000 mg/kg in both spe-
cies and sexes in a 103-week study (NTP 1990a). Two-year oral-gavage studies in four additional rat
strains were also conducted (NTP 1988). Nonneoplastic renal lesions were found in all animals dosed for
2 years. In all strains of rats tested, cytomegaly and karyomegaly of tubular cells in the renal corticome-
dullary region were observed. Frank toxic nephropathy was observed with higher frequency beginning at
52 weeks of exposure. A statistically significant increase in renal-tumor incidence was observed only in
male F344/N rats exposed to TCE at 1,000 mg/kg for 2 years (this was the LOAEL). TCE has been
shown to cause toxicity in proximal renal tubules in vivo; results of in vitro studies have also indicated
toxicity of TCE and its metabolite DCVC in primary cultures of rat tubular cells (Cummings et al. 2000).
Nephrotoxicity was reported in Long-Evans rats after 6 months of inhalation exposure to TCE at
500 ppm (Mensing et al. 2002). The urinary-protein profile reported is consistent with impairment of tu-
bular reabsorption of filtered protein. Inhalation studies were conducted in both sexes of Sprague-Dawley
rats with TCE at 100, 300, and 600 ppm for 2 years and in Swiss mice at 100 and 600 ppm for 78 weeks
(Maltoni et al. 1988a). Renal adenocarcinomas were reported in male rats at 600 ppm (the LOAEL), but
no renal effects were observed in mice. Cytokaryomegaly or megalonucleocytosis was observed at the
end of 2 years of exposure in male rats (77% of the 600-ppm group and 17% of the 300-ppm group) with
no indication of pathologic conditions earlier.
Inconclusive evidence of induction of α2µ-globulin by TCE, formic acid formation, or peroxisome
proliferation as a mechanism or mode of action of TCE as a renal carcinogen was found (Goldsworthy et
al. 1988; Green et al. 2003).
Results of animal studies indicate that kidney cancer occurs at high doses (for example, 1,000
mg/kg and 600 ppm) in male rats and is preceded by nephrotoxicity affecting the proximal tubule. An
analysis by the U.S. Environmental Protection Agency with pooling across strains indicated a modest tu-
mor effect in female rats (EPA 2001). Renal-cell cancers observed in German workers who were highly
exposed to TCE have generally been assumed to be due to an initiating genotoxic effect of DCVC or
DCVC coupled with the promoting effects of recurring cytotoxicity and compensatory hyperplasia (Brun-
ing and Bolt 2000). The complete TCE glutathione conjugation pathway and assumed penultimate
nephrotoxic metabolites are described in Chapter 3. It has been proposed that exposures below nephro-
toxic concentrations or some threshold of exposure probably pose no risk of cancer in that nephrotoxicity
is deemed to be a prerequisite for development of kidney cancer (Bruning and Bolt 2000; Harth et al.
2005). TCE oxidative metabolizing enzymes (such as CYP2E1 and CYP3A5 isoforms) have polymorphic
forms. Known human population diversity in bioactivation and detoxification capabilities is an additional
consideration in determining the exposure concentration below which nephrotoxicity is unlikely. For TCE
inhalation exposure in the occupational setting, the suggested practical threshold below which nephrotox-
icity is unlikely to occur is 250 ppm as an 8-h time-weighted average (Harth et al. 2005).
In humans, inactivation of the von Hippel-Landau (VHL) tumor-suppressor gene is responsible
for the hereditary VHL cancer syndrome. Affected people are predisposed to a variety of tumors; more
than 80% of sporadic renal-cell carcinomas are associated with inactivation of this gene. Brauch et al.
(2004) noted that renal-cell cancer patients unexposed to TCE did not have the somatic VHL gene muta-
tional characteristics of TCE-exposed renal-cell cancer patients. According to Moore and Harrington-
Brock (2000), TCE itself has little if any mutagenic potential, and it is unlikely that any TCE-induced tu-
mors would be mediated by its major oxidative metabolites. TCE recently also yielded negative results
when tested in a Salmonella typhimurium strain (Ames test) that contained DNA coding for cytochrome
P-450 reductase, cytochrome b5, and cytochrome P-450 2E1 (Emmert et al. 2006). TCE glutathione-
conjugated metabolites DCVG and DCVC have, however, been shown to have genotoxic effects in in
vitro test systems.
A recent study provides insight into a TCE renal-carcinogenesis threshold proposal. A strain of
rats (Eker) uniquely susceptible to renal carcinogens was exposed to TCE at an administered dose of 100,
OCR for page 97
Review of Toxicologic Studies 97
250, 500, and 1,000 mg/kg by gavage 5 days/week for 13 weeks (Mally et al. 2006). The Eker rat is a
unique animal model for renal-cell carcinoma, carrying a germ-line alteration of the Tsc-2 tumor-
suppressor gene. Results showed a significant increase in cell proliferation in renal tubular cells but no
increased preneoplastic renal lesions or tumor incidence. In vitro studies were conducted on primary Eker
rat renal epithelial cells by exposing them to the TCE metabolite DCVC dissolved in water at 10-50 µM
for 8, 24, and 72 h. Concentrations of DCVC that reduced rat renal-cell survival to 50% also resulted in
cell transformation. No carcinogen-specific mutations were identified in the VHL or Tsc-2 tumor-
suppressor genes in the transformed cells. Renal-cell carcinomas in the Eker rat have substantial similari-
ties to human renal-cell carcinomas. It is not entirely clear that this or any contemporary experimental
animal model adequately mirrors humans with regard to the effects of TCE-induced mutations in the VHL
gene, but the authors firmly suggest that TCE-mediated renal carcinogenicity may occur only secondarily
to nephrotoxicity and sustained regenerative cell proliferation. The latter findings, coupled with the
aforementioned data of Lash et al. (2005, 2006), suggest that renal-cell cancer may result from prolonged,
high-dose cytotoxicity and sustained cell proliferation but that TCE’s metabolites may lack initiating ac-
tivity.
Both DCVC and its mercapturic acid metabolite N-acetyl-S-(1,2-dichlorovinyl)-L-cysteine have
been found in urine of humans exposed to TCE, and illustrates that the glutathione conjugation pathway is
active (Bernauer et al. 1996). Exposure of volunteers to TCE at 50 or 100 ppm showed that DCVG con-
centrations were 3.4 times higher in males than in females (Lash et al. 1999). Genes associated with
stress, apoptosis, cell proliferation, repair, and DCVC metabolism were up-regulated almost double in
cultured human renal tubular cells exposed to subcytotoxic doses of DCVC for 10 days (Lock et al.
2006). Male rats display higher reduced glutathione conjugation, γ-glutamyl transpeptidase, and cysteine
conjugate β-lyase activity than female rats. Taken together, results in the cited studies indicate that male
humans and male rats both possess significant glutathione conjugation capacity and can produce the criti-
cal TCE metabolite DCVC; renal carcinoma has been observed in male rats and male workers when both
have been exposed to high TCE concentrations for prolonged periods of time. These observations show
data congruence, indicating that the conjugation pathway plays a central role in induction of renal carci-
noma in males of both species. As discussed in Chapter 3, rats have greater capacity to metabolically ac-
tivate TCE by this pathway than humans.
Evaluation of potential risks to human health related to contaminants in water supplies is a central
concern of this project. Given the foregoing, it is sensible to begin to apply recent toxicologic information
to contemporary maximum environmental values. In summary, exposure to high TCE concentrations ap-
pears to lead to saturation of the oxidative metabolic pathway with an attendant pronounced increase in
metabolism via the glutathione-dependent pathway and likely increased production of penultimate toxic
metabolites, such as DCVC sulfoxide, chlorothioketene, and thionoacylchloride from DCVC (Dobrev et
al. 2002). As previously described, substantially larger quantities of these toxic moieties are produced
from TCE by rat kidney than by human kidney. In addition, cultured rat cortical cells have been shown to
be more susceptible to DCVC-induced necrosis than cultured human proximal tubular cells (Lash et al.
2001a). Human kidney cells have the capacity to metabolically activate and to respond adversely to low
concentrations of DCVC, but not to the extent exhibited by male rat kidneys.
Pulmonary Effects
Toxicity
The pulmonary-toxicity potential of TCE has been studied extensively in mice and rats; there ap-
pear to be no reports of TCE-induced lung injury in humans. Forkert et al. (1985) were among the first
scientists to describe lung toxicity in mice. Intraperitoneal injection of very high doses of TCE (2,000 and
2,500 mg/kg) into male CD mice rapidly caused damage of bronchiolar Clara cells and alveolar type II
cells, anesthesia, and a marked reduction in pulmonary cytochrome P-450 content. Female CD-1 mice
OCR for page 98
Contaminated Water Supplies at Camp Lejeune—Assessing Potential Health Effects
98
inhaling TCE at 20-2,000 ppm 6 h/day for up to 5 days exhibited dose-dependent vacuolation of Clara
cells (Odum et al. 1992). Pyknosis of the bronchiolar epithelium also occurred at the higher concentra-
tions. No morphologic changes were seen in the lungs of rats that were exposed to TCE vapor at 500 or
1,000 ppm. Isolated mouse Clara cells metabolized TCE to chloral, trichloroacetate, and trichloroethanol,
but no trichloroethanol glucuronide was detected. It was proposed that the inability of these cells to con-
jugate trichloroethanol with glucuronic acid led to accumulation of chloral to cytotoxic concentrations
(Odum et al. 1992; Green 2000). Forkert et al. (2005) found that oxidation of TCE to chloral was cata-
lyzed in murine lung microsomes by cytchrome P-450s 2E1, 2F2, and 2B1. Forkert et al. (2006) later
demonstrated that bioactivation of TCE by CYP2E1 and CYP2F2 occurred in Clara cells. Dichloroacetyl
lysine adducts were localized in Clara cells in the TCE-treated CD-1 mice, and CYP2E1 and CYP2F2 are
highly concentrated there (Forkert 1995). It is generally accepted that the cytotoxicity and possibly the
weak mutagenicity of chloral and diacetyl chloride contribute to the development of lung tumors in mice.
The mouse appears to be uniquely sensitive to TCE-induced pulmonary toxicity and cancer.
Mice, but not rats, developed lung tumors in the inhalation bioassays conducted by Fukuda et al. (1983)
and Maltoni et al. (1988a). Clara cells are numerous and present throughout the airways of mice. They are
found much less frequently in rats and are rare in humans (Green 2000). Mouse Clara cells contain con-
siderable amounts of smooth endoplasmic reticulum, a membrane network in which cytochrome P-450s
are bound. Human Clara cells are largely devoid of this organelle. Accordingly, metabolic activation of
TCE to chloral is high in mouse, much lower in rat, and undetectable in human microsomes (Green et al.
1997b). Green et al. measured high CYP2E1 concentrations in mouse lung microsomes; concentrations of
CYP2E1 were lower in rats and undetectable in humans. Mace et al. (1998), however, were able to detect
very low concentrations of CYP2E1 mRNA and protein in human peripheral lung tissue. Forkert et al.
(2005) found that male CD-1 mouse lung microsomes efficiently metabolize TCE to chloral hydrate,
whereas the reaction was observed—at low rates—in samples from only three of eight human donors.
Those findings suggest that TCE poses only a minimal risk of pulmonary toxicity in humans.
Cancer
TCE inhalation exposure caused an increased incidence of pulmonary tumors in ICR, Swiss, and
B6C3F1 mice but not in rats or hamsters. When female ICR mice were exposed to TCE at 150 and 450
ppm 7 h/day 5 days/week for 104 weeks followed by an observation period of 3 weeks, lung-tumor inci-
dence increased by a factor of 3 (Fukuda et al. 1983); epichlorohydrin was used as a TCE stabilizer in this
experiment. Female Sprague-Dawley rats exposed at the same concentrations for the same period had no
increase in lung tumors. Male Sprague-Dawley rats had no increase in lung tumors but did have an in-
crease in testicular and renal tumors after exposure to TCE at 600 ppm for 104 weeks but not at 100 or
300 ppm (Maltoni et al. 1986). Excess lung tumors were observed in Swiss mice and B6C3F1 mice ex-
posed to TCE at up to 600 ppm for 78 weeks (Maltoni et al. 1988a). Five gavage studies were also re-
viewed for induction of lung tumors in several strains of rats and mice; no excess lung tumors were found
(NRC 2006). These results, the information presented in the preceding section on pulmonary toxicity, and
the lack of reports of pulmonary injury and cancer in workers suggest that the risk of lung cancer in TCE-
exposed human populations is minimal.
Genotoxicity
TCE is a weak genotoxicant in a number of test systems (Bruning and Bolt 2000; Moore and Har-
rington-Brock 2000; NRC 2006). Genotoxicity generally includes mutational end points, cytogeneticity,
and primary DNA damage, whereas mutagenicity refers to the ability to induce heritable mutations. TCE
oxidative metabolites trichloroacetic acid, dichloroacetic acid, and chloral hydrate generally have shown
weak or no reactivity in mutagenicity tests; the weight of evidence in both in vitro and in vivo test sys-
OCR for page 99
Review of Toxicologic Studies 99
tems indicates that mutations are probably not key events in induction of cancer by these compounds
(Moore and Harrington-Brock 2000). TCE was negative in a Salmonella typhimurium test strain that had
cytochrome P-450 2E1 metabolizing capacity (Emmert et al. 2006).
Neonatal B6C3F1 mice were given chloral hydrate, trichloroacetic acid, and TCE by intraperito-
neal injection at the ages of 8 and 15 days; their livers were examined for 8-hydroxy-2′-deoxyguanosine
24 and 48 h and 7 days after the final dose (Von Tungeln et al. 2002). Mice treated with trichloroacetic
acid or chloral hydrate showed significantly higher DNA-8-hydroxy-2′-deoxyguanosine adduct formation
related to lipid peroxidation or oxidative stress; the authors concluded that male neonatal B6C3F1 mice
are not sensitive to induction of liver cancer by these compounds.
Significant increases in DNA migration in the Comet assay and micronuclei formation were re-
ported in human HepG2 cells after treatment with TCE at 0.5-4 mM (Hu et al. 2008). Increases in both 8-
hydroxy-2′-deoxyguanosine-DNA adducts and thiobarbituric acid-reactive substances were observed;
depletion of glutathione increased susceptibility to TCE-induced effects, whereas cotreatment with N-
acetylcysteine prevented the effects. That indicated that oxidative stress probably played a role in TCE-
induced genotoxic damage in those cells. Hypomethylated DNA was found in both dichloroacetic acid-
promoted and trichloroacetic acid-promoted mouse hepatic tumors in an initiation-promotion experiment
(Tao et al. 2004). Gene expression controlling cell growth, tissue remodeling, and xenobiotic metabolism
was altered in in dichloroacetic acid-induced mouse hepatic tumors (Thai et al. 2003). Overall evidence
indicates that TCE and the oxidative metabolites trichloroacetic acid, dichloroacetic acid, and chloral hy-
drate are unlikely to act primarily by a mutational or genotoxic mechanism as hepatic carcinogens.
The TCE glutathione conjugate DCVC has been shown to have genotoxic effects, including in-
creased reverse mutations in S. typhimurium tester strains, unscheduled DNA synthesis, and formation of
DNA adducts in vitro (Bruning and Bolt 2000; Moore and Harrington-Brock 2000). Genotoxicity meas-
ures in rodent kidneys and primary cultures of human renal cells showed significant dose-dependent in-
creases in results of the Comet assay (DNA single-strand breaks and alkali-labile sites) and in micronuclei
frequency with subtoxic concentrations of TCE (Robbiano et al. 2004). Among the six rodent renal car-
cinogens tested, TCE was among the ones that exhibited the lowest potency for these end points; nonethe-
less, the results indicated that TCE is genotoxic in renal cells isolated from rats and humans. In another
experiment, rats were exposed to TCE by inhalation or to DCVC by oral gavage. Proximal tubules iso-
lated from kidneys of treated rats were assessed for DNA damage with the Comet assay (Clay 2008).
Positive controls were included to demonstrate the sensitivity of the assay. Test results with TCE indi-
cated a negative response in this assay. DCVC showed slight effects in a few animals 2 h after treatment
and at the highest dose tested (10 mg/kg), but the effects were not strong enough to be considered posi-
tive. On the basis of those findings and other published data, the authors concluded that renal tumors seen
in bioassays are nongenotoxic in origin.
Reproductive Effects
Toxicity
Studies in Males
Several studies of the reproductive effects of TCE have been conducted, and many of these were
reviewed by the National Research Council (NRC 2006). Zenick et al. (1984) found reduced copulatory
behavior in male rats after an oral dose of 1,000 mg/kg per day 5 days/week for 6 weeks but indicated that
the changes may have been related to the narcotic effects of TCE. Mice exposed to TCE by inhalation 4
h/day for 5 days (Land et al. 1981) showed an increased percentage of abnormal sperm at 2,000 ppm, the
highest dose tested (about 3,000 mg/kg per day) and no increase at 200 ppm (about 300 mg/kg per day).
Kumar et al. (2000a,b) exposed male Wistar rats by inhalation to 376 ppm for 12 or 24 weeks (4 h/day 5
days/week) and reported decreased epididymal sperm count and motility, reduced testosterone concentra-
OCR for page 100
Contaminated Water Supplies at Camp Lejeune—Assessing Potential Health Effects
100
tions, and lower fertility when the treated rats were mated with untreated females. There were also sig-
nificant reductions in body weight, testicular weight, total cauda epididymal sperm count, and percentage
of motile sperm; the effects were greater after 24 weeks than after 12 weeks of exposure. By 24 weeks,
the testes were atrophied and had smaller seminiferous tubules. Sertoli cells were present, but tubules
contained no spermatocytes, and spermatids and Leydig cells were hypoplastic (Kumar et al. 2001). Xu et
al. (2004) exposed male mice by inhalation to TCE at 1,000 ppm 6 h/day 5 days/week for 1-6 weeks and
found no effects except for a significant reduction in the fertilizing ability of sperm from the TCE-
exposed males when they were combined in vitro with eggs from superovulated control females or when
the males were mated with superovulated control females. A study in male rabbits (Veeramachaneni et al.
2001) reported that a mixture of several agents, including TCE, caused alterations in mating desire and
ability, sperm quality, and Leydig-cell function. The effects were assessed subjectively, and it is difficult
to determine the contribution of TCE to the changes seen.
Forkert et al. (2002) demonstrated that CYP2E1 is involved in the metabolism of TCE to chloral
in Leydig cells and epididymides. Greater sensitivity of the mouse epididymis to high TCE vapor expo-
sures correlated with greater chloral formation and higher concentrations of CYP2E1 in the epididymis
than in the testis. Forkert et al. (2003) later found CYP2E1 in human epididymal epithelium and Leydig
cells. Seminal-fluid samples from eight TCE-exposed mechanics who had diagnoses of clinical infertility
contained TCE and some of its oxidative metabolites. More recently, Kan et al. (2007) evaluated epidi-
dymal damage by TCE at the light-microscopic and electron-microscopic levels in mice after inhalation at
1,000 ppm for 1 day or for 1, 2, 3, or 4 weeks. The study showed epithelial sloughing and degeneration
with separation of the seminal tubules from the basement membrane after exposure for 1 week or more.
Epididymal damage became more severe with increasing duration of exposure. DuTeaux et al. (2003)
found CYP2E1 and dichloroacetyl adducts in the epididymis and afferent ducts, which were indicative of
the formation of reactive cytotoxic metabolites in the cells that were damaged. The absence of mitochon-
drial β-lyase and the lack of formation of protein adducts in the epididymis and afferent ducts of rats
dosed with DCVC suggest that TCE metabolites formed via the glutathione conjugation pathway do not
participate in male reproductive toxicity. DuTeaux et al. (2004a,b) investigated the bioactivation of TCE
and adduct formation in the testis and epididymis. In male rats ingesting TCE at estimated doses of 1.6-
2.0 and 3.4-3.7 mg/kg per day in drinking water for 14 days, there was a dose-dependent reduction in ca-
pacity for in vitro fertilization of ova from untreated females. That effect occurred in the absence of any
apparent alteration in the sperm other than a dose-dependent increase in oxidized proteins. The increase in
lipid peroxidation implicates CYP2E1-mediated formation of reactive metabolites as a mechanism of tox-
icity.
Studies in Females
Manson et al. (1984) exposed female rats orally by gavage to TCE at 10, 100, or 1,000 mg/kg per
day for 2 weeks before mating, 1 week during mating, and throughout gestation. Although high concen-
trations of TCE were measured in fat, adrenal glands, and ovaries, and uterine tissue contained high con-
centrations of trichloroacetic acid, female fertility was not affected. However, 17% of females in the high-
dose group died, and weight gain was significantly reduced. Neonatal survival was also significantly re-
duced at the high dose, particularly in female offspring.
Cosby and Dukelow (1992) conducted a study of oral exposure of pregnant mice to TCE at 24 or
240 mg/kg per day during gestation and in vitro fertilization studies with TCE, trichloroacetic acid, di-
chloroacetic acid, and trichloroethanol. No effects were noted in the in vivo study; in the in vitro studies,
there was a dose-related decrease in the percentage of fertilized embryos with trichloroacetic acid, di-
chloroacetic acid, and trichloroethanol but not with TCE.
Female rats were exposed to several male reproductive toxicants, including TCE, at 0.45% in
drinking water for 2 weeks (Berger and Horner 2003). Oocytes collected after induced ovulation were
incubated with sperm from unexposed males. The percentage of oocytes fertilized, the number of pene-
OCR for page 123
Review of Toxicologic Studies 123
Renal Effects
Toxicity
TCE has little ability to cause renal damage in rodents subjected to high oral or inhalation expo-
sures for extended periods. A LOAEL of 500 mg/kg was found for mild renal injury in rats gavaged daily
for 1 month. LOAELs of 250 and 500 mg/kg for proximal tubular-cell proliferation and karyomegaly,
respectively, have been reported. Those responses were observed in male rats exposed orally five times a
week for 13 weeks. Nephrosis occurs more commonly and is more serious in rats than in mice in lifetime
cancer bioassays. The damage is apparently caused by reactive metabolites of the glutathione conjugation
pathway. That pathway is similar qualitatively, but not quantitatively, in rats and humans (rats metaboli-
cally activate about 10 times as much). Some workers exposed chronically by inhalation and dermally to
TCE sufficient to produce neurologic effects experience renal epithelial toxicity.
Cancer
Chronic exposure to TCE at 1,000 mg/kg per day orally or 600 ppm by inhalation causes satura-
tion of the oxidative metabolic pathway, which leads to increased formation of metabolites via the glu-
tathione pathway. Some of the metabolites are cytotoxic and mutagenic. Male rats, but not female rats and
not mice of either sex, exhibit a low incidence of renal-cell carcinoma when subjected to TCE at the
aforementioned doses for their lifetimes. Increased rates of renal-cell cancer are also reported in some
workers exposed for years to concentrations of TCE high enough to produce CNS effects and renal injury.
The recurring cytotoxicity and compensatory cellular proliferation are thought to be prerequisites for re-
nal-cell carcinoma (that is, coupled with the initiating action of mutagenic glutathione metabolites they
act as promoters).
Pulmonary Effects
Toxicity
Mice appear to be uniquely sensitive to pulmonary injury by TCE vapor. No reports of lung dam-
age after TCE ingestion were located. Vacuolation of Clara cells was observed in mice that inhaled TCE
at concentrations as low as 20 ppm 6 h/day for 5 days. Clara cells are nonciliated bronchiolar mucosal
cells that have high CYP2E1 and CYP2F2 activities. The cytochrome P-450s catalyze the oxidation of
TCE to chloral and diacetyl chloride, two putative cytotoxic and weakly mutagenic metabolites. Clara
cells are numerous and are present throughout mouse airways; they are much less frequent in rats and rare
in humans. CYP2E1 activity and TCE metabolism are undetectable in human lung preparations.
Cancer
Chronic TCE exposure has caused increased incidence of lung cancer in three strains of mice but
not in rats. Lung tumors have not been seen in mice or rats in five oral TCE bioassays. That may be be-
cause presystemic elimination of the orally administered chemical reduced the TCE that reached pulmo-
nary tissues. The TCE-induced mouse lung tumors are not considered relevant to humans since mouse
lung tumors are associated with Clara cells containing high CYP2E1 metabolizing activity and human
lung contains few Clara cells and undetectable CYP2E1 activity.
OCR for page 124
Contaminated Water Supplies at Camp Lejeune—Assessing Potential Health Effects
124
Fertility, Reproductive, and Developmental Effects
Effects of TCE on fertility and reproduction have been seen in several investigations in rodents.
In most cases, there were signs of general toxicity (such as body-weight and organ-weight changes and
CNS depression) at the same exposure concentrations. Male rats exposed to TCE at 376 ppm 4 h/day 5
days/week for 12 or 24 weeks exhibited reduced body-weight gain, spermatoxicity, and reduced fecun-
dity. CYP2E1, chloral formation, and dichloroacetyl adducts were found in testicular Leydig cells and
epididymides of rats and were indicative of production of cytotoxic oxidative metabolites of TCE in the
cells that were damaged. CYP2E1 has been found in human epididymal epithelium and Leydig cells.
Some TCE oxidative metabolites have been identified in seminal fluid of TCE-exposed mechanics, al-
though the relative metabolic capacities of human and rodent tissues have not been established. DuTeaux
et al. (2004a,b) reported a dose-dependent reduction in the ability of sperm from TCE-treated rats to pene-
trate ova from untreated females in vitro. The male rats ingested TCE at estimated doses of 1.6-3.7 mg/kg
per day in drinking water for 14 days. Replication of those findings and further studies of the toxicologic
and human significance of that sperm effect are warranted.
Pregnancy outcomes were generally not affected by exposure to TCE at concentrations high
enough to be maternally toxic, and there was no evidence of second-generation effects. Previously, there
had been reports of cardiovascular defects in offspring of rodents exposed to TCE during gestation. More
recently, well-conducted definitive experiments and a robust database have ruled out such developmental
anomalies. The possibility of developmental neurotoxicity and immunotoxicity was raised in several pub-
lications. Further research is needed to determine whether those results can be duplicated and, if so, to
expand the scope of investigation and assess the human relevance.
Cancer
Leydig cell adenoma has been found in male rats in a 2-year oral and a 2-year inhalation cancer
bioassay of TCE. It is the most frequently encountered testicular tumor in mice and rats. The spontaneous
incidence in old F344 rats is as high as 90%. Most human testicular cancers originate in germ cells or Ser-
toli cells and occur in young or middle-aged men. Leydig cell adenoma is rare in men, so spontaneous or
TCE-induced Leydig cell adenoma is of questionable relevance to humans.
Neurologic Effects
TCE, like many other lipophilic VOCs, inhibits CNS functions as long as it is present at a suffi-
cient concentration in neuronal membranes. Acute effects in humans are usually reversible and range
from fatigue and dizziness to intoxication and anesthesia. A number of studies of human subjects have
concurred that the inhalation LOAEL for impairment of motor or cognitive functions is 100-200 ppm for
several hours. Residual neurotoxic effects (such as trigeminal and olfactory nerve impairment) have been
reported in some workers exposed for years to vapor at concentrations that were probably in that range.
Auditory deficits, reduced performance of tasks, and other effects were observed in more highly exposed
rats, but tolerance usually developed over days or weeks of exposure. LOAELs of 350 and 50 ppm have
been reported for changes in visual evoked potentials in rabbits and decreased wakefulness in rats, respec-
tively. The toxicologic significance of those responses in rodents that inhaled TCE several hours a day for
weeks has not been established. No definitive oral neurologic studies of TCE were located.
Immunologic Effects
TCE causes allergic sensitization in animal studies, including contact dermatitis and exacerbation
of asthma. Some of those effects have been reported in humans after chronic occupational exposure to
OCR for page 125
Review of Toxicologic Studies 125
VOCs by inhalation at relatively high concentrations, but further studies are needed to determine whether
TCE can induce or modulate allergic diseases in humans. Immunosuppression has also been shown in
animal studies after TCE exposure, but it is unclear whether the effects are relevant to humans. Workers
exposed to TCE showed increases in IL-2 and IFN-γ and an increase IL-4, but interpretation of these
changes is difficult, and the data are too sparse to support definitive conclusions. Toxicologic studies have
also shown exacerbation of autoimmune diseases in a genetically modified mouse model (MRL+/+). The
relevance of those findings to humans is unclear, although epidemiologic studies have shown a relation-
ship between solvent exposure and scleroderma, glomerulonephritis, and other immune-related diseases
(see Chapter 5).
Tetrachloroethylene
Hepatic Effects
Toxicity
PCE, like TCE, has little ability to cause acute, subacute, or chronic hepatotoxicity in rodents or
humans. PCE is somewhat more potent because of formation of some additional reactive metabolites. An
acute oral LOAEL of 150 mg/kg was reported by Philip et al. (2007), but the serum concentration of a
liver-specific enzyme in mice progressively declined as the mice were treated over 30 consecutive days.
A NOAEL of 1,440 mg/kg per day was reported in rats that consumed PCE in drinking water for 90 days
(Hayes et al. 1986). As described in Chapter 3, ingestion of a chemical in divided doses over several
hours reduces its potency. In addition, rats are less susceptible than mice because of their lower capacity
for activating PCE metabolically. Humans have even lower capacity than rats.
Cancer
There is clear evidence that near-lifetime inhalation or ingestion of PCE, like that of TCE, results
in increased incidence of liver cancer in B6C3F1 mice. Similarly exposed rats do not develop hepatic tu-
mors. PCE’s LOAEL is 386 mg/kg for 78 weeks compared with TCE’s LOAEL of 1,000 mg/kg for 103
weeks. Trichloroacetic acid, a major metabolite of both PCE and TCE, produces peroxisome proliferation
in mouse liver but not rat or human liver. The very high spontaneous hepatic-tumor incidence in B6C3F1
mice and formation of substantially greater quantities of reactive metabolites suggest that mouse hepatic
tumors may be of little relevance to humans.
Renal Effects
Toxicity
PCE is somewhat more toxic to the kidneys than TCE. A LOAEL of PCE of 600 mg/kg per day
for renal damage was found in rats gavaged for 32 consecutive days. In contrast, consumption of PCE at
up to 1,400 mg/kg per day in drinking water for 90 days failed to damage rats’ kidneys. That discrepancy
can be attributed largely to the kidneys’ receipt of lower tissue doses when exposure was in drinking wa-
ter. A NOAEL of 400 ppm and a LOAEL of 1,000 ppm are described for nephrotoxicity in rats that in-
haled PCE several hours a day for a month or more. Karyomegaly was seen in the renal tubular cells of
mice and rats that inhaled PCE chronically at as low as 100 and 200 ppm, respectively; the nuclear
enlargement may be a predecessor of neoplasia, but a definite link has not been established. Renal effects
of PCE are due primarily to metabolites formed via the glutathione conjugation pathway. Equivalent inha-
OCR for page 126
Contaminated Water Supplies at Camp Lejeune—Assessing Potential Health Effects
126
lation exposures of rats and humans to PCE at 160 ppm for 6 h showed that biotransformation by the glu-
tathione metabolic pathway was 10 times greater in the rats (Volkel et al. 1998).
Cancer
Chronic inhalation of PCE at 200 or 400 ppm produced renal tubular-cell karyomegaly, hyperpla-
sia and a low incidence of tubular-cell adenoma and carcinoma in male rats. Renal tumors did not occur
in female rats or in mice of either sex, although these animals did exhibit karyomegaly.
Pulmonary Effects
Toxicity
There is little evidence of lung injury by inhaled PCE in laboratory animals or humans. Inhalation
experiments with human subjects indicate a NOAEL of 150 ppm and a LOAEL of 200-300 ppm for mild
irritation of nasal passages. Pulmonary-function measurements do not reveal decrements at those concen-
trations. Intermittent inhalation of PCE at 1,600 ppm for 13 weeks produced pulmonary congestion in
rats; 800 pm did not. There is one report (Aoki et al. 1994) of epithelial degeneration in mice that inhaled
PCE at 300 ppm 6 h/day for 5 days. The change was more severe in the olfactory than in the respiratory
mucosa.
Cancer
No increases in proliferative lesions or neoplasms of the respiratory tract have been seen in a
chronic oral or inhalation cancer bioassay in mice and rats. Although CYP2E1 is abundant in mouse lung,
that cytochrome P-450 isozyme is not active as a catalyst of PCE metabolism in the respiratory tract of
other rodents or humans.
Other Cancers
An increased incidence of mononuclear-cell leukemia was found in male and female F344 rats
that inhaled PCE at 200 or 400 ppm for 103 weeks. The increases were not dose-dependent and were
within the incidence range of mononuclear-cell leukemia often seen in control F344 rats. The NTP is no
longer using the F344 strain in its cancer bioassay program, because of its high rates of spontaneous can-
cer of several types. Mononuclear-cell leukemia is rare in people. Thus, that form of leukemia in F344
rats has been judged not to be relevant to humans. Animal cancer bioassay outcomes relevant to human
leukemia, multiple myeloma, and non-Hodgkin lymphoma have not been reported.
Fertility, Reproductive, and Developmental Effects
Information on potential effects of PCE on fertility and reproduction is limited. Inhalation of PCE
for 5 days did not affect sperm morphology in rats but did result in increased incidence of abnormal
sperm heads in mice. The NOAEL and LOAEL for that effect were 100 and 500 ppm, respectively. Long-
term exposure of male and female rats to PCE vapor for two generations resulted in CNS depression, de-
creased body weight during lactation, and nephrotoxicity at 1,000 ppm. There were reductions in live
births, litter size, survival, and body weight in the F2 progeny at that vapor concentration. Those adverse
effects may be secondary to maternal body-weight loss and toxicity. More data are needed to clarify the
effects of PCE on reproductive function.
OCR for page 127
Review of Toxicologic Studies 127
A number of oral and inhalation studies of potential developmental effects of PCE have been
conducted in rodents. Experimental protocols have included inhalation of PCE at 300-1,000 ppm before,
during, or after pregnancy. Manifestations of developmental delay (such as reduced ossification of verte-
brae and soft-tissue dysplasias) have been reported in pups at the relatively high concentration. Ingestion
of PCE at 900 mg/kg per day on days 6-19 of gestation, for example, resulted in increased resorptions,
reduced weight, and microphthalmia or anophthalmia in rat pups. That daily dose was so high that mater-
nal ataxia and weight loss occurred. Developmental effects at lower concentrations were relatively minor
and were not indicative of teratogenicity.
Neurotoxicity
Neurologic Effects
Ingestion and inhalation of sufficient doses of PCE produce CNS depression in rodents and hu-
mans. Because PCE is more lipophilic than TCE, it is moderately more potent as a CNS depressant. Defi-
cits in neurophysiologic functions have been reported in volunteers exposed to PCE at as low as 50 ppm
for 4 h/day for 4 days (Altmann et al. 1990, 1992). A number of animal studies have revealed neurobe-
havioral and neurochemical changes in the brains of animals that inhaled PCE at several hundred parts per
million for various periods. Mattsson et al. (1998), for example, found altered flash-evoked potentials in
rats after 13 weeks of exposure at 800 ppm, but not at 200 ppm. Wang et al. (1993) measured decreases in
regional brain weight, DNA content, and glial proteins in rats exposed continuously to PCE at 600 ppm
for 4 or 12 weeks. Few researchers, however, have evaluated PCE-induced neurobehavioral and neuro-
chemical changes in the same animals, so interpretation of many of the data is difficult.
Neurodevelopmental Effects
Concerns about possible neurodevelopmental effects in children exposed to PCE prompted sev-
eral investigations in animals. Chen et al. (2002), for example, described changes in locomotor activity,
pain threshold, and pentylenetetrazol-induced seizure thresholds in young rats dosed orally with PCE at
50 mg/kg per day for 8 weeks. Exposure of pregnant rats to PCE at 900 ppm resulted in pups with dimin-
ished brain acetylcholine and dopamine concentrations and with neurobehavioral changes on certain days
of testing; inhalation of PCE at 100 ppm was without effect. Such reports suggest that there may be peri-
ods of neurologic development during which sufficiently high PCE exposures are detrimental. Additional
research is needed to determine whether gestational, neonatal, or childhood exposure to such solvents can
impair CNS development and function.
Immunologic Effects
Little information is available on the potential of PCE to suppress the immune system or to in-
duce autoimmune diseases. In one study, PCE was found to suppress natural-killer-cell and T-cell activity
in vitro but to have no effect on rats in vivo. In a second study, inhalation of PCE at 50 ppm reduced bac-
tericidal activity in mice subjected to inhaled microorganisms. Further investigations of PCE are war-
ranted in light of the apparent effects of TCE on the immune system.
HAZARD EVALUATION OF TRICHLOROETHYLENE AND PERCHLOROETHYLENE
EXPOSURE FOR SELECTED END POINTS
The committee used several approaches to consider the health significance of the solvents found
in the water supply at Camp Lejeune. Hazard can be defined as the intrinsic characteristic toxicity of a
OCR for page 128
Contaminated Water Supplies at Camp Lejeune—Assessing Potential Health Effects
128
chemical compound. The hazard evaluation provides information on the inherent toxic potential of an ex-
posure and is not meant to provide a quantitative estimate of risk. This approach compares the lowest
doses of TCE and PCE at which adverse effects were observed in laboratory animals (the LOAELs) with
a range of estimated doses from the Camp Lejeune water supply. It is one line of evidence in assessing
possible relationships between exposure to TCE and PCE in water at Camp Lejeune and potential health
effects.
The lowest dose at which an adverse health effect was observed, the LOAEL, may be subject to
some uncertainty, depending on a number of factors, including the doses that were studied, the end point
chosen, and the method used to assess the end point; for example, death as an observed LOAEL end point
is more certain than a subtle change in an end point that is reversible and of unknown biologic signifi-
cance. LOAELs from animal studies, on average, are associated with a 10% increase in response rate and
can be associated with various risk levels because the statistical power of the studies does not allow ob-
servation of lower levels of exposure. Thus, LOAELs do not define a level below which no adverse ef-
fects can occur. Nevertheless, determination of a LOAEL generally provides a useful measure of toxic
potency. NOAELs are hampered by more uncertainty. A NOAEL is the highest experimental dose at
which an adverse effect did not occur. An experimentally determined NOAEL may be substantially lower
than the actual NOAEL if the doses administered were too low. The present hazard evaluation was based
on LOAELs for selected toxicity end points as described below.
The toxicologic databases on TCE and PCE are extensive, but some data gaps remain for a few
end points. LOAELs observed in animal studies selected for this dose comparison represent a range of
adverse effects and oral doses. The particular end points were chosen in part because it was assumed that
they may be relevant to humans. For TCE, renal tumors in rats were chosen for a chronic high-dose end
point (LOAEL, 1,000 mg/kg per day for lifetime oral exposure [NTP 1990a]), renal toxicity in rats was
chosen for the medium dosage range (LOAEL, 250 mg/kg per day for 13 weeks [Mally et al. 2006]), and
immunosuppression in a sensitive strain of mice was chosen at the lower end of the dosage spectrum
(LOAEL, 22 mg/kg per day in drinking water for 4 or 6 months [Sanders et al. 1982]) (see Figure 4-3 and
Table 4-3). For PCE: renal toxicity in rats (600 mg/kg per day for 32 days [Jonker et al. 1996]) was se-
lected at the upper end of a series of LOAELs, and neurologic changes in young rats (50 mg/kg per day
for 8 weeks [Chen et al. 2002]) at the lower end of LOAEL doses (see Figure 4-4 and Table 4-4).
Uncertainty is associated with the TCE and PCE water concentrations used in the hazard evalua-
tion because they are based on the relatively few mixed water samples analyzed (see Chapter 2). Only a
small set of water-quality measurements are available, and those were taken during the 5 years before the
contaminated wells were closed, so it is unknown how well they represented the conditions during the
preceding decades. In addition, concurrent exposures to organic solvents may have occurred at Camp Le-
jeune. Studies of mechanisms of VOC interactions (see Chapter 3) indicate that such concurrent exposure
is not likely to result in greater than an additive effect. Relatively low doses of multiple VOCs are
unlikely to affect the magnitude of adverse health effects appreciably. Additivity is not formally incorpo-
rated into this appraisal.
The exercise below is not a health risk assessment. Several assumptions (described below) were
used to derive the comparisons, so there is uncertainty and variability in the values. The intent is to pro-
vide general comparisons of the lowest doses at which specific adverse health effects were observed in
experimental toxicologic studies with a range of estimated contaminant concentrations that may have oc-
curred in the Camp Lejeune water supply.
The following describes the assumptions in the evaluation and illustrative calculations. To pro-
vide a standardized basis for comparison, the lowest doses at which a specific adverse effect was seen in
toxicologic studies and the exposure estimates are both expressed in standard terms of milligrams of
chemical per kilogram of body weight per day (mg/kg per day). Standard assumptions commonly used for
hazard evaluations are that adults weigh an average of 70 kg and drink an average of 2 L of water per day
and that children weigh an average of 10 kg and drink 1 L of water per day. Exposure via inhalation and
dermal absorption of VOCs from water during showering, bathing, dishwashing, and other household ac-
OCR for page 129
Review of Toxicologic Studies 129
TABLE 4-3 LOAELs from Animal Studies Used for Comparison with Estimated Daily Human Doses to
TCE Related to Water-Supply Measured Concentrations
Range of Doses End Point LOAEL, mg/kg per day
High Kidney cancer, rats 1,000
Medium Kidney toxicity, rats 250
Low Immunosuppression, mice (sensitive strain) 22
TABLE 4-4 From Animal Studies Used for Comparison with Estimated Daily Human Doses to PCE
Related to Water-Supply Measured Concentrations
Range of Doses End Point LOAEL, mg/kg per day
High Kidney toxicity, rats 600
Low Neurotoxicity, rats 50
tivities has been shown experimentally to account for as much exposure as that from drinking water that
contains the chemicals (see Chapter 3). Therefore, to account for potential inhalation and dermal uptake
in addition to ingestion in drinking water, an intake of 4 L/day is assumed for adults and 2 L/day for chil-
dren. This calculation, therefore, takes into account all three routes of exposure—ingestion, inhalation,
and dermal—of both adults and children. Considerable toxicologic data on VOCs are available from inha-
lation studies. The range of adverse effects is presented in Figures 4-1 and 4-2, but absorbed doses were
usually not determined. Duration of exposure is usually specified in animal studies. A conservative as-
sumption used in this hazard evaluation is that humans receive the stated dose daily, although that is very
unlikely inasmuch as data presented in Chapter 2 indicate that daily exposures were highly variable.
It is important to note that the evaluation has not taken into account uncertainties and additional
considerations (see Chapter 3) related to potentially sensitive populations (such as fetuses and the eld-
erly), possible human interindividual variability in response related to sex and genetic background, such
lifestyle factors as level of exercise , underlying diseases, and VOC interactions. Nevertheless, as dis-
cussed in Chapter 3, rodents absorb a greater fraction of inhaled VOCs and metabolically activate a sub-
stantially greater proportion of their internal dose and are therefore more susceptible than humans to most
adverse effects of TCE and PCE.
Chapter 2 summarizes the water-supply data available from the Tarawa Terrace and Hadnot Point
water systems. Among the measurements with reported values, TCE concentration in mixed water sam-
ples from the Hadnot Point water supply ranged from 1 to 1,400 µg/L (see Table 2-11). Water samples
with detectable PCE from the Tarawa Terrace water supply ranged from 1 to 215 µg/L (Maslia et al.
2007). Given the sparse information regarding the range and magnitude of contaminant concentrations in
the Camp Lejeune water supply, values that correspond to half the highest measured value, the highest
measured value, and twice the highest measured value were selected for this exercise: TCE at 700, 1,400,
and 2,800 µg/L and PCE at 100, 200, and 400 µg/L.
The following calculation was carried out to obtain an estimate of human daily exposure: esti-
mated human daily dose (mg/kg per day) = [mixed water concentration (µg/L) × estimated daily intake
(oral, inhalation, and dermal) (L/day)]/[body weight (kg)]. A sample calculation follows. For Hadnot
Point, the highest measured concentration of TCE in mixed water was 1,400 µg/L. For an adult human,
the daily dose received from water containing TCE at 1,400 µg/L is estimated to be
1,400 µg/L × 4 L/day = 80 µg/kg per day = 0.08 mg/kg per day.
70 kg
Half the highest measured TCE concentration in the water supply (700 µg/L) yields an estimated dose of
0.04 mg/kg per day for adults, and twice the highest measured concentration of TCE (2,800 µg/L) yields
OCR for page 130
Contaminated Water Supplies at Camp Lejeune—Assessing Potential Health Effects
130
an estimated dose of 0.2 mg/kg per day for adults. For a child, the daily dose received from water contain-
ing TCE at 1,400 µg/L is estimated to be
1,400 µg/L × 2 L/day = 280 µg/kg per day = 0.3 mg/kg per day.
10 kg
Half the highest measured TCE concentration in the water supply (700 µg/L) yields an estimated dose of
0.1 mg/kg per day for a child, and twice the highest measured concentration of TCE (2,800 µg/L) yields
an estimated dose of 0.6 mg/kg per day for a child.
Table 4-3 shows the LOAELs from animal studies used to compare with the estimated human
TCE doses related to a range of possible water-supply exposure concentrations. A comparison of
LOAELs for health end points selected from TCE animal studies with the exposure estimates is summa-
rized here:
Kidney cancer. The LOAEL of TCE for lifetime oral exposure leading to kidney cancer in the rat
is 1,000 mg/kg per day (NTP 1990a). The estimated human adult dose at Camp Lejeune is 25,000 times
lower than the LOAEL for exposure at half the highest water-supply concentration, 12,500 times lower
than the LOAEL for exposure at the highest concentration, and 5,000 time lower than the LOAEL for ex-
posure at twice the highest concentration for a lifetime exposure. For a child, the comparable estimates
are 10,000, 3,350, and 1,700 time lower than the LOAEL, respectively.
Renal toxicity. The LOAEL of TCE for renal toxicity in the rat dosed orally for 13 weeks is 250
mg/kg per day (Mally et al. 2006). The estimated human adult dose at Camp Lejeune is 6,250 times lower
than the LOAEL for exposure at half the highest water-supply concentration, 3,125 times lower than the
LOAEL for exposure at the highest concentration, and 1,250 times lower than the LOAEL for exposure at
twice the highest concentration. For a child, the comparable estimates are 2,500, 830, and 415 times lower
than the LOAEL, respectively.
Immunosuppression. The LOAEL of TCE for immunosuppression in a sensitive strain of mouse
ingesting TCE for 4 or 6 months is 22 mg/kg per day (Sanders et al. 1982). The estimated human adult
dose at Camp Lejeune is 550 times lower than the LOAEL for exposure at half the highest water-supply
concentration, 275 times lower than the LOAEL for exposure at the highest concentration, and 110 times
lower than the LOAEL for exposure at twice the highest concentration. For a child, the comparable esti-
mates are 220, 75, and 40 times lower than the LOAEL, respectively. These differences are relatively
smaller than for kidney cancer and kidney toxicity. As stated earlier in the chapter, uncertainties exist re-
garding this end point since there is relatively little toxicologic information on TCE and immune effects.
Additional research may be needed on the potential immunosuppressive effects of TCE.
For PCE, the daily dose received from water at the maximum measured concentration (200 µg/L)
in the water supply for an adult human is estimated to be
200 µg/L × 4 L/day = 0.01 mg/kg per day.
70 kg
Exposure to half the highest measured water supply concentration (100 µg/L) yields a dose of 0.006
mg/kg per day for an adult human and exposure to twice the highest measured water supply concentration
(400 µg/L) yields a dose of 0.02 mg/kg per day. For a child, the daily dose received from water contain-
ing PCE at the maximum measured concentration (200 µg/L) is estimated to be
200 µg/L × 2 L/day = 0.04 mg/kg per day.
10 kg
OCR for page 131
Review of Toxicologic Studies 131
Exposure to half the highest measured water supply concentration (100 µg/L) yields a dose of 0.02 mg/kg
per day for a child and exposure to twice the highest measured water supply concentration (400 µg/L)
yields a dose of 0.08 mg/kg per day.
A comparison of LOAELs for each of the two health end points selected from PCE animal studies
(Table 4-4) with the estimated doses from the water supply is summarized here:
Renal toxicity. The LOAEL for renal toxicity in the rat dosed orally with PCE for 32 days is 600
mg/kg per day (Jonker et al. 1996). The estimated human adult dose at Camp Lejeune is 100,000 times
lower than the LOAEL for exposure at half the highest water-supply concentration, 60,000 times lower
than the LOAEL for exposure at the highest concentration, and 30,000 times lower than the LOAEL for
exposure at twice the highest concentration. For a child, the estimates are 30,000, 15,000, and 7,500 times
lower than the LOAEL, respectively.
Neurotoxicity. The LOAEL of PCE for neurotoxic effects in rats is 50 mg/kg per day for 8 weeks
(Chen et al. 2002). The estimated human adult dose at Camp Lejeune is 8,300 times lower than the
LOAEL for exposure at half the highest water-supply concentration, 5,000 times lower than the LOAEL
for exposure at the highest concentration, and 2,500 times lower than the LOAEL for exposure at twice
the highest concentration. For a child, the comparable estimates are 2,500, 1,250, and 625 times lower
than the LOAEL, respectively. As noted earlier in this chapter, there is a need for additional research to
clarify the neurotoxic effects of PCE.
The comparisons above included health end points observed in animals that were considered
relevant to humans. Renal toxicity and cancer, neurotoxicity, and immune-related effects have been re-
ported in some epidemiology studies and in clinical reports. The dose comparisons1 suggest considerable
differences between the estimated doses from human exposure to contaminated water supplies at Camp
Lejeune under conservative assumptions of exposure and the lowest doses associated with the develop-
ment of renal toxicity, kidney cancer, neurotoxicity, and immunosuppression in rodents. The drinking-
water doses at Camp Lejeune are substantially lower. As pointed out in this section, however, each and
1
One member, Lianne Sheppard, objected to inclusion of the hazard evaluation in the report as written and
offered the following explanation: “Comparison of toxicology-based LOAEL values with estimated exposures to the
Camp Lejeune population uses questionable logic to support inference that adverse health effects are unlikely to
have occurred. Although LOAEL estimates give evidence about the presence of a hazard, they should not be used to
make inference about the absence of hazard at lower doses. The absence of evidence of a hazard (e.g., at levels be-
low the LOAEL) cannot be equated with evidence of the absence of hazard (Altman and Bland 1995; Fleming
2008). Because of their small sample size, animal studies are only able to identify hazards that induce high levels of
response (on average 10% increase in response for the LOAEL). Moreover, levels of excess response considered
acceptable in humans are much lower than 1 in 10, typically on the order of 1 in 10,000 to 1 in 1 million (EPA
2005). While low-dose extrapolation involves additional untestable assumptions, dividing the LOAELs by 1,000 to
100,000 provides an alternative approach to the informal hazard evaluation presented above. This second approach
compares Camp Lejeune exposures with an acceptable hazard in humans, as extrapolated from toxicologic studies.
The results lead to strikingly different conclusions because they yield acceptable hazards that are both larger and
smaller than the estimated exposures; indeed, some are several orders of magnitude lower than Camp Lejeune expo-
sures. Alternatively, standard practice would replace informal hazard evaluation with a formal risk assessment,
although this task was outside the committee charge. Despite my reservations on this one area of the assessment, I
support the overarching findings and recommendations of the report.”
Other members disagree with Dr. Sheppard’s characterization that the hazard evaluation is based on questionable
logic. The reasons for this are stated in the text. The validity of results using the approach she outlines above is ques-
tioned by some committee members. There were varying views among committee members on the value of the in-
formation generated by the hazard evaluation effort, ranging from members who found it quite useful because it
provided a rough benchmark for speculating about the likelihood of adverse health effects, to members who placed
less reliance on results, given limited exposure information and their uncertainty about the applicability of toxi-
cologic information. Regardless of the approach taken to the hazard evaluation, however, all committee members
strongly support the overarching findings and recommendations of the report.
OCR for page 132
Contaminated Water Supplies at Camp Lejeune—Assessing Potential Health Effects
132
every source of uncertainty (e.g., interindividual variability, lifestyle, genetic background, exposure as-
sessment, completeness of the database) has not been factored into this estimate since it is a hazard
evaluation procedure and not a health risk assessment.
ALLOWABLE LIMITS OF VOLATILE ORGANIC COMPOUNDS IN DRINKING WATER
Current regulatory standards termed maximum contaminant levels (MCLs) for several VOCs in
drinking water, including TCE and PCE, were developed by EPA in the middle 1980s (50 Fed. Reg.
46880 [1985]; 52 Fed. Reg. 25690 [1987]; Cotruvo 1988). Under the U.S. Safe Drinking Water Act, the
public-health goal or maximum contaminant level goal (MCLG) for a compound was initially determined.
The MCLG is the concentration that would result in “no known or anticipated adverse effect on health”
with a large margin of safety. Second, an MCL, or enforceable standard, was set as close as feasible to the
MCLG; technical and economic factors were taken into consideration. EPA consulted the International
Agency for Research on Cancer guidelines when assessing epidemiologic and animal cancer data and in
its own qualitative weight-of-evidence scheme for determining the potential for a compound to increase
cancer risk in humans. TCE and PCE fell into category I in the latter scheme, in which the MCLG by
definition equals zero as an aspirational goal. Economic considerations for water treatment were also de-
liberated. Technical feasibility focused on analytic considerations; the lowest concentrations that can be
reliably detected within specified limits of precision and accuracy during routine laboratory operations
(practical quantitation limits) were determined. With that approach, an MCL of 0.005 mg/L (5 µg/L or 5
ppb) was set for selected VOCs, including TCE and PCE.
In 2005, EPA issued new guidelines for carcinogen risk assessment in which incorporation of in-
creased scientific understanding of the biologic mechanisms that can cause cancer was supported for in-
clusion in risk assessments with other improved risk-assessment practices (EPA 2005). In the more than
20 years since the original MCLs were established, considerable kinetic and biologic mechanism-of-
action information on TCE and PCE has been published, as reviewed in the present report. There are dif-
ferent approaches to risk assessment that yield different results. At least one recent study has explored
different approaches, including the use of contemporary published elements of TCE’s biologic mode of
action and a cancer-risk model that was the best fit to the data (Clewell and Andersen 2004). The latter
approach yielded a TCE concentration of 265 µg/L in drinking water; below this concentration, a car-
cinogenic hazard to human health was deemed unlikely. This is one example of the possible application
of toxicologic and mechanistic biologic data to a cancer health risk assessment for TCE, which yields a
value greater than one based on analytical limits of detection. EPA is currently updating its risk assess-
ments on TCE and PCE and is considering new data and different assessment approaches as part of its
reassessments. In summary, the few TCE and PCE measurements available from mixed drinking-water
samples at Camp Lejuene (see Chapter 2) indicated that some samples exceeded the MCLs derived as
briefly described above.
CONCLUSIONS
TCE and PCE are well-studied compounds compared with most other compounds of environ-
mental concern. On the basis of the review presented above, the committee concludes that the strongest
evidence of health effects of relevance to humans are renal toxicity, kidney cancer, neurobehavioral ef-
fects, and immunologic effects, which have generally been observed at high concentrations in a work-
place setting and in exposure to tens to thousands of milligrams per kilogram of body weight in animal
studies. Discussion of the toxicologic evidence in context with the epidemiologic evidence on TCE and
PCE (presented in Chapter 5) is provided in Chapter 7. The evidence on renal toxicity and cancer is par-
ticularly convincing because concordance has been found in the bioactivation of TCE and PCE and in
their modes of action in rodents and humans. However, gaps in the toxicologic database preclude drawing
conclusions about some other health effects related to the nervous system and the immune system, par-
OCR for page 133
Review of Toxicologic Studies 133
ticularly with regard to potential effects on the developing or young animal. Implicit inherent limitations
of toxicologic studies are that relatively homogeneous populations of laboratory animals are used and ex-
posures are typically to single chemicals. On average, the lowest increase in effect that can usually be de-
tected (LOAEL) is around 10% due to statistical power related to the number of animals that can be tested
in any one study. In the instances of TCE and PCE, however, rodents are more susceptible to toxic ef-
fects.
A central issue in toxicology (and at Camp Lejeune) is whether doses were sufficient to produce
specific adverse effects. The lowest doses at which adverse health effects have been seen in animal or
clinical studies are many times higher than the worst-case (highest) assumed exposures at Camp Lejeune.
However, that does not rule out the possibility that other, more subtle health effects that have not been
well studied could occur, although it somewhat diminishes their likelihood.
Another important issue is whether any adverse effects that may have occurred were reversible or
permanent and (still) detectable when an epidemiology study might be conducted. Observations in animal
studies indicate that very high acute or chronic doses of TCE or PCE are necessary to injure renal proxi-
mal tubular cells. Results of occupational-exposure studies indicate that relatively high, chronic exposures
result in modest, reversible changes in the most sensitive indexes of renal injury in workers. Thus, it is
unlikely that renal toxicity would be a useful end point to examine in future epidemiology study of Camp
Lejeune residents. A similar conclusion can be drawn with regard to the occurrence and detection of he-
patic toxicity. Reproductive and developmental effects in rodents were quite modest and often secondary
to general toxicity, decreased food intake, and reduced body-weight gain resulting from high maternal
doses of TCE and PCE. The toxicologic data provide strong evidence that neither solvent is associated
with congenital malformations in rats. Thus, on the basis of this review, reproductive effects and hepa-
torenal toxicity are probably not of great concern at Camp Lejeune.
There is reasonable interspecies concordance between rats and humans in the bioactivation of
TCE and PCE and in their mode of induction of kidney cancer. A low incidence of kidney cancer has
been seen in workers exposed for many years to TCE at concentrations high enough to cause dizziness,
headache, and other reversible neurologic effects. The background incidence of kidney cancers in unex-
posed persons is minimal. Nevertheless, there is little likelihood of identifying any increased incidence of
renal tumors in the relatively small population that may be available for study at Camp Lejuene.
Irreversible neurobehavioral effects associated with solvent exposure generally are chronic and
result from high doses. Solvent abusers and workers chronically exposed to high vapor concentrations
may exhibit various neurobehavioral effects and residual brain damage. Fetuses, infants, and young chil-
dren exposed to such organic solvents as TCE and PCE at lower concentrations may experience subtle
neurodevelopmental effects, but no relevant investigations were identified. There are few data from ani-
mal studies on this topic.
Immune suppression and autoimmunity related to TCE exposure have been demonstrated in some
sensitive animal models. TCE-induced glomerulonephritis and scleroderma occur in low incidences in
highly exposed worker populations. Much less is known about the potential immunologic effects of PCE
(particularly as related to exposures during development), which may warrant further consideration for
inclusion in studies of populations exposed to TCE or PCE.
