Perchloroethylene (PCE) Toxicology and Neurotoxicity

1. Methodology Overview Chemical Identity and Background

Dr. Rosenfeld of RoseyFields conducts perchloroethylene (PCE) risk assessments using the following methodology: we characterize hazard by synthesizing the peer-reviewed animal toxicology and human epidemiology literature on PCE; we identify points of departure (PODs) for the most sensitive, well-supported non-cancer endpoints across organ systems and apply standard uncertainty-factor methodology, expressed through a margin-of-exposure (MOE) framework, to derive health-protective reference values; we evaluate carcinogenic potential using the weight of the animal and human evidence and apply a linear low-dose slope factor derived from the most robust rodent bioassay; and we combine these hazard values with site- or scenario-specific exposure estimates to calculate margins of exposure and excess lifetime cancer risk for the populations of concern. The summary below documents the toxicokinetic, toxicological, and epidemiological evidence base for PCE and explains how RoseyFields applies the resulting reference values in practice.

2. Chemical Identity and Background

Perchloroethylene (PCE, also known as tetrachloroethylene, CASRN 127-18-4, molecular formula C2Cl4, molecular weight 165.83 g/mol) is a colorless liquid with a chloroform-like odor (Lewis, 2007). It has a melting point of -22.3°C, a boiling point of 121.3°C, a vapor pressure of 18.5 mmHg at 25°C, water solubility of 206 mg/L at 20°C, and a log octanol-water partition coefficient (log Kow) of 3.40 (Lide, 2007). PCE is somewhat less volatile and more lipophilic than trichloroethylene, its close structural analog, and these properties drive both its environmental persistence and its tendency to accumulate in adipose tissue.

Nearly 65% of PCE production volume is used as an intermediate in the manufacture of fluorinated compounds; the remainder is used predominantly in dry cleaning and vapor degreasing operations, along with smaller-volume uses. Occupational exposure occurs primarily via inhalation in dry cleaning facilities and metal degreasing operations; environmental and consumer exposure occurs through contaminated groundwater and drinking water, indoor air (including vapor intrusion and residence near dry cleaners), and residual PCE in dry-cleaned garments.

3. Toxicokinetics (ADME)

Absorption

Inhalation is considered the major exposure route for PCE; human and animal studies confirm rapid and ready pulmonary uptake, with equilibrium occurring after several hours. Oral absorption through the gut is essentially complete (100%) in mice, rats, and dogs. Dermal absorption of PCE vapor is comparatively minimal — approximately 1% of the absorbed dose compared with inhalation of the same vapor concentration — but direct skin contact with liquid PCE, particularly under conditions of impeded evaporation or full immersion, can result in absorption approaching 100% of the delivered dermal dose.

Distribution

PCE is broadly distributed throughout the body and crosses both the blood-brain barrier and the placenta. Because of its lipophilicity, the highest tissue concentrations occur in adipose tissue, with correspondingly elevated concentrations in the brain and liver relative to other tissues; skeletal muscle contains the lowest concentration of any tissue measured. PCE also concentrates in human breast milk — detectable levels were found in seven of eight breast-milk samples from urban U.S. mothers in one study (Pellizzari et al., 1982) — and PBPK modeling predicts breast-milk concentrations ranging from roughly 857 to 8,440 µg/L for women exposed under occupational conditions. Long residence time in adipose tissue allows body burden to increase with continuous or repeated exposure.

Metabolism

PCE is metabolized through the same two general pathways as trichloroethylene: (1) oxidative metabolism via cytochrome P450 (CYP), predominantly CYP2E1, yielding trichloroacetic acid (TCA) as the major urinary metabolite along with oxalic acid and dichloroacetic acid; and (2) glutathione (GSH) conjugation to form S-(1,2,2-trichlorovinyl)glutathione (TCVG), which is further processed to the cysteine conjugate TCVC and ultimately bioactivated in the kidney by beta-lyase, or oxidized by flavin-containing monooxygenase 3 (FMO3), to a reactive thioketene species implicated in PCE nephrotoxicity. Overall metabolism of PCE is comparatively limited relative to other chlorinated solvents, particularly at higher exposure concentrations, and oxidative metabolism dominates in rodents. Notably, the glutathione-conjugation pathway appears to carry a substantially larger share of overall metabolism for PCE than for TCE — Luo et al. (2018a) reported roughly 21-fold higher metabolic flux through this pathway for PCE than for TCE at equimolar doses in mice — a finding directly relevant to PCE’s comparative nephrotoxic and renal carcinogenic potential (Cichocki et al., 2016). Species- and strain-related differences in beta-lyase and related enzyme activities may help explain sex- and species-specific renal carcinogenicity, with as much as a 29-fold difference in glutathione-pathway metabolite exposure reported across 45 mouse strains (Luo et al., 2019).

Elimination

PCE is eliminated primarily through pulmonary excretion of the unchanged parent compound, regardless of exposure route, while metabolites are eliminated predominantly in urine. Half-lives from blood-rich tissues, muscle, and adipose tissue are approximately 12–16 hours, 30–40 hours, and 55–65 hours, respectively. As in the case of TCE, PCE metabolite half-life differs sharply by species: the terminal half-life of oxidative metabolites is approximately 144 hours in humans compared with 10 hours or less in rodents, a difference attributable to the long residence time of PCE in human adipose tissue and continued release and metabolism of parent compound over time.

Because internal dose at the target organ, rather than external exposure concentration, drives PCE toxicity — and because the relative contribution of the oxidative and glutathione pathways differs by species, dose, and target tissue — EPA’s PCE risk evaluation relies on a physiologically based pharmacokinetic (PBPK) model (based on Chiu and Ginsberg, 2011) to convert applied animal doses into human-equivalent internal dose metrics before dose-response modeling. This PBPK-based framework underlies the derivation of essentially all PODs discussed below.

4. Toxicological Evidence Base (Animal Studies)

Animal toxicology on PCE spans inhalation, oral gavage, and drinking-water exposures across rats and mice, and has consistently identified the central nervous system, kidney, liver, immune/hematological system, and reproductive/developmental systems as sensitive targets, in addition to conclusive evidence of carcinogenicity across multiple tissues.

Kidney

Jisa (1993) and NTP (1986b) both conducted chronic (2-year) inhalation bioassays in rats and mice that identified nuclear enlargement (karyomegaly) and cytomegaly of the renal proximal tubules in both sexes, along with tubular cell hyperplasia and nephrosis. Benchmark dose modeling of the Jisa (1993) data — which scored High in data quality and produced a lower composite uncertainty factor than the available human study — yielded a human-equivalent concentration (HEC) of 9.0 ppm in rats and 2.1 ppm in mice for nuclear enlargement of the proximal tubules; because no discernible difference was found between species, the more sensitive mouse-derived POD was selected to represent the kidney domain.

Liver

Three studies provided adequate dose-response data for liver effects in mice: Jisa (1993), NTP (1986b), and Buben and O’Flaherty (1985). Jisa (1993) identified increased angiectasis in the liver at an HEC of 31 ppm (High quality, NOAEL-based, reduced uncertainty). NTP (1986b) identified increased liver degeneration/necrosis at a substantially higher HEC of 310 ppm. Buben and O’Flaherty (1985) reported increased liver-to-body-weight ratio in mice at an oral HED of 40 mg/kg-bw/day, but this endpoint was judged less clearly adverse on its own, potentially reflecting PPAR-alpha activation of limited relevance to humans, and the study scored only Medium in data quality. The Jisa (1993) POD for increased angiectasis was selected to represent the liver domain.

Central Nervous System

Altmann et al. (1990) exposed 10 human volunteers to PCE at 50 ppm for 4 hours/day over 4 days and identified increased latencies for pattern-reversal visual-evoked potentials — a marker of delayed neurological signaling — providing the principal acute non-cancer POD used in this risk evaluation (HEC of 5 ppm at 8 hr/day, adjusted across several exposure durations). For chronic exposure, Cavalleri et al. (1994) identified color-vision loss (color confusion) among PCE-exposed workers at an HEC of 2.2 ppm, and Echeverria et al. (1995) — in a cohort of 65 Michigan dry-cleaning workers with a time-weighted-average exposure of 41 ppm — identified statistically significant deficits in visual pattern recognition, pattern memory, and reaction time at an inferred HEC of 8.3 ppm. Both chronic studies scored Medium in data quality with equivalent cumulative uncertainty factors; EPA therefore used the midpoint of the two PODs (5.2 ppm) as the best-overall representative value for the chronic neurotoxicity domain.

Immune and Hematological System

Emara et al. (2010) evaluated Egyptian dry-cleaning workers and identified reduced erythrocyte counts and hemoglobin along with increased immune cell markers, IgE, and IL-4 — findings indicative of possible anemia and immune dysregulation — at an HEC of 6.4 ppm. This High-quality human study was used to represent both the immune and hematological hazard domains.

Reproductive and Developmental Effects

Beliles et al. (1980) exposed mice to PCE by inhalation for 5 days and identified reduced sperm quality at an HEC of 21 ppm; although the exposure duration was short, it encompassed the full window of active spermatogenesis and sperm production, and the study scored High in data quality. For developmental toxicity, three animal studies were considered: Nelson et al. (1979) reported decreased offspring weight gain and developmental neurotoxicity (altered behavior and changes in brain acetylcholine) at an HEC of 29 ppm, though this study scored only Low in data quality; Tinston (1994), a two-generation reproductive toxicity study scoring High in data quality, identified increased neonatal (F2 generation) pup mortality and CNS depression in F1 and F2 offspring at an HEC of 18 ppm; and Carney et al. (2006), also High quality, identified decreased fetal and placental weight along with skeletal effects at an HEC of 16 ppm. The Tinston (1994) POD was selected to represent the developmental domain because its two-generation design was judged most likely to capture the full range of potential developmental outcomes.

Carcinogenicity Bioassays

Chronic bioassays provide conclusive evidence of PCE carcinogenicity in both rats and mice by both inhalation and ingestion. Statistically significant increases in liver tumors (hepatocellular adenomas and/or carcinomas) were observed in male and female B6C3F1 and Crj:BDF1 mice following inhalation exposure (Jisa, 1993; NTP, 1986b) and in male and female B6C3F1 mice following oral gavage exposure (Nci, 1977). Significant increases in mononuclear cell leukemia (MCL) were observed in male and female F344/N and F344/DuCrj rats exposed by inhalation (Jisa, 1993; NTP, 1986b). Additional findings potentially related to treatment included increased testicular interstitial cell tumors and renal tubular adenomas/adenocarcinomas in male F344/N rats, brain gliomas in male and female F344/N rats (NTP, 1986b), and hemangiosarcomas and Harderian gland adenomas in male Crj:BDF1 mice (Jisa, 1993). The mouse hepatocellular tumor data from Jisa (1993) — observed consistently across all three chronic bioassays — form the quantitative basis for the cancer slope values discussed in Section 8.

5. Epidemiological Evidence Base (Human Studies)

Human evidence on PCE derives substantially from studies of dry-cleaning and other solvent-exposed occupational cohorts, together with a smaller body of drinking-water and biomonitoring studies.

Neurotoxicity

Altmann et al. (1990), Cavalleri et al. (1994), and Echeverria et al. (1995) — described in Section 4 — together constitute the principal human quantitative evidence base for PCE neurotoxicity, consistently identifying visual and neurocognitive deficits (increased visual-evoked potential latency, color vision loss, and impaired pattern recognition/memory) in PCE-exposed workers across independent cohorts, an unusual degree of consistency that supported selection of neurotoxicity as one of the best-overall chronic non-cancer endpoints.

Nephrotoxicity

Mutti et al. (1992) evaluated dry-cleaning workers and reported urinary markers consistent with proximal tubular injury following long-term occupational PCE exposure; this finding was corroborated by several additional epidemiological studies of dry-cleaning cohorts. The Mutti et al. (1992) POD (a LOAEL, HEC of 5.0 ppm) carried greater uncertainty than the corresponding rodent-derived POD from Jisa (1993) and so was not selected as the primary kidney-domain reference value, but it independently corroborates the kidney as a sensitive human target.

Immunological and Hematological Studies

Emara et al. (2010), described in Section 4, provides the principal human evidence for PCE immune and hematological effects, identifying reduced erythrocyte counts, reduced hemoglobin, and altered immune markers among Egyptian dry-cleaning workers.

Reproductive and Developmental Epidemiology

Beyond the animal evidence described above, epidemiological studies have reported associations between PCE exposure and adverse pregnancy outcomes in women, including an increased risk of spontaneous abortion, based on evidence summarized in the 2012 EPA IRIS Assessment (U.S. EPA, 2012c).

Cancer Epidemiology

No available human cancer studies were judged suitable for quantitative dose-response assessment; the human epidemiological evidence for PCE carcinogenicity is characterized as suggestive rather than conclusive, in contrast with the conclusive rodent bioassay evidence described in Section 4. As a result, the quantitative cancer risk values discussed in Section 8 are derived entirely from rodent bioassay data rather than human data.

6. Health Outcomes Associated with PCE Exposure, by Body System

The following overview organizes the illnesses and health outcomes associated with PCE exposure by body system, distinguishing findings with strong, quantitatively modeled dose-response support from those with suggestive or less consistent evidence.

Central Nervous System

Strong, quantitatively modeled: Visual and neurocognitive deficits — increased visual-evoked potential latency (Altmann et al., 1990), color vision loss (Cavalleri et al., 1994), and impaired pattern recognition, pattern memory, and reaction time (Echeverria et al., 1995) — are supported by multiple independent human occupational cohorts and form the basis of both the acute and chronic best-overall non-cancer PODs.

Kidney

Strong, quantitatively modeled: Renal tubular karyomegaly/cytomegaly and related nephropathy are demonstrated in high-quality rodent bioassays (Jisa, 1993; NTP, 1986b) and corroborated by human urinary biomarker evidence in dry-cleaning workers (Mutti et al., 1992).

Moderate, incorporated via combined cancer estimate: Renal tubular adenomas/adenocarcinomas were observed in male rats in one bioassay (NTP, 1986b) as a treatment-related finding, though the primary quantitative cancer POD is instead based on liver tumors, which were observed most consistently across bioassays.

Liver

Strong, quantitatively modeled: Non-cancer hepatotoxicity (increased angiectasis, liver degeneration/necrosis) is demonstrated across multiple high-quality rodent studies (Jisa, 1993; NTP, 1986b). Hepatocellular adenomas and carcinomas are the most consistently observed tumor type across all three chronic bioassays in both mice and rats, by both inhalation and oral routes (Jisa, 1993; NTP, 1986b; Nci, 1977), and form the quantitative basis of the cancer slope factor and inhalation unit risk.

Immune and Hematological System

Strong, quantitatively modeled: Reduced erythrocyte counts and hemoglobin, and altered immune cell markers (IgE, IL-4), are demonstrated in a High-quality human study of dry-cleaning workers (Emara et al., 2010).

Moderate, incorporated via combined cancer estimate: Mononuclear cell leukemia (MCL) was consistently observed in rat bioassays (Jisa, 1993; NTP, 1986b), though it is not the primary tumor type driving the quantitative cancer slope values.

Reproductive System

Strong, quantitatively modeled: Reduced sperm quality following inhalation exposure is demonstrated in a High-quality rodent study (Beliles et al., 1980).

Suggestive/less consistent: Human epidemiological evidence links PCE exposure to adverse pregnancy outcomes, including spontaneous abortion, but this association has not been quantitatively modeled into a reference value.

Developmental Effects

Strong, quantitatively modeled: Neonatal mortality and CNS depression in offspring are demonstrated in a High-quality two-generation rodent study (Tinston, 1994), the study selected to represent the developmental domain. Decreased fetal/placental weight and skeletal effects are also quantitatively modeled (Carney et al., 2006).

Suggestive/less consistent: Developmental neurotoxicity (altered behavior, changes in brain acetylcholine) was reported in a Low-quality rodent study (Nelson et al., 1979); it remains the only available dataset on this specific functional/molecular endpoint and is treated as supportive but not the primary basis for the developmental reference value.

Cancer (Systemic)

Strong, quantitatively modeled: Hepatocellular tumors in mice, observed consistently across three independent bioassays and both major exposure routes (Jisa, 1993; NTP, 1986b; Nci, 1977), anchor the inhalation unit risk and oral slope factor for PCE.

Moderate, supportive but not independently modeled: Mononuclear cell leukemia in rats (Jisa, 1993; NTP, 1986b) and additional treatment-related findings — testicular interstitial cell tumors, renal tubular adenomas/adenocarcinomas, and brain gliomas in rats, and hemangiosarcomas and Harderian gland adenomas in mice — support the overall weight-of-evidence conclusion of multisite carcinogenicity but are not separately quantified.

Suggestive: Human epidemiological evidence for PCE carcinogenicity, while consistent with a carcinogenic hazard, is not considered adequate for quantitative dose-response modeling, and no specific human cancer site has been identified as a robust quantitative anchor.

7. Derivation of Non-Cancer Reference Values (Points of Departure and Uncertainty Factors)

As with EPA’s parallel TSCA risk evaluation for trichloroethylene, the PCE risk evaluation derives PODs on an organ-system and endpoint-specific basis using PBPK-derived human-equivalent concentrations (HECs) rather than a single classical oral reference dose (RfD). Non-cancer risk is characterized using a margin-of-exposure (MOE) framework, in which the MOE is calculated as the ratio of the non-cancer POD (HEC or, for dermal pathways, a route-to-route-extrapolated human-equivalent dose, HED) to the estimated human exposure. The resulting MOE is then compared against a benchmark MOE equal to the total composite uncertainty factor applied in deriving that POD; an MOE below the benchmark indicates the margin of safety is inadequate, while an MOE at or above the benchmark indicates the exposure is not expected to result in the associated adverse effect with an adequate margin of safety.

Uncertainty Factors Applied

Consistent with EPA’s RfD/RfC Guidance, standard composite uncertainty factors were applied to each POD:

• UFA (interspecies): applied to account for residual toxicodynamic uncertainty in extrapolating from animals to humans, typically reduced (e.g., to 1 or 3) when PBPK modeling has already addressed toxicokinetic differences.
• UFH (intraspecies/human variability): commonly 10 across the PCE non-cancer PODs, reflecting the absence of the HEC99-type variability adjustment used in the parallel TCE assessment.
• UFL (LOAEL-to-NOAEL): applied when a study’s POD is a LOAEL rather than a NOAEL, ranging from 1 (NOAEL-based PODs, e.g., Jisa, 1993 liver and kidney endpoints) to 10 (LOAEL-based PODs, e.g., Cavalleri et al., 1994; Echeverria et al., 1995; Mutti et al., 1992; Emara et al., 2010; NTP, 1986b liver).
• UFS (subchronic-to-chronic): generally 1 for studies of adequate chronic duration, but as high as 10 for the Buben and O’Flaherty (1985) subchronic gavage study, contributing to that study’s high cumulative UF of 3,000.

Composite uncertainty factors across the selected PCE endpoints ranged from 10 (Altmann et al., 1990, acute neurotoxicity) to 300 (NTP, 1986b, liver degeneration/necrosis), with most chronic non-cancer endpoints falling around a total UF of 30 to 100.

Selected Best-Overall Non-Cancer Endpoints

EPA identified the acute neurotoxicity finding of Altmann et al. (1990) as the best-overall acute non-cancer POD, and selected a best-overall representative POD within each chronic health-effect domain, summarized below.

RoseyFields uses these HEC values directly as the working inhalation reference concentrations for PCE non-cancer risk characterization by domain, applying the corresponding benchmark MOE (equal to the total UF) as the acceptability threshold in each case, and applies the route-to-route-extrapolated dermal PODs from the underlying risk evaluation when dermal exposure is the relevant pathway. Because the chronic PODs span roughly a 150-fold range across all evaluated endpoints (though within about 10-fold for most), RoseyFields selects the domain-specific POD most relevant to the exposure scenario and sensitive subpopulation under evaluation rather than defaulting to a single value across all assessments.

8. Carcinogenicity Classification and Cancer Slope Factor

Under EPA’s 2005 Guidelines for Carcinogen Risk Assessment, PCE is characterized as “likely to be carcinogenic in humans by all routes of exposure,” based on conclusive evidence of carcinogenicity in rats and mice and suggestive, but not independently quantifiable, evidence in humans. Because no human cancer dataset was judged suitable for dose-response modeling, the quantitative cancer risk values are derived entirely from rodent bioassay data using a linear low-dose extrapolation approach, consistent with EPA’s default position that a linear approach is appropriate when “mode of action information is supportive of linearity or mode of action is not understood” — reflecting the complex, multi-mechanism mode of action believed to underlie PCE tumorigenesis across different tissue types.

Inhalation Unit Risk (IUR)

The inhalation unit risk for PCE is 2 × 10⁻³ per ppm (3 × 10⁻⁴ per mg/m³), derived from male mouse hepatocellular tumor data in Jisa (1993) — the tumor type observed most consistently across all three chronic bioassays and both major exposure routes. This IUR should not be applied to exposures exceeding 60 ppm (400 mg/m³, the ambient-equivalent exposure corresponding to the underlying POD), because the dose-response relationship becomes non-linear above this level and use of the linear IUR would tend to overestimate risk at higher exposures. An alternative IUR based on combined male and female rat mononuclear cell leukemia (MCL) data is also available and may be used as a secondary or comparative estimate.

Oral Slope Factor (OSF) and Dermal Slope Factor

An oral slope factor of approximately 2.0 × 10⁻³ per mg/kg-day was derived by PBPK-model-based route-to-route extrapolation from the same Jisa (1993) mouse hepatocellular tumor dose-response data used for the IUR. A corresponding dermal slope factor, also approximately 2.0 × 10⁻³ per mg/kg-day, was derived using the relationship: dermal slope factor = IUR (per mg/m³) ÷ inhaled volume (m³) × body weight (kg), reflecting the assumption of essentially complete (100%) absorption for both oral and appropriately modeled dermal exposure scenarios.

As with TCE, EPA elected not to use the IUR or OSF to estimate theoretical cancer risk from single, short-term, or intermittent PCE exposures, because methodology for extrapolating lifetime cancer risk to acute exposures carries substantial uncertainty and the relationship between a single short-term PCE exposure and cancer induction in humans has not been established. Cancer risk estimates for PCE are therefore based exclusively on lifetime-averaged exposure, represented as the lifetime average daily concentration or dose (LADC/LADD).

9. Uncertainties and Sensitive Populations

Key Uncertainties

• Cancer risk basis: because no human cancer dataset was adequate for dose-response modeling, the PCE cancer slope values rely entirely on interspecies extrapolation from rodent bioassay data, in contrast with the TCE risk evaluation, which anchors its cancer values in human epidemiological data — introducing comparatively greater interspecies extrapolation uncertainty for PCE.
• Complex, incompletely understood mode of action: PCE tumorigenesis is believed to involve multiple contributing mechanisms of varying importance across tissue types, and the relative contribution of the oxidative versus glutathione-conjugation metabolic pathways in humans remains uncertain, particularly for kidney effects.
• Route-to-route and dermal extrapolation: because the PBPK model used for PCE does not include a dermal compartment, dermal PODs and the dermal slope factor were derived by extrapolation from inhalation data using simplified exposure assumptions (e.g., assumed ventilation rate and body weight), introducing additional uncertainty specific to dermal risk estimates.
• Wide range across non-cancer endpoints: chronic non-cancer PODs span roughly a 150-fold range across all evaluated endpoints (though generally within 10-fold for most), reflecting differences in study design, species, and endpoint sensitivity.
• Non-linear dose-response above the cancer POD range: the inhalation unit risk is explicitly not valid above 60 ppm (400 mg/m³), requiring careful attention to the applicable exposure range in any cancer risk calculation.

Sensitive and Susceptible Subpopulations

• Infants, via PCE detected in and modeled for human breast milk, with predicted concentrations in the range of hundreds to thousands of µg/L under occupational maternal exposure conditions.
• Pregnant women and the developing fetus, given the epidemiological association with spontaneous abortion and the quantitative animal evidence for neonatal mortality, CNS depression, and skeletal/developmental effects.
• Children of employees present at dry-cleaning facilities, a subpopulation EPA specifically evaluated using lifestage-adjusted PODs given differences in body weight and inhalation rate relative to adults.
• Individuals with pre-existing hematological or immune vulnerabilities, given the quantitative reduction in erythrocytes, hemoglobin, and altered immune markers identified in occupationally exposed workers.
• Long-tenure dry-cleaning and degreasing workers, given that both the primary human neurotoxicity data and the nephrotoxicity biomarker data derive from chronically exposed occupational cohorts.
• Individuals with elevated CYP2E1 activity or genetic variability in glutathione-pathway enzymes, given documented 10- to 50-fold interindividual variability in CYP2E1 activity and substantial strain-dependent variability in glutathione-pathway metabolite formation.

10. Application of Reference Values in RoseyFields Risk Assessments

Non-Cancer Risk: Margin of Exposure

For each relevant PCE exposure pathway and averaging time, RoseyFields calculates a margin of exposure (MOE) as the ratio of the domain-appropriate POD (HEC, or route-to-route-extrapolated dermal HED) to the estimated human exposure:

MOE = Non-Cancer Hazard Value (POD) ÷ Human Exposure

The resulting MOE is compared against the benchmark MOE for that endpoint (i.e., the total composite uncertainty factor used to derive the POD, ranging from 10 to 300 depending on the domain). An MOE at or above the benchmark indicates negligible concern for the associated adverse effect; an MOE below the benchmark indicates an inadequate margin of safety warranting further evaluation or risk mitigation. Where multiple exposure pathways or domains are relevant to the same population, RoseyFields evaluates MOEs by target organ/effect domain (CNS, kidney, liver, immune/hematological, reproductive, developmental) rather than combining dissimilar endpoints into a single index.

Cancer Risk: Excess Lifetime Cancer Risk

For scenarios involving chronic or lifetime exposure, RoseyFields calculates excess lifetime cancer risk (ELCR) using the lifetime average daily concentration or dose (LADC/LADD) and the applicable slope factor:

ELCR = LADC × IUR (2 × 10⁻³ per ppm), or LADD × OSF/dermal slope factor (2.0 × 10⁻³ per mg/kg-day)

RoseyFields confirms that estimated exposure concentrations fall within the valid range of the linear IUR (below 60 ppm / 400 mg/m³) before applying this calculation, and compares the resulting ELCR against applicable regulatory benchmarks (commonly in the 1 × 10⁻⁶ to 1 × 10⁻⁴ range depending on the regulatory and population context) to characterize whether the estimated risk falls within an acceptable range.

Dose Reconstruction

Where direct exposure monitoring data are unavailable, RoseyFields reconstructs PCE exposure using validated modeling approaches appropriate to the setting — occupational inhalation modeling for dry-cleaning and degreasing operations, the Dermal Exposure to Volatile Liquids Model or the Consumer Exposure Model’s permeability and fraction-absorbed sub-models for dermal pathways (selected according to whether evaporation is impeded or unimpeded), and fate-and-transport or vapor-intrusion modeling for groundwater- or soil-vapor-impacted sites — combined with population-specific behavioral parameters (exposure duration, frequency, body weight, ventilation rate, and, for cancer risk, total years of lifetime exposure). Reconstructed exposure estimates are then evaluated against the appropriate domain-specific MOE benchmark or cancer slope value described above, with particular attention to lifestage-specific adjustments where children or other sensitive subpopulations are present.

11. Conclusion

PCE presents a well-characterized, multi-organ hazard profile, with the central nervous system standing out as an unusually consistent target across independent human occupational cohorts, alongside quantitatively supported effects on the kidney, liver, immune/hematological system, and reproductive/developmental systems. PCE is characterized as likely to be carcinogenic in humans by all routes of exposure, based on conclusive rodent bioassay evidence of multisite tumorigenesis — most consistently hepatocellular tumors — and suggestive but not independently quantifiable human evidence. RoseyFields applies the resulting reference values — domain-specific HEC/HED PODs evaluated through a margin-of-exposure framework, together with the inhalation unit risk and oral/dermal slope factor for cancer risk — within a standard MOE and excess lifetime cancer risk framework, informed by scenario-specific dose reconstruction, to characterize risk for the populations and exposure pathways relevant to each site-specific assessment.

Bibliography