Perfluorooctanoic Acid (PFOA) Proposition 65 Dose Assessment and Regulatory Toxicology Evaluation
Project Overview
RoseyFields conducted a technical toxicology and regulatory risk assessment to evaluate a health-protective exposure threshold for perfluorooctanoic acid (PFOA) under California’s Proposition 65 framework. The project involved reviewing the toxicological literature, identifying the most sensitive developmental and reproductive endpoints, and evaluating how those endpoints should be translated into a Maximum Allowable Dose Level (MADL) for human exposure. The work focused on whether the available animal and human evidence supported a defensible reproductive toxicity benchmark for PFOA, and how that benchmark should be calculated for multiple potential exposure routes, including ingestion and dermal exposure. The assessment was designed for a setting in which the scientific dispute was not merely whether PFOA is hazardous, but rather which toxicological endpoint should control, what dose metric should be used, and how a protective Proposition 65 benchmark should be derived from the available evidence.
PFOA is one of the most studied per- and polyfluoroalkyl substances (PFAS) because of its widespread historical use, environmental persistence, bioaccumulation potential, and long half-life in humans. It has been used in industrial and commercial applications associated with surfactants, fluoropolymer manufacturing, and stain- and water-resistant product formulations. Toxicological studies have linked PFOA exposure to a range of noncancer health effects, including developmental, hepatic, endocrine, immune, and mammary gland effects, with some of the most sensitive endpoints arising from prenatal and early-life exposures. In addition, human exposure may occur through contaminated drinking water, food, dust, and consumer-product contact, and prenatal transfer can occur through placental and lactational pathways. These characteristics make PFOA a recurring subject of environmental litigation, product warning disputes, and exposure assessments involving historical releases, drinking-water contamination, and consumer exposure scenarios.
The RoseyFields project was therefore structured as a chemical-specific regulatory toxicology analysis. Rather than beginning with a preexisting regulatory value and applying it mechanically, the work evaluated the underlying science used to derive a proposed PFOA MADL and tested whether the derivation method was consistent with California Office of Environmental Health Hazard Assessment (OEHHA) practice, EPA health-advisory methodology, and the broader Proposition 65 framework for reproductive toxicants. The project drew on a draft PFOA MADL report, meeting slides describing alternate dose-calculation approaches, peer-reviewed developmental toxicity studies, dermal penetration research, and comparator OEHHA MADL documents for other chemicals. Together, those sources provided the basis for a defensible evaluation of how a PFOA dose limit could be derived, how route-specific assumptions affect the final value, and how uncertainty factors and human-equivalent-dose adjustments can materially change the resulting benchmark.
Chemicals of Concern
The primary chemical of concern in this project was perfluorooctanoic acid (PFOA), an eight-carbon perfluoroalkyl acid that has historically been used in the production of fluorotelomers and fluorinated surfactants. PFOA is environmentally persistent, resistant to degradation, and detectable in human serum, cord blood, milk, and multiple environmental media. Because it is slowly eliminated from the human body, repeated or chronic exposures can result in long-term internal body burdens even when environmental concentrations appear relatively low. The project focused on PFOA as a reproductive and developmental toxicant, particularly in relation to prenatal exposure and postnatal developmental outcomes in offspring.
Although the immediate objective was to evaluate a PFOA-specific Proposition 65 dose level, the assessment also considered the broader PFAS toxicology context. Comparator PFAS literature and OEHHA materials were useful because they showed how California agencies approach uncertainty factors, reference-dose derivation, and health-protective concentrations for PFAS chemicals when human evidence is limited and animal toxicology drives the benchmark. For example, OEHHA’s Notification Level Recommendation for Perfluorobutane Sulfonic Acid (PFBS) describes a standard framework for identifying the most sensitive well-conducted toxicology studies, selecting a critical endpoint, and applying uncertainty factors to derive a health-protective concentration. That framework was relevant here because the PFOA project similarly required selecting among multiple animal studies and determining whether a simple MADL conversion was adequate or whether a more protective adjustment for interspecies and intraspecies uncertainty was warranted.
Methods
The first component of the project was a review of California Proposition 65 dose-setting methodology. Under Proposition 65, a MADL for reproductive toxicity is intended to represent an exposure level that is 1,000 times lower than the no-observed-effect level for the relevant reproductive or developmental endpoint. OEHHA MADL documents for comparator chemicals such as N-methylpyrrolidone (NMP), n-hexane, and acrylamide were reviewed to understand how the agency typically identifies the critical study, selects the NOEL/NOAEL, converts the animal dose to a human dose level, and derives a daily microgram-per-day threshold. Those documents also helped establish the “conventional” Proposition 65 calculation pathway in which a NOEL expressed in mg/kg-day is multiplied by an assumed human body weight and then divided by 1,000 to produce the MADL.
This comparator review mattered because one of the central questions in the PFOA project was whether a standard Proposition 65 MADL calculation was sufficiently protective for a persistent PFAS with complex toxicokinetics, or whether the toxicology supported a more refined approach using a human equivalent dose (HED) and additional uncertainty factors. The draft PFOA materials and associated slides explicitly contrasted these two approaches, showing that the choice of methodology could produce markedly different candidate MADLs. RoseyFields therefore analyzed not only the toxicology endpoint itself, but also the methodological consequences of using a conventional Proposition 65 conversion versus an HED- and uncertainty-factor-based approach modeled on EPA and OEHHA risk-assessment practices.
The second component of the project was the identification of the most sensitive toxicology studies suitable for use as the basis of a PFOA MADL. The draft report materials indicated that no relevant human studies were identified for purposes of deriving a reproductive or developmental MADL, making the analysis dependent on animal studies and supporting mechanistic information. RoseyFields reviewed the studies identified in the draft report as the primary basis for the proposed PFOA MADL: Tucker et al. (2015) and Macon et al. (2011), along with Franko et al. (2012) for dermal absorption. The project also considered the extent to which the EPA 2016 Drinking Water Health Advisory for PFOA supported the use of these studies as the most sensitive, sufficiently robust data sources for dose derivation.
Tucker et al. (2015) evaluated perinatal PFOA exposure in CD-1 and C57Bl/6 mice and found that the mammary gland was a sensitive pubertal target following low-dose exposure. Prenatal exposure caused significant mammary developmental delays in female offspring, and those delays began during puberty and persisted into young adulthood. The study reported that the mammary gland appeared to be more sensitive than other pubertal endpoints and that developmental effects were dose-dependent. In the PFOA MADL draft materials, Tucker et al. was treated as one of the two most important developmental studies because it identified adverse mammary developmental effects at comparatively low dose levels, including effects in CD-1 mice at 0.01 mg/kg-day and in C57Bl/6 mice at 0.3 mg/kg-day, with a NOAEL identified at 0.1 mg/kg-day for C57Bl/6 mice.
Macon et al. (2011) examined prenatal PFOA exposure in CD-1 mice and reported developmental effects and internal dosimetry findings following both full-gestation and late-gestation dosing. Exposure increased offspring relative liver weights, and the study also found significant mammary developmental effects at low doses. Importantly, the study showed that internal PFOA concentrations remained elevated in offspring serum and liver for weeks after exposure, underscoring the toxicokinetic persistence of PFOA and the biological plausibility of developmental effects arising from gestational exposure. The draft PFOA materials treated Macon et al. as an additional critical study because it provided a conservative NOAEL for liver-related developmental effects and supported the view that developmental endpoints can occur at relatively low doses in mice.
Once the candidate toxicological endpoints were identified, RoseyFields evaluated the calculation of an ingestion-based MADL. The draft PFOA report materials described a conventional Proposition 65 approach in which the selected animal NOAEL is converted to a human daily dose by multiplying by an assumed body weight and then dividing by 1,000 to derive a daily allowable dose. The meeting slides also summarized this as “common practice OEHHA practice,” using a 70 kg body weight and no explicit HED or uncertainty-factor adjustment in the basic calculation. Under this approach, the draft materials indicated that the resulting proposed MADL for PFOA would be 7 µg/day if based on the selected NOAEL and conventional Proposition 65 scaling alone.
RoseyFields did not treat that calculation as self-validating. Instead, the project evaluated whether a simple NOAEL-to-human scaling exercise adequately reflected the way PFAS toxicology is often interpreted in health-protective regulatory contexts. PFOA has unusual toxicokinetic characteristics compared with many conventional Proposition 65 chemicals: it is persistent, bioaccumulative, transferred prenatally and lactationally, and associated with sensitive developmental endpoints that may arise at doses lower than more traditional systemic toxicity endpoints. For that reason, the project examined whether a human equivalent dose adjustment and an additional uncertainty factor should be applied before the Proposition 65 conversion step, particularly where the point of departure was derived from animal developmental studies rather than direct human data.
A major analytical issue in the project was whether the proposed PFOA MADL should incorporate a human equivalent dose (HED) and uncertainty factor (UF) rather than relying solely on the conventional Proposition 65 arithmetic. The draft PFOA slides indicated that when HED and uncertainty-factor adjustments were incorporated—drawing from EPA’s PFOA health advisory and OEHHA-style methodology—the resulting proposed MADL became substantially lower, approximately 3.3 ng/day rather than 7 µg/day. RoseyFields evaluated this alternative because it directly affected the defensibility and conservatism of the benchmark.
The logic for this adjustment was consistent with the broader OEHHA and EPA practice of deriving health-protective concentrations for noncancer effects from animal studies by accounting for interspecies extrapolation, human variability, and other uncertainty inherent in translating experimental dose-response findings into real-world human risk benchmarks. The PFBS notification-level document, while addressing a different PFAS, describes a standard OEHHA approach that applies a composite uncertainty factor to animal-based points of departure in order to account for interspecies and intraspecies variability. RoseyFields reviewed this framework because it provided a defensible analogue for evaluating whether a PFOA MADL derived from mouse developmental studies should similarly include explicit adjustments rather than relying only on the 1,000-fold Proposition 65 divisor.
The final methodological component was the evaluation of dermal exposure. Proposition 65 matters frequently involve multiple exposure pathways, and the PFOA draft materials specifically contemplated the derivation of a dermal MADL. EPA has noted that the health effects of dermal exposure to PFOA have not been as extensively quantified as oral exposure, but the draft materials relied on Franko et al. (2012) to evaluate whether dermal absorption could nonetheless represent a meaningful exposure route. According to the draft report, Franko et al. found that PFOA can permeate both mouse and human skin, with a reported 23.4% ± 1.9% penetration through full-thickness human skin and 45.6% ± 1.9% retention in the dermis, which together were used as the basis for a conservative 69% dermal absorption fraction.
RoseyFields evaluated whether it was scientifically reasonable to apply that absorption fraction to the ingestion-based point of departure to derive a dermal MADL. That analysis required examining the underlying toxicology endpoint, the biological relevance of dermal uptake for a persistent PFAS, and the extent to which a route-to-route extrapolation was consistent with Proposition 65 practice for other chemicals. Comparator MADL documents, including the OEHHA NMP dermal and inhalation MADL document, were useful in understanding how California has historically treated route-specific reproductive toxicity benchmarks and when separate route-specific values are warranted.
Results and Technical Findings
The project’s principal finding was that PFOA presents a credible basis for a Proposition 65 reproductive/developmental toxicity benchmark derived from animal studies, but the resulting value depends heavily on the methodological choices used to translate those studies into a daily allowable dose. The toxicological literature reviewed in the project supports the conclusion that developmental endpoints—particularly mammary gland developmental delays and increased offspring liver weights following prenatal exposure—are among the most sensitive effects available for dose derivation. Tucker et al. and Macon et al. both support the proposition that low-dose prenatal PFOA exposure can produce adverse developmental outcomes in mice, and the draft PFOA materials reasonably identified those studies as candidate bases for a reproductive or developmental MADL.
A second major finding was that the choice of dose-derivation methodology materially affects the proposed MADL. A conventional Proposition 65 calculation using a selected NOAEL, a standard human body weight, and the 1,000-fold MADL divisor yielded a candidate value in the microgram-per-day range. By contrast, a more health-protective approach incorporating human-equivalent-dose adjustment and uncertainty factors yielded a value in the nanogram-per-day range, orders of magnitude lower than the conventional result. This difference is not a minor technicality; it changes the practical significance of the benchmark for product-warning, environmental, and consumer-exposure cases. RoseyFields’ review therefore showed that the core scientific issue is not only which study supplies the point of departure, but also whether the selected methodology properly accounts for cross-species toxicokinetic differences, persistent body burden, and uncertainty associated with translating animal developmental effects to human exposure limits.
A third finding was that dermal exposure cannot be dismissed categorically for PFOA, even though oral ingestion is the more commonly evaluated route in drinking-water and environmental contamination contexts. The dermal absorption evidence reviewed in the project indicates that PFOA can penetrate and be retained in human skin to a nontrivial degree, which supports the scientific plausibility of a dermal exposure benchmark in appropriate circumstances. That does not mean dermal exposure will control every case, but it does mean that product-contact, occupational, or consumer-use scenarios involving repeated skin contact may warrant explicit route-specific evaluation rather than automatic reliance on an oral benchmark alone.
Finally, the project demonstrated that PFAS-specific risk assessment often requires a more nuanced regulatory toxicology approach than conventional single-chemical Proposition 65 analyses. PFOA’s persistence, placental transfer, long half-life, developmental sensitivity, and complex toxicokinetics create a stronger argument for careful scrutiny of point-of-departure selection, HED adjustment, uncertainty factors, and route-specific exposure assumptions. RoseyFields’ analysis therefore treated the MADL not as a rote mathematical output, but as a chemical-specific scientific conclusion that had to be defended through toxicology, pharmacokinetics, and regulatory methodology together.
Conclusions
This project illustrates RoseyFields’ ability to perform high-level regulatory toxicology and chemical-specific dose assessment for emerging contaminants and litigated exposure scenarios. The PFOA MADL evaluation required more than a summary of PFAS hazards. It required the integration of developmental toxicology, dose-response analysis, route-to-route extrapolation, pharmacokinetic reasoning, and Proposition 65 regulatory methodology into a single defensible benchmark analysis. By reviewing the most sensitive developmental studies, evaluating alternate dose-derivation methods, and assessing whether dermal absorption should be incorporated into the final benchmark, RoseyFields developed a framework for determining not only whether PFOA poses reproductive or developmental risks, but also what level of daily exposure should be considered health protective under California law.
The work also demonstrates a broader capability that is increasingly important in environmental and product-liability litigation: the ability to move from hazard identification to benchmark derivation. In many disputes involving PFAS and other persistent chemicals, the critical issue is not whether a chemical has been associated with adverse effects in the abstract, but whether the exposure at issue exceeds a scientifically supportable threshold. That inquiry requires the ability to select the right toxicological endpoint, evaluate competing dose-conversion approaches, understand the role of uncertainty factors, and determine whether alternate exposure pathways such as dermal uptake need to be addressed. RoseyFields’ PFOA analysis shows how those questions can be answered using a structured, evidence-based approach grounded in regulatory toxicology and forensic risk assessment.
More broadly, this type of project is representative of the firm’s work at the intersection of PFAS science, human health risk assessment, and environmental litigation support. Whether the issue involves Proposition 65 warnings, drinking-water contamination, product exposure, or historical industrial releases, RoseyFields can evaluate the toxicological record, translate it into practical exposure benchmarks, and explain the scientific basis for those conclusions in a form suitable for expert support, technical reporting, and strategic case development.
References
California Office of Environmental Health Hazard Assessment (OEHHA). (2003, March). Proposition 65 maximum allowable dose level (MADL) for reproductive toxicity for N-methylpyrrolidone for dermal and inhalation exposures.
California Office of Environmental Health Hazard Assessment (OEHHA). (2010, February). Proposed maximum allowable dose level (MADL) for reproductive toxicity for acrylamide.
California Office of Environmental Health Hazard Assessment (OEHHA). (2021, January). Notification level recommendation for perfluorobutane sulfonic acid (PFBS) in drinking water. California Environmental Protection Agency.
Franko, J., Meade, B. J., Frasch, H. F., Barbero, A. M., Anderson, S. E., & Abdel-Rahman, A. A. (2012). Dermal penetration potential of perfluorooctanoic acid (PFOA) in human and mouse skin. Journal of Toxicology and Environmental Health, Part A.
Macon, M. B., Villanueva, L. R., Tatum-Gibbs, K., Zehr, R. D., Strynar, M. J., Stanko, J. P., White, S. S., Helfant, L., & Fenton, S. E. (2011). Prenatal perfluorooctanoic acid exposure in CD-1 mice: Low-dose developmental effects and internal dosimetry. Toxicological Sciences, 122(1), 134–145.
Tucker, D. K., Macon, M. B., Strynar, M. J., Dagnino, S., Andersen, E., & Fenton, S. E. (2015). The mammary gland is a sensitive pubertal target in CD-1 and C57Bl/6 mice following perinatal perfluorooctanoic acid (PFOA) exposure. Reproductive Toxicology, 54, 26–36.
U.S. Environmental Protection Agency. (2016, May). Drinking water health advisory for perfluorooctanoic acid (PFOA).