EPA Releases Methanol Noncancer Draft Assessment

Product Liability Update

Date: June 14, 2011

Overview

The U.S. Environmental Protection Agency (EPA) recently released its draft human health assessment for the noncancer risks of methanol ("Toxicological Review of Methanol (Non-Cancer): In Support of Summary Information on the Integrated Risk Information System [IRIS]") for peer review by independent agencies and a 60-day public comment period. An EPA listening session to allow all interested parties to present scientific and technical comments on the draft was held on May 26, at which representatives of the methanol industry sought additional time for the publication of the draft until additional scientific information can be collected. The public comment period ends on June 17.

The draft proposes a lower allowable oral exposure level for methanol than the previous EPA assessment published in 1991. The draft assessment also proposes, for the first time, to establish a daily inhalation concentration level of methanol that purportedly is safe for humans. Previous EPA assessments of methanol did not set an inhalation exposure level, as the available health effects data was thought to be inadequate for determining an allowable safe inhalation exposure level for humans. The current methanol draft assessment has been met with controversy because it is based in part on Italian cancer studies that have been questioned by the National Toxicology Program (NTP), which resulted in EPA withholding the cancer risk portion of the methanol assessment pending further review. The assessment also has been criticized because the proposed allowable concentration for exposure overlaps at its lower end with the amount of methanol that naturally occurs in background levels regularly encountered by humans through consuming methanol-containing products such as orange juice.

Methanol in Everyday Life

Methanol, also known as methyl alcohol, wood alcohol, methyl hydroxide, wood naphtha, and wood spirit, is a clear, colorless liquid that has an alcoholic odor. Methanol is naturally occurring and one of the most basic, common, and simply structured organic chemicals. Not surprisingly, it is among the highest production volume chemicals reported in EPA's Toxic Release Inventory. It has many commercial uses and is a basic building block for hundreds of chemical products with an estimated annual production of 12 billion gallons globally and 1.3 billion gallons in the United States. Methanol is used as a feedstock for formaldehyde, which in turn is used in the production of fiberboard, plywood, and particle board for the construction and automotive industries. It is also used as a feedstock in the production of methyl t-butyl ether, a gasoline additive for use in automobiles and small engines. Consumer products that contain methanol include varnishes, shellacs, paints, windshield washer fluid, antifreeze, de-icers, fuel additives, and adhesives.

In addition to its commercial applications, methanol is naturally found in humans as a result of both metabolism and through dietary sources such as fruit, fruit juices, vegetables, certain artificial sweeteners, and alcoholic beverages. Methanol can be measured in exhaled breath and body fluids.

Methanol as a Potential Risk to Humans

Acute toxicity from methanol can result from relatively low exposure due to a metabolic process that predominantly affects the nervous system, with potentially lasting effects such as blindness, Parkinson-like symptoms, and cognitive impairment. Acute symptoms, which generally include nausea, dizziness, and headache, have been noted in adult human subjects exposed to as little as 200 ppm (262 mg/m3) methanol - the current threshold limit value published by the American Conference of Governmental Industrial Hygienists. Methanol is easily absorbed through dermal, inhalation, and oral routes of exposure and is rapidly metabolized into formaldehyde in both animals and humans. Further, because children generally have higher baseline breathing rates and are more active than adults, there is concern that they may be subject to higher methanol doses than adults exposed to equivalent concentrations.

Several case reports cited by EPA in its draft report suggest an association between the methanol-induced Parkinson-like symptoms and lesions in the brain similar to those seen in Parkinson's disease. Despite the existence of many case reports on acute human exposures, the knowledge base for long-term, low-level exposure of humans to methanol is limited.

What the Draft Report Says

EPA's draft assessment provides estimates of noncancer risks from oral and inhalation exposure to methanol beyond naturally occurring background levels. Based almost solely on animal studies that show changes in brain, skeletal, and liver weights in laboratory rats exposed to methanol, EPA has made a proposed determination that methanol exposure poses a developmental and reproductive risk to humans. It proposes this determination despite acknowledging the "limited data regarding the reproductive or developmental toxicity in humans."

Based on animal studies, EPA proposes lowering the oral reference dose (RfD) (expressed in units of milligrams per kilogram per day [mg/kg/day]) to 0.4 mg/kg/day. The RfD is an estimated acceptable level for daily oral methanol exposure for humans (including susceptible subgroups such as children) where there will likely be no risk of harm. The previous IRIS assessment for methanol included an RfD of 0.5 mg/kg/day that was derived from a 1986 EPA subchronic oral study of Sprague-Dawley rats. EPA's reassessment of the RfD is based largely on acute animal studies since there are few studies of chronic oral exposure to methanol in animals and there are no studies discussing long-term methanol exposure in humans.

Moreover, EPA for the first time set an inhalation reference concentration (RfC) (expressed in units of milligrams per cubic meter [mg/m3]) of 2 mg/m3. The RfC is analogous to the oral RfD in that it sets an allowable level of exposure to methanol via inhalation exposure. Previously, EPA did not believe that there was sufficient data to set a methanol RfC for humans. In its draft report, EPA derived the proposed RfC from an inhalation study performed on rats and extrapolated results from existing oral exposure studies. Based on lack of data, EPA used a remarkable human exposure modeling to set the new RfC.

EPA seeks comment on the scientific soundness of the model used in the assessment, the scientific justification for the subtraction of background levels of methanol from the data in relation to the quantification of noncancer risks, and the scientific justification of the extrapolation approach from rats to humans in relation to inhalation exposures.

Why Industry is Concerned

The methanol industry has questioned EPA's methods and conclusions. Gregory Dolan, executive director of the Methanol Institute, concluded that the current assessment is "so fundamentally flawed that to proceed with peer review of the [draft assessment] would be meaningless, and in fact, counterproductive to the scientific process." This criticism is based, in large part, on EPA's reliance on data from the Italian Ramazzini Institute, which EPA used to calculate its methanol cancer risk assessment. Following the NTP's review of some of Ramazzini's original methanol preliminary study results last spring, EPA announced that it was "holding" four draft IRIS assessments and two published IRIS assessments that reference Ramazzini data until it could further investigate. However, EPA relies on the same Italian cancer data in the current methanol noncancer risk assessment because, EPA says, this data is the only oral study of methanol available.

Since data for subchronic, chronic, or in utero human exposures to methanol are very limited or nonexistent, EPA's determinations regarding the longer-term effects of methanol are based primarily on rat studies, including some in which the lab rats inhaled methanol, to calculate human oral exposure levels. There is substantial controversy regarding the applicability of rat models to calculate human exposure estimates, and some have suggested that rabbits are a better predictor of human effects than rats.

Finally, EPA's draft largely fails to consider human background levels of methanol that range from 0.4 mg/L to 4.0 mg/L - overlapping the proposed oral exposure level. Based on EPA's analysis, a person who routinely drinks a glass of orange juice, which, according to a 1993 study by the UK Department of Health, contains as much as 420 mg/kg of methanol, would be consuming levels of methanol that EPA considers harmful.

Because of its reliance on the questionable Italian preliminary study results and its extrapolation of limited animal studies to assess the risk of harm to humans, EPA's draft assessment of the noncancer risks of methanol exposure has been questioned by the NTP and industry alike.