Language selection

Government of Canada / Gouvernement du Canada

Search

Proposed health-based air quality objectives for arsenic

Draft document for public consultation

Consultation period ends: July 21, 2026

On this page

  1. Preamble
  2. 1.0 Proposed health-based air quality objectives (HBAQOs) for inorganic arsenic (As)
  3. 2.0 Executive summary
    1. 2.1 Health effects
    2. 2.2 Short-term health-based air quality objective
    3. 2.3 Long-term health-based air quality objective
  4. 3.0 Pollutant properties, sources and exposures
    1. 3.1 Substance identity
    2. 3.2 Sources and environmental fate
    3. 3.3 Exposure and biomonitoring
    4. 3.4 Populations who may be disproportionally impacted by exposure
  5. 4.0 Health effects
    1. 4.1 Toxicokinetics
    2. 4.2 Short-term health effects
    3. 4.3 Long-term health effects
    4. 4.4 Mode of action
    5. 4.5 Considerations for presence of a threshold for cancer
    6. 4.6 Populations who may be disproportionally impacted due to health
    7. 4.7 Selected key studies
  6. 5.0 Derivation of the short-term health-based air quality objective
  7. 6.0 Derivation of the long-term health-based air quality objective
  8. 7.0 References
  9. Appendix A: List of acronyms and abbreviations
  10. Appendix B: International considerations

Preamble

Health Canada is committed to helping people in Canada maintain and improve their health. As a part of this mandate, Health Canada has initiated the development of health-based air quality objectives (HBAQOs) for pollutants that may be found in ambient air and that may be harmful to human health. HBAQOs are intended to assist all levels of government and other air partners in assessing and communicating risk, and in informing the need for risk management actions.

HBAQOs are non-regulatory values that represent a concentration of a pollutant, averaged over a recommended timeframe, below which health effects are not expected. They take into consideration short-term and long-term exposure to pollutants, and that specific populations may be disproportionally impacted due to their health status, the community they live in or other factors. For particular health effects, certain pollutants do not have concentrations that are considered without any health risk, in which case proposed values resulting in negligible health risk are provided.

This document provides an overview of the sources of and exposure to arsenic from ambient air in Canada, reviews the most up-to-date information on the health effects of arsenic, and outlines the approach used to derive proposed short-term and long-term HBAQOs. HBAQOs undergo expert peer-review prior to public consultation to ensure that they are protective and based on sound science.

This document is available for a 60-day public consultation period. All comments must be received before July 21, 2026. Please send comments (with rationale, where required) or questions to Health Canada via email: air@hc-sc.gc.ca.

1.0 Proposed health-based air quality objectives (HBAQOs) for inorganic arsenic (As)

Proposed HBAQO (as As) Recommended averaging time Health effect
0.2 µg/m3 1 hour Fetal developmental effects
0.001–0.01 µg/m3Table footnote * Annual Lung cancerTable footnote **

2.0 Executive summary

Arsenic is a natural element found in its inorganic form within the Earth's crust. Releases of arsenic to ambient air, predominantly as inorganic arsenic trioxide (As2O3) dusts, can occur during activities such as mining, smelting, energy production, incineration of waste products and during wildfire events. Proximity to sources is an important consideration for exposure to arsenic in ambient air. The proposed HBAQOs are applicable to all inorganic forms of arsenic (as elemental As) and does not include organic arsenic compounds or arsine gas.

2.1 Health effects

Short-term inhalation of arsenic irritates the respiratory tract and impacts the developing fetus as shown by reduced fetal weight and congenital malformations in offspring of exposed animals.

Long-term inhalation of arsenic causes respiratory and lung cancer as shown in smelter workers and experimental animals. Arsenic is a known human carcinogen listed on the Canadian Environmental Protection Act, 1999 (CEPA) Toxic Substances List under Schedule 1.

2.2 Short-term health-based air quality objective

The proposed short-term HBAQO for arsenic in ambient air is 0.2 µg/m3, with a recommended averaging time of 1 hour.

2.3 Long-term health-based air quality objective

The proposed long-term HBAQO for arsenic in ambient air is a range of 0.001–0.01 µg/m3, with a recommended averaging time of one year (annual).

3.0 Pollutant properties, sources and exposures

3.1 Substance identity

Arsenic is a metalloid occurring naturally in the Earth's crust. It exists in 4 oxidation states (0, -3, +3 and +5) with trivalent arsenic (arsenite; As+3) and pentavalent arsenic (arsenate; As+5) being the most common. Arsenic is rarely found in its pure form but rather occurs as part of inorganic compounds within rock and mineral ore formations, or organic compounds in fish and seafood. Smelting of these ores releases inorganic arsenic as a fine dust in the form of arsenic trioxide (As2O3). As2O3 is the most common form of arsenic in ambient air and is recognized as one of the most toxicologically relevant forms of arsenic (US EPA, 1995; Styblo et al., 2000; Kuivenhoven and Mason, 2023).

Inorganic arsenic compounds in general are commonly grouped together for human health risk assessment purposes, given similar health effects, potency and metabolic processes, as well as due to a lack of detailed speciation data on individual compounds contributing to potential exposures in air. As such, this document focuses exclusively on inorganic arsenic species, and the resulting HBAQOs are relevant to all inorganic forms of arsenic. This assessment does not cover organic arsenic compounds or arsine gas, both of which have different toxicological profiles and are not considered relevant in the context of exposure from ambient air.

3.2 Sources and environmental fate

Arsenic is introduced into the environment through natural sources such as the erosion and weathering of rock and soil containing arsenic, as well as from anthropogenic activities including mining and quarrying, metal production such as steel manufacturing, base-metal smelting, energy production from fossil fuels such as coal, incineration of waste products, and the historical use of arsenicals for wood preservation (Hindmarsh and McCurdy, 1986; Hutton and Symon, 1986; ATSDR, 2007; IARC, 2012). Key sources to ambient air are generally limited to anthropogenic point sources including mining, base-metal smelting activities, burning of fossil fuels (coal) and waste products, and steel manufacturing (ATSDR, 2007; ECCC, 2017), where approximately 2-thirds of the atmospheric flux of arsenic is of anthropogenic origin (Patel et al., 2023). Data from the 2023 National Pollutant Release Inventory (NPRI, 2023) reporting year describes a release of 15,413 kg of arsenic into ambient air from 191 facilities, with the majority emitted from one smelting operation. Risk management actions under CEPA have focused on mitigating releases of arsenic into the environment from power generation, mining and base-metal smelting, wood preservation and steel manufacturing (ECCC, 2017).

When released to air, arsenic mainly exists as a mixture of arsenite and arsenate, and is commonly adsorbed onto particulate matter (PM) and dispersed by the wind. Its residence time in the atmosphere is dependent on the size of the PM. For example, arsenic associated with coarse particulate matter (more than 2.5–100 µm diameter) generated from mechanical processes, such as the crushing and grinding of ore, has a relatively short atmospheric residence time of minutes to hours due to a larger settling velocity (Csavina et al., 2012). However, arsenic adsorbed to PM10 (less than 10 µm diameter) may be resuspended via wind erosion and mechanical disturbance (Seinfeld and Pandis, 2006; Csavina et al., 2012). Processes resulting in the release of PM2.5, such as those produced during smelting and combustion through the condensation of high-temperature vapours, diffusion and coagulation (Seinfeld and Pandis, 2006; Csavina et al., 2012), may remain in the atmosphere between 7 to 10 days (Rahn, 1976, Matschullat, 2000), and can travel long distances before settling by dry and wet deposition (ATSDR, 2007).

Climate change is considered to be an emerging factor impacting the chemistry and increasing mobility of arsenic in weathered mine wastes (Martin et al., 2014). Similarly, climate change continues to increase both annual area burned and soil burn severity. For example, modelling estimates suggest that, since 1972, wildfires around Yellowknife alone have potentially led to the release of 141–562 tonnes of arsenic, with 61–381 tonnes emitted to the atmosphere, previously contained in forests, exposed rocks, topsoil and peatlands (Sutton et al., 2024).

Arsenic in the vapour phase is only released by high-temperature volatilization, and thus ambient air concentrations of vapour-phase arsenic are generally low and may only be of concern in the vicinity of some high-temperature industrial processes (WHO, 2024).

3.3 Exposure and biomonitoring

Point sources are significant contributors of arsenic emissions to ambient air although, for the general population, exposure from air is typically low and rather occurs primarily through other exposure routes such as food (rice, fruit products and cereal) and drinking water (in regions with arsenic deposits) (CFIA, 2011; Health Canada, 2021a, 2022, 2025). Exposure data from theNational Air Pollution Surveillance (NAPS) program reported that 24-hour average arsenic concentrations associated with PM across 16 urban and rural stations within Canada ranged from less than 0.000016–0.74 µg/m3 (mean 0.0009 µg/m3; median 0.0004 µg/m3) between 2009 and 2013, across 4,128 samples (Galarneau et al., 2016). Exposure can be considerably higher near activities that release arsenic into the air. For example, annual average concentrations of arsenic in PM measured between 1996-1998 from predominantly residential areas in the vicinity of smelting and refining facilities in Quebec (Rouyn-Noranda; Murdochville and Montreal), Ontario (Sudbury and Timmins), Manitoba (Flin Flon) and British Columbia (Trail) ranged from 0.006–0.589 µg/m3 compared to a composite background concentration of 0.00062 +/- 0.00028 µg/m3 (annual average) from sites with no point sources (Newhook et al., 2003).

More comprehensive data sets are also available that outline spatial and temporal trends of ambient arsenic concentrations in air near local smelting, refining and/or electric power generation operations. In the Belledune (New Brunswick) and Rouyn-Noranda communities, concentrations of arsenic in air have generally shown a downward trend in annual average concentrations over time, and with distance from the facilities. Between 1967 and 2005, estimated concentrations in Belledune declined from 0.2 µg/m3 to 0.0009 µg/m3 at the site most impacted by the emissions. In Rouyn-Noranda, data from 1993 to 2005 indicates annual average arsenic concentrations ranging from approximately 0.2–1.04 µg/m3 at the monitoring station nearest to the facility and 0.06–0.26 µg/m3 at a station farther away and, post-2005 concentrations typically ranged from 0.11–0.2 µg/m3 at the station adjacent to the smelter to less than 0.05–0.068 µg/m3 at stations farther away (CISSS-AT, 2019; Valcke et al., 2022). Regarding average daily concentrations, measurements from 2018 indicate a similar concentration gradient relative to distance from the facility (minimum: 0.0005 µg/m3; maximum: 1.04 µg/m3) with significantly higher short-term peak concentrations, as qualitatively described by the authors (CISSS-AT, 2020). An air monitoring survey between October 2003 to September 2004 at various locations in the Greater Sudbury area near local mining, smelting and refining operations reported daily average arsenic concentrations of 0.0024–0.061 µg/m3 (Sudbury Soils Study, 2008a), and were considered elevated relative to the reference value of 0.001 µg/m3 representative of typical concentrations (Sudbury Soils Study, 2008b).

Due to the persistent nature of arsenic and atmospheric settling, historical and current emissions can also impact concentrations in soil and other media, thus also contributing to exposures. For example, elevated concentrations of arsenic in soil have also been reported in communities living within the reach of the emission plume or raised dust material (Valcke et al., 2022; MELCCFP, 2023). As well, legacy arsenic contamination in soil may impact air quality following environmental mobilization from wildfire events. Air monitoring data collected from 2023–24 in the vicinity of Giant Mine, an abandoned mine site in Yellowknife (Northwest Territories) with elevated levels of arsenic in soil, sediment and water from historical emissions of arsenic trioxide to air between 1948 and 2004 (CIRNAC, 2018), indicated that local and regional wildfires were associated with intermittent increases in arsenic measured in PM (CIRNAC, 2024). Specifically, arsenic concentrations in 24-hour PM10 samples were above 0.3 µg/m3 over a period of 4 days during the 2023 wildfire season, due to wildfire smoke, or a combination of smoke and wind-blown dust.

It is also noted that arsenic in ambient air can infiltrate the indoor environment and settle with dust, with one group of studies demonstrating homes located near industrial zones had higher dust and metal loading rates as compared to homes in non-industrial zones (for example, via tracked-in soil, wind-blown dust) (Rasmussen et al., 2013, 2018; CISSS-AT, 2019). Arsenic found on settled particles may be ingested due to physical contact, especially in children presenting hand-to-mouth behaviours, or resuspended in dust and inhaled.

Total exposure from all routes is typically measured as arsenic and arsenic metabolites in urine, blood, hair or fingernails (ATSDR, 2007; CISSS-AT, 2020; Health Canada, 2021b). Of these measurements, urinary levels are considered to be the most reliable measurements representing short-term arsenic exposure (that is, exposure over the last few days), whereas levels of arsenic in hair and fingernails produce exposure estimates representing the last 6 to 12 months. In comparison, measurement of blood arsenic is not generally considered to be a reliable means of monitoring human populations for arsenic exposure (ATSDR, 2007).

Urinary levels of arsenic have been monitored through the Canadian Health Measures Survey, providing an integrated snapshot of population exposure from all sources. Concentrations of inorganic arsenic are relatively unchanged in the general Canadian population for age groups 3 to 79, between 2009 and 2019 (Health Canada, 2021b). However, it is acknowledged that individuals living in areas of higher active emissions to air may have higher body burdens of arsenic. For example, recent data investigating nail arsenic levels in residents of Rouyn-Noranda, in comparison to a non-exposed reference population, shows levels approximately 4 times higher in adults and children residing near the smelter (CISSS-AT, 2019, 2020).

Other Canadian biomonitoring initiatives measuring arsenic in target populations include the Maternal-Infant Research on Environmental Chemicals (MIREC) study and the First Nations Biomonitoring Initiative (FNBI). MIREC is a national-level prospective biomonitoring study carried out in pregnant people aged 18 years and older recruited from 10 cities across Canada between 2008 and 2011. The FNBI focused on biomonitoring of First Nations people (20 years of age and older) living on-reserve south of the 60° parallel in 2011. In both the MIREC study and FNBI, arsenic concentrations were found to be generally lower relative to national figures (that is, general Canadian population living in the 10 provinces, from the Canadian Health Measures Survey) or for women of child-bearing age and the overall population, respectively (Health Canada, 2021b). For both sets of results, investigators recommended further study to elucidate factors contributing to these differences (Assembly of First Nations, 2013; Ettinger et al., 2016).

3.4 Populations who may be disproportionally impacted by exposure

Certain populations are disproportionally exposed to arsenic in contrast to the general population. This includes people residing in areas with higher natural background concentrations and/or living or working near current or historical industrial releases of arsenic (Gump et al., 2023). Here, exposure to higher levels of arsenic relative to the general population may occur via ambient air, indoor dust, drinking water and country foods.

It should also be noted that individuals residing near point sources of arsenic may also be disproportionately exposed to other pollutants of concern, which may impact overall health risk from environmental pollutants.

4.0 Health effects

Health Canada considers the available scientific information to develop informed opinions based on a weight-of-evidence approach when deriving HBAQOs. The toxicological and epidemiological database for inorganic arsenic is extensive and spans various associations and/or causal relationships such as developmental effects, neurological effects, diabetes, cardiovascular disease and cancer. Humans appear to be more sensitive to chronic arsenic toxicity than laboratory animals, most likely due to differences in metabolism. However, for effects related to acute toxicity, less data is available that show differences across species. Thus, while this section focuses primarily on human data involving exposures via inhalation, animal data is included where relevant. It is also important to note that many of the effects described in this section are also observed following oral exposures, as arsenic exerts similar effects once absorbed in the systemic circulation (Smith et al., 2009). For effects following both short-term and long-term exposures, trivalent arsenic is considered to be the more toxic form compared to the pentavalent state (Styblo et al., 2000; Kuivenhoven and Mason, 2023).

Health Canada commissioned reports (RSC, 2019; RSI, 2022, 2023a) to evaluate the causal relationship between arsenic exposure and non-cancer endpoints, to investigate the mode of action (MOA) leading to the development of cancer from both inhalation and oral exposure to inorganic arsenic, and to examine the statistical relationships between exposure and cancer mortality in select occupational cohorts. As well, complementary literature searches focusing on peer-reviewed articles and international agency assessments of adverse health effects specific to inhalation exposures were carried out to identify key primary studies for in-depth analysis. The methods used in each published review article were critically evaluated to assess the degree of confidence in study conclusions, ensuring that only the strongest reviews from the literature were consulted as sources for identifying key primary studies. This current document focuses on these key cancer and non-cancer health endpoints.

4.1 Toxicokinetics

The major routes of arsenic absorption in the general population are ingestion and inhalation. Most inorganic arsenic compounds are well absorbed (more than 80%) from the gastrointestinal tract (IARC, 2012), while estimates of absorption via inhalation remain varied, ranging between 40% and 90% (Vahter et al., 1986; Yip and Dart, 2001; ATSDR, 2007). For both inhalation and ingestion, soluble forms are more readily absorbed than insoluble forms (Maud and Rumsby, 2008). Because inorganic arsenic in air is associated with PM, absorption will depend on particle deposition within the lung (Sturm, 2012). Larger airborne arsenic-containing particulates deposited in the upper airways may be swallowed following mucociliary clearance and eventually absorbed in the intestines (Mutlu et al., 2018), with particles under 2 µm reaching the alveoli being available for systemic absorption (Maud and Rumsby, 2008).

Data on distribution after inhalation is limited, but it appears that arsenic is available to nearly all tissues, as it is distributed in the bloodstream bound to hemoglobin within erythrocytes (Axelson, 1980). Arsenic accumulates in keratin-rich tissues and are therefore at highest concentrations in tissue such as hair, nails and skin, with the highest absolute amounts found in muscles, bones, kidneys, liver and lungs. Arsenic in cord blood can cross the placenta and has been found in fetal tissues. It can also be detected in breast milk.

Metabolism is the same by inhalation and oral exposure routes. Briefly, arsenate is reduced in the body to arsenite by various reducing enzymes, which is then oxidatively methylated in the liver by arsenite methyltransferase (AS3MT). The methylation reaction yields organic forms of arsenic, monomethylarsonic acid (MMA) and dimethylarsinic acid (DMA), which facilitates urinary excretion of arsenic due to an increase in solubility. While MMA and DMA are much less toxic than arsenate and arsenite (Sattar et al., 2016), it has been noted that the MMA/DMA ratio may be used as an indicator of cancer risk in arsenic-exposed populations; high levels of urinary MMA correlated to increased cancer risk, and increased urinary DMA showed an inverse risk correlation (Minatel et al., 2018). Due to the strength of the arsenic-carbon bond, inorganic arsenic is not significantly re-formed during metabolism (ATSDR, 2007).

Most inorganic arsenic is rapidly excreted in urine, although small amounts can be eliminated via feces, skin, hair, nails, breast milk and sweat. In humans, the relative proportions of arsenic species in the urine are usually about 10% to 30% inorganic arsenic, 10% to 20% MMA and 60% to 70% DMA, with proportions of inorganic arsenic increasing at higher doses as the methylation capacity becomes saturated (Vahter, 2000; Caldwell et al., 2009). Genetic polymorphisms of enzymes associated with methylation can lead to increased total arsenic retention time in the body, with greater elimination of inorganic arsenic and MMA and reduced elimination of DMA. This may manifest as varied susceptibility to arsenic toxicity across the human population. Some arsenic may remain bound to tissue, depending inversely on the rate of methylation, which varies considerably across tissues (European Commission, 2001; ATSDR, 2007).

The time course of excretion in humans exposed by inhalation has not been thoroughly investigated. The available data suggests that, in general, excretion of arsenic from the body is fairly rapid, with both human and animal studies reporting significant decreases within days (Vahter et al., 1986). That said, non-absorbed arsenic particles deposited in the lung may have residence times spanning years, with a pulmonary clearance half-life of up to 6 years estimated in some occupationally exposed populations (Hazelton et al., 2001), and post-mortem investigations have indicated elevated levels of arsenic in the lungs, liver and kidneys of deceased workers (Brune et al., 1980; Wester et al., 1981; Gerhardsson et al., 1988). For example, autopsy data from smelter workers obtained several years after retirement showed arsenic levels in lung tissue at 6 times higher than in a control group (that is 0.047 µg/g vs 0.008 µg/g, respectively) (Brune et al., 1980).

4.2 Short-term health effects

The occupational and epidemiological toxicity data describing effects from short-term exposures by inhalation of inorganic arsenic in humans is limited; rather, most of the available data representing short-term effects are from ingestion poisoning scenarios. In addition, there is an inadequate reporting of exposure parameters in many of the available inhalation studies (that is durations, concentrations, co-exposures).

Acute inhalation of high concentrations of arsenic dusts is associated with irritation to the mucous membranes of the nose and throat, as described in case reports of arsenic poisoning. This may cause bronchitis, rhinitis or laryngitis, with potential for tracheal and bronchial hemorrhage and, in severe cases, perforation of the nasal septum (UKPID, 1997). In most cases, inhaled quantities are usually considered insufficient to cause overt symptoms of acute systemic toxicity (ATSDR, 2005). However, for those short-term exposure scenarios where systemic effects are observed following acute exposures to trivalent arsenic (both oral and by inhalation), toxicity is thought to be attributable to its ability to bind to cellular proteins containing sulfhydryl groups. This inhibits the production of energy needed to maintain tissue functions, and results in a decrease in glutathione, which is necessary for the metabolic detoxification of arsenic. The primary target organs are the gastrointestinal tract, heart, brain and kidney. Eventually, the skin, bone marrow and peripheral nervous system are also affected. Arsenic trioxide has direct toxic effects on endothelial cells, increasing the permeability of small blood vessels (Kuivenhoven and Mason, 2023). As well, pentavalent arsenic may replace phosphate in several biological reactions, impacting metabolic processes involved in generating energy (Kuivenhoven and Mason, 2023).

In laboratory animals, adverse effects to the respiratory system and immune system have also been reported in various investigations following acute inhalation exposures (ATSDR, 2007). Studies typically exhibited dose-dependent markers of non-specific lung toxicity, including increased pulmonary bactericidal activity (presumably as a result of injury to alveolar macrophages) and a corresponding concentration-related increase in susceptibility to introduced respiratory bacterial pathogens, as witnessed in mice after single 3-hour exposures to 270, 500 and 940 µg/m3 As2O3 (Aranyi et al., 1985). Additionally, rats exposed to 0.16, 1.60 and 16 μg/kg As2O3 via intratracheal instillation displayed statistically significant changes, including increased lung water content ratio, increased protein in the bronchoalveolar lavage fluid, pulmonary interstitial thickening, cell membrane edema and increased inflammatory cytokines (Mingxing et al., 2019). Following a 14-day study in mice limited to 2 exposure concentrations (50 µg/m3 or 1,000µg/m3 As2O3), the antibody response in a plaque assay was decreased by ∼70% post-immunization in both exposure scenarios, with no change observed in lipopolysaccharide-induced B-cell proliferation, suggesting an inhibition of T cell–dependent antibody response by B-lymphocytes (Burchiel et al., 2009). These results, as well as other studies employing other routes of exposure, suggest that arsenic can act as an immunotoxicant, interfering in both innate and adaptive immune response functions (Giles and Mann, 2022).

Short-term exposure to inorganic arsenic across multiple routes, including inhalation, may also lead to adverse effects on the brain and nervous system. Data from occupational settings and poisoning scenarios have reported neurological symptoms such as light-headedness, delirium, encephalopathy and peripheral neuropathy, as described in a case study of a worker spending several days working within a smelting furnace without the required respiratory protection (Beckett et al., 1986). A report of 4 patients with arsenic poisoning confirmed by laboratory tests (for example, urine levels, fingernail levels) exhibited progression to distal axonal degeneration, causing weakness and tingling in arms and legs (Donofrio et al., 1987). The MOA for neurological effects has been postulated to involve broad perturbation of multiple cellular pathways via induction of oxidative stress, thiamine deficiency and decreased acetyl cholinesterase activity (Mochizuki, 2019). One study in mice found that exposure to 3 and 10 mg As2O3/kg/day orally for 14 days resulted in changes in the concentrations of metabolites of norepinephrine, dopamine and serotonin in several brain regions, including the cerebral cortex, hippocampus, hypothalamus and corpus striatum (Tadanobu et al., 1990).

Although often attributed to repeated exposures, developmental toxicity from exposure in utero can be considered an acute endpoint, due to the unknown and potentially small window of time in which adverse developmental effects can occur during gestation, and the unique susceptibility of the fetus at discrete times (Davis et al., 2009; OECD, 2011). As such, repeated exposure is not a necessary prerequisite for developmental toxicity to manifest. Although the arsenic exposures were not reported, developmental effects associated with occupational and environmental exposures have been noted in a series of studies at and near the Rönnskär copper smelter in Sweden (Nordström et al., 1978a, 1978b, 1979a, 1979b). Here, pregnant employees of the smelter were found to have an increased incidence of spontaneous abortion (Nordström et al., 1979a), their children had an increased incidence of congenital malformations (Nordström et al., 1979b) and decreased average birth weight (Nordström et al., 1978a), with all associations being statistically significant. For pregnant women living in close proximity to the smelter, an increased incidence of spontaneous abortion and decreased average birth weight was also observed (Nordström et al., 1978b, 1979b). As an example, the children of mothers who had not been employed during pregnancy exhibited significantly decreased (p< 0.01) birth weights only when the mother had lived close to the smelter (under 10 km) during pregnancy (Nordström et al., 1979b). A case-control study of stillbirths in the vicinity of a Texas arsenic pesticide factory using an atmospheric dispersion model over a 10-year period showed a statistically significant increase in the risk of stillbirth in the highest exposure category (over 100 ng As/m3) only among Hispanics, which was not observed in white non-Hispanics or African Americans (Ihrig et al., 1998). Here the authors hypothesize that a genetic polymorphism affecting folate metabolism in Hispanics may be behind the increased risk, but the mechanism remains unclear.

Effects of exposure to arsenic during pregnancy have also been investigated in animal toxicity studies. In contrast to epidemiological data, these studies had better characterization of concentration-response relationships, particularly at low levels of exposure. Maternal inhalation of arsenic in rodents during gestation has demonstrated developmental toxicity in the offspring. These studies examined various fetal, offspring and maternal endpoints, and also suggest that mice tend to be more sensitive than rats to developmental toxicity after arsenic exposure. It is noted that 2 studies (Nagymajtényi et al., 1985; Holson et al., 1999) were featured prominently in the derivation of short-term air quality objectives by other international agencies (see Appendix 1). Nagymajtényi et al. (1985) exposed pregnant mice to 0, 0.26, 2.9 or 28.5 mg As2O3/m3 for 4 hours on days 9–12 of gestation and examined several fetotoxic endpoints. In all 3 exposure groups, a statistically significant decrease in fetal weight was observed with a 23%, 9.8% and 3.5% reduction reported, from the high-exposure to the low-exposure groups, respectively. At the highest concentration, a statistically significant increase in the number of fetuses with retarded growth and skeletal malformations were observed. The percentage of dead fetuses increased with higher doses but was not statistically significant. Additionally, the frequency of fetal chromosomal aberrations displayed a statistically significant increase in the highest exposure group. In comparison, results from the study of Holson et al. (1999), which employed multidose whole-body inhalation exposure (0.3, 3.0 and 10.0 mg/m3 As2O3) to female rats 6 hours/day from 14 days prior to mating, through mating and until gestation day 19, displayed respiratory distress (rales), decrease in net body weight gain and a decrease in food intake. Here, maternal toxicity was reported at a lowest-observed-adverse-effect level (LOAEL) of 10 mg/m3, and developmental toxicity in pups was not observed at any concentration.

4.3 Long-term health effects

The epidemiological database for effects related to long-term inhalation of inorganic arsenic included numerous primary studies and reviews in the peer-reviewed literature, along with several assessments by regulatory agencies and authoritative bodies.

4.3.1 Cancer

Inorganic arsenic is well recognized as a human carcinogen across numerous government agencies and recognized authorities including

The above classifications reflect the results of epidemiological studies, carcinogenicity studies in experimental animals, investigations describing its metabolism in the body and modes of action of carcinogenicity, and apply to the group of inorganic arsenic compounds as a whole, given the shared metabolic pathway of elemental and inorganic species (IARC, 2012).

4.3.1.1 Carcinogenicity in animals

Historically, the carcinogenic potential of arsenic in animals was not as clearly demonstrated as it has been in humans, in part due to negative findings in experimental animals, likely reflecting incomplete (for example, lack of full life-stage studies) and a limited number of studies (Huff et al., 2000; Tokar et al., 2010). More recent studies across multiple routes of exposure have provided findings of carcinogenicity in rodents, with IARC (2012) concluding that there was sufficient evidence overall that arsenic and arsenic compounds cause cancer in experimental animals, although the evidence specific to arsenic trioxide was considered limited. Specifically, data from exposures via inhalation is limited to a study of rats and mice exposed to gallium arsenide (NTP, 2000). Here, female rats exposed to 0, 0.01, 0.1 and 1 mg/m3 for 6 hours/day 5 days/week over approximately 2 years exhibited a dose-related increase in lung adenoma, lung adenoma/carcinoma (but not carcinoma alone), mono-nuclear cell leukemia and adrenal medulla pheochromocytomas (last tumour type is not likely relevant to humans). However, no significant effect on development of tumours was observed in male rats using the same exposure protocol. Male and female mice exposed to several doses of gallium arsenide also did not develop dose-dependent increases in tumours.

Limited conclusions can be drawn from intratracheal instillation studies regarding their relevance to inhalation. One study demonstrated a non–statistically significant increase in respiratory tumours of hamsters exposed to intratracheal instillations of 3 or 6 mg/kg body wt of arsenic trioxide weekly over 15 weeks (Pershagen et al., 1984). However, 2 studies using 3 or 3.75 mg/kg body wt of calcium arsenate in male Syrian hamsters exposed weekly over 15 weeks developed a statistically significant increase in respiratory tumourigenicity (p<0.05) (Pershagen and Björkland, 1985; Yamamoto et al., 1987), whereas a similar exposure regimen with arsenic trioxide was found to be inconclusive for respiratory tumourigenicity (Yamamoto et al., 1987).

4.3.1.2 Carcinogenicity in humans

Despite limited available data in animals, particularly related to exposures by inhalation, several epidemiological studies have been reported in arsenic-exposed smelter workers that consistently indicate that inhalation exposure to inorganic arsenic (primarily to arsenic trioxide dust) increases the risk of cancer, particularly respiratory and lung cancers. Indeed, IARC (2012) concluded that there is sufficient evidence in humans for the carcinogenicity of mixed exposure to inorganic arsenic compounds, including arsenic trioxide and other arsenite and arsenate compounds. Of the available data, 3 copper smelting occupational cohorts (that is, the Asarco Tacoma cohort in Washington, the Anaconda cohort in Montana and the Rönnskär cohort in Sweden) provided the strongest evidence of the relationship between inhalation of inorganic arsenic and respiratory and lung cancer mortality, and are discussed in detail below.

4.3.1.3 Asarco Tacoma Copper Smelter

Several analyses examined cancer mortality for the Asarco Tacoma cohort of Tacoma, Washington (Pinto et al., 1977; Enterline and Marsh, 1982; Enterline et al., 1987a; Enterline et al., 1995). This cohort included 2,802 primarily Caucasian male workers (mean age of 30 at first exposure), who worked at the smelter for at least 1 year between 1940 and 1964, with follow-up until 1986. As described in the most recent analysis by Enterline et al. (1995), cumulative exposures (expressed as µg/m3-yr) were estimated retrospectively using measurements of arsenic trioxide in air collected between 1938 and 1957, job histories, personal air measurements between 1971 and 1984 and urinary arsenic levels from workers beginning in 1948. Information on smoking was not available. Mean cumulative exposures of the lowest (less than 750 µg/m3-yr) and highest (over 45,000 µg/m3-yr) groups were 405 µg/m3-yr and 58,957 µg/m3-yr, respectively. The standardized mortality ratio (SMR) representing the ratio of observed deaths in the exposed cohort to the expected number of deaths in the reference population (in this case, cancer mortality rates for Caucasian men for the State of Washington) was used to measure the effect of exposure. With 1,583 deaths observed in the cohort, there were statistically significant increases in mortality for all malignant neoplasms taken together, cancer of the large intestine, cancer of the respiratory system, and bone cancer when compared to the reference population (Enterline et al., 1995). The SMRs for respiratory cancer mortality (for example, 2.1 for the total cohort [p<0.01]) were considered to have the strongest correlation with cumulative exposure to arsenic (relative to other sites) and confirmed a previously-reported supra-linear exposure-response relationship (that is, steeper at lower doses) found in the same cohort. Respiratory cancer mortality represented the only relevant, statistically-elevated SMR for exposures occurring over less than 20 years.

4.3.1.4 Anaconda Copper Smelter

An elevated risk of respiratory cancer mortality among workers from the Anaconda copper smelter in Montana (operating from 1884 to 1977) was originally reported by Lee and Fraumeni (1969). The cohort included 8,045 primarily Caucasian male workers employed for at least 1 year between 1938 and 1956 and was followed through to 1989 (mean duration of 10.3 years), where a total of 4,930 (63%) were deceased, including 446 from respiratory cancer. Information on smoking was not available. Updates and further cohort and nested case-referent analyses have since been published (Lubin et al., 1981, 2000, 2008; Welch et al., 1982; Brown and Chu, 1983a, 1983b; Lee-Feldstein, 1983, 1986, 1989).

Exposure was estimated using 702 measurements of airborne arsenic between 1943 and 1958 to describe time-weighted average exposures for light, medium and heavily exposed areas. In more recent analyses, these respective estimates were refined to also incorporate workers' exposure times within the different areas and respiratory protection, resulting in time-weighted averages of 290, 580 and 11,300 µg/m3, with an overall mean concentration of 350 µg/m3. All evaluations showed statistically significant increases in SMRs for respiratory cancer mortality, with Lubin et al. (2000) illustrating that risk associated with cumulative arsenic exposure was consistent with a linear dose-response relationship. The authors also concluded that it was unlikely that smoking confounded the assessment of lung cancer risk based on an analysis of the proportion of cigarette smokers relative to the extent of exposure to arsenic.

In the analysis by Lubin et al. (2008), an updated SMR for respiratory cancer mortality was calculated as 1.56 (p<0.001) based on the age-specific and calendar year–specific US mortality rates for respiratory cancer in White males as the reference population. However, to minimize effects of unmeasured exposures in workers after terminating employment (commonly assumed to be zero in epidemiological studies and potentially underestimating risk), a restricted sub-cohort of workers was analyzed where the SMR was 1.87 (p<0.001). This analysis also reconfirmed the linear relationship of the dose response and showed that risks for respiratory cancer mortality were greater for cumulative exposure of shorter duration at higher concentrations, compared with exposures of longer duration at lower concentrations (that is, there was a concentration-rate effect) (Lubin et al., 2008).

4.3.1.5 Rönnskär Copper Smelter

The Rönnskär Sweden cohort included 3,916 male workers employed from 1928 to 1967 and followed for mortality until the end of 1981 (1,275 deaths reported). It should be noted that this cohort is separate from the women working at or living near the Rönnskär smelter, described in other studies and presented in section 4.2 (Nordström et al., 1978a, 1978b, 1979a, 1979b). In contrast to other smelter cohorts, this cohort includes workers with shorter exposures, with an inclusion criterion of having worked at least 3 months. Although the mean duration of employment was not described, it was noted that some workers had more than 30 years of exposure. Exposure histories were determined via air concentration data and detailed work history documentation (from 1945 onward). Prior to this, exposures were estimated via production figures, changes in production methods, and sick leave data. In the most frequently cited analysis by Järup et al. (1989), a statistically significant increase in lung cancer mortality was shown with increasing cumulative exposures (ranging from less than 250 μg/m3-yr to over 10,000μg/m3-yr). Corresponding SMRs ranging from 2.7 to 11.4, with an overall SMR of 3.7 (3.0–4.5, 95% confidence interval [CI]), were calculated using age-specific mortality rates from lung cancer in Sweden. While the authors did not provide the mean exposure levels within each exposure category, or provide formal dose-response modelling, Viren and Silvers (1994) fit the linear model to the summary data from the Sweden cohort and concluded that a linear dose-response model was the most appropriate. Similar to the Lubin et al. (2008) analysis of the Anaconda cohort, the intensity of arsenic exposure was considered more important than the duration for lung cancer risk (Järup et al., 1989).

4.3.1.6 Other studies

Other studies include a United Kingdom tin smelter cohort (Jones et al., 2007), which described statistically significant excess lung cancer mortality in a cohort of 1,462 male workers employed for at least 1 year between 1967 and 1995 (SMR of 1.6; p<0.001). However, these workers experienced significant co-exposures to other pollutants including lead, cadmium, antimony and radio-nucleotides (polonium-210), potentially confounding the results. A prospective study of a cohort of tin mining and smelter workers in China (Fan et al., 2016) investigated the association between prior diagnosis of pulmonary conditions and lung cancer incidence. However, the workers in this study were co-exposed to underground radon, which showed a stronger association with lung cancer than arsenic. In addition, several studies suggested that some residents living near heavy industrial areas emitting arsenic and other metals (that is, near smelters or arsenical pesticide plants) may have increased risk of lung cancer (Matanoski et al., 1981; Cordier et al., 1983; Brown et al., 1984; Pershagen, 1985; Frost et al., 1987; Bessö et al., 2003). However, these observed increases in risk were small, often not clearly linked to arsenic exposure levels, and in some populations not detectable. Finally, while mortality related to other cancers was also noted (for example, renal, digestive tract, lymphatic), this data was less consistent and SMRs were less than those for lung and respiratory cancers.

4.3.2 Non-cancer

Chronic exposure to arsenic by inhalation has been associated with numerous non-cancer health effects, with varying degrees of strength. Non-cancer effects with the strongest associations to long-term arsenic exposure include vascular and cardiovascular effects, as well as neurological effects. These are discussed below, in addition to diabetes, which is considered to represent an emerging health concern associated with chronic inhalation. All have been well documented following chronic oral exposures, strengthening the potential for their association with exposures via inhalation, given that arsenic exerts similar effects once absorbed and in the systemic circulation (Smith et al., 2009).

4.3.2.1 Vascular and cardiovascular effects

Cardiovascular disease (CVD), that is, disease of the heart or blood vessels, includes coronary heart disease (CHD), atherosclerosis, myocardial infarctions, stroke and heart failure among other conditions. Meta-analyses have shown associations between high exposure to arsenic in drinking water and CVD (Navas-Acien et al., 2005; Moon et al., 2012, 2013; Chowdhury et al., 2018). In comparison, most occupational studies of arsenic in which inhalation is the primary route of exposure have shown excess mortality only for respiratory cancer, and not mortality from cardiovascular or cerebrovascular diseases (Lubin et al., 2000). However, it has been proposed that the high background of CVD mortality (constituting about one-third of total mortality in the general population), together with the healthy worker effect, confound and obscure the analysis of CVD risks from occupational inhalation exposure to arsenic (Hertz-Picciotto et al., 2000). For the Tacoma smelter cohort, Enterline et al. (1995) found an association between ischemic heart disease (SMR = 1.2, p<0.01) and cumulative exposure to arsenic. This association was strengthened when adjustments were made addressing healthy worker survivor effect, although no statistically significant increase in CVD mortality was found (Hertz-Picciotto et al., 2000). For the Anaconda cohort, no statistically significant increases in CVD mortality were observed (Lubin et al., 2000; Lubin and Fraumeni, 2000). However, in a reanalysis accounting for the healthy worker bias, Keil and Richardson (2016) concluded that the cumulative incidence of heart disease was 3 times larger than respiratory cancer by the age of 70.

Limited data also indicate the potential for cardiovascular effects to occur later in life following exposure during childhood. Increased hazard ratios for ischemic heart disease were observed in adult males that resided near the Tacoma smelter during childhood between 1907 to 1932, for more than10 years and within 1.6 km of the facility (1.77; 95% CI = 1.21, 2.58) (Tollestrup et al., 2003). Subclinical CVD (measured as carotid intima media thickness) was significantly increased with increasing total arsenic concentrations in children residing in Syracuse, New York (Gump et al., 2023), where the exposure was largely from air and soil contamination associated with historical industrial pollution.

Results from 3 additional cohorts of tin and copper workers exposed to inhaled arsenic found a consistently increased risk of stroke mortality, but results were inconsistent for CVD and CHD mortality (Binks et al., 2005; Chen et al., 2006; March et al., 2009). Alterations in serum lipoprotein and apolipoprotein levels, considered early CVD risk markers, were significantly associated with urinary arsenic (p<0.001) in 57 plant reclamation workers (duration of exposure to arsenic of 14.72±7.3 years) exposed to arsenic in air estimated to be at or slightly above the US Occupational Safety and Health Administration (OSHA) occupational limit of 10 μg/m3 (Ledda et al., 2017).

No data describing CVD risks from inhalation of arsenic in animal models was identified. However, a recent study involving repeated intratracheal instillation of arsenic in mice (0.06, 0.31, 1.57, 8.51, and 42.5 μg/kg body wt) reported concentration-dependent alterations in markers of cardiac function (that is, changes to heart rate and blood pressure), markers of oxidative stress and cardiac tissue inflammation, considered significantly different at the highest 2 doses (Qi et al., 2024).

With regard to vascular effects, a subset of workers of the Rönnskär copper smelter (47 workers) demonstrated increased incidences of Raynaud's phenomenon (that is, decreased blood flow to the fingers resulting in numbness) at arsenic exposures estimated to be near the applicable Swedish occupational limits (500 μg/m3 from the late 1940s to 1975, and 50 μg/m3 thereafter, reflecting changes in the allowable exposure standard over the course of the study period) (Lagerkvist et al., 1986, 1988). It has been postulated that these functional alterations in vessels may be related to the manifestation of blackfoot disease — an arsenic-associated peripheral vascular disease leading to gangrenous change of the foot and the lower extremities, endemic to 2 confined areas in Taiwan characterized by chronic high arsenic exposure from drinking well water (Tseng, 2008).

4.3.2.2 Neurological effects

Arsenic can permeate the blood-brain barrier and accumulate in many parts of the brain, with the highest concentrations found in the pituitary gland (Tyler and Allan, 2014). Neurological effects have frequently been reported following both oral and inhalation exposure to arsenic. Peripheral neuropathy is the most consistent finding, reported in both oral and inhalation studies. A subset of workers of the Rönnskär smelter (47 workers, as described above by Lagerkvist et al., 1986, 1988) demonstrated a significant correlation between cumulative inhalation exposure and reduced nerve conduction velocity in peripheral nerves (Blom et al., 1985). In a follow-up examination of the same workers, Lagerkvist and Zetterlund (1994) noted that, in addition to peripheral neuropathy, the workers had developed additional neurological symptoms (for example, numbness, muscle pain). A subgroup of Asarco Tacoma smelter workers demonstrated similar peripheral neuropathies (26/61, versus 4/33 in controls), but, unlike the Rönnskär smelter workers, the difference in mean nerve conduction velocities was not statistically significant from controls (Feldman et al., 1979). Other studies of smelter workers exposed primarily via inhalation reported various subjective neurological effects, such as mood and sleep disorders, irritability, fatigue and pain (Hałatek et al., 2009; Sińczuk-Walczak et al., 2010, 2014). In a 40-year retrospective descriptive study of medical records from patients living in proximity to an arsenic mine in Japan, 98% of patients experienced dysesthesia (abnormal sensations of the skin), hearing loss (50%), peripheral neuropathy (29%) and disordered smell perception (28%) (Ishii et al., 2017). For this study, exposure levels were unknown and likely included multiple routes of exposure. Evidence from animal models indicates that neurological damage from arsenic is largely a result of oxidative stress, mitochondrial dysfunction and impaired synaptic activity, which drives cell damage and ultimately, cell death (Vázquez-Cervantes et al., 2023).

Arsenic is associated with neurological impairments in children, largely based on studies of contaminated drinking water (Tyler and Allan, 2014; Tsuji et al., 2015). Less information is available describing neurological effects resulting from inhalation. A significant reduction in cognitive performance scores (for example, verbal IQ, long term memory) related to increasing levels of arsenic in urine was observed in children living near metal smelters with active emissions (Calderón et al., 2001; Rosado et al., 2007); however, concentrations in air were not reported in either study. A follow-up study of children living near one of the smelters did not report significantly different changes in parent ratings of their behaviour (for example, oppositional behaviour, cognitive problems, hyperactivity) (Roy et al., 2011). Of concern is also the potential for effects in the immature brain being not immediately apparent, and instead reflected later in life as permanent effects such as cognitive impairment and motor and memory alterations (Vázquez-Cervantes et al., 2023).

4.3.2.3 Diabetes

Numerous studies have detected connections between prolonged exposure to arsenic via drinking water and an elevated likelihood of diabetes mellitus development, although the underlying mechanisms are unclear (Liu et al., 2023). Significant associations between the total urinary concentration of arsenic and the prevalence of diabetes and prediabetes in the Canadian population has been demonstrated, with adjusted odds ratios of 1.81 (95% CI: 1.12–2.95) and 2.04 (95% CI: 1.03–4.05), respectively, although causal inference is limited in part due to the lack of a long-term exposure assessment (Feseke et al., 2015). Additionally, an association with diabetes has been observed in a review of epidemiological studies of general populations exposed to relatively high elevated levels of arsenic in drinking water (≥ 150 µg arsenic/litre), but no association was found in lower exposures (less than 150 µg arsenic/litre) (Maull et al., 2012). Much less data exists from inhalation studies. In an analysis of a small cohort of workers from the Rönnskär smelter, a higher prevalence of diabetes was determined after long-term follow-up and analysis of death certificates and clinical information from the health care unit of the company, with increasing odds ratios (2.0, 4.2, 7.0) for increasing long-term occupational exposures, estimated as well below, close to, or above 500 μg/m3, respectively (Rahman and Axelson, 1995).

Conversely, no significant association with diabetes was noted from occupational exposures to arsenic in ambient air across 9 cohorts of workers from the pesticide, smelting and glass industries (Navas-Acien et al., 2006). Here, the authors cited the uncertainty in the comparability of study participants with the general population used as reference, limitations in exposure assessment, lack of information on concomitant exposures, lack of information on major diabetes risk factors, and the possibility of a healthy worker survivor effect as factors confounding the assessment. Similarly, a meta-analysis of 6 studies of copper smelter workers, wood workers and pesticide workers, all with an inhalation route of exposure, did not find a statistically significant increase in diabetes risk (relative risk [RR] = 1.08; 95% CI: 0.79–1.46) with none of the studies having data on exposure levels, despite a significant association noted via studies that featured oral exposures (RR = 1.57; 95% CI: 1.27–1.93) (Sung et al., 2015).

4.3.3 Summary of health effects

Taken together, while the above non-cancer health effects are acknowledged as potential adverse health endpoints of concern from long-term inhalation exposures to arsenic, a weight-of-evidence approach demonstrated that respiratory and lung cancer represent the most relevant, consistent and well-documented adverse effects for risk assessment purposes. This endpoint is supported by multiple lines of evidence, along with several well-defined occupational cohorts demonstrating dose-dependent cancer mortality. Indeed, data from these latter cohorts are the basis of all cancer-based long-term exposure air quality objectives published by other agencies for arsenic.

4.4 Mode of action

Studies of the MOA of arsenic have largely focused on its carcinogenicity, which is the focus of this section, given that lung cancer is considered to represent the most sensitive and relevant effect with long-term exposure. Although arsenic is a well-established carcinogen, the exact MOA(s) leading to this effect remain to be fully elucidated. Of note, it is unclear whether inorganic arsenic–induced lung cancer is the result of direct contact of lung tissue with inhaled arsenic (that is, portal-of-entry effect) or redistribution of systemically absorbed arsenic to lung tissue (that is, lung as a target tissue of arsenic in blood), or both.

Extensive literature reviews and analyses of the available MOA data for arsenic were commissioned by Health Canada (RSC, 2019; RSI, 2023a) given the implications to the derivation of the proposed long-term HBAQO. Briefly, the Risk Science Centre (RSC) (2019) utilized epidemiological, animal and in vitro data of exposure to arsenic in a weight-of-evidence approach to arrive to an integrated MOA for different levels of biological organization.

The toxicity of inorganic arsenic is thought to result from 2 primary molecular initiating events: (1) binding to cysteines (sulphydryl groups) in regulatory proteins leading to widespread disruption of a variety of crucial biological functions, and (2) disruption of normal reactive oxygen species (ROS)–mediated cell signaling, oxidative stress and damage to macromolecules (RSI, 2023a). Both molecular initiating events are highly interconnected, often appearing to affect the same or similar downstream events. Although the relative contributions of each are unclear, both appear important in arsenic-induced carcinogenesis (RSI, 2023a). These pathways result in multifaceted genotoxicity, epimutagenicity and dysfunctional DNA repair. This information highlights 2 critical considerations:

With regard to MOA, there is consensus that dysregulation of key genomic pathways, ultimately impacting genomic instability, leads to a loss of growth suppression together and a resistance to apoptosis, resulting in sustained cell survival. These steps are considered to represent the main pathways regulating arsenic-induced lung cancer, resulting in the appearance of abnormal cells, hyperplasia to dysplasia and cancer in situ. Ultimately, arsenic may facilitate the transition to malignant cancer, via mechanisms involving the escape from immune surveillance and destruction, acquisition of replicative immortality, increased angiogenesis, invasion and metastasis (RSI, 2023a). Together, this leads to the progression of clinically evident tumours, metastatic cancer and increased incidence of cancer at the population level. A summary of the available evidence on the MOA and key events sorted by level of biological organization (simplest to most complex) is outlined in List 1.

List 1. MOAs and key events associated with arsenic exposure sorted by level of biological organization. Adapted from RSI (2023a).

Level 1: Toxicokinetics

Level 2: Molecular Initiating Events (MEIs)

Level 3: Biochemical Response from MEIs

Level 4: Cellular Response

Level 5: Tissue Response

Level 6: Organism Response

Briefly, both molecular initiating events appear important in downstream arsenic-induced carcinogenesis, although the relative contributions of each are unclear, as both often appear to cause the same or similar downstream events. For example, chromosomal instability has been shown to result from both direct binding and damaging cysteine-rich telomerase, as well as destabilizing 8-oxo-7,8-dihydroguanine, a key nucleobase in the telomer, due to ROS-mediated oxidation (Smith, 2008). Other examples of important altered functions involving both initiating events (that is, cysteine-binding that alters ROS levels) include metabolism of arsenic by binding to arsenic(+3) methyltransferase (Stýblo et al., 2021), and quenching and detoxification of ROS via superoxide dismutase and glutathione (Xu et al., 2017).

4.5 Considerations for presence of a threshold for cancer

Although there are data in the literature to support a potential MOA involving a threshold dose-response for cancer, there are significant uncertainties which require consideration when choosing the appropriate concentration-response approach. Arsenic-induced cancer is a complex process due to multiple forms of arsenic and metabolites having distinctive potencies and actions with numerous molecular, biochemical and cellular pathway targets. Indeed, there are significant nuanced factors to consider in the development of a long-term cancer-based HBAQO regarding treating inorganic arsenic as a threshold or non-threshold carcinogen: (1) the complex MOA, (2) background cancer risk, (3) and interindividual variability.

First, as described in section 4.4, the arsenic-induced cancer MOA is a complex series of cellular and molecular events, some of which have yet to be fully understood. Predicting cancer incidence based on a single key event is difficult, and there are remaining uncertainties on how molecular mechanisms interact in cancer development at this time.

Moreover, the biological pathways disrupted by arsenic inhalation are the same that are involved in idiopathic lung carcinogenesis in the general population, highlighting the importance of background cancer risk as an important consideration. In other words, similar pathways are triggered in the development of lung cancer, suggesting that arsenic can accelerate pathogenesis thus adding incrementally to the preexisting background of response. This additivity to background paradigm is associated with a linear dose response at low doses (Crump et al., 1976; US EPA, 2005; White et al., 2009; EFSA Scientific Committee et al., 2017). This is significant, as lung cancer is the leading cause of cancer death in Canada, with approximately 32,000 new cases of lung cancer diagnosed each year (Brenner et al., 2024)

Interindividual variability in response to arsenic exposure is another important consideration. As discussed in section 4.6, responses to arsenic are known to vary considerably across individuals due to extrinsic and intrinsic factors (for example, smoking status, pre-existing disease, nutritional deficiencies, life stage and early life exposure, reproductive status, sex and genetic polymorphisms affecting arsenic metabolism and clearance rate) (Minatel et al., 2018; Sanyal et al., 2020). At the population level, it is understood that some attributes tend to smooth and linearize dose-response relationships (White et al., 2009), precluding the assumption of a "safe" or "no-risk" level of chemical exposure in the diverse general population (Woodruff et al., 2023).

Taking into consideration the complex MOA, the potential additivity of other exposures to ongoing background levels of key events (leading to lung cancer), and the substantial interindividual variability due to the presence of numerous risk modifiers, it is difficult to estimate a population threshold. Furthermore, this threshold, which would likely be indistinguishable from zero, would carry a low level of confidence in terms of providing adequate health protection.

4.6 Populations who may be disproportionally impacted due to health

Susceptibility to arsenic-related adverse health effects is based on numerous factors affecting arsenic metabolism and DNA damage-repair pathways, as well as lifestyle and genetic variations (Minatel et al., 2018). Polymorphisms in genes impairing arsenic reduction and methylation reactions (for example, genetic changes to AS3MT) can alter ratios of urinary MMA to DMA, thus increasing cancer risk (Chi et al., 2018). Considering the arsenic MOA, polymorphisms in genes impacting DNA damage-repair pathways, particularly those involving oxidative stress and DNA damage repair (for example, DNA strand breaks and base modifications from ROS damage) may also impact susceptibility to the effects of arsenic.

Dietary deficiencies in proteins and other nutrients (for example, folic acid, methionine, choline, vitamins B6 and B12) have been shown to impact arsenic metabolism, particularly impairing methylation and excretion (Gamble et al., 2007; Minatel et al., 2018). Pre-existing disease presence or co-exposure to therapeutic compounds may also differentially impact individuals exposed to arsenic (Minatel et al., 2018).

Sex and age may also impact susceptibility. Some studies suggest that males may be more susceptible to arsenic-induced effects relative to females (Waalkes et al., 2014; Ferrario et al., 2016), with some limited data suggesting that childhood exposure may lead to increased morbidity in later life for exposed boys (Tollestrup et al., 2003). Another study demonstrated increases in mortality from lung cancer and bronchiectasis (chronic inflammation and infection of the bronchi) in adulthood, associated with exposure to high concentrations of arsenic in drinking water in utero and during childhood (Smith et al., 2006). It should be noted that it remains possible that a subset of the elderly may be more susceptible to the effects of arsenic, in part due to the accumulation of arsenic in tissues in cases of high exposures over a lifetime; and, as noted previously, exposure during pregnancy may bring added risk to the developing fetus and children later in life, although the critical windows during pregnancy remain poorly understood (Bommarito and Fry, 2016).

Of note, there is some evidence of smoking being additive, synergistic or multiplicative with arsenic co-exposures. Briefly, in an analysis of 16 studies evaluating arsenic-smoking statistical interaction (Folesani et al., 2023), eleven studies did not reveal any interaction, whereas 5 studies revealed a degree of synergism between arsenic exposure and cigarette smoking leading to lung carcinoma. Of these 5 studies, 4 were from drinking water exposures (Chen et al., 2010; Wadhwa et al., 2011; Ferreccio et al., 2013; Steinmaus et al., 2013), with one study from inhaled arsenic, which described the synergism as being most likely a sub-multiplicative interaction (Su et al., 2022). Overall, Folesani et al. (2023) concluded that the interaction between systemic arsenic exposure and tobacco smoke appears to be negligible at low doses, while there may be a synergistic effect at high arsenic concentrations.

4.7 Selected key studies

4.7.1 Short term exposure

A weight-of-evidence approach identified developmental toxicity as the most critical endpoint of acute inhalation exposure to arsenic. As reviewed earlier in the document, short-term inhalation studies in animals have demonstrated developmental toxicity (Nagymajtényi et al., 1985; Holson et al., 1999). Of these, the Nagymajtényi et al. (1985) study featured the most endpoints and the lowest level of exposure at which developmental effects occurred. However, as maternal toxicity was not co-examined in this study, and is noted as a limitation, it remains a possibility that such developmental effects may be secondary to maternal toxicity. Nonetheless, Nagymajtényi et al. (1985) found multiple statistically significant adverse effects, including number of fetuses with retarded growth, increased skeletal malformations and fetal liver cell chromosomal aberrations, with the most sensitive effect being reduced fetal weight (LOAEL = 0.26 mg As2O3/m3). A change in fetal body weight is considered a sensitive indicator of developmental toxicity (OECD, 2008), with such changes often observed at doses below those producing other signs of developmental toxicity (US EPA, 1991). This endpoint is relevant to humans as low birth weight is among the most commonly studied health outcomes in environmental epidemiology, as it predisposes physical and mental growth failure and premature death among infants (Kamai et al., 2019). The Nagymajtényi et al. (1985) study was carried out in mice, which are thought to be more sensitive than rats to developmental toxicity following arsenic exposure. Given the potential severity of latent downstream effects compromising the perinatal viability and growth of the offspring associated with a decrease in fetal weight, coupled with the sensitivity of the endpoint, the Nagymajtényi et al. (1985) study was selected as the key study for acute HBAQO derivation.

Studies demonstrating developmental effects in pregnant female employees were considered as potential key studies, but their inadequate reporting of exposure concentrations and durations, in addition to potential confounders, limited their use for quantitative risk assessment purposes (Nordström et al., 1978a, 1978b, 1979a, 1979b; Ihrig et al., 1998). However, they do lend support to the selection of developmental toxicity as the critical endpoint.

4.7.2 Long term exposure

A weight-of-evidence approach was used to identify the critical effect of long-term exposure to arsenic. Overall, there is strong and consistent evidence, with corresponding monotonic concentration-response trends, demonstrating that inhalation of arsenic in humans is associated with an increased risk of respiratory and lung cancer mortality. Various non-cancer effects were investigated, including cardiovascular effects, diabetes, peripheral neuropathy and other neurological effects. However, these latter associations were either equivocal or displayed a weaker weight of evidence than respiratory and lung cancer mortality.

Of the epidemiological data reporting increased respiratory and lung cancer mortality reviewed by Health Canada, the Asarco Tacoma, Anaconda Montana (both from the US) and Rönnskär (Sweden) cohorts were determined to be the most appropriate cohorts for derivation of a long-term HBAQO due to their robust concentration-response relationships between inhalation exposure to arsenic and cancer mortality. Data from these 3 cohorts, in various combinations, are the basis of all cancer-based long-term exposure air quality objectives published by other organizations for arsenic (see Appendix 2.). Here, various published exposure assessments of the 3 cohorts were identified as candidates for HBAQO derivation, and the study of greatest confidence was selected as the key study to represent each of the 3 cohorts.

Regarding the Asarco Tacoma cohort, Enterline et al. (1995) was selected as the most appropriate study for concentration-response assessment, as it was the most recent study investigating respiratory cancer in this cohort and it represents updates of previous investigations, with longer follow-up than Enterline et al. (1987a) and Enterline and Marsh (1982). Lubin et al. (2008) was selected as the most appropriate study for concentration-response assessment of the Anaconda cohort, as it represents the most recent analysis with adequate exposure information, and incorporates an adjustment factor for heavily exposed workers to account for the use of respirators. Smoking status for the Anaconda cohort was not initially collected, but was examined by Welch et al. (1982), demonstrating a higher smoking prevalence among the workers than in the general population; while the proportion of cigarette smokers only varied slightly across the exposure categories, smoking was determined unlikely to be a significant confounder. Järup et al. (1989) remains the analysis that best describes the Rönnskär cohort, with appropriate data for quantitative risk assessment and no other studies identified. Together, these 3 key studies were determined to describe the relationship between chronic inhalation exposure to arsenic and respiratory and lung cancer mortality adequately, as well as to feature large worker cohorts, lengthy follow-ups, and reliable and sufficient exposure estimations. As these 3 studies were considered of similar quality, all were retained as key studies for concentration-response characterization.

Despite their strengths, there are some notable limitations to utilizing these cohort assessments, including the retrospective nature of the studies, the potential for healthy worker biases, the lack of representation of a diverse population, the potential for co-exposures to other substances as well as the potential for exposure to arsenic by ingestion and in non-occupational settings, such as the home environment. In particular, there is some uncertainty related to the methodologies employed in the analyses of all 3 cohorts to estimate worker exposures. For the Asarco Tacoma cohort, Enterline et al. (1995) used urinary levels of arsenic to estimate exposures in some workers while, for the Anaconda cohort, mean airborne arsenic concentrations were assumed to be constant over time (Lubin et al., 2008). With the Rönnskär Sweden cohort, mean exposure levels for each category are not given, and some years are estimates from sick leave and production figures. As well, it is noted that a related nested-case control study of the Rönnskär data demonstrated a significant interaction with smoking status that likely led to an underestimation of lung cancer risk from arsenic alone (Järup and Pershagen, 1992). Despite the noted limitations to the use of these occupational studies, they remain the strongest description of concentration-response relationships observed within the literature.

Chronic animal data was not considered appropriate for derivation of health-based values due to notable differences in responses that do not reflect human exposures (NTP, 2021), combined with the availability of multiple robust human studies.

5.0 Derivation of the short-term health-based air quality objective

The Nagymajtényi et al. (1985) mouse developmental study was selected as the key study for acute HBAQO derivation. The most sensitive endpoint was reduced fetal weight and, although not amenable to benchmark dose (BMD) modelling, the lowest dose (0.26 mg/m3, or 260 µg/m3) was identified as the LOAEL. Time adjustments were not considered appropriate, given the nature of the effect and uncertainty surrounding mechanism of action, the window of temporal vulnerability, and relative importance of total dose versus peak tissue concentration as a determinant of toxicity.

Thus, starting with the point of departure (POD) as a LOAEL of 260 µg/m3 for decreased fetal weight, the HBAQO was calculated such that:

HBAQO (μg/m3) = POD (μg/m3)/ total uncertainty

Thus,

HBAQO (μg/m3) = 260 As2O3 μg/m3/ 1000 = 0.26 As2O3 μg/m3

Where total uncertainty factors = 1,000, composed of an uncertainty factor for

Finally, as As2O3 is 76% by molecular weight arsenic, a conversion is performed to determine the proposed short-term HBAQO for arsenic.

Short-term HBAQO for arsenic = 0.26 x 76% = 0.2 As μg/m3

In recommending an averaging time associated with the HBAQO value, the general approach is to select averaging times based on the duration of exposure needed to cause the selected health effect. Here, a 1-hour averaging time is the recommended metric to assess potential health risks for the short-term arsenic HBAQO, based on the short exposure duration employed in the critical study (4 hours/day) and short time frame during which developmental effects may occur. The derivation of the proposed short-term (acute) HBAQO for arsenic in ambient air is summarized in Table 1.

Table 1: Summary of the derivation of the acute HBAQO for arsenic.
Criterion Summary
Key study Nagymajtenyi et al., 1985
Study population 4 groups of CFLP mice; 11, 8, 8 and 8 litters (female)
Exposure methods 4 hr/day per day on gestation days 9–12; As2O3 at 0, 260, 2900, 28,500 µg/m3
Critical effects Fetal developmental effects (decreased fetal weight)
LOAEL 260 µg/m3
NOAEL None
Extrapolation to 1 hour Not applicable, due to developmental endpoint considerations
POD 260 µg/m3 (LOAEL)
Total Uncertainty Factors (UFs) 1,000
Interspecies UF 10
Intraspecies UF 10
LOAEL-to-NOAEL UF 10
Conversion factor (As2O3 to inorganic As) 0.76 (As2O3 is 76% by molecular weight arsenic)
Acute HBAQO (As) 0.2 μg/m3

Thus, the proposed short-term (acute) HBAQO for arsenic (As) is 0.2 μg/m3, with a recommended 1-hour averaging time.

6.0 Derivation of the long-term health-based air quality objective

Enterline et al. (1995), Lubin et al. (2008) and Järup et al. (1989), which described increased lung or respiratory cancer mortality in the Asarco Tacoma, Anaconda Montana, and Rönnskär Sweden copper smelting cohorts, respectively, were retained as the key studies. All 3 studies are used in lieu of a single key study. In general, meta-analysis of multiple studies of comparable quality increases precision and accuracy in the estimated PODs, as well as the robustness of the conclusions.

The pooling of lung cancer (Järup et al., 1989) and respiratory cancer (Enterline et al., 1995; Lubin et al., 2008) mortality data was considered appropriate as 96% of the observed deaths in the 2 US studies were due to lung cancer, as described by Erranguntla et al. (2012). Moreover, lung cancer mortality was considered as an appropriate surrogate to estimate lung cancer incidence risk, as lung cancer has a poor prognosis, and historical lung cancer survival has been low, with 5-year survival rates in the US being approximately 10% in the early 1970s, rising to approximately 20% in 2010 (Lu et al., 2019).

The meta-analysis commissioned by Health Canada (RSI, 2022) involved converting cumulative exposure metrics to concentrations in air using a lifetable analysis for lung cancer, and fitting concentration-response modelling against multiplicative RR values from the key studies, noting that a RR approach compares death rates within the cohort group (that is, among those workers exposed to different arsenic levels), and not versus a reference population. Multiplicative RR models were determined to be appropriate, given that risk of lung cancer increases rapidly with age.

Using BMD modelling, benchmark concentrations (BMCs) and lower limit benchmark concentrations (BMCLs) corresponding to benchmark responses (BMRs) for lung cancer of 1%, 5% and 10% were calculated for each study. Modelling required comparison of cohort data to that of Canadian age-specific all-cause mortality rates and age-specific lung cancer mortality rates obtained from Statistics Canada 2015 data (that is, the most recent validated set), which demonstrated that lifetime risk of dying due to lung cancer without exposure to arsenic was estimated to be 6.83%. A weighted approach was used to create a single combined value from individual PODs representing all 3 cohorts, where more weight was given to the cohorts whose data better fit the model. Together, the combined excess risk–based BMCLs for 1%, 5% and 10% BMRs were 10.2 µg/m3, 52.2 µg/m3 and 108.4 µg/m3, respectively. The 1% BMCL is retained due to the severity of the effect and in accordance with US EPA recommendations when modelling epidemiological data (US EPA, 2012).

It is acknowledged that the key studies did not include adjustment for smoking history and that smoking is a known risk factor for developing lung and respiratory cancers. Thus, Health Canada commissioned a report to conduct a quantitative bias analysis to adjust the effect estimates from the 3 cohorts for the potential confounding effect of smoking, and to provide results that address uncertainty about the influence of smoking on the derived POD for arsenic inhalation exposure (RSI, 2023b). Here, the analysis determined that smoking was unlikely to be a significant confounder, and thus the POD was not modified. This conclusion is predicated on 2 key assumptions: (1) similar smoking prevalence existed across arsenic exposure levels within each cohort, which was supported by cohort analyses (Pergashen et al., 1981, 1985; Welch et al., 1982; Enterline et al., 1987b; Järup et al., 1992); and (2) the interaction between smoking and arsenic exposure in the induction of lung cancer was assumed to have an additive effect rather than synergistic or multiplicative relationship. As discussed earlier, there is some evidence that supports the possibility of an interaction effect between smoking habits and arsenic exposure. However, the Risk Sciences International (RSI) (2023b) analysis indicated that large uncertainties would be introduced in any PODs adjusted from quantitative sensitivity analysis assuming the presence of an interaction between smoking and effects from arsenic exposure, owing to the nature of small-sample case-control and case-cohort data and the degree of assumptions that would need to be taken. Thus, it was concluded that any adjustment for unmeasured smoking history in the occupational cohorts would not strengthen the confidence of the derived PODs for arsenic inhalation exposure. Taken together, the POD underpinning the proposed long-term HBAQO was not modified.

In deriving the proposed HBAQO for lung cancer, both threshold and non-threshold approaches were carefully considered in light of data suggesting the existence of a threshold for arsenic induced lung cancer (see section 4.6). Overall, uncertainty remained surrounding which metabolic pathways are essential for causing arsenic-mediated cancer, concerns over background additivity due to similarities between arsenic induced and idiopathic lung cancers, as well as substantial interindividual variability across the Canadian population due to numerous risk modifiers that can alter one's response to arsenic exposure. These considerations suggest that the response is likely linear at low concentrations and that, should a threshold exist, it would likely be indistinguishable from zero. As such, derivation of an IUR value through a non-threshold extrapolation approach is considered most appropriate in estimating lung cancer risk from inhalation of arsenic.

The IUR value is an estimate of the increased cancer risk from inhalation exposure to a concentration of 1 µg/m3 for a lifetime of exposure. Using the POD and the BMR, the IUR value is:

IUR = BMR / POD

Thus,

IUR = 0.01/ 10.2 μg/m3 = 9.8 x 10-4 per μg/m3

Using the IUR value, various risk-specific concentrations can be determined as follows:

Risk-specific concentration = risk level / IUR

Thus, the range of estimated lifetime risks (from one new cancer above background per million people, to one new cancer above background per 10,000 people (10-6 to 10-4) over a lifetime are,

10-4= 0.0001/ 9.8 x 10-4 per μg/m3 = 0.1 μg/m3

10-5= 0.00001/ 9.8 x 10-4 per μg/m3 = 0.01 μg/m3

10-6= 0.000001/ 9.8 x 10-4 per μg/m3 = 0.001 μg/m3

For the long-term HBAQO, where the critical effect occurs after many years of exposure and for which shorter-term variations in concentrations are not expected to influence lifetime cancer risk, a yearly temporal scale (that is, annual average) is recommended.

The derivation of the proposed long-term (chronic) HBAQO for arsenic in ambient air is summarized in Table 2.

Table 2. Summary of the derivation of the long-term HBAQO for arsenic.
Criterion Summary
Key studies

Enterline et al., 1995

Lubin et al., 2008

Järup et al., 1989
Study population

3 occupational cohorts (copper smelters):

Anaconda (Montana, US); 2,802 males

Tacoma (Washington, US); 8,014 males

Rönnskär (Sweden); 3,916 males
Exposure methods Varied estimates of cumulative exposure in μg/m3-yrs
Exposure duration more than 3 months to working lifetime
Critical effects Lung cancer
Risk model Multiplicative relative risk meta-analysis (RSI, 2022)
Extrapolation model Linear non-threshold
BMR 1% BMR
POD 10.2 µg/m3
Inhalation Unit Risk (IUR) 9.8 x 10-4 per μg/m3
Estimated lifetime risk 10-4 0.1 μg/m3 (recommended annual averaging time)
Estimated lifetime risk 10-5 0.01 μg/m3 (recommended annual averaging time)
Estimated lifetime risk 10-6 0.001 μg/m3 (recommended annual averaging time)

As Health Canada considers 10-6 to 10-5 estimated lifetime risk as "essentially negligible", the proposed long-term HBAQO is a range of 0.001–0.01 µg/m3, with a recommended annual averaging time.

Considering that people in Canada may be exposed to arsenic through multiple sources (such as food, drinking water, air and soil), that any exposure is considered to be associated with some level of risk, that there is considerable variability in biological responses to arsenic, and that there may be potential co-exposures within disproportionally impacted populations affected by point sources, every effort should be made to maintain arsenic levels in ambient air ALARA.

7.0 References

  1. Agency for Toxic Substances and Disease Registry (ATSDR). 2005. Medical Management Guidelines for Arsenic (As) and Inorganic Arsenic Compounds. Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service.
  2. Agency for Toxic Substances and Disease Registry (ATSDR). 2007. Toxicological profile for Arsenic. Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service.
  3. Aranyi C, Bradof JN, O'Shea WJ, Graham JA, Miller FJ. 1985. Effects of arsenic trioxide inhalation exposure on pulmonary antibacterial defenses in mice. J Tox Environ Health. 15:163–172.
  4. Assembly of First Nations. 2013. First Nations Biomonitoring Initiative: National Results. 2011. Ottawa, ON, Canada. https://www.fehncy.ca/wp-content/uploads/2020/07/AFN-biomonitoring-initiative-_fnbi_en_-_2013-06-26.pdf
  5. Axelson O. 1980. Arsenic compounds and cancer. J Toxicol Environ Health. 6: 1229.
  6. Beckett WS, Moore JL, Keogh JP, Bleecker ML. 1986. Acute encephalopathy due to occupational exposure to arsenic. Br J Ind Med. 43:66–67.
  7. Belledune Area Health Study. 2005a. Appendix A – Human Health Risk Assessment. Prepared for the Department of Health and Wellness, Government of New Brunswick.
  8. Belledune Area Health Study. 2005b. Summary Report. Prepared for the Department of Health and Wellness, Government of New Brunswick.
  9. Bessö A, Nyberg F, Pershagen G. 2003. Air pollution and lung cancer mortality in the vicinity of a nonferrous metal smelter in Sweden. Int J Cancer. 107:448–52.
  10. Binks K, Doll R, Gillies M, Holroyd C, Jones SR, McGeoghegan D, Scott L, Wakeford R, Walker P. 2005. Mortality experience of male workers at a UK tin smelter. Occup Med. 55: 215–226.
  11. Blom S, Lagerkvist B, Linderholm H. 1985. Arsenic exposure to smelter workers: clinical and neurophysiological studies. Scandinavian Journal of Work, Environment and Health. 11(4): 265–69.
  12. Bommarito PA, Fry RC. 2016. Developmental windows of susceptibility to inorganic arsenic: A survey of current toxicological and epidemiological data. Toxicol res. 5:1503–1511.
  13. Brenner DR, Gillis J, Demers AA, Ellison LF, Billette J-M, Zhang S, Liu J, Woods RR, Finley C, Fitzgerald N, et al. 2024. Projected estimates of cancer in Canada in 2024. CMAJ. 196:E615-E623.
  14. Brown CC, Chu KC. 1983a. Approaches to epidemiologic analysis of prospective and retrospective studies: example of lung cancer and exposure to arsenic. In: Risk Assessment Proc. SIMS Conf. on Environmental Epidemiology June 28–July 2, 1982.
  15. Brown CC, Chu KC. 1983b. Implications of the multistage theory of carcinogenesis applied to occupational arsenic exposure. Journal of the National Cancer Institute. 70(3): 455–463.
  16. Brown CC, Chu KC. 1983c. A new method for the analysis of cohort studies: implications of the multistage theory of carcinogenesis applied to occupational arsenic exposure. Environ Health Perspect. 50: 293–308.
  17. Brown LM, Pottern LM, Blot WJ. 1984. Lung cancer in relation to environmental pollutants emitted from industrial sources. Environ Res. 34:250–261.
  18. Brune D, Nordberg G, Wester PO. 1980. Distribution of 23 elements in the kidney, liver and lungs of workers from a smeltery and refinery in North Sweden exposed to a number of elements and of a control group. Sci Total Environ. 16:13–35.
  19. Burchiel SW, Mitchell LA, Lauer FT, Sun X, McDonald JD, Hudson LG, Liu K. 2009. Immunotoxicity and biodistribution analysis of arsenic trioxide in C57Bl/6 mice following a 2-week inhalation exposure. Toxicol Appl Pharmacol. 241:253–259.
  20. Calderón J, Navarro ME, Jimenez-Capdeville ME, Santos-Diaz MA, Golden A, Rodriguez-Leyva I, Borja-Aburto V, Dı́az-Barriga F. 2001. Exposure to arsenic and lead and neuropsychological development in Mexican children. Environmental Research. 85:69–76.
  21. Caldwell KL, Jones RL, Verdon C, Jarrett J, Caudill S, Osterloh JD. 2009. Levels of urinary total and speciated arsenic in the US population: National Health and Nutrition Examination Survey 2003– 2004. J Expos Sci Environ Epidemiol. 19:59–68.
  22. Canadian Environmental Protection Act, 1999 (CEPA). 1999. S.C. c. 33. https://laws-lois.justice.gc.ca/eng/acts/c-15.31/FullText.html
  23. Canadian Food Inspection Agency (CFIA). 2011. Food safety action plan. Report. 2010-2011 targeted surveys. Chemistry. TS-CHEM-10/11. https://publications.gc.ca/collections/collection_2024/acia-cfia/A104-493-2010-eng.pdf
  24. Centre intégré de santé et de services sociaux de l'Abitibi-Témiscamingue (CISSS-AT). 2019. Rapport de l'étude de biosurveillance menée à l'automne 2018 sur l'imprégnation au plomb, au cadmium et à l'arsenic des jeunes enfants du quartier Notre-Dame de Rouyn-Noranda. Direction de la santé publique de l'Abitibi-Témiscamingue, unité de la santé environnementale.
  25. Centre intégré de santé et de services sociaux de l'Abitibi-Témiscamingue (CISSS-AT). 2020. Rapport de l'étude de biosurveillance menée à l'automne 2019 sur l'imprégnation à l'arsenic de la population du quartier Notre-Dame de Rouyn-Noranda. Direction de la santé publique de l'Abitibi-Témiscamingue, unité de la santé environnementale.
  26. Chen C, Chiou H, Hsu L, Hsueh Y, Wu M, Chen C. 2010. Ingested arsenic, characteristics of well water consumption and risk of different histological types of lung cancer in northeastern Taiwan. Environ Res. 110: 455–462.
  27. Chen W, Yang J, Chen J, Bruch J. 2006. Exposures to silica mixed dust and cohort mortality study in tin mines: exposure-response analysis and risk assessment of lung cancer. Am J Ind Med. 49:67–76.
  28. Cheung J, Hu X, Parajuli R, Rosol R, Torng A, Mohapatra A, Lye E, Chan H. 2020. Health risk assessment of arsenic exposure among the residents in Ndilǫ, Dettah, and Yellowknife, Northwest Territories, Canada. Int J of Hygiene and Env Health. 230, 113623.
  29. Chi L, Gao B, Tu P, Liu C, Xue J, Lai Y, Ru H, Lu K. 2018. Individual susceptibility to arsenic-induced diseases: the role of host genetics, nutritional status, and the gut microbiome. Mammalian Genome. 29 : 63-79.
  30. Chowdhury R, Ramond A, O'Keeffe L, Shahzad S, Kunutsor S, Muka T, Gregson J, Willeit P, Warnakula S, Khan H, et al. 2018. Environmental toxic metal contaminants and risk of cardiovascular disease: Systematic review and meta-analysis. BMJ. 362:k3310.
  31. Cohen SM, Arnold L, Beck B, Lewis A, Eldan M. 2013. Evaluation of the carcinogenicity of inorganic arsenic. Crit Rev Toxicol. 43: 711-752.
  32. Cordier S, Thériault G, Iturra H. 1983. Mortality patterns in a population living near a copper smelter. Environ Res. 31:311-22.
  33. Crown-Indigenous Relations and Northern Affairs Canada (CIRNAC). 2018. Crown-Indigenous Relations and Northern Affairs Canada and Government of the Northwest Territories. Plain Language Summary of the Risk Assessment for Giant Mine. Project No. 2385. https://www.rcaanc-cirnac.gc.ca/eng/1540244275340/1618400678875
  34. Crown-Indigenous Relations and Northern Affairs Canada (CIRNAC). 2024. Crown-Indigenous Relations and Northern Affairs Canada and Government of the Northwest Territories. Giant Mine Remediation Project. 2024. Giant Mine Remediation Project Annual Report 2023–24. https://gmob.ca/wp-content/uploads/2025/01/2024-12-18-GMRP-Annual-Report-2023-2024.pdf
  35. Crump K, Hoel D, Langley C, Peto R. 1976. Fundamental carcinogenic processes and their implications for low dose risk assessment. Cancer research. 36: 2973-2979.
  36. Csavina J, Field J, Taylor MP, Gao S, Landazuri A, Betterton E, Saez A. 2012. A review on the importance of metals and metalloids in atmospheric dust and aerosol from mining operations. Sci Total Environ. 433: 58–73.
  37. Davis A, Gift J, Woodall G, Narotsky M, Foureman G. 2009. The role of developmental toxicity studies in acute exposure assessments: Analysis of single-day vs. multiple-day exposure regimens. Regulatory Toxicology and Pharmacology, 54: 134–142.
  38. Donofrio P, Wilbourn A, Albers J, Rogers L, Salanga V, Greenberg H. 1987. Acute arsenic intoxication presenting as Guillain-Barre-like syndrome. Muscle Nerve. 10:114–120.
  39. Enterline PE, Day R, Marsh GM. 1995. Cancers related to exposure to arsenic at a copper smelter. Occup Environ Med. 52: 28–32.
  40. Enterline PE, Henderson VL, Marsh GM. 1987a. Exposure to arsenic and respiratory cancer. A reanalysis. Am J Epidemiol. 125: 929–38.
  41. Enterline PE, Marsh GM. 1982. Cancer among workers exposed to arsenic and other substances in a copper smelter. Am J Epidemiol. 116: 895–911.
  42. Enterline PE, Marsh GM, Esmen NA, Henderson VL, Callahan CM, Paik M. 1987b. Some effects of cigarette smoking, arsenic, and SO2 on mortality among US copper smelter workers. J Occup Med. 29: 831–8.
  43. Environment and Climate Change Canada (ECCC). 2017. Toxic substances list: inorganic arsenic compounds. https://www.canada.ca/en/environment-climate-change/services/management-toxic-substances/list-canadian-environmental-protection-act/inorganic-arsenic-compounds.html
  44. Environment Canada / Health Canada (EC/HC). 1993. Canadian Environmental Protection Act – Priority Substances List Assessment Report: Arsenic and its compounds – PSL1. https://www.canada.ca/en/health-canada/services/environmental-workplace-health/reports-publications/environmental-contaminants/canadian-environmental-protection-act-priority-substances-list-report-arsenic-compounds.html#a0
  45. Erraguntla, NK, Sielken RL, Valdez-Flores C, Grant R. 2012. An updated inhalation unit risk factor for arsenic and inorganic arsenic compounds based on a combined analysis of epidemiological studies. Reg Toxicol Pharmacol. 64: 329–341.
  46. Ettinger AS, Arbuckle TE, Fisher M, Liang C, Davis K, Cirtiu C, Bélanger P, LeBlanc A, Fraser WD. 2016. Arsenic levels among pregnant women and newborns in Canada: Results from the Maternal-Infant Research on Environmental Chemicals (MIREC) cohort. Environ Res. 153: 8–16.
  47. European Commission. 2001. European Commission Working Group on Arsenic, Cadmium and Nickel Compounds. Ambient air pollution by As, Cd and Ni compounds. Position Paper, October 2000. https://www.aces.su.se/reflab/wp-content/uploads/2016/11/as_cd_ni_position_paper.pdf
  48. European Food Safety Authority (EFSA) Scientific Committee, Hardy A, Benford D, Halldorsson T, Jeger MJ, Knutsen KH, More S, Mortensen A, Naegeli H, Noteborn H, Ockleford C. 2017. Update: Use of the benchmark dose approach in risk assessment. EFSA Journal, 15:e04658.
  49. Fan Y, Jiang Y, Hu P, Chang R, Yao S, Wang B, Li X, Zhou Q, Qiao Y. 2016. Modification of association between prior lung disease and lung cancer by inhaled arsenic: a prospective occupational-based cohort study in Yunnan, China. J Exp Sci Environ Epidemiol. 26: 464–470.
  50. Feldman RG, Niles C, Kelly-Hayes M, Sax D, Dixon WJ, Thompson D, Landau E. 1979. Peripheral neuropathy in arsenic smelter workers. 29: 939.
  51. Ferrario D, Gribaldo L, Hartung T. 2016. Arsenic exposure and immunotoxicity: A review including the possible influence of age and sex. Curr Environ Health Rep 3: 1–12.
  52. Ferreccio C, Yuan Y, Calle J, Benítez H, Parra RL, Acevedo J, Smith AH, Liaw J, Steinmaus C. 2013. Arsenic, tobacco smoke, and occupation: Associations of multiple agents with lung and bladder cancer. Epidemiology. 24: 898–905.
  53. Feseke S, St-Laurent J, Anassour-Sidi E, Ayotte P, Bouchard M, Levallois P. 2015. Arsenic exposure and type 2 diabetes: results from the 2007-2009 Canadian Health Measures Survey. Health Promot Chronic Dis Prev Can. 35: 63–72.
  54. Folesani G, Galetti M, Petronini PG, Mozzoni P, La Monica S, Cavallo D, Corradi M. 2023. Interaction between occupational and non-occupational arsenic exposure and tobacco smoke on lung carcinogenesis. A systematic review. Int J Environ Res Public Health. 20: 4167.
  55. Frost F, Harter L, Milham S, Royce R, Smith AH, Hartley J, Enterline P. 1987. Lung cancer among women residing close to an arsenic emitting copper smelter. Archives of Environmental Health. 42: 148–152.
  56. Galarneau E, Wang D, Dabek-Zlotorzynska E, Siu M, Celo V, Tardif M, Harnish D, Jiang Y. 2016. Air toxics in Canada measured by the National Air Pollution Surveillance (NAPS) program and their relation to ambient air quality guidelines. Journal of the Air & Waste Management Association. 66: 184–200.
  57. Gamble MV, Liu X, Slavkovich V, Pilsner JR, Ilievski V, Factor-Litvak P, Levy D, Alam S, Islam M, Parvez F, Ahsan H, Graziano JH. 2007. Folic acid supplementation lowers blood arsenic. Am J Clin Nutrit. 86: 1202–1209.
  58. Gerhardsson L, Brune D, Nordberg GF, Wester PO. 1988. Multi-elemental assay of tissues of deceased smelter workers and controls. The Science of the Total Environment. 74: 97–110.
  59. Giles BH, Mann KK. 2022. Arsenic as an immunotoxicant. Tox and Appl Pharm. 454: 116248.
  60. Gump BB, Heffernan K, Brann LS, Hill DT, Labrie-Cleary C, Jandev V, MacKenzie JA, Atallah-Yunes NH, Parsons PJ, Palmer CD, et al. 2023. Exposure to arsenic and subclinical cardiovascular disease in 9- to 11-Year-Old children, Syracuse, New York. JAMA Netw Open. 6: e2321379.
  61. Hałatek T, Sińczuk-Walczak H, Rabieh S, Wasowicz W. 2009. Association between occupational exposure to arsenic and neurological, respiratory and renal effects. Tox and Appl Pharm. 239: 193–199.
  62. Hazelton WD, Luebeck EG, Heidenreich WF, Moolgavkar SH. 2001. Analysis of a Historical Cohort of Chinese Tin Miners with Arsenic, Radon, Cigarette Smoke, and Pipe Smoke Exposures Using the Biologically Based Two-Stage Clonal Expansion Model. Radiat Res. 156: 78–94.
  63. Health Canada. 2018. Weight of Evidence: General Principles and Current Applications at Health Canada. Prepared for: Task Force on Scientific Risk Assessment. Prepared by: Weight of Evidence Working Group.
  64. Health Canada. 2021a. Health Canada's Proposal to Update the Maximum Level for Total Arsenic in Fruit Juice and Fruit Nectar. Notice of Proposal - List of Contaminants and Other Adulterating Substances in Foods. Reference Number: NOP/ADP C-2021-2. April 9, 2021. Available: https://www.canada.ca/en/health-canada/services/food-nutrition/public-involvement-partnerships/proposal-update-maximum-level-total-arsenic-fruit-juice-fruit-nectar/document.html
  65. Health Canada. 2021b. Arsenic in Canadians. Ottawa, ON. Available: https://www.canada.ca/en/health-canada/services/environmental-workplace-health/reports-publications/environmental-contaminants/human-biomonitoring-resources/arsenic-canadians.html
  66. Health Canada. 2022. Assessment in support of risk management for arsenic in rice-based foods intended for infants and young children. August 2022. https://open.canada.ca/data/en/info/3b62274a-b926-4eea-9bee-9bb3a5bb2377
  67. Health Canada. 2025. Draft guidelines for Canadian drinking water quality, arsenic. Guideline technical document for public consultation. https://www.canada.ca/en/health-canada/programs/consultation-draft-guidelines-canadian-drinking-water-quality-arsenic/document.html
  68. Hertz-Picciotto I, Arrighi HM, Hu SW. 2000. Does arsenic exposure increase the risk for circulatory disease? Am J Epidemiol. 151:174–181.
  69. Higgins ITT, Oh MS, Kryston KL, Burchfiel CM, Wilkinson NM. 1986. Arsenic exposure and respiratory cancer in a cohort of 8044 Anaconda smelter workers: A 43-year follow-up study. Prepared for the Chemical Manufacturers' Association and the Smelters Environmental Research Association. (not published). As described in US EPA 1995.
  70. Higgins I, Welch K, Burchfield C. 1982. Mortality of Anaconda smelter workers in relation to arsenic and other exposures. University of Michigan, Dept. Epidemiology, Ann Arbor, MI. (not published). As described in US EPA 1995.
  71. Hindmarsh JT, McCurdy RF. 1986. Clinical and environmental aspects of arsenic toxicity. Crit Rev Clin Lab Sci. 23: 315.
  72. Holson JF, Stump DG, Ulrich CE, Farr CH. 1999. Absence of prenatal developmental toxicity from inhaled arsenic trioxide in rats. Toxicological sciences: an official journal of the Society of Toxicology. 51: 87–97.
  73. Huff J, Chan P, Nyska A. 2000. Is the Human Carcinogen Arsenic Carcinogenic to Laboratory Animals? Toxicological Sciences. 55: 17–23.
  74. Hutton M, Symon C. 1986. The quantities of cadmium, lead, mercury and arsenic entering the U.K. environment from human activities. Sci Total Environ. 57: 129.
  75. Ihrig M, Shalat SL, Baynes C. 1998. A hospital-based case-control study of stillbirths and environmental exposure to arsenic using an atmospheric dispersion model linked to a geographic information system. Epidemiology. 9: 290–294.
  76. International Agency for Research on Cancer (IARC). 2012. Arsenic, metals, fibres and dusts. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Volume 100C.
  77. Ishii N, Mochizuki H, Ebihara Y, Shiomi K, Nakazato M. 2017. Incidence of clinical symptoms and neurological signs in patients with chronic arsenic exposure in Miyazaki, Japan: A 40-year retrospective descriptive study. J Neuro Sci. 381: 1378.
  78. Järup L, Pershagen O, Wall S. 1989. Cumulative arsenic exposure and lung cancer in smelter workers: a dose-response study. Am Journal Indus Med. 15: 31–41.
  79. Järup L, Pershagen G. 1992. Arsenic exposure, smoking, and lung cancer in smelter workers – a case-control study. Am J Epidemiol. 134: 545–551.
  80. Jomova K, Alomar SY, Nepovimova E, Kuca K, Marian V. 2025. Heavy metals: Toxicity and human health effects. Arch of Tox. 99: 153–209.
  81. Jones SR, Atkin P, Holroyd C, Lutman E, Batlle J, Wakeford R, Walker P. 2007. Lung cancer mortality at a UK tin smelter. Occup Med. 57: 238–245.
  82. Kamai EM, McElrath TF, Ferguson KK. 2019. Fetal growth in environmental epidemiology: mechanisms, limitations, and a review of associations with biomarkers of non-persistent chemical exposures during pregnancy. Environ Health. 18: 43–73.
  83. Keil AP, Richardson DB. 2016. Reassessing the link between airborne arsenic exposure among anaconda copper smelter workers and multiple causes of death using the parametric g-formula. Environ Health Perspec. 125: 608–14.
  84. Kuivenhoven M, Mason K. 2023. Arsenic Toxicity. Treasure Island (FL): StatPearls Publishing.
  85. Lagerkvist B, Linderholm H, Nordberg GF. 1986. Vasospastic tendency and Raynaud's phenomenon in smelter workers exposed to arsenic. Environ Res. 39: 465–474
  86. Lagerkvist B, Linderholm H, Nordberg GF. 1988. Arsenic and Raynaud's phenomenon. Vasospastic tendency and excretion of arsenic in smelter workers before and after the summer vacation. Int. Arch Occup Environ Health. 60: 361–364.
  87. Lagerkvist BJ, Zetterlund B. 1994. Assessment of exposure to arsenic among smelter workers: a five‐year follow‐up. Am J Indus Med. 25: 477–488.
  88. Ledda C, Iavicoli I, Avola R, Senia P, Santarelli L, Pomara C, Rapisarda V. 2017. Serum lipid, lipoprotein and apolipoprotein profiles in workers exposed to low arsenic levels : lipid profiles and occupational exposure. Tox Letts. 282: 49–56.
  89. Lee A, Fraumeni J. 1969. Arsenic and respiratory cancer in man: An occupational study. J Natl Cancer Inst 42: 1045–1052.
  90. Lee-Feldstein A. 1983. Arsenic and respiratory cancer in man. Follow-up of an occupational study: In Arsenic: Industrial, biomedical, and environmental perspectives. Lederer and Fensterheim (ed.). New York. 245–265.
  91. Lee-Feldstein A. 1986. Cumulative exposure to arsenic and its relationship to respiratory cancer among copper smelter employees. J Occup Med. 28: 296–302.
  92. Lee-Feldstein A. 1989. A comparison of several measures of exposure to arsenic. Matched case-control study of copper smelter employees. Am J Epidemiol. 129: 112–124.
  93. Liu J, Hermon T, Gao X, Dixon D, Xiao H. 2023. Arsenic and diabetes mellitus: a putative role for the immune system. All life.16: 2167869.
  94. Lu T, Yang X, Huang Y, Zhao M, Li M, Ma K, Yin J, Zhan C, Wang Q. 2019. Trends in the incidence, treatment, and survival of patients with lung cancer in the last four decades. Cancer Manag Res. 11: 943–953.
  95. Lubin J, Fraumeni J. 2000. Re: 'Does arsenic exposure increase the risk for circulatory disease?'. Am J Epidemiol. 152: 290–293.
  96. Lubin J, Moore LE, Fraumeni J, Cantor KP. 2008. Respiratory cancer and inhaled inorganic arsenic in copper smelter workers: a linear relationship with cumulative exposure that increases with concentration. Environ Health Perspect. 116: 1661–1665.
  97. Lubin J, Pottern LM, Blot WJ, Tokudome S, Stone BJ, Fraumeni J. 1981. Respiratory cancer among copper smelter workers: Recent mortality statistics. J Occup Med. 23: 779–784.
  98. Lubin J, Pottern LM, Stone BJ, Fraumeni J. 2000. Respiratory cancer in a cohort of copper smelter workers: results from more than 50 years of follow-up. Am J Epidem. 151: 554–565.
  99. Marsh GM, Esmen N, Buchanich J, Youk A. 2009. Mortality patterns among workers exposed to arsenic, cadmium, and other substances in a copper smelter. Am J Indust Med. 52: 633–44.
  100. Martin R, Dowling K, Pearce D, Sillitoe J, Florentine S. 2014. Health Effects Associated with Inhalation of Airborne Arsenic Arising from Mining Operations. Geosciences. 4: 128–175.
  101. Matanoski GM, Landau E, Tonascia J, Lazar C, Elliott EA, McEnroe W, King K. 1981. Cancer mortality in an industrial area of Baltimore. Environ Res. 25: 8–28.
  102. Matschullat J. 2000. Arsenic in the geosphere—A review. Sci Total Environ. 249:297–312.
  103. Maud J, Rumsby P. 2008. A review of the toxicity of arsenic in air. Science Report – SC020104/SR4. Bristol, UK: Environment Agency.
  104. Maull EA, Ahsan H, Edwards J, Longnecker MP, Navas-Acien A, Pi J, Silbergeld EK, Styblo M, Tseng C, Thayer K, Loomis D. 2012. Evaluation of the association between arsenic and diabetes: A National Toxicology Program workshop review. Environ Health Perspect.120: 1658–1670.
  105. Minatel BC, Sage AP, Anderson C, Hubaux R, Marshall EA, Lam WL, Martinez VD. 2018. Environmental arsenic exposure: From genetic susceptibility to pathogenesis. Environ Int. 112: 183–197.
  106. Mingxing S, Haiying W, Congsong S, Chunyu Y, Liu C, Wang Q. 2019. Acute toxicity of intratracheal arsenic trioxide instillation in rat lungs. J Appl Toxicol. 39: 1578–1585.
  107. Ministère de l'Environnement, de la Lutte contre les Changements Climatiques, de la Faune et des Parcs (MELCCFP). 2023. Certificat d'analyse: École 50 Ave Murdoch, Rouyn-Noranda Québec. Numéros de l'échantillons: L062308-01 and L062308-02.
  108. Mochizuki H. 2019. Arsenic neurotoxicity in humans. Int J Mol Sci. 20: 3418.
  109. Moon K, Guallar E, Navas-Acien A. 2012. Arsenic exposure and cardiovascular disease: an updated systematic review. Current Atherosclerosis Report. 14: 542–555.
  110. Moon KA, Guallar E, Umans JG, Devereux RB, Best LG, Francesconi KA, Goessler W, Pollak J, Silbergeld EK, Howard BV, Navas-Acien, A. 2013. Association between exposure to low to moderate arsenic levels and incident cardiovascular disease. Annals of Internal Medicine. 159: 649–59.
  111. Mutlu EA, Comba IY, Cho T, Engen PA, Yazici C, Soberanes S, Hamanaka RB, Niğdelioğlu R, Meliton AY, Ghio AJ, Budinger G, Mutlu GM. 2018. Inhalational exposure to particulate matter air pollution alters the composition of the gut microbiome. Environ Pollut. 240: 817–830.
  112. Nagymajtényi L, Selypes A, Berencsi G. 1985. Chromosomal aberrations and fetotoxic effects of atmospheric arsenic exposure in mice. J Appl Toxicol. 5: 61–3.
  113. National Pollutants Release Inventory (NPRI). 2023. Arsenic. https://pollution-waste.canada.ca/national-release-inventory/
  114. National Toxicology Program (NTP). 2000. Toxicology and Carcinogenesis Studies of Gallium Arsenide (CAS No. 1303-00-0) in F344/N Rats and B6C3F1Mice (Inhalation Studies) (NTP Technical Report 492). Research Triangle Park, NC
  115. National Toxicology Program (NTP). 2021. Report on Carcinogens, Fifteenth Edition. Research Triangle Park, NC. U.S. Department of Health and Human Services, Public Health Service.
  116. Navas-Acien A, Sharrett AR, Silbergeld EK, Schwartz BS, Nachman KE, Burke TA, Guallar E. 2005. Arsenic exposure and cardiovascular disease: a systematic review of the epidemiological evidence. American J Epidemiol. 162: 1037–1049.
  117. Navas-Acien A, Silbergeld EK, Streeter RA, Clark JM, Burke TA, Guallar E. 2006. Arsenic exposure and type 2 diabetes: a systematic review of the experimental and epidemiological evidence. Environ Health Perspect. 114: 641–648.
  118. Newhook R, Hirtle H, Byrne K, Meek ME. 2003. Releases from copper smelters and refineries and zinc plants in Canada: human health exposure and risk characterization, Science of The Total Environment. 301: 23–41.
  119. Nordstrom S, Beckman L, Nordenson I. 1978a. Occupational and environmental risks in and around a smelter in northern Sweden. I. Variations in birth weight. Hereditas. 88: 43–46.
  120. Nordstrom S, Beckman L, Nordenson I. 1978b. Occupational and environmental risks in and around a smelter in northern Sweden III. Frequencies of spontaneous abortion. Hereditas. 88: 51–54.
  121. Nordstrom S, Beckman L, Nordenson I. 1979a. Occupational and environmental risks in and around a smelter in northern Sweden. V. Spontaneous abortion among female employees and decreased birth weight in their offspring. Hereditas. 90: 291–296.
  122. Nordstrom S, Beckman L, Nordenson I. 1979b. Occupational and environmental risks in and around a smelter in northern Sweden. VI. Congenital malformations. Hereditas. 90: 297–302
  123. Office of Environmental Health Hazard Assessment California (OEHHA). 2011. Technical support document for cancer potency factors. Appendix B. Chemical-specific summaries of the information used to derive unit risk and cancer potency values.
  124. Office of Environmental Health Hazard Assessment California (OEHHA). 2014. Technical supporting document for noncancer RELs, Appendix D. Individual acute, 8-hour, and chronic reference exposure level summaries (updated in 2014). https://oehha.ca.gov/media/downloads/crnr/appendixd1final.pdf
  125. Organisation for Economic Co-operation and Development (OECD). 2008. Guidance Document on Mammalian Reproductive Toxicity Testing and Assessment. Series on Testing and Assessment. No. 43. https://one.oecd.org/document/ENV/JM/MONO(2008)16/en/pdf
  126. Organisation for Economic Co-operation and Development (OECD). 2011. Guidance Document for the Derivation of an Acute Reference Concentration (ARFC). Series on Testing and Assessment. No. 153. https://web-archive.oecd.org/2012-06-14/93080-48542016.pdf
  127. Patel KS, Pandey PK, Martín-Ramos P, Corns WT, Varol S, Bhattacharya P, Zhu Y. 2023. A review on arsenic in the environment: contamination, mobility, sources, and exposure. RSC Adv. 13: 8803–8821.
  128. Pershagen G. 1985. Lung cancer mortality among men living near an arsenic-emitting smelter. Am J Epidemiol 122: 684–694.
  129. Pershagen G, Björkland NE. 1985. On the pulmonary tumorigenicity of arsenic trisulfide and calcium arsenate in hamsters. Cancer Letters. 27: 99–104.
  130. Pershagen G, Nordberg G, Björkland NE. 1984. Carcinomas of the respiratory tract in hamsters given arsenic trioxide and/or benzo[a]pyrene by pulmonary route. Environ Res. 34: 227–241.
  131. Pershagen G, Wall S, Taube A, Linnman L. 1981. On the interaction between occupational arsenic exposure and smoking and its relationship to lung cancer. Scand J Work Environ Health. 7: 302–9.
  132. Pinto SS, Enterline PE, Henderson V, Varner MO. 1977. Mortality experience in relation to a measured arsenic trioxide exposure. Environ Health Perspect. 19: 127–130.
  133. Qi Z, Zhao Q, Yu Z, Yang Z, Feng J, Song P, He X, Lu X, Chen X, Li S, et al. 2024. Assessing the impact of PM2.5-bound arsenic on cardiovascular risk among workers in a non-ferrous metal smelting area: Insights from chemical speciation and bioavailability. Environ Sci Technol. 58: 8228–8238.
  134. Rahman M, Axelson O. 1995. Diabetes mellitus and arsenic exposure: a second look at case-control data from a Swedish copper smelter. Occup Environ Med. 52: 773–774.
  135. Rahn KA. 1976. The Chemical Composition of the Atmospheric Aerosol. Technical Report. University of Rhode Island: Kingston, RI, USA.
  136. Rasmussen PE, Levesque C, Chénier M, Gardner HD, Jones-Otazo H, Petrovic S. 2013. Canadian House Dust Study: Population-based concentrations, loads and loading rates of arsenic, cadmium, chromium, copper, nickel, lead and zinc inside urban homes. Sci Total Environ. 433: 520–529.
  137. Rasmussen PE, Levesque C, Chénier M, Gardner HD. 2018. Contribution of metals in resuspended dust to indoor and personal inhalation exposures: Relationships between PM10 and settled dust. Build Environ. 143: 513-522.
  138. Rosado JL, Ronquillo D, Kordas K, Rojas O, Alatorre J, Lopez P, Garcia-Vargas G, Del Carmen Caamaño M, Cebrián ME, Stoltzfus RJ. 2007. Arsenic exposure and cognitive performance in Mexican schoolchildren. Environ Health Perspect. 115: 1371–1375.
  139. Roy A, Kordas K, Lopez P, Rosado JL, Cebrian ME, Vargas GG, Ronquillo D, Stoltzfus RJ. 2011. Association between arsenic exposure and behavior among first-graders from Torren, Mexico. Environ Res. 111: 670–676.
  140. Risk Sciences International (RSI). 2022. Arsenic review and dose response modeling. Risk Sciences International, Ottawa, Ontario. Contract report for Health Canada. Available upon request.
  141. Risk Sciences International (RSI). 2023a. Arsenic-induced lung cancer mode of action analysis and point of departure derivation. Ottawa, Ontario. Contract report for Health Canada. Available upon request.
  142. Risk Sciences International (RSI). 2023b. Inhaled arsenic POD analysis – sensitivity analysis on bias from unmeasured smoking. Ottawa, Ontario. Contract report for Health Canada. Available upon request.
  143. Risk Science Center (RSC), University of Cincinnati. 2019. Evaluation of inorganic arsenic: Causality analysis, mode of action and determination of dose-response approach draft. Contract report for Health Canada. Available upon request.
  144. Sanyal T, Bhattacharjee P, Paul S, Bhattacharjee P. 2020. Recent advances in arsenic research: Significance of differential susceptibility and sustainable strategies for mitigation. Front pub health. 8: 1–16.
  145. Sattar A, Xie S, Hafeez MA, Wang X, Hussain HI, Iqbal Z, Pan Y, Iqbal M, Shabbir MA, Yuan Z. 2016. Metabolism and toxicity of arsenicals in mammals. Environ Toxicol Pharmacol. 48: 214–224.
  146. Seinfeld JH, Pandis SN. 2006. Atmospheric Chemistry and Physics: From Air Pollution to Climate Change. New York, NY, USA.
  147. Sińczuk-Walczak H, Janasik BM, Trzcinka-Ochocka M, Stanisławska M, Szymczak M, Hałatek T, Walusiak-Skorupa J. 2014. Neurological and neurophysiological examinations of workers exposed to arsenic levels exceeding hygiene standards. Int J Occup Med Environ Health. 27: 1013–25.
  148. Sińczuk-Walczak H, Szymczak M, Hałatek T. 2010. Effects of occupational exposure to arsenic on the nervous system: clinical and neurophysiological studies. Int J Occup Med Environ Health. 23: 347–55.
  149. Smith AH, Ercumen A, Yuan Y, Steinmaus CM. 2009. Increased lung cancer risks are similar whether arsenic is ingested or inhaled. Journal of Exposure Science and Environmental Epidemiology. 19: 343–348.
  150. Smith AH, Marshall G, Yuan Y, Ferreccio C, Liaw J, von Ehrenstein O, Steinmaus C, Bates MN, Selvin S. 2006. Increased mortality from lung cancer and bronchiectasis in young adults after exposure to arsenic in utero and in early childhood. Environ Health Perspect. 114: 1293–1296.
  151. Smith S. 2018. Telomerase can't handle the stress. Genes Dev. 32: 597–599.
  152. Steinmaus CM, Ferreccio C, Romo JA, Yuan Y, Cortes S, Marshall G, Moore LE, Balmes JR, Liaw J, Golden T, Smith AH. 2013. Drinking water arsenic in Northern Chile: High cancer risks 40 years after exposure cessation. Cancer Epidemiol Biomark Prev. 22: 623–630.
  153. Sturm R. 2012. Theoretical models of carcinogenic particle deposition and clearance in children's lungs. J Thorac Dis. 4: 368–376.
  154. Stýblo M, Del Razo LM, Vega L, Germolec DR, LeCluyse E, Hamilton G, Reed W, Wang C, Cullen WR, Thomas DJ. 2000. Comparative toxicity of trivalent and pentavalent inorganic and methylated arsenicals in rat and human cells. Arch Toxicol. 74: 289–299.
  155. Stýblo M, Venkatratnam A, Fry RC, Thomas DJ. 2021. Origins, fate, and actions of methylated trivalent metabolites of inorganic arsenic: progress and prospects. Arch Toxicol. 95: 1547–1572.
  156. Su Z, Wei M, Jia X, Fan Y, Zhao F, Zhou Q, Taylor P, Qiao Y. 2022. Arsenic, tobacco use, and lung cancer: An occupational cohort with 27 follow-up years. Environ Res. 206: 112611.
  157. Sudbury Soils Study. 2008a. Appendix O. Sudbury Area Risk Assessment (SARA) Group. http://www.sudburysoilsstudy.com/
  158. Sudbury Soils Study. 2008b. Volume II – Human Health Risk Assessment Chapter 5: Results and Discussion. Sudbury Area Risk Assessment (SARA) Group. http://www.sudburysoilsstudy.com/
  159. Sung T, Huang J, Guo H. 2015. Association between arsenic exposure and diabetes: a meta-analysis. BioMed Int. p. 368087
  160. Sutton O, McCarter C, Waddington J. 2024. Globally-significant arsenic release by wildfires in a mining-impacted boreal landscape. Environ Res Letts. 19: 064024.
  161. Tadanobu I, Zhang Y, Shigeo M, Hiroko S, Hiromichi N, Hiroki M, Yuichi S, Eiichi A. 1990. The effect of arsenic trioxide on brain monoamine metabolism and locomotor activity of mice. Toxicology Lett. 54: 345–353.
  162. Texas Commission on Environmental Quality (TCEQ). 2012. Arsenic and Inorganic Arsenic Compounds. CAS Registry Numbers: 7440-38-2 (Arsenic). Office of the Executive Director.
  163. Tokar E, Benbrahim-Tallaa L, Ward JM, Lunn R, Sams RL, Waalkes MP. 2010. Cancer in experimental animals exposed to arsenic and arsenic compounds. Critical Reviews in Toxicology 40: 912–927.
  164. Tollestrup K, Frost FJ, Harter LC, McMillan GP. 2003. Mortality among children residing near the American Smelting and Refining Company (ASARCO) Copper Smelter in Ruston, Washington. Arch Environ Health. 58: 683–691.
  165. Tseng CH. 2008. Cardiovascular disease in arsenic-exposed subjects living in the arseniasis-hyperendemic areas in Taiwan. Atherosclerosis. 199: 12–18.
  166. Tsuji JS, Garry MR, Perez V, Chang ET. 2015. Low-level arsenic exposure and developmental neurotoxicity in children: A systematic review and risk assessment. Toxicology. 337 :91–107.
  167. Tyler CR, Allan AM. 2014. The effects of arsenic exposure on neurological and cognitive dysfunction in human and rodent studies: A review. Curr Environ Health Rep. 1: 132–147.
  168. United Kingdom Poison Information Documents (UKPID). 1997. Arsenic Trioxide. National Poisons Information Service Centre.
  169. United States Environmental Protection Agency (US EPA). 1991. Guidelines for Developmental Toxicity Risk Assessment. Risk Assessment Forum. Federal Register 56(234):63798-63826.
  170. United States Environmental Protection Agency (US EPA). 1995. Arsenic, inorganic; CASRN 7440-38-2. Integrated Risk Information System (IRIS) Chemical Assessment Summary. National Center for Environmental Assessment.
  171. United States Environmental Protection Agency (US EPA). 2005. Guidelines for Carcinogen Risk Assessment. March 2005. Risk Assessment Forum, Washington DC, EPA/630/P-03/001B -p18.
  172. United States Environmental Protection Agency (US EPA). 2012. Benchmark Dose Technical Guidance. Risk Assessment Forum. Washington, DC. 20460 EPA/100/R-12/001 June 2012.
  173. United States Environmental Protection Agency (US EPA). 2025. IRIS Toxicological Review of Inorganic Arsenic. CASRN 7440-38-2. January 2025. National Center for Environmental Assessment. Integrated Risk Information System. Center for Public Health and Environmental Assessment, Office of Research and Development, U.S. Environmental Protection Agency, Washington
  174. Vahter M. 2000. Genetic polymorphisms in the biotransformation of inorganic arsenic and its role in toxicity. Toxicol Lett. 112–113: 209–217.
  175. Vahter M, Friberg L, Rahnster B, Nygren Å, Nolinder P. 1986. Airborne arsenic and urinary excretion of metabolites of inorganic arsenic among smelter workers. Int Arch Occup Environ Health. 57: 79–91.
  176. Valcke M, Ponce G, Bourgault MH. 2022. Évaluation du risque cancérigène attribuable aux concentrations d'arsenic et de cadmium dans l'air de la ville de Rouyn-Noranda. Institut national de santé publique du Québec Gouvernement du Québec. www.inspq.qc.ca/sites/default/files/publications/2875-risque-cancerigene-concentrations-arsenic-cadmium-air-rouyn-noranda.pdf
  177. Vázquez Cervantes GI, González Esquivel DF, Ramírez Ortega D, Blanco Ayala T, Ramos Chávez LA, López-López HE, Salazar A, Flores I, Pineda B, Gómez-Manzo S, et al. 2023. Mechanisms associated with cognitive and behavioral impairment induced by arsenic exposure. Cells. 12: 2537.
  178. Viren JR, Silvers A. 1994. Unit risk estimates for airborne arsenic exposure: an updated view based on recent data from two copper smelter cohorts. Regulatory Toxicology and Pharmacology. 20: 125–138.
  179. Waalkes MP, Qu W, Tokar EJ, Kissling GE, Dixon D. 2014. Lung tumours in mice induced by "whole-life" inorganic arsenic exposure at human-relevant doses. Arch Toxicol. 88: 1619–1629.
  180. Wadhwa SK, Kazi TG, Kolachi NF, Afridi HI, Khan S, Chandio AA, Shah AQ, Kandhro GA, Nasreen S. 2011. Case-control study of male cancer patients exposed to arsenic-contaminated drinking water and tobacco smoke with relation to non-exposed cancer patients. Hum Exp Toxicol. 30: 2013–2022.
  181. Welch K, Higgins I, Oh M, Burchfiel C. 1982. Arsenic exposure. Smoking, and respiratory cancer in copper smelter workers. Archives of Environmental Health: An International Journal. 37: 325–335.
  182. Wester PO, Brune D, Nordberg G. 1981. Arsenic and selenium in lung, liver, and kidney tissue from dead smelter workers. British J Indus Med. 38: 179–184.
  183. White RH, Cote I, Zeise L, Fox M, Dominici F, Burke T, White P, Hattis DB, Samet JM. 2009. State-of-the-science workshop report: Issues and approaches in low-dose-response extrapolation for environmental health risk assessment. Environ Health Perspect. 117: 283–7.
  184. Woodruff TJ, Rayasam SD, Axelrad DA, Koman PD, Chartres N, Bennett DH, Birnbaum LS, Brown P, Carignan CC, Cooper C, et al. 2023. A science-based agenda for health-protective chemical assessments and decisions: overview and consensus statement. Environ Health. 12: 132.
  185. World Health Organization (WHO). 2000. Air Quality Guideline for Europe. 2nd ed. WHO Regional Publication European Series. No.91.
  186. World Health Organization (WHO). 2024. Human health effects of benzene, arsenic, cadmium, nickel, lead and mercury: report of an expert consultation. Copenhagen. WHO Regional Publication European Series.
  187. Xu M, Rui D, Yan Y, Xu S, Niu Q, Feng G, Wang Y, Li S, Jing M. 2017. Oxidative damage induced by arsenic in mice or rats: A systematic review and meta-analysis. Biol Trace Elem Res. 176: 154–175.
  188. Yamamoto A, Hisanaga A, Ishinishi LN. 1987. Tumorigenicity of inorganic arsenic compounds following intratracheal instillations to the lungs of hamsters. Int J Cancer. 40: 220–223.
  189. Yip L, Dart R. 2001. Arsenic. In: Sullivan J and Kreiger G, eds. Clinical environmental health and toxic exposures. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins. p. 858–865.

Appendix A: List of acronyms and abbreviations

As
arsenic
As2O3
arsenic trioxide
AS3MT
arsenite methyltransferase
ALARA
as low as reasonably achievable
ATSDR
Agency for Toxic Substances and Disease Registry
BMC
benchmark concentration
BMCL
benchmark concentration level
BMR
benchmark response
CEPA
Canadian Environmental Protection Act, 1999
CFIA
Canadian Food Inspection Agency
CHD
coronary heart disease
CI
confidence interval
CISSS-AT
Centre intégré de santé et de services sociaux de l'Abitibi-Témiscamingue
CVD
cardiovascular disease
DMA
dimethylarsinic acid
DNA
deoxyribonucleic acid
EC
Environment Canada
ECCC
Environment and Climate Change Canada
FNBI
First Nations Biomonitoring Initiative
HBAQO
health-based air quality objective
HC
Health Canada
hr
hour
IARC
International Agency for Research on Cancer
iAs
inorganic arsenic
IUR
inhalation unit risk
kg
kilogram
LOAEL
lowest-observed-adverse-effect level
m3
cubic metre
MEI
molecular initiating events
MELCCFP
Ministère de l'Environnement, de la Lutte contre les Changements Climatiques, de la Faune et des Parcs (Québec)
mg
milligram
MIREC
Maternal-Infant Research on Environmental Chemicals
MMA
monomethylarsonic acid
MOA
mode of action
NAPS
National Air Pollution Surveillance
ng
nanogram
NOAEL
no-observed-adverse-effect level
NOAELADJ
no-observed-adverse-effect level, adjusted
NOAELHEC
no-observed-adverse-effect level, adjusted to a human equivalent concentration
NPRI
Nationals Pollutant Release Inventory
NTP
National Toxicology Program
OECD
Organisation for Economic Co-operation and Development
OEHHA
Office of Environmental Health Hazard Assessment (California)
OHSA
Occupational Safety and Health Administration
p
p-value
PM
particulate matter
POD
point of departure
REL
reference exposure level
ROS
reactive oxygen species
RR
relative risk
RSC
Risk Services Center
RSI
Risk Sciences International
SMR
standardized mortality ratio
TC
tumorigenic concentration
TCEQ
Texas Commission on Environmental Quality (Texas)
UF
uncertainty factor
UKPID
United Kingdom Poison Information Documents
US
EPA United States Environmental Protection Agency
WHO
World Health Organization
wt
weight
µg
microgram
µm
micrometre
yr
year

Appendix B: International considerations

Appendix 1: Short-term inhalation reference exposure levels (RELs). Summaries of the key studies used in the development of acute health-based reference exposure levels from TCEQ (2012) and OEHHA (2014) assessments are described.

Criterion TCEQ, 2012 OEHHA, 2014
Key Study Holson et al., 1999 Nagymajtenyi et al., 1985
Study population 4 groups of Crl:CD(SD)BR rats; 24 per group (female) 4 groups of CFLP mice; 11, 8, 8 and 8 litters (female)
Exposure Methods 6 hr/day on 14 days prior to mating to gestation day 19; As2O3 at 0, 300, 3000, 10,000 µg/m3 4 hr/day on gestation days 9–12; As2O3 at 0, 260, 2900, 28,500 µg/m3
Critical Effects Maternal toxicity (rales) Fetal developmental effects (decreased fetal weight)
LOAEL n/a 260 µg/m3 (197 µg/m3 as arsenic)
NOAEL 3000 µg/m3 n/a
Extrapolation to 1 hour 5451 µg/m3 (NOAELADJ; modified Haber's law;) N/A, due to developmental endpoint considerations
POD 3891.3 µg/m3 (NOAELHEC) (US EPA default dosimetry adjustment) 260 µg/m3 (LOAEL)
Total Uncertainty Factors (UFs) 300 1000
Interspecies UF 3 10Footnote †
Intraspecies UF 10 10Footnote †
LOAEL-to-NOAEL UF n/a 10
Database Deficiencies UF 10 n/a
Conversion factor (As2O3 to inorganic As) 0.76 (As2O3 is 76% by molecular weight arsenic) 0.76 (As2O3 is 76% by molecular weight arsenic)
Acute HBAQO (As) 9.9 μg/m3 0.2 μg/m3

Appendix 2: Inhalation unit risks for carcinogenic effects. All agencies utilized 1 to 3 cohorts of occupational inhalation exposure to arsenic in smelter workers, and all identified the critical effect as respiratory and/or lung cancer. Key studies represent the exposure analyses utilized. Note: Estimated lifetime risk levels (µg/m3) have been generated for comparison purposes only, to one significant digit.

Criterion EC/HC, 1993 US EPA, 1995Footnote † WHO, 2000 OEHHA, 2011Footnote †† TCEQ, 2012
Key Studies
  • Higgins et al., 1986
  • Enterline and Marsh, 1982
  • Brown and Chu, 1983a;b;c;
  • Higgins et al., 1982
  • Lee-Feldstein, 1983
  • Brown and Chu, 1983a;b;c
  • Higgins et al., 1982
  • Lee-Feldstein, 1983
  • Järup et al., 1989
  • Viren and Silvers, 1994
  • Enterline et al., 1987a
  • Enterline et al., 1987a
  • Lubin et al., 2008
  • Enterline et al., 1995
  • Järup et al., 1989
  • Viren & Silvers, 1994
Study population
  • Anaconda cohort
  • Anaconda cohort
  • Tacoma cohort
  • Anaconda cohort
  • Tacoma cohort
  • Rönnskär cohort
  • Tacoma cohort
  • Anaconda cohort
  • Tacoma cohort
  • Rönnskär cohort
Unit Risk Derivation TC05 values were calculated for the 3 individual cohorts; Anaconda cohort was selected, as it was the most health protective value (TC05 = 7.83 µg/m3) Geometric mean of the 2 cohort geometric means, calculated as average unit risk of individual studies for each cohort (that is, not a weighted average) Geometric mean of the 3 cohort geometric means, calculated as average unit risk of individual studies for each cohort (that is, not a weighted average) Evaluated separately by sex and for 4 smoking exposure groups, with IUR representing best estimate of upper 95% confidence limit Inverse variance weighted average of IURs calculated for each cohort
Dosimetric Adjustments Adjusted for between occupational and environmental exposures (using Canadian mortality rates) Utilized US 1976 lung cancer mortality rates and survival probabilities Not described Adjusted for smoking (Welch et al., 1982) Texas-specific mortality rates for lung cancer and survival rates were used to calculate PODs and IURs
Inhalation Unit Risk (IUR) 6.4 × 10-3 per µg/m3 4.3 × 10-3 per µg/m3 1.5 × 10-3 per µg/m3 3.3 × 10-3 per µg/m3 1.5 x 10-4 per μg/m3
Estimated lifetime risk 10-4 0.02 μg/m3 0.02 μg/m3 0.07 μg/m3 0.03 μg/m3 0.7 μg/m3
Estimated lifetime risk 10-5 0.002 μg/m3 0.002 μg/m3 0.007 μg/m3 0.003 μg/m3 0.07 μg/m3
Estimated lifetime risk 10-6 0.0002 μg/m3 0.0002 μg/m3 0.0007 μg/m3 0.0003 μg/m3 0.007 μg/m3

Page details

2026-05-22