Royal Canadian Mounted Police
Symbol of the Government of Canada

Common menu bar links

2.0 Methods

Forensic Identification Services Chemical Carcinogenicity Evaluation

The objective of the methods outlined below was to identify the carcinogenic potential and to classify or categorize this potential of chemicals that have been, may have been, or are in current use or development for use by FIS personnel.

Cantox was provided a list of 66 chemicals used by FIS personnel in their laboratory work. The primary sources of information used to evaluate carcinogenicity potential was to identify the chemicals that have already been evaluated and classified by government and non-government agencies, such as Canadian Centre for Occupational Health and Safety (CCOHS), the International Agency for Research on Cancer (IARC) monographs, the National Toxicology Program’s Report on Carcinogens (NTP RoC), European Commission’s Institute for Health and Consumer Protection (IHCP), California Environmental Protection Agency - Proposition 65: List of Chemicals Known to Cause Cancer (CalEPA Prop 65), American Conference of Governmental Industrial Hygienists – Threshold Limit Values for Chemical Substances and Physical Agents & Biological Exposure Indices (ACGIH, 2009), and Agency for Toxic Substances & Disease Registry (ATSDR). If a chemical was not listed in any of these databases, further searches of publicly available scientific databases, such as PubMed, ToxNet or other databases provided by CCOHS were conducted to locate pertinent data to characterize the potential for a substance to cause cancer.

The strength of data in determining carcinogenic potential in descending order, includes: a) the results of studies in humans, called epidemiology studies, often occupationally exposed to the chemicals in question, b) the results of animal studies, generally 2-year bioassays conducted in rats and mice, c) the results of genetic toxicity testing to assess DNA-damaging potential, and d) structure-activity relationship analyses (Goodman and Wilson, 1991; ICPEMC Membership Committee, 1982; Anderson, 1993; Richard, 1994; Ashby, 1996; van den Brandt et al., 2002; Cogliano et al., 2004; U.S. EPA, 2005; IARC, 2006; Mayer et al., 2008; Benigni and Bossa, 2008; Guyton et al., 2009; SCHER, 2009). New technologies involving the assessment of changes in gene and protein expression in relation to tumour development, use of transgenic animals, and of enhanced computer simulations are also emerging for application in cancer hazard identification (Barlow et al., 2002; Benfenati et al., 2009; Guyton et al., 2009).

The notion of human epidemiology and animal testing is reasonably straightforward and there exist a number of guidelines regarding their design, conduct, and interpretation (OECD, 1981a,b; Cogliano et al., 2004; Pocock et al., 2004; IARC, 2006). However, the concept of genetic toxicity testing may be a little less familiar. The assessment of the potential for chemicals to cause damage to DNA is generally conducted in a tiered fashion (Anderson, 1993; Brusick, 1994) whereby the initial assessment is through the use of (bench-top type) tests, often in bacteria or cultured mammalian cells. Chemicals that produce negative results in these tests are generally not assessed further. If the results of the test(s) are positive (i.e., suggesting mutagenic activity), chemicals are usually then assessed in one or more in vivo tests involving whole animals whereby the exposure to the chemical of interest incorporates pharmacokinetics as well as appropriate biotransformation products (Anderson, 1993; Ashby, 1996). As such, the results of in vivo tests are generally preferable to those of test as they most likely better reflect the mutagenic potential of the chemical in whole organisms. Tests for genotoxic effects are essentially screening level assays used to predict potential for carcinogenic activity (ICPEMC, 1982; Anderson, 1993; Ashby, 1996).

It has been well documented that the genetic toxicity, and by extrapolation the carcinogenic potential, of chemicals is determined by their structure and to a lesser extent their physical-chemical properties. A number of functional chemical groups have been identified which are likely to produce positive results in test(s) of mutagenic potential (Ashby and Tennant, 1991; Ashby, 1996; Benigni and Bossa, 2008). For example, nitrosoureas, nitrosoamides, epoxides, polyaromatic hydrocarbons, aromatic amines, and others have been clearly identified as possessing mutagenic potential. Many of these have also been documented to be potent carcinogens, particularly in animal bioassays. From such structure-activity relationships, the mutagenic potential of classes of compounds can be inferred (i.e., chemicals containing nitrosamine groups are almost invariably mutagenic). Similarly, there exists chemical structures for which there is little evidence of mutagenic activity. Given the utility of chemical structure in assessing genetic toxicity potential, the assessment of a chemical for which mutagenicity test data are not available can involve the analysis of the functional groups present and/or the use of data from similarly-structured compounds which contain the same functional groups as the chemical of interest. This structure activity approach was applied to those chemicals that either had no identifiable human or animal carcinogenicity data and which have not been subjected to actual genetic toxicity testing. In addition the SAR approach was applied to those chemicals where only limited genetic toxicity testing data were available.

The chemicals assessed were categorized according to the weight-of-evidence available from public databases developed by authoritative regulatory bodies and from the published scientific literature. The categories and their levels of evidence were considered according to a modification to IARC’s (IARC, 2006) and Health Canada’s (Health Canada, 1994) classification criteria. The specific IARC model was not used since many of the chemicals on the list have not been specifically evaluated by this agency. The classification categories included: "Known to be human carcinogens", "Reported to produce tumours in experimental animals but for which human evidence is either lacking or inconclusive", and "Not reported to have carcinogenic potential". The description of these categories is presented below.

  • Known to be human carcinogens: Chemicals classified in this category have been demonstrated to human carcinogens on the basis of human experience, results of human epidemiology data, and on the classification of chemicals as human carcinogens by appropriate competent authorities.
  • Reported to produce tumours in experimental animals, but for which human evidence is either lacking or inconclusive: Chemicals classified in this category have shown some evidence of carcinogenic potential in animal studies that were conducted according to established protocols (e.g., OECD, NTP, EPA, NTP), used appropriate routes of exposure (i.e., oral, dermal, or inhalation), and were reported in adequate detail. As a conservative procedure some chemicals were classified in this category on the basis of weak evidence of carcinogenic activity in animals.
  • Not reported to have carcinogenic potential: Chemicals in this category have not been reported to be associated with the development of tumours in humans or in studies conducted with animals. This category was further subdivided into:
    • Chemicals with theoretical risk:  These chemicals have been reported to show evidence of genetic toxicity or have structural alerts present identifying carcinogenic or mutagenic potential.
    • Chemicals without theoretical risk:   These chemicals have not been reported to cause genetic toxicity and do not have any identifiable structural alerts present to indicate genetic or carcinogenic concern.

It should be noted that these classification systems are to some extent arbitrary and do not necessarily take into account the fact for some chemicals that could be placed in category "B", in that they have weak or moderate evidence of carcinogenic activity in experimental animals, actually could be considered in category "C" since the manner in which the tumours in animals develop in response to chemical treatment is not relevant to humans (i.e., due to excessively high doses, occurring by a route of exposure not relevant to occupational exposure, or are due to species-specific toxic mechanisms). The classification systems and associated nomenclature used in this report, however, do not allow for this level of interpretation of the experimental data.

It is critical to note that the classification system is based on hazard alone. It does not take into account the nature and magnitude of potential human exposures, including those of FIS personnel. The likelihood of a given chemical causing cancer in human populations, even those classified as "Known to be human carcinogens" requires careful consideration of the exposure circumstance and of the conditions under which the chemical was classified. As a result, as part of this evaluation, where data exist, limits of "safe" exposure to chemicals considered either "Known to be human carcinogens" (category "A") or "Reported to produce tumours in experimental animals, but for which human evidence is either lacking or inconclusive" (category "B") were identified so as to provide a measure of the carcinogenic potency of the chemical involved. Chemicals with higher allowable daily exposures are less "potent" in terms of toxicity or carcinogenic activity. "Safe" levels of exposure to chemicals that have carcinogenic potential and which have DNA-damaging ability, or for which the way in which the chemical causes cancer is not known, are calculated so as to provide de minimis or negligible amount of risk. As a result, for those chemicals a "safe" level of exposure may be stated for a risk of cancer of 1 in 100,000 or 1-in-1,000,000. These types of exposure limits, often called "risk-specific-doses" (RsD), "Virtually Safe Doses (VsD), or Non-Significant Risk Level (NSRL as used by the CalEPA), are generally based on very conservative (i.e., health protective) assumptions and are derived from mathematical modeling of either cancer in incidence rates in humans exposed to the chemical or tumour data from rodent studies (Krewski et al., 1990; U.S. EPA, 2005; Barlow et al., 2006; SCHER, 2009). These types of exposure limits are associated with chemicals, in terms of our classification scheme, considered "Known to be human carcinogens". In some cases, these types of exposure limits are also associated with chemicals considered "Reported to produce tumours in experimental animals, but for which human evidence is either lacking or inconclusive".

The "safe" levels of exposure to chemicals that show no evidence of cancer causation in humans, but are associated with cancer development in rodents, but by mechanisms not involving DNA damage, are generally calculated differently than for those chemicals that cause cancer in humans and/or in animals and show evidence of DNA damaging potential. For these chemicals, doses that are not associated with tumour development (usually from animal studies since human data for chemicals in this category are generally lacking), a dose termed a no-observable-adverse-effect level or NOAEL, is used as starting point to which "safety" or "uncertainty" factors are applied to arrive at an exposure limit (U.S. EPA, 2005). The starting point could also be dose determined through mathematical modelling of tumour data to establish a dose level, known as the "Benchmark Dose Level" (BMDL)" associated with a pre-determined degree of risk (e.g., the statistical determination from tumour data the dose associated with a 10% additional risk from background) (U.S. EPA, 2005; EFSA, 2009). Exposure limits derived from NOAELs or BMDLs to which safety factors are applied are often called an Acceptable Daily Intake (ADI) or Reference Dose (RfD) (Bolt and Degen, 2004; U.S. EPA, 2005; EFSA, 2009; SCHER, 2009). It is worthy to note that exposures below the ADI or RfD would be considered to have "zero" risk for cancer development. In our scheme, ADI or RfD type exposure limits would typically be applied to those chemicals considered "Not reported to have carcinogenic potential".