Choice of a Marker Compound for Polycyclic Aromatic Hydrocarbon Mixtures

29. While it is accepted that the increased risk of cancer observed in occupational studies is likely to be the result of the carcinogenic activity of the whole PAH mixture present, for practical reasons a marker compound for this carcinogenic activity is desirable. The use of B[a]P as a marker compound for total PAH exposure was briefly noted in previous paragraphs. We now discuss this use in more detail. B[a]P is a potent carcinogen in laboratory animals, it is very extensively documented and is readily measured. It is often used as a marker for total PAH exposure in industry and in the environment, and the Panel has examined the potential use of B[a]P to set an air quality standard for PAH mixtures in the UK.

30. The Panel considered that results from a recent Canadian study provide detailed support for the use of B[a]P as a marker for total PAH levels in the industrial context of ambient air in an aluminium smelting plant (Farant and Gariepy, 1998). In particular, the relation of B[a]P levels to the levels of 18 other individual PAHs was shown to be relatively stable. Together, these 19 PAH compounds constituted 90-95% by weight of the PAHs measured in the air from the plant.

31. The Panel then went on to consider whether B[a]P levels would serve as a reliable marker for total PAH levels in ambient air in the general environment as opposed to an industrial setting. It was found that although the mixtures were broadly similar in composition there were subtle differences. The analysis was therefore taken a stage further to determine whether these variations were likely to be reflected in differences in total carcinogenic activity.

32. The approach used by the Panel was to compare the sum of the potential carcinogenic contribution of 7 individual PAHs, listed in Table 5, that are present in ambient air with that of B[a]P, which served as a reference compound. The 7 PAHs were chosen because they had been classified as ‘probably' or ‘possibly' carcinogenic to humans, by IARC or the Department of Health's Committee on Carcinogenicity of Chemicals in Food, Consumer Products and the Environment (COC) and are measured in the TOMPs network. Contributions to total carcinogenicity from other PAH compounds are expected to be small although some PAHs, e.g. anthanthrene, with similar levels of carcinogenic activity to the chosen seven compounds, have not been measured at all sites and the carcinogenic potency of some PAHs has not been tested. Quantitative information on the relative carcinogenic potency of PAH compounds manifestly does not exist for humans and so an approach based upon data derived from animal studies was developed. The results of such studies have been interpreted cautiously but they do provide a means of ranking carcinogenic potency, within the limits of the test procedure, for a series of chemically similar compounds.

33. The potential contribution of each PAH compound was assessed by multiplying its concentration in air by its carcinogenic potency, both relative to B[a]P: i.e. Relative Concentration (RC) x Relative Potency (RP). Data for carcinogenic potency were obtained from animal experiments in which known concentrations of six of the seven PAH compounds were implanted in the lungs of rats (Deutsch-Wenzel et al., 1983; Wenzel-Hartung et al., 1990). Dose-response relationships were derived on the basis of the number of tumours that developed, and the carcinogenic potency of each compound was determined relative to B[a]P. The remaining compound, benzo(a)anthracene, has apparently not been tested in this system but its carcinogenic potency has been assessed, relative to that of B[a]P, by repeated application to the skin of mice. The results are shown in Table 5. It can be seen that, for example, benz[a]anthracene is present in London air at a higher concentration than B[a]P (Relative Concentration = 1.16), but that its estimated potency relative to B[a]P is low (Relative Potency = 0.1). Thus the overall contribution of benz[a]anthracene is only about one-tenth of that of B[a]P.

34. The results of the calculation are shown in bold type in the last line of the table. The estimated contribution of B[a]P to the total carcinogenicity of the 7 chosen PAH compounds is similar in the three locations considered: London: 44.6%; Middlesbrough: 37.5%; aluminium smelter: 49.3%. Data on relative carcinogenic potency can also be obtained from animal experiments in which the different PAH compounds are applied to the skin (United States Environmental Protection Agency, 1993). Use of these figures rather than those from the lung implantation studies gives similar estimates for the percentage contribution of B[a]P to the total carcinogenicity of the seven chosen PAH compounds. The overall conclusion from this approach was that occupational studies of workers at aluminium smelters that used B[a]P as a marker of PAH exposure formed a suitable basis for recommending an environmental standard, so long as major changes in the ambient mixture of PAH compounds do not occur in the future, and that an air quality standard for PAH mixtures could be expressed in terms of the ambient concentration of B[a]P.

Table 5. The estimated contribution of selected PAHs (particulate and gaseous) to total carcinogenic activity of PAH mixtures from different sites of exposure

Site Type	London: average 1991-95			Middlesbrough: average 1993-95			Aluminium Smelter
Source	DETR, 1997			DETR, 1997			Farant and Gariepy, 1998
PAHa	RC	RP	RCxRP	RC	RP	RCxRP	RC	RP	RCxRP
Benzo[a]pyrene	1	1	1	1	1	1	1	1	1
Benz[a]anthracene	1.2	0.1b	0.1	1.3	0.1	0.1	1.8	0.1	0.2
Dibenzo[ah]anthracene	0.3c	1.91	0.6	0.5c	1.91	1.0	0.1	1.91	0.2
Benzo[b]fluoranthene	2.4	0.11	0.3	2.4	0.11	0.3	3.8	0.11	0.4
Benzo[k]fluoranthene	2.1	0.03	0.1	1.9	0.03	0.1	1.2	0.03	0.1
Indeno[123cd]pyrene	1.8	0.08	0.1	1.5	0.08	0.1	0.7	0.08	0.1
Chrysene	2.0	0.03	0.1	2.3	0.03	0.1	3.5	0.03	0.1
% Contribution of B[a]P to total carcinogenic activity	44.6			37.5			49.3

Notes

^a Contributions from other PAHs are expected to be small although some PAHs with similar levels of carcinogenic activity have not been measured at all sites (e.g. anthanthrene) and the carcinogenic activity of some PAHs has not been tested.

^b The figure for benzo[a]anthracene is based on studies of its carcinogenic effect on the skin, see paragraph 33.

^c Also includes dibenz[ac]anthracene.

RC = Relative Concentration RP = Relative Potency. In both cases, relative to benzo[a]pyrene