REGULATORY
TOXICOLOGY
AND
16,223-244
PHARMACOLOGY
Environmental ALAN 2828 Birchwood
Drive,
Tobacco
Smbke
RODGMAN
Winston-Salem,
Received
(1992)
May
North
Carolina
27103
9, 1992
In 1992, the U.S. Environmental Protection Agency (EPA) issued a “draft” assessmentof ETS and lung cancer in adults and respiratory disorders in children. Relying on weak and inconclusive epidemiological data, the supposed similarity between ETS and MS, the presence of “known or suspected carcinogens” in MS and by extrapolation in ETS, and the “biological plausibility” of an adverse relationship between ETS and health, the EPA recommended that ETS be classified as a “Group A (known human) carcinogen.” Fundamental physical and quantitative chemical differences among ETS, MS, and SS and human exposure to each smoke were disregarded: The three are not equivalent nor is ETS exposure a quantitative variant of cigarette smoking. A substantial difference in retention percentage overlaysthe huge dosimetric difference between exposures. As a result, the “dosage” of ETS retained is miniscule relative to MS. Also, conclusions reached by the EPA and the use of tenuous relationships as basesfor Group A classification are unwarranted because of failure to consider the data upon which the “tumorigenicity” of the ETS components was based, questions on the presence and/or levels of these components in MS, and data indicating that a 25- to 30-fold decrease of a high-level dose of MS or MS condensate diminished the effects observed in bioassays from pronounced to zero, i.e., a threshold was demonstrated. Finally, EPA overlooked the more than 100 tobacco smoke components known to inhibit the tumorigenic action of many of the listed “tumorigens.” 0 1992 Academic PISS, Inc.
INTRODUCTION The U.S. Environmental Protection Agency (EPA) (1990a,b) issued two draft documents in which it defined ETS as a carcinogen and designated it as a “Group A carcinogen” (EPA, 1990b) in an effort to encourage regulation of smoking in the workplace. The EPA relied on data from various epidemiological studies on lung cancer incidence in ETS-exposed nonsmokers as an indication that ETS was causally related to lung cancer even though these studies have been severely criticized for various inadequacies (Layard, 1990). Also, EPA extrapolated SS (and MS) qualitative and quantitative composition data to ETS with little regard for either the profound physical and quantitative chemical differences among MS, SS, and the highly diluted ETS system or the biological implications of these differences. Of prime concern to the EPA were 43 smoke and tobacco components (IARC, 1985, 1986; Hoffmann and 223 0273-2300/92
$5.00
Copyright 0 I992 by Academic Press, Inc. All rights of reproduction in any fom resxved.
224
ALAN
ROJIGMAN
Hecht, 1989) which, in one biological system or other and at doses far in excess of those encountered in MS or SS (EPA, 1990a), had been reported as “tumorigenic.” The EPA assessedthe health consequences with regard to lung cancer of the listed 43 components as follows: “Of the 99 compounds in tobacco smoke that have been studied in detail, at least 43 are complete carcinogens,’ each able on its own to cause the development of cancer in humans or animals.” This assessment is incorrect since most have not been shown to be (1) tumorigenic to any human tissue or (2) tumorigenic to lung tissue in laboratory animals. These two points are addressed by Hoffmann and Hecht (1989) in the text accompanying their “List of 43.” In addition, the few that have produced lung tumors in laboratory animals have done so at dose levels far in excess of those encountered in MS, SS, or ETS. The EPA also disregarded evidence indicating that ETS is an extremely dilute system to which a nonsmoker is exposed compared to MS inhaled by a smoker. The EPA ignored the wealth of data available from inhalation studies involving lifetime exposure of laboratory animals to cigarette MS: No squamous cell carcinomas, the tumor type reportedly associated with cigarette smoking in humans, have been observed in animals exposed via inhalation to MS (cf. Huber, 1989). Claims-based on these negative findings-that the rodent is an inappropriate model for such inhalation studies, particularly those involving MS, have been shown to be invalid. As a matter of Agency policy, the EPA overlooks the fact that progressive reduction of the dose of an administered material which produced a given effect in a bioassay eventually leads to a level at which no effect is observed. Despite much data to contradict it, the theory to which the EPA subscribes is that the only safe threshold limit value for a material tumorigenic in animals is a “zero” dose for humans. This concept is applied to compounds such as benzo[a]pyrene and similar polycyclic aromatic hydrocarbons (PAHs) for which there is either no hard evidence of tumorigenicity in man or, if the theory were correct, cancer in man would be even more extensive than it currently is. Many PAHs are components of foodstuffs in the average diet. Except for 5-methylchrysene, PAHs listed by Hoffmann and Hecht (1989) as tobacco and tobacco smoke tumorigens have been identified in many foodstuffs. Maga (1988) listed 65 common PAH-containing foodstuffs. When many foodstuffs are heated during preparation, their PAH content increased dramatically, e.g., a single serving of charcoal-broiled meat contains more than 600 times the benzo[a]pyrene level in the MS from one cigarette. Maga also reported the dietary intake of benzo[a]pyrene (charcoal-broiled meat excluded) averaged about 500 rig/day. From their analysis of human exposure to benzo[a]pyrene, Waldman et al. (199 1) reported that “the range and magnitude of dietary exposures” ranged from 2 to 500 rig/day and “were much greater than for inhalation (10 to 50 ng/[day]).” For some subjects, however, they found a dietary maximum of 1149 rig/day, despite omission of the contribution of benzo[a]pyrenecontaining beverages. These benzo[a]pyrene-intake estimates (Maga, 1988; Waldman et al., 1991) were lower than that reported by Hattemeyer-Frey and Travis (1991): 2200 rig/day (97% from diet; 3% from inhalation and water contamination). If benzo[u]pyrene were tumorigenic in man and its threshold limit value were zero, the ’ The citation referred to the Surgeon General’s 1989 report which, in turn, reproduced the table eventually presented in Hoffmann and Hecht (I 989).
ENVIRONMENTAL
TOBACCO
225
SMOKE
incidence of digestive tract cancer would be substantially higher than it is. When there is evidence to the contrary that a threshold does indeed exist for a compound under investigation, the EPA is required to consider such evidence and not adhere to the presumption of “no threshold.” This article presents examples of such “evidence to I,, the contrary.” THE PROPERTIES
OF MS, SS, AND
ETS
ETS comprises diluted and aged: (A) exhaled MS (i.e., MS not retained by the smoker), (B) SS generated during the puff (including, for the cigarette, vapor-phase components diffusing from the tobacco rod through the cigarette paper), and (C) SS generated during the smolder period between puffs. By weight, C B A > B. The SSs, B + C, contribute 85 to 90% of ETS; exhaled MS, A, constitutes the remainder. Studies (Laskowski, 195 1; Dalhamn et al., 196 1) indicate that no inhaled MS component is 100% absorbed and retained in the smoker’s respiratory tract; i.e., every inhaled MS component is considered to be present in exhaled MS. With no evidence to the contrary, it is highly probable that the compositions of inhaled MS, exhaled MS, SS, and ETS are qualitatively the same but quantitatively different. Steenland (1992) noted: [T]he constituents of sidestream smoke are different from those of inhaled mainstream smoke. . . These differences imply that it is difficult to determine the relative toxicity of sidestream smoke YSmainstream smoke. Consequently, arguments inferring ETS health effectsbased on known health effects of mainstream smoke . are not appropriate.
Just as profound quantitative differences exist among the chemical compositions of fresh and aged MS, fresh and aged SS, and ETS, the physical properties of these smokes differ significantly. One physical property important in these smoke types, their inhalation, and their retention is particle size, frequently described as mean mass aerodynamic diameter (MMAD). Table 1 summarizes the physical properties of MS, SS, and ETS and their differences. When freshly generated MS is inhaled during smoking, the aerosol particles are exposed in the respiratory tract to a high-humidity atmosphere. As a result, inhaled particles absorb water and increase in size. Exhaled particles are, on average, 20-25% larger than inhaled particles (Ingebrethsen, 1986, 1989; Ingebrethsen and Sears, 1989; Ingebrethsen et al., 1988, 1990). When these water-saturated exhaled MS particles are released to the atmosphere, they cool and immediately undergo several evaporative processes which are completed in a few milliseconds. Components, usually gaseous under ambient conditions, and components with modest vapor pressures evaporate from the particle. Water, incorporated in the particle either during the smoke formation process in the tobacco rod or during its residence time in the highly humid confines of the respiratory tract, also evaporates. THE DESIGNATION
BY EPA OF ETS AS A “GROUP
A CARCINOGEN”
Hoffmann and Hecht (1989) listed 43 tobacco and/or tobacco smoke components as “tumorigenic agents” and the per cigarette MS deliveries of each. Examination
226
ALAN
RODGMAN TABLE
1
PHYSICAL PROPERTIESOF MS, SS, AND ETS AND THEIR EFFECTON SMOKE PROPERTIES ~OpefiY Number of identified components
MS”
SS”
ETS
4 100 in particulate phase (PP); 1000 in vapor phase (VP). Some components are present in both the PP and VP, e.g., HCN. simple phenols, volatile Nnitrosamines.
Composition assumed to be qualitatively similar to that of MS; i.e., the number and identity of the SS and ETS components are the same as those in MS. Quantitative diierences in component levels are substantial. The distribution of a component between PP and VP depends on the nature (acid, base, neutral) and the physical properties (vapor pressure, etc.) of the particular component. The decay (decrease) of an individual ETS component is also dependent on numerous factors such as its nature, its physical properties, and the temperature, relative humidity (RH), ventilation, and nature of surfaces (carpets, drapes, upholstered furnishings, etc.) in the smoke space.
850-950°C 500-65O”C Low oxygen-containing atmosphere 30-40
500-600°C 500-6OO”C Reducing atmosphere
Fresh whole MS particles have MMAD = 0.3-0.4 pmc and contain volatile components which readily vaporize from the particles.
Fresh SS particles are
Approximate temperature of Fire cone Smoke formation Smoke formation atmosphere l l
Approximate ‘70of tobaccorod consumedb Particle size, pm
Particle concentration, No./cm’ Retention of particulate matter in respiratory tract
Because of coagulation, hydration, evaporative transfer and other physical processes, e.g., the cloud effect, MS particles behave as though they have a MMAD in the micrometer range: 109 to 10’0
50-60
about the samesizeas MS particles; within a short time (
TOXICOLOGY
AND
16,223-244
PHARMACOLOGY
Environmental ALAN 2828 Birchwood
Drive,
Tobacco
Smbke
RODGMAN
Winston-Salem,
Received
(1992)
May
North
Carolina
27103
9, 1992
In 1992, the U.S. Environmental Protection Agency (EPA) issued a “draft” assessmentof ETS and lung cancer in adults and respiratory disorders in children. Relying on weak and inconclusive epidemiological data, the supposed similarity between ETS and MS, the presence of “known or suspected carcinogens” in MS and by extrapolation in ETS, and the “biological plausibility” of an adverse relationship between ETS and health, the EPA recommended that ETS be classified as a “Group A (known human) carcinogen.” Fundamental physical and quantitative chemical differences among ETS, MS, and SS and human exposure to each smoke were disregarded: The three are not equivalent nor is ETS exposure a quantitative variant of cigarette smoking. A substantial difference in retention percentage overlaysthe huge dosimetric difference between exposures. As a result, the “dosage” of ETS retained is miniscule relative to MS. Also, conclusions reached by the EPA and the use of tenuous relationships as basesfor Group A classification are unwarranted because of failure to consider the data upon which the “tumorigenicity” of the ETS components was based, questions on the presence and/or levels of these components in MS, and data indicating that a 25- to 30-fold decrease of a high-level dose of MS or MS condensate diminished the effects observed in bioassays from pronounced to zero, i.e., a threshold was demonstrated. Finally, EPA overlooked the more than 100 tobacco smoke components known to inhibit the tumorigenic action of many of the listed “tumorigens.” 0 1992 Academic PISS, Inc.
INTRODUCTION The U.S. Environmental Protection Agency (EPA) (1990a,b) issued two draft documents in which it defined ETS as a carcinogen and designated it as a “Group A carcinogen” (EPA, 1990b) in an effort to encourage regulation of smoking in the workplace. The EPA relied on data from various epidemiological studies on lung cancer incidence in ETS-exposed nonsmokers as an indication that ETS was causally related to lung cancer even though these studies have been severely criticized for various inadequacies (Layard, 1990). Also, EPA extrapolated SS (and MS) qualitative and quantitative composition data to ETS with little regard for either the profound physical and quantitative chemical differences among MS, SS, and the highly diluted ETS system or the biological implications of these differences. Of prime concern to the EPA were 43 smoke and tobacco components (IARC, 1985, 1986; Hoffmann and 223 0273-2300/92
$5.00
Copyright 0 I992 by Academic Press, Inc. All rights of reproduction in any fom resxved.
224
ALAN
ROJIGMAN
Hecht, 1989) which, in one biological system or other and at doses far in excess of those encountered in MS or SS (EPA, 1990a), had been reported as “tumorigenic.” The EPA assessedthe health consequences with regard to lung cancer of the listed 43 components as follows: “Of the 99 compounds in tobacco smoke that have been studied in detail, at least 43 are complete carcinogens,’ each able on its own to cause the development of cancer in humans or animals.” This assessment is incorrect since most have not been shown to be (1) tumorigenic to any human tissue or (2) tumorigenic to lung tissue in laboratory animals. These two points are addressed by Hoffmann and Hecht (1989) in the text accompanying their “List of 43.” In addition, the few that have produced lung tumors in laboratory animals have done so at dose levels far in excess of those encountered in MS, SS, or ETS. The EPA also disregarded evidence indicating that ETS is an extremely dilute system to which a nonsmoker is exposed compared to MS inhaled by a smoker. The EPA ignored the wealth of data available from inhalation studies involving lifetime exposure of laboratory animals to cigarette MS: No squamous cell carcinomas, the tumor type reportedly associated with cigarette smoking in humans, have been observed in animals exposed via inhalation to MS (cf. Huber, 1989). Claims-based on these negative findings-that the rodent is an inappropriate model for such inhalation studies, particularly those involving MS, have been shown to be invalid. As a matter of Agency policy, the EPA overlooks the fact that progressive reduction of the dose of an administered material which produced a given effect in a bioassay eventually leads to a level at which no effect is observed. Despite much data to contradict it, the theory to which the EPA subscribes is that the only safe threshold limit value for a material tumorigenic in animals is a “zero” dose for humans. This concept is applied to compounds such as benzo[a]pyrene and similar polycyclic aromatic hydrocarbons (PAHs) for which there is either no hard evidence of tumorigenicity in man or, if the theory were correct, cancer in man would be even more extensive than it currently is. Many PAHs are components of foodstuffs in the average diet. Except for 5-methylchrysene, PAHs listed by Hoffmann and Hecht (1989) as tobacco and tobacco smoke tumorigens have been identified in many foodstuffs. Maga (1988) listed 65 common PAH-containing foodstuffs. When many foodstuffs are heated during preparation, their PAH content increased dramatically, e.g., a single serving of charcoal-broiled meat contains more than 600 times the benzo[a]pyrene level in the MS from one cigarette. Maga also reported the dietary intake of benzo[a]pyrene (charcoal-broiled meat excluded) averaged about 500 rig/day. From their analysis of human exposure to benzo[a]pyrene, Waldman et al. (199 1) reported that “the range and magnitude of dietary exposures” ranged from 2 to 500 rig/day and “were much greater than for inhalation (10 to 50 ng/[day]).” For some subjects, however, they found a dietary maximum of 1149 rig/day, despite omission of the contribution of benzo[a]pyrenecontaining beverages. These benzo[a]pyrene-intake estimates (Maga, 1988; Waldman et al., 1991) were lower than that reported by Hattemeyer-Frey and Travis (1991): 2200 rig/day (97% from diet; 3% from inhalation and water contamination). If benzo[u]pyrene were tumorigenic in man and its threshold limit value were zero, the ’ The citation referred to the Surgeon General’s 1989 report which, in turn, reproduced the table eventually presented in Hoffmann and Hecht (I 989).
ENVIRONMENTAL
TOBACCO
225
SMOKE
incidence of digestive tract cancer would be substantially higher than it is. When there is evidence to the contrary that a threshold does indeed exist for a compound under investigation, the EPA is required to consider such evidence and not adhere to the presumption of “no threshold.” This article presents examples of such “evidence to I,, the contrary.” THE PROPERTIES
OF MS, SS, AND
ETS
ETS comprises diluted and aged: (A) exhaled MS (i.e., MS not retained by the smoker), (B) SS generated during the puff (including, for the cigarette, vapor-phase components diffusing from the tobacco rod through the cigarette paper), and (C) SS generated during the smolder period between puffs. By weight, C B A > B. The SSs, B + C, contribute 85 to 90% of ETS; exhaled MS, A, constitutes the remainder. Studies (Laskowski, 195 1; Dalhamn et al., 196 1) indicate that no inhaled MS component is 100% absorbed and retained in the smoker’s respiratory tract; i.e., every inhaled MS component is considered to be present in exhaled MS. With no evidence to the contrary, it is highly probable that the compositions of inhaled MS, exhaled MS, SS, and ETS are qualitatively the same but quantitatively different. Steenland (1992) noted: [T]he constituents of sidestream smoke are different from those of inhaled mainstream smoke. . . These differences imply that it is difficult to determine the relative toxicity of sidestream smoke YSmainstream smoke. Consequently, arguments inferring ETS health effectsbased on known health effects of mainstream smoke . are not appropriate.
Just as profound quantitative differences exist among the chemical compositions of fresh and aged MS, fresh and aged SS, and ETS, the physical properties of these smokes differ significantly. One physical property important in these smoke types, their inhalation, and their retention is particle size, frequently described as mean mass aerodynamic diameter (MMAD). Table 1 summarizes the physical properties of MS, SS, and ETS and their differences. When freshly generated MS is inhaled during smoking, the aerosol particles are exposed in the respiratory tract to a high-humidity atmosphere. As a result, inhaled particles absorb water and increase in size. Exhaled particles are, on average, 20-25% larger than inhaled particles (Ingebrethsen, 1986, 1989; Ingebrethsen and Sears, 1989; Ingebrethsen et al., 1988, 1990). When these water-saturated exhaled MS particles are released to the atmosphere, they cool and immediately undergo several evaporative processes which are completed in a few milliseconds. Components, usually gaseous under ambient conditions, and components with modest vapor pressures evaporate from the particle. Water, incorporated in the particle either during the smoke formation process in the tobacco rod or during its residence time in the highly humid confines of the respiratory tract, also evaporates. THE DESIGNATION
BY EPA OF ETS AS A “GROUP
A CARCINOGEN”
Hoffmann and Hecht (1989) listed 43 tobacco and/or tobacco smoke components as “tumorigenic agents” and the per cigarette MS deliveries of each. Examination
226
ALAN
RODGMAN TABLE
1
PHYSICAL PROPERTIESOF MS, SS, AND ETS AND THEIR EFFECTON SMOKE PROPERTIES ~OpefiY Number of identified components
MS”
SS”
ETS
4 100 in particulate phase (PP); 1000 in vapor phase (VP). Some components are present in both the PP and VP, e.g., HCN. simple phenols, volatile Nnitrosamines.
Composition assumed to be qualitatively similar to that of MS; i.e., the number and identity of the SS and ETS components are the same as those in MS. Quantitative diierences in component levels are substantial. The distribution of a component between PP and VP depends on the nature (acid, base, neutral) and the physical properties (vapor pressure, etc.) of the particular component. The decay (decrease) of an individual ETS component is also dependent on numerous factors such as its nature, its physical properties, and the temperature, relative humidity (RH), ventilation, and nature of surfaces (carpets, drapes, upholstered furnishings, etc.) in the smoke space.
850-950°C 500-65O”C Low oxygen-containing atmosphere 30-40
500-600°C 500-6OO”C Reducing atmosphere
Fresh whole MS particles have MMAD = 0.3-0.4 pmc and contain volatile components which readily vaporize from the particles.
Fresh SS particles are
Approximate temperature of Fire cone Smoke formation Smoke formation atmosphere l l
Approximate ‘70of tobaccorod consumedb Particle size, pm
Particle concentration, No./cm’ Retention of particulate matter in respiratory tract
Because of coagulation, hydration, evaporative transfer and other physical processes, e.g., the cloud effect, MS particles behave as though they have a MMAD in the micrometer range: 109 to 10’0
50-60
about the samesizeas MS particles; within a short time (
