Toxic Chemicals in Cigarette Mainstream Smoke - Hazard and Hoopla

These are curious times. The Canadian government has passed legislation that requires cigarette manufacturers to routinely test and publish the amounts of 44 toxic substances in cigarette mainstream smoke (MSS). Following in the footsteps of their northern neighbor, various US legislators and regulators are considering modifications to their cigarette testing and reporting programs that will also list toxicants in MSS. Across the Atlantic Ocean, the European Commission has passed a directive that may also follow the North American lead for public disclosure of MSS toxic chemicals for each brand of cigarette sold in the marketplace. United Kingdom authorities have also expressed their intention to follow this mandate.

It is difficult to understand the motivation and value of these existing or potentially forthcoming legislative actions. Although there is nearly total agreement among the world's scientists that cigarette smoking is a health hazard, few are bold enough to say with credibility which smoke chemicals or classes of chemicals are responsible for the adverse effects. Therefore, if the specialists are unable to interpret the smoke toxicant data, how is the general public to use their newfound knowledge?

The posting of smoke chemical toxicant data is also problematic for the Tobacco Industry for several reasons. First, no standard analytical methods exist for most suspected toxicants. Second, the listing of smoke toxicant yields may ignite a 21st Century version of the ‘tar’ wars in the USA during the 1960s; we have already seen evidence of such competition beginning in the US. Third, and most important of all, no one knows whether or not reducing the yield of one or more publicized MSS toxicant will result in a ‘less hazardous’ cigarette.

Assuming that the current situation is approximately as described above, the authors of this paper critically examined the existing lists of MSS toxicants. They discarded chemicals that are no longer relevant, e.g., DDT, N-nitrosodiethanolamine, added known smoke constituents that are glaringly absent, e.g., dioxins, and replaced the existing 1950-60s era nonfiltered cigarette MSS yields with those more representative of the present-day marketplace. Data for the Kentucky reference 1R4F cigarette smoked under standardized smoking conditions, i.e., those established by the International Organization for Standardization (ISO) and the Federal Trade Commission (FTC), are used as a surrogate for the modern-day cigarette whenever possible.

A list of smoke toxicants and their approximate concentrations in today's cigarettes is nearly useless without an appropriate ranking of their relative toxicity. Unfortunately, the toxicological data for ranking importance are available for fewer than 5% of the approximately 4800 reported smoke constituents. Although neither of this paper's authors presumes to be a toxicologist, we cite in our discussion several published attempts at ranking smoke toxicants. Specifically, ranking by US Occupational Safety and Health Administration (OSHA) permissible workplace exposure levels, use of US Environmental Protection Agency (EPA) toxicity criteria supplemented with California EPA criteria, and use of the Human Exposure - Rodent Potential methodology and database developed by AMES et al. when data are available. There appears to be a wide divergence in the permissible exposures allowable in the workplace and those advocated by environmental regulators. Thus, it is expected that rankings such as those presented herein will ultimately form the basis of MSS toxic chemical prioritization for either attempts at reduction by product developers or development of standardized analytical methods.

This review of MSS toxicants also explores the limitations of toxicological evaluations. The toxicity data used in the above ranking are derived wholly from studies of pure compounds. It is highly improbable that extrapolation of bioassay results determined on an individual compound to that compound when it is a component of a mixture as complex as cigarette MSS is valid. For example, several decades of research involving numerous investigators reported that the benzo[a]pyrene (BaP) content of cigarette smoke condensate (CSC) accounts for only a few percent of the tumor-bearing animals in the skin-painting bioassay. Subsequently they asserted that the tumorigenic polycyclic aromatic hydrocarbons (PAHs) in CSC could account for no more than 3 to 4% of the tumor-bearing animals. Inclusion of promoters, e.g., phenols, raises the level to about 5%. However, several of the same investigators recently claimed that BaP is one of two smoke components responsible for lung cancer in cigarette smokers.

While much is written about the hundred or so toxic components in cigarette smoke, little is published about the numerous nontoxic smoke components that have been shown in various bioassays to counteract the effects of the toxic ones. In some cases the inhibiting components are also listed as toxic, e.g., nicotine inhibits the mutagenicity of N-nitrosodimethylamine; the promoter phenol inhibits the tumorigenicity of BaP; the weakly tumorigenic benz[a]anthracene negates the potent tumorigenicity of BaP. On a one-to-one molar basis, many bicyclic, tricyclic, and tetracyclic nontumorigenic PAHs counteract the tumorigenicity of BaP and dibenz[a,h]anthracene.

To further illustrate this murky toxicological situation, the history and current knowledge of the importance of tobacco-specific nitrosamines (TSNAs) to the hazards of smoking is reviewed. In brief, these compounds were discovered in tobacco products and found to transfer to MSS (and sidestream smoke). Toxicological evaluations on the pure compounds demonstrated that they are potent carcinogens. Some public health scientists believed that if the levels of TSNAs could be reduced or lowered in MSS, then this would lead to a ‘less hazardous’ cigarette. Once given this assignment, agronomists discovered that at least for flue-cured tobaccos, the levels of TSNAs can be greatly reduced through the use of indirect heating in the curing barns. This was wonderful news. However, toxicologists soon conducted experiments comparing the toxicity of MSS from flue-cured cigarettes containing high and ultra-low concentrations of TSNAs. It must have been a surprise to these investigators when they could find no significant difference between the toxicities of the two smokes.

Some public health scientists have asserted that the reduction of the per cigarette ‘tar’ delivery below 15 mg/cig does not reduce the risk from smoking because of the hazard resulting from the higher levels of additives used to maintain consumer acceptability. Although no data in support of this assertion have ever been offered, much data generated during the past decade contradict the assertion. Ingredient addition at the usual level or at levels several times greater than normal does produce some minor changes in the smoke chemistry, but these changes do not result in any adverse biological response as measured in various bioassays to determine mutagenicity, tumorigenicity, etc.

From our review of the literature gathered to prepare this paper, we have come to several conclusions. These include the following:

1. It is possible to prepare a list of the known toxicants in MSS and to prioritize some of them based upon existing biological data. However, for more than 95% of the known constituents in MSS, there are no biological data.

2. Even if there were biological data for most MSS components, extrapolation of this pure-compound knowledge to the biological properties of a mixture containing them is beyond our scientific ability.

3. At our current state of scientific knowledge, no one will ever be able to legitimately claim the development of a ‘less hazardous’ cigarette based solely on the reduction of known toxic chemicals in MSS.

4. The approach of reducing ‘tar’ yields of cigarettes appears in retrospect to be the most practical means of producing a ‘less hazardous’ cigarette, because when product developers reduce ‘tar', both the known and unknown toxicants are reduced.

5. The ranked toxicants in MSS contain both gas-phase and semi-volatile constituents that appear to be important determinants of toxicity. Some of these constituents, e.g., N-nitrosodimethylamine, phenols, are reduced by triacetin-plasticized cellulose acetate filters. These filters also reduce ‘tar'. Additionally, it is well known that charcoal-containing filters have a high efficiency for removing carbonyl compounds from MSS. Development of more consumer-acceptable products that reduce gas-phase toxicants appears to be another route to a ‘less hazardous’ cigarette.