Decades of research on the health effects of tobacco use have generated overwhelming evidence to support the conclusion that tobacco use causes disease. An inference of causality requires evidence along the causal pathway from exposure to disease, including evidence on the effects of tobacco from experimental and observational study designs, and from investigations into the biological mechanisms of disease. A widely cited criteria for making a causal inference in epidemiology and public health are the Hill Criteria (Weed, 2000). The judgment that tobacco use causes diseases such as lung cancer and heart disease has been based on evidence from a wide range of investigations that fulfill the requirements of the Hill Criteria. This has been thoroughly reviewed and documented in reports of the Surgeon General on tobacco, such as the 2004 and 2010 reports (HHS, 2004a, 2010).
The evaluation of the health effects and mechanisms of modified risk tobacco products (MRTPs) is a closely related enterprise. Development of many MRTPs will be based on existing evidence and knowledge of the mechanisms of tobacco-related disease. In general, MRTPs are designed to remove or block a step in the causal pathway between tobacco exposure and disease. As such, evidence about how an MRTP intervenes on the causal pathways for tobacco-related disease will be critical. However, inferences about the health effects of an MRTP based on prior knowledge of the causal pathways of tobacco disease, while relevant, will not be sufficient to inform regulatory decisions. Independent evidence on the health effects of the MRTP will be necessary. The study of the health effects of tobacco use can provide an illustrative precedent for the evaluation
of MRTPs. The same range of research methods employed to establish a causal relationship between tobacco and disease will be needed to provide evidence on the health effects of MRTPs on both individual and public health. This chapter discusses that evidence and provides guidance on how the Food and Drug Administration (FDA) should consider different types of that evidence in its decision-making process. The chapter begins with a discussion of the composition of modified tobacco products. The committee then discusses biomarkers of MRTPs, including biomarkers of exposure and biomarkers of effects. Next, it discusses preclinical and clinical studies, including the advantages and disadvantages of those studies, and what evidence the various study types can provide to inform the FDA’s decisions on MRTPs.
Smokeless tobacco products, such as oral snuff, and combusted tobacco products, such as cigarettes, are the main types of tobacco products used in the United States (SAMHSA, 2007). The composition of tobacco and tobacco smoke has been the subject of intense study for at least the past 60 years, and studies have identified more than 8,000 constituents of tobacco and tobacco smoke (Rodgman and Perfetti, 2009). Validated methods are available to quantify many constituents of tobacco and tobacco smoke (Borgerding and Klus, 2005; Rodgman and Perfetti, 2009), and the chemical composition can have a large effect on the potential health risks of a given product. Product composition, including how the constituents compare to other products, therefore, is an important aspect of any new product. Although different tobacco products continue to be introduced, this section discusses the types of tobacco products currently available, the methods for analyzing them, and the commonly reported constituents. Smokeless products are discussed first, followed by a discussion of combusted products.
Smokeless Tobacco Products
Types of Smokeless Products
Smokeless tobacco products used in the United States include moist snuff and chewing tobacco (for oral use), and dry snuff (for nasal use). Types of chewing tobacco include plug, twist, and loose leaf varieties. The use of chewing tobacco and dry snuff has declined over time. Oral moist snuff is by far the most popular kind of smokeless tobacco in the United States (Federal Trade Commission, 2007). Oral moist snuff is used by placing the tobacco—either loose or packaged in a tea bag-like sachet—
in the space between the cheek and gum, or lip and gum. Generally, oral moist snuff is not chewed. Brands such as Copenhagen and Skoal, manufactured by Altria Group, Inc., and Grizzly and Kodiak, marketed by Reynolds American, Inc., are common.
The use of any form of smokeless tobacco has declined substantially between 1986 and 2003 (Nelson et al., 2006); in this time period, there was an approximately 5 percent decrease in overall smokeless tobacco sales (in pounds) (Federal Trade Commission, 2007). However, the use of moist snuff or dip increased by approximately 87 percent over the same period (Nelson et al., 2006). In 2005, total dollar sales for moist snuff accounted for more than 80 percent of total sales for smokeless tobacco (Federal Trade Commission, 2007). In 2008, 3.5 percent of Americans aged 12 or older (0.4 percent of women aged 12 or older and 6.8 percent of men aged 12 or older) had used a smokeless tobacco product in the previous month (SAMHSA, 2011).
Moist snuff for oral use contains both high salt and high moisture content (Stepanov et al., 2010). When placed in the oral cavity, the product generates excess saliva, usually requiring spitting. Recently, the tobacco industry has introduced and promoted spit-free smokeless tobacco products. These new products, such as Camel Snus and Marlboro Snus, contain low moisture content and are distributed in small pouches of flavored tobacco. The products have been marketed to current cigarette smokers for situations where smoking is prohibited (Hatsukami et al., 2007a). These products have design features in common with snus products that have been used in Sweden for many years. Users of Swedish Snus place the product between the gum and upper lip; it does not usually stimulate salivation. Other new smokeless tobacco products continue to appear. These include dissolvable products such as Camel Orbs (a pellet), Camel Sticks (a twisted toothpick-size stick), and Camel Strips (a film strip placed on the tongue). All of those new products are made from finely ground flavored tobacco (Rainey et al., 2011).
Methods of Analysis
Methods of analysis of the components of smokeless tobacco are standardized (IARC, 2007; Richter and Spierto, 2003; Richter et al., 2008; Song and Ashley, 1999; Stepanov and Hecht, 2005; Stepanov et al., 2008, 2010). Smokeless tobacco analyses include analyses for moisture content, pH, and components. Moisture content can be determined by the difference in weight before and after drying. For measurement of pH, the tobacco is extracted with water and the pH is determined with a pH meter. Nicotine can be determined by extraction of the tobacco and analysis by combined gas chromatography-mass spectrometry (GC-MS) or high-performance
liquid chromatography-mass spectrometry (LC-MS). Minor tobacco alkaloids such as nornicotine and anatabine are extracted, derivatized by reductive alkylation, and determined by gas chromatography-tandem mass spectrometry (GC-MS/MS). Tobacco-specific N-nitrosamines (TSNAs) are extracted and analyzed by either gas chromatography with nitrosamine selective detection or by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Both conventional and supercritical fluid extractions have been used. Polycyclic aromatic hydrocarbons can be quantified by extraction with cyclohexane followed by solid-phase extraction and GC-MS. Aldehydes are measured by extraction, derivatization with 2,4-dinitrophenylhydrazine, and GC-MS. Anions such as nitrate, nitrite, and chloride are determined by anion exchange with conductivity detection.
Laboratory analysis of constituents in these products would be a standard first step in the initial evaluation of any new product. These analyses are generally quite straightforward involving standard methods of extraction, sample cleanup, analyte identification, and quantitation. Data from diverse laboratories involved in the analysis of various products give comparable results for most analytes. There are differences in the literature in the manner in which the analytical data are expressed. Some investigators have expressed their data per dry weight of product, while others use wet weight, or even portion size. Because traditional moist snuff products typically contain about 50 percent water, it is crucial to recognize the manner in which the data are being expressed and to take this into consideration when making judgments on constituent levels. The expression of constituent levels per dry weight of product, with inclusion of data on water content is standard (Stepanov et al., 2008). Because portion sizes are fixed in the products encased in tea bag-like sachets, it is also important to report constituent levels per portion size for these products.
Laboratory analysis of constituents, however, may not reflect constituent uptake under conditions of use. Biomarker of exposure studies, described below, provide a more realistic indication of exposure.
Commonly Reported Constituents
Thousands of compounds have been identified in unburned tobacco (Rodgman and Perfetti, 2009), but routine analyses of smokeless tobacco have focused on relatively few of these compounds thought to be critical in its biological activities (IARC, 2007; Richter and Spierto, 2003; Richter et al., 2008; Song and Ashley, 1999; Stepanov and Hecht, 2005; Stepanov et al., 2008, 2010). Commonly reported constituents include TSNAs, nicotine and minor tobacco alkaloids, nitrite, nitrate
and other anions, metals, aldehydes, and polycyclic aromatic hydrocarbons. Nicotine is generally reported as protonated and unprotonated (determined by measuring pH of the product). This is important because unprotonated nicotine is absorbed more readily through the oral mucosa than protonated nicotine. Plasma nicotine levels are directly related to pH of the product: higher pH values lead to higher levels of plasma nicotine (IARC, 2007). Minor tobacco alkaloids might, along with nicotine, contribute to addiction. Unlike cigarette smoke, the most common strong carcinogens in smokeless tobacco products are TSNAs. Extensive data demonstrating their presence in parts per million quantities, greater than nitrosamine concentrations in any other consumer product intended for oral use, are available (IARC, 2007; Richter et al., 2008; Stepanov et al., 2008). Levels of polycyclic aromatic hydrocarbons and aldehydes have been less frequently reported (Stepanov et al., 2008, 2010).
There is solid evidence that nicotine is addictive, but little evidence of addictive potential for other constituents of smokeless tobacco products. With respect to the induction of cancer, it is suspected but not proven that TSNAs play a major role, while other compounds such as polycyclic aromatic hydrocarbons and aldehydes may also contribute. There may be other unidentified or unrecognized compounds in smokeless tobacco that contribute in important ways to its adverse health effects. Among the thousands of identified compounds in smokeless tobacco products, the 28 currently identified carcinogens represent only a small fraction (IARC, 2007; Rodgman and Perfetti, 2009). Furthermore, seemingly innocuous compounds such as sodium chloride, which occurs in amounts more than 5 percent in some smokeless tobacco products (IARC, 2007), could exacerbate the effects of carcinogens by leading to local irritation, among other effects (Stepanov et al., 2008).
Types of Products
Cigarettes are by far the most used combusted tobacco product. In 2009, there were more than 46 million cigarette smokers in the United States, about 20.6 percent of the adult population (CDC, 2010). Between the mid-1960s and 2004, cigarette smoking among adults decreased from approximately 42 percent to 21 percent; however, prevalence has not changed substantially since then (CDC, 1999, 2011b). Additionally, after substantial declines (66 percent) in cigar consumption from 1964 to 1993, consumption rates for cigars increased by close to 50 percent from 1993 to 1997 (NCI, 1998). In 2010, 5.2 percent of Americans aged 12 or older
had smoked cigars in the past month (SAMHSA, 2011). Other combusted products include pipes and water pipes.
Methods of Analysis
Because combusted products are burned, their constituents cannot simply be extracted as with smokeless tobacco products. Various machine methods attempt to simulate the smoking of tobacco products, and the smoke is collected and analyzed (IARC, 2004). Different organizations use different methods for generating smoke. For example, the International Organization for Standardization and the U.S. Federal Trade Commission smoking regimen uses a 35 mL puff every 60 seconds, and a puff duration of 2 seconds, with the filter ventilation holes (if present) open. Health Canada uses an intense smoking regimen with a 55 mL puff every 30 seconds, and a puff duration of 2 seconds, with the filter ventilation holes completely blocked. The Massachusetts Department of Health method has a 45 mL puff every 30 seconds, and a puff duration of 2 seconds, with the filter ventilation holes 50 percent blocked. It is widely recognized that none of these methods accurately reproduces the many ways smokers actually use cigarettes, but the methods can be used for comparison of one product to another (IARC, 2004).
Researchers can collect and analyze both mainstream smoke, which emanates from the filter end of the cigarette, and sidestream smoke, which emanates mainly from the burning tip of the product. For collection, a glass fiber filter separates arbitrarily gas phase constituents from total particulate matter, which collects on the filter (Adam et al., 2006). Once the combusted material is collected, the methods of analysis of the various constituents of cigarette smoke have some similarities to those used for smokeless tobacco. Because the products of combustion are generally more complex than those obtained by extraction of unburned tobacco, multiple extraction or purification steps are often necessary before the analysis can be completed, usually by GC-MS or LC-MS/MS techniques (IARC, 2004).
Laboratory analyses by machine smoking would be a standard first step in the initial evaluation of any new product, even though it is widely recognized that this approach has limitations. Machine smoking methods do not replicate human smoking conditions because smokers may vary their way of smoking a cigarette depending on many factors. Important among these is the well-established phenomenon of compensation, in which smokers may alter their method of smoking in order to compensate for lower machine measured amounts of nicotine and other constituents. They accomplish this in a number of different ways including increasing puff number or volume and blocking filter vents (NCI, 2001). Under a
given set of machine smoking conditions, analyses of particular constituents are generally well standardized leading to reasonable agreement in constituent levels among different laboratories. However, formalized interlaboratory comparisons have only been carried out for a few constituents. When reporting constituent levels for any product, it is crucial to describe the type of smoking regimen that has been used.
There is no proven method to replicate the many ways humans smoke cigarettes. The World Health Organization, under the Framework Convention on Tobacco Control, has adopted the approach of expressing machine-measured constituents per mg of nicotine for use in regulation, because this would presumably mitigate some of the effects of compensation (Burns et al., 2008). However, this approach is untested in a regulatory setting.
The measurement of smoke constituents can be challenging. Even measurement of parameters seemingly as simple as pH and free nicotine have led to controversy (Chen and Pankow, 2009; Pankow et al., 2003).
Commonly Reported Constituents
The FDA has developed a list of “harmful and potentially harmful constituents in tobacco products and tobacco smoke” that includes more than 100 constituents from various classes of chemicals (FDA, 2011a, 2011c). These include “tar,” nicotine and minor tobacco alkaloids, carbon monoxide (CO), nitrogen oxides, polycyclic aromatic hydrocarbons, TSNAs, volatile nitrosamines, aldehydes, aromatic amines, metals, phenols, ketones, volatile hydrocarbons such as benzene and butadiene, ethylene and propylene oxide, furan, hydrazine, hydrogen cyanide, heterocyclic aromatic amines, nitrogen compounds, pyridine, vinyl chloride, polonium-210, and others. The majority of these constituents have been routinely analyzed, and extensive data are available on their concentrations in tobacco smoke (Chen and Molduveanu, 2003; Counts et al., 2004; Ding et al., 2006, 2007; Gregg et al., 2004; Hammond and O’Connor, 2008; IARC, 2004; Roemer et al., 2004).
Furthermore, the same considerations discussed above with respect to smokeless tobacco apply to combusted products. It is not certain that the current list of harmful and potentially harmful constituents is complete. There may be other constituents among the more than 8,000 in tobacco and tobacco smoke (Rodgman and Perfetti, 2009) that are important but currently unrecognized. It is also known that there are interactions between carcinogens and tumor promoters or cocarcinogens that may not be recognized when simply analyzing a list of compounds (HHS, 2010; IARC, 2004).
Summary of Product Composition
Analysis of smokeless tobacco products or combusted products can be achieved using standardized and validated methods for a variety of constituents. Although there could be some inter-laboratory differences in results of these analyses, most data are generally comparable for a given product. In the analysis of smokeless tobacco products, the method of extraction and the method of expressing the results need to be taken into account when comparing data. In the analysis of combusted products, the method of machine smoking is critical when comparisons are to be made. None of the standard machine smoking methods replicate human smoking conditions, but these methods can be useful for comparison of different products under comparable conditions.
Studies of tobacco and tobacco-related diseases use a number of different biomarkers, and the validity of those biomarkers are key to the validity of the studies; biomarkers will continue to play an important role in the FDA’s regulation of MRTPs. The FDA will be making regulatory decisions about the marketing of MRTPs in the immediate future, but the latency period between tobacco exposure and the development of major clinical adverse health consequences is usually quite long. Validated biomarkers and other surrogates of tobacco-related disease outcomes that can provide information over a shorter time frame, therefore, will play a critical role in the evaluation of MRTPs. The Family Smoking Prevention and Tobacco Control Act of 2009 (FSPTCA) highlights the importance of addressing biomarkers and surrogates when it specifies that regulations or guidance issued by the Agency shall “include validated biomarkers, intermediate clinical endpoints, and other feasible outcome measures, as appropriate.”1
Terminology around biomarkers can be a controversial issue. Over the course of evaluating both the statutory language and the prevailing literature, the committee encountered inconsistencies in the definitions for terms central to this discussion, including the terms “biomarker,” “surrogate,” “intermediate endpoint,” and “endpoint.” The committee also found it important to differentiate between biomarkers of exposure and biomarkers of effect or risk. In this report, the committee broadly categorizes biomarkers as biomarkers of exposure and biomarkers of risk, and further distinguishes among specific types of biomarkers of risk. Specifically, the committee adopts the definitions articulated in the Institute
1 Family Smoking Prevention and Tobacco Control Act of 2009, Public Law 111-31, 123 Stat. 1776 (June 22, 2009)
Biomarker: A characteristic that is objectively measured and evaluated as an indicator of normal biological responses, pathogenic processes, or pharmacologic responses to an intervention.
Biomarker of exposure: The chemical, or its metabolite, or the product of an interaction between a chemical and some target molecule or cell, that is measured in a compartment in an organism.
Biomarker of risk: A biomarker that indicates a risk factor for a disease.
Clinical endpoint: A characteristic or variable that reflects how a patient or consumer feels, functions, or survives.
Surrogate endpoint: A biomarker that is intended to substitute for a clinical endpoint. A surrogate endpoint is expected to predict clinical benefit (or harm or lack of benefit or harm) based on epidemiologic, therapeutic, pathophysiologic, or other scientific evidence.
SOURCE: Adapted from IOM (2010).
of Medicine’s (IOM’s) 2010 report, Evaluation of Biomarkers and Surrogate Endpoints in Chronic Disease (IOM, 2010). Relevant definitions from that report are presented in Box 3-1. Biomarkers of exposure and biomarkers of risks are discussed below.
Biomarkers of Exposure
Biomarkers of human exposure to specific constituents of tobacco products may be the constituents themselves; metabolites of the constituents in urine, blood, breath, saliva, nails, or hair; or protein- or DNA-binding products (adducts) of the constituents or their metabolites. These biomarkers have the potential to bypass many of the uncertainties in product analysis and provide a realistic and direct assessment of carcinogen and toxicant dose in an individual. It should be emphasized however that the biomarkers discussed here are virtually all biomarkers of exposure to specific tobacco or tobacco smoke constituents. In most cases, they have not been validated as biomarkers of risk. Furthermore, these biomarkers are derived from specific constituents of tobacco products thought to be harmful to the consumer, but there may be unknown
or unmeasured constituents that are also harmful, or there may be combination effects of individual constituents that cannot be recognized by measurement of individual biomarkers of exposure. Presently, there is no single accepted biomarker that predicts the risk of disease in people who use tobacco products.
These biomarkers of exposure to tobacco toxicants and carcinogens are most frequently quantified by LC-MS/MS, GC-MS/MS, and related techniques. The first step in validation is analytical validation. This topic has been previously discussed in detail in a recent IOM report (IOM, 2010). Chapters of this 2010 report are provided in Appendix B.
Validation with Respect to Product Use
The second step in validation of a biomarker of exposure to tobacco toxicants and carcinogens is demonstrating that the biomarker is actually related to tobacco product exposure. The most reliable method of demonstrating this relationship is to assess levels of the biomarker after a research participant has stopped using the tobacco product. In one representative study, researchers assessed at various times (3, 7, 14, 21, 28, 42, and 56 days) the persistence of eight tobacco smoke carcinogens and toxicant biomarkers in the urine of 17 people who had stopped smoking. The biomarkers were metabolites of 1,3-butadiene, acrolein, crotonaldehyde, benzene, ethylene oxide, pyrene (a representative polycyclic aromatic hydrocarbon), and nicotine-derived nitrosamine ketone (NNK), a TSNA. These biomarkers, which are described in more detail below, include some of the major carcinogens and toxicants present in cigarette smoke. Levels of all these biomarkers—except for 1,3-butadiene metabolites (called dihydroxybutyl mercapturic acid)—decreased significantly after 3 days of smoking cessation (P <.001). The rates of decrease for most of the biomarkers were rapid, reaching nearly their ultimate levels (81-91 percent reduction) after 3 days, while that of the NNK metabolite (called 4-[methylnitrosamino]-1-[3-pyridyl]-1-butanol and its glucuronides [total NNAL]) was gradual, reaching a 92 percent reduction after 42 days. The decrease in the pyrene metabolite was variable among research participants, reaching about 50 percent of baseline, consistent with other common environmental sources of pyrene, such as diet. These results demonstrated that all biomarkers investigated in this study except dihydroxybutyl mercapturic acid were related to cigarette smoking (Carmella et al., 2009). A similar study carried out in smokeless tobacco users demonstrated the reduction of total NNAL after cessation of product use (Hecht, 2002).
Another method of validating tobacco carcinogen and toxicant biomarkers with respect to tobacco product exposure is to compare their levels in smokers and nonsmokers. Numerous studies of this type have been reported, and individual biomarkers are described in an upcoming section and presented in Table 3-1. Biomarkers of exposure of tobacco-specific compounds such as NNK, N-nitrosonornicotine (NNN), and nicotine are not found in non-tobacco users unless they have been exposed to secondhand tobacco smoke (see Table 3-1). Other biomarkers, such as those related to combustion products such as pyrene, are detected in both smokers and nonsmokers because they occur not only in tobacco products, but also in the diet and polluted air. Therefore, some of the ranges of values overlap between smokers and nonsmokers, as shown in Table 3-1. However, biomarker levels are consistently higher in smokers compared to those in nonsmokers in individual studies (Hecht et al., 2010). Biomarkers of the tobacco-specific compounds are similar in smokers and smokeless tobacco users, while those of some of the volatile organic combustion products are considerably lower in smokeless tobacco users (Hecht, 2002; Hecht et al., 2010).
Exposure to secondhand cigarette smoke can contribute to biomarker levels in nonsmokers, but the levels are generally small, about 1-5 percent of the levels in smokers (Hecht et al., 2010). Some biomarkers that are consistently elevated in nonsmokers exposed to secondhand tobacco smoke are cotinine, a major metabolite of nicotine, and NNAL and its glucuronides, metabolites of NNK (Hecht, 2002, 2003b; HHS, 2006). Cut points in these biomarkers for distinguishing light smokers from non-smokers exposed to secondhand smoke have been discussed (Goniewicz et al., 2011).
Validation with Respect to Disease Risk
One approach to determining the relationship of exposure biomarkers to disease risk is to carry out prospective epidemiologic studies, or cohort studies. In these studies, samples from healthy research participants are collected and stored, and demographic and lifestyle data are obtained using questionnaires. The participants are then followed for years, and eventually diseases such as cancers will occur in some of them. The stored samples from these research participants are retrieved, along with samples from appropriately matched controls that remain disease free, to form a nested case-control study. These samples can be analyzed for the biomarkers to determine their relationship to disease. The magnitude of the relationship to disease risk for each biomarker or their combinations can be evaluated using standard statistical analysis methods. Although there are certain limitations of this approach, which have been discussed
|Range of Recent Mean Values or Concentrations|
|Biomarker||Source||Smokers||Nonsmokers||References (smokers)||References (nonsmokers)|
|Nicotine equivalents||Nicotine||70.4-154 μmol/24hr||N/Ab||(Lowe et al., 2009; Mendes et al., 2008; Roethig et al., 2007, 2009; Scherer et al., 2006, 2007a; Zedler et al., 2006)|
|Total NNAL||NNK||1.1-2.9 nmol/24hr||N/Ab||(Carmella et al., 2009; Kavvadias et al., 2009b; Lowe et al., 2009; Melikian et al., 2007; Mendes et al., 2008; Roethig et al., 2009; Sarkar et al., 2008; Scherer et al., 2007a; Stepanov and Hecht, 2005)|
|Total NNN||NNN||0.049-0.24 nmol/24hr||N/Ab||(Kavvadias et al., 2009a, 2009b; Stepanov and Hecht, 2005; Stepanov et al., 2009)|
|1-HOP||Pyrene||0.50-1.45 nmol/24hr||0.18-0.50 nmol/24hr||(Carmella et al., 2009; Feng et al., 2006a; Mendes et al., 2008; Roethig et al., 2007, 2009; Sarkar et al., 2008; Scherer et al., 2007a; Suwan-ampai et al., 2009)||(Feng et al., 2006a; Roethig et al., 2007, 2009; Scherer et al., 2007a; Suwan-ampai et al., 2009)|
|MHBMA||1,3-butadiene||15.5-322 nmol/24hr||0.65-7.5 nmol/24hr||(Carmella et al., 2009; Roethig et al., 2009; Sarkar et al., 2008; Scherer et al., 2006)||(Carmella et al., 2009; Roethig et al., 2009; Sarkar et al., 2008)|
|SPMA||Benzene||3.2-32.1 nmol/24hr||0.17-3.14 nmol/24hr||(Carmella et al., 2009; Ding et al., 2009; Feng et al., 2006a; Mendes et al., 2008; Roethig et al., 2007; Sarkar et al., 2008; Scherer et al., 2006, 2007a)||(Carmella et al., 2009; Ding et al., 2009; Feng et al., 2006a; Roethig et al., 2007; Sarkar et al., 2008; Scherer et al., 2006, 2007a; Suwan-ampai et al., 2009)|
|HPMA||Acrolein||5,869- 11,190 nmol/24hr||1,131-1,847 nmol/24hr||(Carmella et al., 2009; Ding et al., 2009; Mendes et al., 2008; Roethig et al., 2007, 2009; Sarkar et al., 2008; Scherer et al., 2006, 2007a)||(Carmella et al., 2009; Roethig et al., 2007, 2009; Sarkar et al., 2008; Scherer et al., 2006, 2007a)|
|HBMA||Crotonaldehyde||9,800- 26,000 nmol/24hr||1,200-3,200 nmol/24hr||(Carmella et al., 2009; Scherer et al., 2006, 2007b)||(Carmella et al., 2009; Scherer et al., 2006, 2007b)|
|HEMA||Ethylene oxide||19.1-102 nmol/24hr||6.51-38.8 nmol/24hr||(Carmella et al., 2009; Ding et al., 2009)||(Carmella et al., 2009; Ding et al., 2009)|
|Cd||Cadmium||2.3-12.8 nmol/24hr||1.34-8.04 nmol/24hr||(Batariova et al., 2006; Hoffmann et al., 2000; McElroy et al., 2007; Paschal et al., 2000)||(Batariova et al., 2006; Hoffmann et al., 2000; McElroy et al., 2007; Paschal et al., 2000)|
|Cyanoethylvaline||Acrylonitrile||112±81 pmol/g globin||6.5±6.4 pmol/g globin||(Scherer, 2005; Scherer et al., 2007a)||(Scherer, 2005; Scherer et al., 2007a)|
|Range of Recent Mean Values or Concentrations|
|Biomarker||Source||Smokers||Nonsmokers||References (smokers)||References (nonsmokers)|
|Carbamoylethylvaline||Acrylamide||84.1±41.8 pmol/g globin||27.8±7.1 pmol/g globin||(Scherer, 2005; Scherer et al., 2007a)||(Scherer, 2005; Scherer et al., 2007a)|
|Hydroxyethylvaline||Ethylene oxide||132±92 pmol/g globin||21.1±12.7 pmol/g globin||(Scherer, 2005; Scherer et al., 2007a)||(Scherer, 2005; Scherer et al., 2007a)|
|4-aminobiphenyl-globin||4-aminobiphenyl||0.26±0.006d pmol/g globin||0.067±0.009d pmol/g globin||(Roethig et al., 2009; Scherer, 2005)||(Roethig et al., 2009; Scherer, 2005)|
|Exhaled CO||Carbon monoxide||17.4-34.4 ppm||2.6-6.5 ppm||(Scherer, 2006; Scherer et al., 2007a)||(Scherer, 2006; Scherer et al., 2007a)|
|Carboxyhemoglobin||Carbon monoxide||3.4-7.1%||0.35-1.45%||(Lowe et al., 2009; Roethig et al., 2009; Scherer, 2006; Scherer et al., 2007a)||(Lowe et al., 2009; Roethig et al., 2009; Scherer, 2006; Scherer et al., 2007a)|
(Rundle and Ahsan, 2008), such epidemiologic studies with prospective study design and objective measurements of biomarkers in biospecimens would provide a direct link of the disease of interest to the biomarker and its parent compound. The relationship of tobacco carcinogens and toxicant biomarkers to cancer and other diseases has been examined in only limited prospective studies to date. Examples are cotinine and total NNAL with respect to lung cancer. In one prospective study, serum cotinine was related linearly to lung cancer risk, with no suggestion of a plateau at high exposure levels (Boffetta et al., 2006). Two molecular epidemiologic studies related total NNAL to lung cancer risk. In the first study, researchers saw a dose-dependent association between urinary levels of total NNAL and risk of lung cancer (Yuan et al., 2009). In relation to the lowest quartile of total NNAL, the risk of lung cancer associated with the second and third tertiles were 1.43 (95% CI, 0.86-2.37) and 2.11 (95% CI, 1.25-3.54), respectively (P for trend = .005) after adjustment for number of cigarettes smoked per day, number of years of cigarette use, and total cotinine (cotinine plus its glucuronide). Smokers in the highest tertiles of total urinary NNAL and cotinine displayed an 8.5-fold increased risk for developing lung cancer as compared to smokers in the lowest tertiles of these measures but otherwise similar in smoking history. A second study also showed this association using prospective measurements of total NNAL in serum, although no relationship with cotinine was seen (Church et al., 2009). Prospective measures have also been used to evaluate the association between baseline cotinine and cardiovascular disease (Whincup et al., 2004).
Description of Some Widely Used Biomarkers of Exposure
This section provides greater discussion on the common biomarkers of exposure. “Nicotine equivalents,” the combination of nicotine, cotinine, 3’-hydroxycotinine, and their glucuronides, represent 73-96 percent of the nicotine levels delivered to a user of tobacco products (Hukkanen et al., 2005). This combination is a widely accepted biomarker of nicotine uptake that directly measures, to a high percentage, the nicotine dose.
Total NNAL, the sum of free and glucuronidated NNAL, and total NNN, the sum of free and glucuronidated NNN, are biomarkers of uptake of the carcinogenic tobacco-specific nitrosamines NNK and NNN (Hecht, 2008). NNK and NNN always occur together in tobacco products, and they are potent carcinogens in laboratory animals (IARC, 2007). Nicotine equivalents, total NNAL, and total NNN are unique biomarkers because of their tobacco specificity. They are only detected in people exposed to tobacco products or (for nicotine equivalents and occasionally NNN) in people who use nicotine replacement products
(Stepanov et al., 2009). As indicated in Table 3-1, the levels of these biomarkers in nonusers of tobacco exposed to secondhand tobacco smoke are generally considerably low compared to those observed in users of tobacco products.
1-HOP is a biomarker of exposure to polycyclic aromatic hydrocarbons, tobacco smoke particulate phase constituents, and products of incomplete combustion. These compounds are also commonly found in polluted air and the diet. Many polycyclic aromatic hydrocarbons are potent carcinogens in laboratory animals. The most widely studied polycyclic aromatic hydrocarbon carcinogen is benzo[a]pyrene (BaP). Polycyclic aromatic hydrocarbons always occur as mixtures, and 1-HOP, which is a metabolite of the noncarcinogen pyrene, an ever-present component of these mixtures, is a widely accepted biomarker of exposure to this class of compounds.
The mercapturic acids MHBMA, SPMA, HPMA, HBMA, and HEMA are biomarkers of the tobacco smoke gas phase constituents 1,3-butadiene, benzene, acrolein, crotonaldehyde, and ethylene oxide, respectively (Carmella et al., 2009). 1,3-butadiene, benzene, and ethylene oxide cause tumors in multiple organs of mice and rats (HHS, 2004b; IARC, 2008). Both acrolein and crotonaldehyde are associated with lipid peroxidation and perhaps inflammation (Chung et al., 1999; Thompson and Burcham, 2008). Acrolein reacts with the p53 gene at codons associated with lung cancer, a phenomenon also observed in studies of polycyclic aromatic hydrocarbon diol epoxide metabolites (Feng et al., 2006b). Acrolein is an intense irritant and cilia-toxic compound (IARC, 1995). Acrylonitrile, acrylamide, and 4-aminobiphenyl are also well-established carcinogens (HHS, 2004b; IARC, 1987, 1999b; Klaunig, 2008).
CO competes with oxygen for binding to hemoglobin and hinders the ability of oxygen to be released from hemoglobin. Although smokers are unlikely to experience acute CO-related symptoms (Scherer, 2006), CO is believed to impair oxygen delivery and cause complications of atherosclerosis and other cardiovascular diseases in smokers (HHS, 2004a).
Among the compounds related to the biomarkers, NNK and NNN, BaP, 1,3-butadiene, benzene, ethylene oxide, cadmium, and 4-aminobiphenyl are considered “carcinogenic to humans” by the International Agency for Research on Cancer (IARC, 1987, 1999a, 2006, 2007, 2008; Straif et al., 2005) and potentially may be involved in causing different types of cancer in tobacco users (Hecht, 1999, 2003a, 2010). Many of these compounds also have considerable toxic effects. Additionally, NNK, NNN, BaP, 1,3-butadiene, benzene, acrolein, acetaldehyde, formaldehyde, and CO were recommended for regulation under the World Health Organization’s Framework Convention on Tobacco Control (Burns et al., 2008).
These and other widely used biomarkers of exposure are presented
in Table 3-1, which presents urinary biomarkers, hemoglobin adduct biomarkers, and others. Recent data on the range of values for these biomarkers of exposure are given for both smokers and nonsmokers. Although the ranges of values for smokers and nonsmokers overlap for certain biomarkers in the table, biomarker levels are consistently elevated in smokers in the individual studies referenced in the table.
Examples of Other Biomarkers
Examples of some other exposure biomarkers include urinary or plasma phenanthrene tetraol and phenanthrols (Church et al., 2010; Hecht et al., 2005), 3-hydroxybenzo[a]pyrene and BaP tetraols (Forster et al., 2008; Zhong et al., 2010), and hydroxyfluorenes (Jacob et al., 2007) for polycyclic aromatic hydrocarbons; DNA adducts of various compounds in white cells and various tissues (Phillips, 2002); 3-ethyladenine in urine (Feng et al., 2006a); and 2-cyanoethylmercapturic acid in urine for acrylonitrile (Scherer et al., 2010).
A group of biomarkers related to inflammation, oxidative stress, and other conditions that could be influenced by tobacco products have been termed “biomarkers of potential harm” by authors from Altria, and these might be considered as risk biomarkers. Some of these, such as markers of oxidative damage, straddle the border between exposure and effect markers because they are caused by exposure to tobacco products but do not directly result from a known measured constituent of these products. One example is 8-epi-prostaglandin F2a, an established biomarker of oxidative damage, which is significantly higher in smokers than in nonsmokers (Frost-Pineda et al., 2011).
“Biomarkers of potential harm” also include those biomarkers related to inflammation (such as white blood cell count, high-sensitivity C-reactive protein [CRP], fibrinogen, and von Willebrand’s factor) and platelet activation (such as 11-dehydrothromboxane B2) as well as triglycerides and alkaline phosphatase, all of which were significantly elevated in smokers, including in one recent study that examined the relationship between these biomarkers, machine-measured tar yields, and biomarkers of exposure to cigarette smoke constituents in more than 3,500 smokers and more than 1,000 nonsmokers (Frost-Pineda et al., 2011; Liu et al., 2011). Body mass index, smoking duration, cigarette tar category, and some biomarkers of exposure were significant factors in multiple regression models for the biomarkers of potential harm. Body mass index was the highest ranking factor in the models for white blood cell count, high-sensitivity CRP, fibrinogen, and 8-epi-prostaglandin F2a, while gender and smoking duration influenced 11-dehydrothromboxane B2 and von Willebrand’s factor. Overall, the relationship between cigarette smoking, biomarkers of
exposure, other factors, and these biomarkers of potential harm was quite complex (Liu et al., 2011).
Analysis of spent filters has also been used to estimate exposure. Examples include the measurement of solanesol, nicotine, NNK, and acrolein in filters. In some studies, these measurements correlated with urinary exposure biomarkers (Mariner et al., 2010; Morin et al., 2010; Pauly et al., 2009).
Examples of Biomarker Application in Product Evaluation
Exposure biomarkers are useful in evaluating new products that, according to laboratory analyses, have lower levels of certain constituents. Studies of this type have been reviewed (Hatsukami et al., 2007b). Some typical results are presented here.
Omni cigarettes were advertised as having reduced carcinogens, including nitrosamines and polycyclic aromatic hydrocarbons. Decreases of 53 percent in levels of NNK and 20 percent in levels of pyrene in smoke were advertised, based on machine measurements. Smokers were randomized to use either the Omni cigarette or medicinal nicotine, and exposure biomarkers were assessed for a 4-week period. The reductions in total NNAL were only 21 percent in those who used the Omni cigarette compared to baseline levels with their usual brand, while there was no significant reduction in 1-HOP (Hatsukami et al., 2004).
Quest cigarettes were available with deliveries of 0.3 mg nicotine per cigarette or 0.05 mg nicotine per cigarette. When smokers switched from their customary brand to the 0.05 mg nicotine yield cigarette for 6 weeks, they experienced significant reductions in cotinine (96 percent), total NNAL (78 percent), total NNN (67 percent), 1-HOP (36 percent), HPMA (56 percent), and SPMA (69 percent). In addition to these reductions, the 0.05 mg cigarette was associated with relief from withdrawal symptoms from the users’ usual cigarette (Hatsukami et al., 2010).
In a 4-week study of smokeless tobacco users who switched from their usual conventional brand of smokeless tobacco to either Swedish Snus or the nicotine patch, total NNAL levels decreased significantly, although the overall mean total NNAL level was significantly lower for research participants who switched to the nicotine patch than for research participants who switched to snus. These results are consistent with the lower levels of NNK in Swedish Snus than in conventional moist snuff products available in the United States (Hatsukami et al., 2004).
In a recent study, smokers were randomized to receive the smokeless tobacco products Camel Snus, Taboka, or medicinal nicotine over a 4-week period in which they quit smoking. Significant reductions of exhaled CO, urinary cotinine, and total NNAL were observed in all groups. A significant
reduction of total NNN was also observed in the treatment groups, except for the Camel Snus group. Total NNAL levels were greater in the Camel Snus group than in those who used medicinal nicotine (Kotlyar et al., 2011). These results reflect the lower levels of NNK and NNN in these products compared to the amounts delivered in cigarette smoke.
Cross-sectional studies of biomarkers and product use have also been reported. In one study, data from the U.S. National Health and Nutrition Examination Survey (1999-2008) were used to evaluate levels of biomarkers of a variety of toxicants and carcinogens in smokers compared to smokeless tobacco users. Smokeless tobacco users had higher levels of several polycyclic aromatic hydrocarbon biomarkers, as well as higher levels of total NNAL, than did nonusers of tobacco. Of 33 biomarkers analyzed, 18 were significantly lower in smokeless tobacco users than in smokers, while 10 of the 33 biomarkers were not different. The levels of the other five biomarkers, including total NNAL, were higher in smokeless tobacco users than in smokers (Naufal et al., 2011).
In summary, biomarkers can provide a more realistic assessment of the consumer’s exposure to carcinogens and toxicants in tobacco products than simple analyses of the products because laboratory analyses cannot fully duplicate human use conditions. In most cases, the general trend of laboratory results is reflected in the biomarker data.
Summary of Biomarkers of Exposure
Validated tobacco carcinogen and toxicant biomarkers of exposure for a variety of compounds are now available. Measurement of a panel of these biomarkers in an appropriately conducted study can provide a realistic assessment of human uptake of a variety of toxicants and carcinogens in tobacco products. Many studies of this type show a relationship between product constituent levels and biomarker levels, but the relationship is not always straightforward. If the panel of biomarkers presented were decreased to the levels found in nonsmokers, it is likely that there would be a beneficial effect on health, but this has not been proven. Some tobacco carcinogen and toxicant biomarkers such as cotinine and total NNAL have been related to cancer risk in molecular epidemiologic studies, but most of the biomarkers discussed here would still be best described as exposure biomarkers, pending the availability of more data.
In summary, the evaluation of new products would always include standard laboratory analyses of constituents as a first step. Whether differences in constituent levels translate to differences in exposure to tobacco carcinogen and toxicant biomarkers requires testing in an appropriately designed clinical study.
Although many studies have shown a relationship between individual
constituents of tobacco products and chronic diseases, there is no proof that any individual constituent or group of constituents is responsible for a given disease. Therefore, it is possible that constituents that play a decisive role in disease causation are simply not being measured, or that there are interactive effects among constituents that are critical in disease etiology but are not taken into account in the analyses. There may also be interactions between particular constituents and biological processes such as inflammation that are not fully captured by biomarker analyses. There are also limited dose response data relating constituents such as TSNAs, polycyclic aromatic hydrocarbons, volatile organic compounds, or heavy metals to specific diseases, and therefore reductions in the levels of a particular chemical or class of chemicals cannot be reliably generalized to a reduction in disease.
A particularly important question is whether a given measurement has evolved from being a “biomarker of exposure” to a “biomarker of risk” or “surrogate endpoint for disease.” The committee recognizes that this question could be critical in the design of studies on MRTPs. For example, studies on smokeless tobacco products would produce significantly lower biomarkers of volatile combustion products (such as CO, acrolein, or benzene) than studies on combusted products because smokeless tobacco products do not deliver significant quantities of these materials. Epidemiologic studies demonstrate that the risk for lung cancer is higher in smokers than in smokeless tobacco users. Furthermore, when smokers stop smoking, their risk for lung cancer gradually decreases over a period of years. Based only on these facts, one might propose for example that exhaled CO is a biomarker of risk because it would clearly decrease when one stopped smoking and presumably when a smoker switched to smokeless tobacco. But there is no biological rationale for this observation, because CO is not known to be involved as a causative agent for lung cancer. Therefore, the committee believes that, for a biomarker of exposure to be accepted as a biomarker of risk or a surrogate endpoint for disease, there should be a strong biological rationale as well as compelling data from clinical or epidemiologic studies. Presently, there are only limited data on the relationship of exposure biomarkers to chronic disease.
There is no standard approved design for clinical trials in which new products would be evaluated with respect to biomarker outcomes. This topic has been reviewed recently, and further studies are required (Hatsukami et al., 2009).
Biomarkers of Risk
The validity of a study that uses a biomarker of risk is only as good as the validity of the biomarker. The utility of biomarkers of risk ultimately
hangs on the assumption that they not only correlate to the clinical endpoint of interest, but also that the biomarker will fully capture the complete effect of an intervention on the clinical endpoint (Prentice, 1989).
Biomarkers of risk can include blood, other bodily fluid, or tissue markers and risk factors that relate to the natural history and progression of specific diseases and conditions. However, they cannot be considered as markers of disease occurrence on their own. Further, a single biomarker could be a predictor of many diverse conditions, such as markers of systemic inflammation or other immune system dysfunction (e.g., cytokines or CRP, blood immunoglobulin A levels or eosinophil counts). Another example of a biomarker that could predict many conditions is high levels of oxidative stress. Other biomarkers are applied to particular conditions such as cardiovascular disease (e.g., high-density lipoprotein or low-density lipoprotein cholesterol) or adult-onset diabetes (e.g., glucose intolerance, intermediate fasting blood glucose levels, glycosylated hemoglobin levels), but again, they are not indicators of the disease per se. Some biomarkers of disease can be extremely complex at the cellular or molecular level, such as rates of nuclear DNA repair, which can predict the occurrence of various cancers or other systemic conditions, but do not necessarily indicate disease presence.
Although more speculative, another related issue relevant to future use of biomarkers is the concomitant use of pharmacological (“chemopreventive”) interventions that may be used to prevent cancer or other conditions. There are currently, for example, several candidate pharmacological interventions for human cancer chemoprevention. These are based mostly on basic research, but are also subject to human testing, including certain vitamins, resveratrol, polyamines, and flavanoids. Although none of these are fully proven in humans at the present time, in the future it is conceivable that as these products emerge as proven preventive entities, tobacco products that contain some of these agents may emerge, and complicate the regulation of health claims. As is currently the case, some of these may be marketed as dietary supplements or under the umbrella of the “nutraceuticals” movement. Regulators should be alert to the emergence of such combination products, and refer to Section 201 (rr)(4) of the Federal Food, Drug, and Cosmetic Act (as amended by the FSPTCA), which states that a “tobacco product shall not be marketed in combination with any other article or product regulated under this Act (including a drug, biologic, food, cosmetic, medical device, or a dietary supplement).”2
2 Family Smoking Prevention and Tobacco Control Act of 2009, Public Law 111-31, 123 Stat. 1776 (June 22, 2009).
Surrogate Endpoints in the Study of Disease Outcomes
Surrogate endpoints are a set of predisease measures that are not clinically overt conditions but nonetheless represent nascent or early pathological processes for many subsequent clinical conditions. The presumption, with varying amount of evidence, is that some portion of these early processes progress over time to produce overt clinical illness. For many years, there has been substantial concern about adopting surrogate endpoints as the sole measure of therapeutic efficacy in clinical trials, particularly because there are very important counterexamples in the history of drug regulation where surrogate endpoint control did not lead to disease prevention or amelioration; such intermediate endpoints included blood pressure control, antiarrhythmic treatments, and cholesterol-lowering agents.
The standards for using biomarkers of risk as surrogate endpoints are even more stringent as “the surrogate endpoints should be a perfect proxy for the effect of an intervention on the recipient’s risk of important clinical outcomes” (IOM, 2010). It is not uncommon, however, for potential surrogate endpoints to fail to predict clinical outcomes. According to Fleming and DeMets (1996), such failures often occur because: (1) the surrogate endpoint does not affect the same pathophysiologic pathway that leads to the clinical outcome of interest; (2) there are multiple causal pathways linked to a particular clinical outcome, but the intervention in question only affects one pathway mediated through the surrogate among several causal pathways linked to the disease; (3) the surrogate under study is insensitive to or is not a part of the causal pathway of the intervention’s effect, or is insensitive to its effect; or (4) the intervention results in additional mechanisms of action independent of the disease process. The pharmacologic suppression of ventricular arrhythmias in the postmyocardial infarction setting well illustrates this. Premature ventricular contractions in the presence of ongoing myocardial damage or ischemia confer a poor prognosis, and it was thought that the pharmacologic suppression of these would result in clinical benefit. In fact, however, the Cardiac Arrhythmia Suppression Trial demonstrated that the suppression was harmful (CAST II Investigators, 1992).
The ideal setting for the use of a surrogate endpoint is when the surrogate endpoint lies along the only causal pathway of the clinical endpoint’s process, and the surrogate’s effect mediates the intervention’s entire effect on the clinical outcome. Ideally, one should have a thorough understanding of the disease process and causal pathways, as well as a deep appreciation of the intervention’s mechanisms of action. Admittedly, that is unlikely to occur with MRTPs that have a multiplicity of biological and organ system effects. It is important, therefore, to validate a surrogate used to assess the health effects of MRTPs. To be validated,
it is essential that its effect be simultaneously, prospectively, and directly assessed against the desired clinical endpoint. In 2010, the IOM published a comprehensive report on the evaluation of biomarkers and surrogates, Evaluation of Biomarkers and Surrogate Endpoints in Chronic Disease; the committee refers the reader to that report for a detailed description of standards for the evaluation of biomarkers and surrogates. Appendix B presents the framework developed by that committee.
Because most chronic disease progression occurs over a pathogenic continuum, it is possible that in some circumstances surrogate endpoints that are close to overt clinical illness may have value in MRTP assessment. There may be instances where endpoints, such as coronary calcification levels or abnormal bone architecture and density, may be adequate to be considered in product evaluation studies. Endpoints will require a thorough evidence review and explicit specification, and could possibly improve the MRTP evaluation process. Furthermore, some of the outcomes can only be obtained with invasive procedures, and may not be suitable for all research studies.
It should be noted that with respect to reflecting true disease outcomes, biomarkers have been controversial. In general, because there have been many documented instances where pharmacological alteration of biomarker levels has not led to disease progression in the predicted direction, biomarkers have received limited credibility as disease endpoints (Hatsukami et al., 2006; Hecht et al., 2010). In general, they are not acceptable alternatives to true disease endpoints.
Preclinical assessment is an established step in the evaluation of any new product. In the case of a potential MRTP, the first step would be the analysis of harmful and potentially harmful constituents, as discussed in previous sections. This would be followed by in vitro toxicity and genetic toxicology tests in bacterial and mammalian systems. In these tests, extracts or fractions of the MRTP would be compared to standard conventional products. Although all in vitro tests have limitations, the collective results can nevertheless provide potentially useful information. If the results of these tests signaled decreased activity compared to a standard conventional product, the evaluation would proceed to the next stage of studies with laboratory animals. The potential MRTP would again be compared to a standard conventional product using a suitable animal model system. Finally, the evaluation would proceed to short-term genetic toxicology tests in people who used the new or conventional product. The choice of the comparison product in all of these studies is clearly important. Generally, initial comparisons should be between products of
the same class, either combusted or noncombusted. The committee discusses the selection of comparison products further in Chapter 6.
Preclinical studies of the effects of smokeless tobacco products and combusted tobacco products are discussed below.
Smokeless Tobacco Products
Reviews of in vitro assays (Johnson et al., 2009) and animal models (IARC, 1985, 2004, 2007; Secretan et al., 2009) for the evaluation of smokeless tobacco or smokeless tobacco extracts have been published. Table 3-2 summarizes preclinical studies for the evaluation of harms from smokeless tobacco products.
In Vitro Studies
In vitro laboratory assays include the Ames test, and tests on cytotoxicity, proliferation, and programmed cell death (apoptosis); these tests provide routine tandem toxicology analyses. Mutation induction by Salmonella typhimurium in the Ames test or toxicologic effects noted in various human or animal cells are evaluated after exposure to smokeless tobacco extracts. As depicted in Table 3-2, smokeless tobacco extracts are a product of physical (e.g., grinding, freeze drying) or chemical methods (e.g., organics: dimethylsulfoxide, methylene chloride, methanol, acetone, ethanol; or inorganics: buffered salt solutions [Hanks, phosphate buffered saline, saline], and water or artificial saliva) (Bernzweig et al., 1998; Lindemann and Park, 1988; Merne et al., 2004; Rohatgi et al., 2005; Shirname-More, 1991; Yildiz et al., 1999).
Further assessment of genotoxic activities of smokeless tobacco extracts requires standardization not only for the method of extraction, but also for levels of moisture and humectant content.
With widespread use of these aforementioned cell assays, molecular expression patterns for epithelial and mesenchyme cells are also routinely determined using multispectral cytometric instruments. However, this approach should not ignore normal physical adherence characteristics of experimental cell targets such as epithelial and mesenchymal cells that adhere to tissue culture surfaces; false positive or negative results may be obtained in comparison to nonadherent immune cells that are also examined using flow cytometry. Furthermore, introduction of new smokeless tobacco products will require incorporation of additional cell laboratory designs to evaluate genotoxic potential.
An important cell culture design improvement is a raft three-dimension assay. This design uses mimicry of human mucosa structure to assess genotoxic responses. In addition, commercial molecular kits are available
|Type of Assay||Summary of Results||References for Assay|
|In Vitto Assays|
S. typhimurium mutagenesis
S9 (+/-); strains
Base change mutations
(rfa, uvrB, pkM101)
|Some carcinogens and ST products have produced a range of results due to experience and spontaneous revision of mutations in S. typhimurium strains and intralaboratory variability.
In general STE use demonstrated a positive Ames test but use of S9, dual strains, and base change mutations are infrequent but they add to specificity
|(Hakura et al., 2005; Jansson et al., 1991; Johnson et al., 2009; Niphadkar et al., 1996; Stamm et al., 1994; Whong et al., 1984, 1985)|
|Cell Assays||An increase in cell assays that evaluate genotoxicity as determined by cytotoxicity, proliferation, and apoptosis has shown STE to function in a dose- and time-dependent manner in some cell targets.
Variables in cell number, cell density, time of incubations, and form of STE create variability.
A standardization of these features will aid in determinations. Critical for evaluation is the origin of the cell used and the type of cell: primary, immortalized, or transformed/immortalized.
|Cytotoxicity||This assay uses trypan blue dye exclusion, tertrazolium salt assays (MTT), nuclear identification using DAPI staining, and propidium iodide nuclear identification.
Extraction methods also affect results in a dose-dependent manner to increase cytotoxicity with DMSO compared to aqueous extractions (DMEM, H20, PBS, artificial saliva).
Number of ST products tested included American moist snuffs, a commercial Swedish moist snuff, and 11 ST products. Other ST products used were Kentucky moist ST, a loose-leaf form of ST, and reference 1S3.
|(Rickert et al., 2008)|
|IKentucky moist ST extracted using PBS produced a time-dependent response.
Assays used included trypan blue dye exclusion and a MTT assay with reduction in tetrazolium salt to formazan dye.
|(Bagchi et al., 2001, 2002)|
Human cell lines
|The absence of normal cells and the use of only transformed or
immortalized cells will not permit a cytotoxic association with a strong
relationship to human exposure to ST. It is unfortunate that there is only
limited analysis of ST products to determine validity of responses.
|(Copp et al., 2009; Gregory and Gfell, 1996; Rohatgi et al., 2005; Shirname-More, 1991)|
Rodent cell lines:
||Cytotoxicity assays for rodent cell lines used a concentration and time
dependence to assess activity. However, except for limited use of oral
epithelial cells, no other primary cells were examined. The other cell
targets are either immortalized or transformed. These assays used DAPI
staining, a nuclear fluorescent dye; lactic dehydrogenase release as a
function of membrane integrity; neutral red stain to observe Jysosomes;
tetrazolium salt to formazan to determine mitochondrion viability (MTT/MTS);
and Trypan dye exclusion to test for viability.
|(Bagchi et al., 1996; Hasseus et al., 1997; Mangipudy and Vishwanatha, 1999; Muns et al., 1994; Rickert et al., 2008; Yildiz et al., 1999)|
|These studies used the proliferation marker BrdU, 3HdT, or colony
formation counts. MTT/MTS assays were also used. Exposures to assess
this assay used STE prepared from HBSS or DMEM. The extract was
derived from Kentucky loose-leaf ST, dry and moist snuffs, Kentucky
reference chewing tobacco, commercial Swedish snuff, or commercial
khaini. The results showed a dose- and time-dependent response for
normal human oral epithelial cells, but other cells produced no effect
as noted for hamster carcinoma cells, inhibition for rat cells, or variable
responses for low versus high doses for human oral keratinocytes and fibroblasts.
An extract of dichloromethane showed a reduction in proliferation of mouse tongue epithelial cells.
|(Bagchi et al., 2001; Gijare et al., 1989; Hasseus et al., 1997; Wang et al., 2001; Yildiz et al., 1999)|
|Type of Assay||Summary of Results||References for Assay|
• Human cells
• Normal human oral epithelial cells
• Rodent cell lines
• Golden Syrian
• Rat or mouse
|Programmed cell death distinguishes cell death from an internal cell death and an inflammatory imposed cell death. Assays assess membrane blebbing; flipping of membrane phosphatidyl-serine, using Annexin V, which is a calcium-binding protein; activated receptors, TRADD or Fas; cytosol cysteine proteases designated caspases; or nuclear fragmentation by denoting nucleosome formation as noted by 180 to 200bp band ladder in an electrophoretic gel. Nuclear TUNEL of cleaved ends of DNA is another routinely used marker.
Assays used a DMEM or PBS extraction and exposed cells to reference STs (2S1,2S3).
A positive apoptotic result requires at least two assays, which has not been uniformly performed. Furthermore, STE produces apoptosis in normal human oral keratinocytes and hamster epidermoid carcinoma cells in a dose dependent manner.
|(Bagchi et al., 1999; Banerjee
et al., 2007; Mangipudy and
|Other Genotoxic Assays|
|Chromosome Aberrations Sister Chromatid Exchange Micronuclei||The number and types of cell targets have been limited, and these assays are not as often reported.||(Jaju et al., 1992; Jansson et al., 1991; Patel et al., 1994; Trivedi et al., 1993)|
|Preclinical models use either ST or STE, but they are not applied at the frequency humans with which use an ST product per day. They are conducted on hamsters, rats, and mice to produce a few tumors on an inconsistent basis. Tumors are noted in the oral cavity, lip canal, and forestomach but not in a significant frequency. There is a need to evaluate levels of carcinogen before placement and after placement of ST product into animals. There is also a lack of examination of oral tissues for pathology changes similar to human users of ST. There is a concern that continual placement of ST will produce false positive results because of local irritation. Therefore improvement in preclinical animal evaluations of ST harms rests upon use of various tissues.||(Grasso and Mann, 1998; IARC, 2007; Johansson et al., 1989, 1991b)|
|Hamster Buccal Pouch Studies||The hamster buccal pouch permits placement of ST in close contact with mucosa. This tissue lacks lymphatic drainage and robust T lymphocyte responses, but Langerhan cell numbers can be increased as integrity of mucosa is reduced. ST either by placement, ligation, or with bees wax plugs for up to 2 years produced no tumors. Carcinogens, polycyclic aromatic hydrocarbon, 7,12 DMBA, MCA, or quinone, exemplified by 4-nitroquinoline-N-oxide or virus (herpes simplex virus), were required to facilitate tumorigenesis of ST. TSNA injections similar to rats were also used to produce tumors in nasal, lung, trachea, and adrenal tissues.||(Ashrafi et al., 1992; Correa et al., 1990; Dunham et al., 1966, 1975; Herrold and Dunham, 1962; Hoffmann et al., 1981; Homburger, 1971; Jorquera et al., 1992; Papageorge et al., 1996; Park et al., 1986; Peacock and Brawley, 1959; Peacock et al., 1960; Schuller et al., 1993, 1994; Schwartz and Gu, 2002; Shklar et al., 1985; Worawongvasu et al., 1991)|
|Type of Assay||Summary of Results||References for Assay|
|Rat Lip Canal Studies||ST and STE has been placed into surgically formed lip canals, swabbed into the oral cavity, and placed into the rat diet. Changes in mucosa are recorded (e.g., hyperplasia, hyperothokeratosis, hyperchromatic nuclei, nuclear and cytoplasmic reversal, and mitotic figures) with stroma fibrosis and hyperplasia of forestomach. Small and inconsistent numbers of tumors are noted. Exposure levels of TSNA derived from ST in the saliva are generally higher than a single application of ST by humans (e.g., 11-55 ppm compared to 0.2 ppm). The concentration of ST used in studies is five times the exposure by humans from a single application. ST products with a high TSNA produce more mucosa pathology than ST products with low TSNA levels. Combined exposure of ST with 4-NQO increased tumorigenesis, but HSV-1 inoculation did not increase tumors with snuff administration.||(Hirsch and Johansson, 1983; Hirsch and Thilander, 1981; Hirsch et al., 1984; Hoffmann and Adams, 1981; Hoffmann et al., 1992; Johansson et al., 1989, 1991a, 1991b; Larsson et al., 1989; Palladino et al., 1986; Schwartz et al., 2010)|
|Rat Dietary Studies||Dietary incorporation with TSNA produced several different types of malignant tumors: lung, liver, pancreas, prostate, mammary glands, leukemia, and lymphoma. Dried snuff fed to rats or mice produced a tumor of the kidney, and one rat and three mice developed leukemia, but these latter neoplasias are likely spontaneous||(DiPaolo, 1962; Hecht et al., 1986)|
|Mouse Dietary Studies||Dietary exposure of ST or injection of TSNA in mice or transgenic mice (INS-GAS/FVB) produced respectively chronic pancreatitis, or lung adenomas.||(IARC, 2007; Stenstrom et al., 2007)|
to facilitate examination for genotoxic change among target cells. Some of these kits permit tagging and identification of chemical substances in cell sites, silencing of specific RNAs, transfection of genetic material to modify specific cellular pathways, or immortalization of epithelial cells, which facilitates cell culture growth (Andrei, 2006; Andrei et al., 2010; Singh and Nalwa, 2011).
A consistency of cell number, type, and differentiation of cell type is achievable. However, specific attention to cell features of primary cells in comparison to immortalized (or transformed immortalized, malignant cell lines) is suggested. It is also a practical conclusion that persistent genotoxic cell harm will result in a redesign of the smokeless tobacco product.
Furthermore, it is also expected that assays will address loss of normal cell physiology as reflected in regards to not only cancer, but also infection, inflammation, respiratory, or cardiovascular disease processes. These latter pathologies are often neglected in cell studies, but they are suspected to be indirect targets for ST-derived substances.
Animal models for the evaluation of harm from smokeless tobacco or smokeless tobacco extracts products have included Syrian hamster buccal pouch; various strains of rats that are exposed through the diet; or various strains of rats that have undergone surgery to produce a lip canal that allows placement of smokeless tobacco into a tube of mucosa (e.g., F344, Sprague Dawley, Wistar, and SD). Dietary exposure among transgenic mice has also been reported.
A concentration and use pattern consistent with human exposure to smokeless tobacco products should be employed in animal models, but this has not been achieved. Previous studies used smokeless tobacco extracts or derivative concentrations several fold above the single self-administered exposure by humans (Hoffmann and Adams, 1981; Palladino et al., 1986). Moreover, smokeless tobacco and smokeless tobacco extracts handling and storage under carefully controlled conditions are required to prevent inappropriate formation of TSNAs (Brunnemann et al., 2002; Djordjevic et al., 1993).
At best, animal models mimic human tissue responses. However, in our present situation with the introduction of new spit-less smokeless tobacco products (e.g., additives flavorings), there is an increased difficulty to achieve this goal and evaluate chemical and biologic interplay in animals. Furthermore, attention needs to be focused upon direct contact of pathology sites in the oral cavity, gingiva/periodontum, and in nondirect contact disease tissues in respiratory and cardiovascular sites,
which have been reported to be under the oral tissue’s influence (Fisher et al., 2005; Ismail et al., 1983).
Animal models provide avenues to assess direct tissue damage. Additionally, animal models also offer opportunities to determine—prior to tumor induction in the oral cavity—infection, inflammation, or major organ damages in locations other than the site of smokeless tobacco application.
Epithelial oral pathology changes, benign tumors, and malignant tumors are observed in animal models and humans after exposures to smokeless tobacco or smokeless tobacco extracts products under various study conditions. However, under identical experimental conditions, not every study produced tumors (described in Table 3-2). In response to this observation and to enhance tumorigenesis, a combination of smokeless tobacco exposure with chemical promoters (in comparison to only smokeless tobacco exposures) was used to produce more local and distant tumors. A reevaluation of smokeless tobacco with promoters is still required to include human carcinogens such as BaP or viral infection patterns similar to human exposure.
It is also recognized that differences between animal species as reflected by liver microsome activity, cytochrome P450 expressions, or disposition of smokeless tobacco- or smokeless tobacco extract-derived substances result in a variability of tumor induction. However, a consistent tissue response trend is expected to determine genotoxicity or tissue harm (Leslie et al., 2007; Wu et al., 2002).
Persistent observed formation of tumors or pathologies associated with increased infection, inflammation, or respiratory or cardiovascular harm will be causes for redesign of smokeless tobacco products.
A number of in vitro and animal studies have investigated the effects of combusted tobacco products on cancer- and noncancer-related endpoints. Table 3-3 summarizes the key models used in those studies.
Evaluation of Oxidative and Nitrosative Stress Oxidative and nitrosative stress is produced by combustible tobacco products, and these reactive oxygen species and reactive nitrogen species (ROS/RNS) lead to modifications of DNA, proteins, and lipids. Extract of combustible products can be made by collecting the particles or passing the smoke through a saline solution. Detection of some individual ROS/RNS components can be measured directly, such as superoxide (luminol, dihydroethidium)
|Type of Assay||Summary of Results||References for Assay|
|Cell Based Assays|
|Oxidative/Nitrosative Stress||CSE has been shown to induce ROS/RNS in cell culture assays. CSE increases superoxide and nitric oxide generation in cells and also increases oxidative modifications of macromolecules (8-hydroxy-deoxyguanosine, 4-hydroxynonenal, 3-nitrotyrosine), which can be detected by antibody-based methods. Additionally, the antioxidant activity (i.e., glutathione), which can be measured by colorimetric reactions, is reduced after CSE exposure, resulting in enhanced expression of antioxidant genes, which can be assessed by quantitative PCR.||(Bertram et al., 2009; Bond et al., 1989; Malhotra et al., 2008; Peluffo et al., 2009; Sussan et al., 2009)|
|Inflammation||Inflammation is assessed in cell-based models via expression of cytokines (measured by ELISA and Western blot) and proinflammatory signaling cascade proteins (e.g., NF-kB, MAPK).|
|Epithelial cells||CSE has been shown to induce expression of proinflammatory cytokines IL-8 and MCP-1.||(Starrett and Blake, 2011)|
|Smooth muscle cells||CSE has been shown to induce secretion of IL-8, eotaxin, and VEGF-A in smooth muscle cells.||(Baarsma et al., 2011)|
|Inflammatory cells||Dendritic cells and monocytes secrete cytokines (measured by ELISA and Western blot) and up-regulate NF-kB (measured by DNA-binding activity assay).||(Mortaz et al., 2009; Zhou et al., 2011)|
|Mucus Production||CSE results in increased mucin synthesis (antibody-based) and glycoprotein secretion in tracheal epithelial cells.||(Lin et al., 1989; Wu et al., 1991)|
|Biphasic culture||Mucus-secreting epithelial cells are maintained in an air-liquid interface to mimic in vivo conditions.||(Whitcutt et al., 1988)|
|Endothelial Activation||CSE leads to enhanced expression of cell adhesion proteins, such as ICAM, VCAM, and E-selectin, and coagulating factors such as thrombomodulin and von Willebrand’s factor, which are measured by ELISA.||(Chen et al., 2004, 2009; Furie et al., 2000; Guarino et al., 2011)|
|Type of Assay||Summary of Results||References for Assay|
|Monocyte adhesion||Activated endothelial cells adhere to monocytes, which can be assessed by coculture experiments in which monocytes are added to a monolayer of endothelial cells.||(Reilly et al., 2004)|
|Inflammation and Emphysema|
|Whole-body exposure||This CS exposure system subjects the entire animal to CS for a duration of several hours per day. In this model, CS induces oxidative stress, inflammation, apoptosis, and airspace enlargement.||(Clauss et al., 2011; Ma et al., 2005; Sussan et al., 2009; Yoshida et al., 2010)|
|Nose-only exposure||This exposure model requires the placement of the rodent’s nose in a small chamber for relatively short periods of time. This exposure results in a relatively potent exposure that is shorter in duration than the whole-body exposure. This CS exposure model results in oxidative stress, inflammation, apoptosis, and airspace enlargement comparable to the whole-body exposure.||(Churg et al., 2009; Hautamaki et al., 1997)|
|Innate Immune Function|
|Bacterial infection||Exposure to CS results in innate immune dysfunction, resulting in increased bacterial exacerbations. CS increases inflammation and decreases bacterial clearance and killing in response to P. aeruginosa, H. influenzae, S. aureus, and others.||(Harvey et al., 2011; Huvenne et al., 2011)|
|Viral infection||CS exposure heightens the inflammatory responses in lungs to influenza H1N1 and rhinoviruses. Additionally, CS enhances viral-induced emphysema caused by the viral PAMP poly(I:C).||(Bauer et al., 2010; Kang et al., 2008; Mallia et al., 2011)|
or nitric oxide (2,3-diaminonapthalene) (Bertram et al., 2009; Peluffo et al., 2009). Additionally, many of the oxidative modifications to macromolecules can also be detected, including oxidatively modified DNA (8-hydroxy-deoxyguanosine) (Bond et al., 1989), lipids (malodialdehyde, 4-hydroxynonenal), and proteins (3-nitrotyrosine). Mammalian cells contain large concentrations of the antioxidant glutathione, which scavenges ROS/RNS, resulting in oxidation of glutathione. The ratio of reduced/oxidized glutathione can be quantified to determine the antioxidant capacity of the cells (Sussan et al., 2009). Cells that are undergoing oxidative stress have a decline in their pool of reduced glutathione and an increase in oxidized glutathione. Furthermore, cells respond to oxidative stress by up-regulating a large number of stress response antioxidant and phase II detoxification genes that are aimed at removing the stress and restoring homeostatic glutathione levels. The expression or activity of many of these proteins, including NRF2, SOD1, NQO1, and HMOX-1, can be quantified by quantitative polymerase chain reaction (PCR), Western blots, or commercially available activity assays (Malhotra et al., 2008).
Evaluation of Inflammation In vitro measures of inflammation are primarily based on production of cytokines and chemokines by epithelial, smooth muscle, and inflammatory cells (Baarsma et al., 2011; Mortaz et al., 2009; Starrett and Blake, 2011). Individual cytokines can be measured at both the protein and messenger RNA (mRNA) level. Enzyme-linked immunosorbent assays (ELISA) and Western blot assays can be used to measure protein levels, while quantitative PCR can be used to measure mRNA levels. Additionally, nuclear factor kappa B (NF-kB) represents a major pro-inflammatory transcription factor, and its activity often correlates with inflammation (Zhou et al., 2011). Thus, transcriptional activity of NF-kB can be quantified via a DNA-binding assay. Other signaling cascades, such as mitogen-activated protein kinase (MAPK) signaling, can also result in proinflammatory responses, and activation of these pathways can be detected via specific antibodies that detect the phosphorylated forms of key effector proteins (Cheng et al., 2009).
Evaluation for Mucus Production (Biphasic Culture) Mucus is secreted by airway epithelial cells. A recent advance in culturing airway epithelial cells in vitro is the development of a biphasic culture system in which epithelial cells are maintained in an air-liquid medium interface (Whitcutt et al., 1988). This culture system reflects the in vivo situation and allows further cell differentiation. Quantifying airway mucin synthesis in culture often relies on the characteristics of several biochemical properties of mucin, such as amino acid and carbohydrate compositions, molecular size and enzymatic characterization, and the presence of O-glycosidic bonds
in the isolated molecules (Kim et al., 1985; Wu et al., 1985, 1991). However, these characteristics cannot be used practically in the routine quantification of mucin synthesis and mucous cell population in culture. Several monoclonal antibodies that are useful in the identification of mucous cell population and the quantification of mucin synthesis have been developed (Basbaum et al., 1986; Lin et al., 1989; St. George et al., 1985).
Evaluation for Endothelial Activation Recent studies suggested the involvement of endothelial cells in the pathogenesis of cigarette smoke-induced diseases like emphysema, chronic obstructive pulmonary disease (COPD), and cancer. Extracts of smokeless tobacco also induce proinflammatory changes in cultured human vascular endothelial cells. Activation of endothelial cells following exposure with cigarette smoke extract can be assessed by measuring several different biochemical markers (Chen et al., 2004, 2009; Furie et al., 2000; Guarino et al., 2011). For example, expression of adhesion molecules, such as intercellular adhesion molecule-1 (ICAM-1), E-selectin, and vascular cell adhesion molecule 1 (VCAM-1) can be assessed by Western blot or ELISA. Other markers of endothelial activation, including von Willebrand’s factor and thrombomodulin can also be assessed by ELISA. Activated endothelial cells also express cytokines, such as interleukin 8 (IL-8) and monocyte chemotactic protein-1 (MCP-1), and can be measured as described above. Expression of these activation markers result in enhanced binding of endothelial cells to leukocytes, which can be observed in coculture experiments where monocytes are added to a monolayer of endothelial cells. In these experiments, adhesion is determined through quantification of bound monocytes.
Experiments exposing animals to tobacco smoke have been conducted in hamsters, rats, mice, dogs, rabbits, nonhuman primates, and ferrets. While it is informative to observe the effects of tobacco products in live animal models, it is not possible to mimic human use patterns of combusted products in laboratory animals. This necessarily introduces some artificiality to the experiments, and limits meaningful extrapolation of the findings from animal models to human effects.
Non-cancer Disease Rodent Models for Combusted Products Combusted tobacco products present a risk for pulmonary inflammation and COPD that needs to be evaluated in preclinical models. Multiple animal models of emphysema exist, although the only true inhalation model is the cigarette smoke model of emphysema (Harvey et al., 2011; Rangasamy et al., 2004). There are a variety of commercially available exposure systems,
which consist primarily of either whole-body exposure systems or nose-only exposure systems. Whole-body exposure systems are advantageous in their ability to more carefully regulate the concentration of smoke in the exposure chamber over a period of hours. On the other hand, nose-only exposures typically expose individual mice directly to the smoke from one or a small number of cigarettes, resulting in a potent, although relatively short, exposure. Comparisons of the two methods demonstrate increased levels of carboxyhemoglobin in the rodents exposed via the nose-only method compared to whole-body exposure (Mauderly et al., 1989). Both methods are widely used, and emphysema has been demonstrated after 6 months in whole-body (Clauss et al., 2011; Ma et al., 2005; Sussan et al., 2009; Yoshida et al., 2010) and nose-only exposure systems (Churg et al., 2009; Hautamaki et al., 1997). Both exposure models result in increased oxidative stress, inflammation, and apoptosis in the lungs, and also result in alveolar destruction and airspace enlargement. These responses are all hallmarks of emphysema. However, chronic bronchitis cannot be replicated in rodents. Thus, the combustible products can be assessed for oxidative stress, inflammation, apoptosis, and emphysema in lungs of rodent models.
Chronic exposure to combusted products also causes defects in pulmonary innate immune response that increases bacterial and viral exacerbations in COPD and other diseases (Anzueto et al., 2007; Brusselle et al., 2011). Exposure to chronic cigarette smoke causes immune dysfunction in mice leading to bacterial exacerbations (Harvey et al., 2011). A 1-month cigarette smoke exposure and staphylococcus enterotoxin-induced exacerbation mouse model has also been established that shows heightened T-cell and B-cell responses (Huvenne et al., 2011).
In addition, elastase-induced emphysema has been used as a model to determine the effects of bacterial colonization and emphysematous lesion formation and inflammation in hamsters (Wang et al., 2010). Cigarette smoke exposure heightens inflammatory responses in lungs of mice infected with H1N1 (Bauer et al., 2010). Studies have also established rhinovirus infections as a mediator of viral exacerbations in COPD patients (Mallia et al., 2011). Enhanced secretion of chemokines and proteases were seen in each model. Thus, assessment of inhalable tobacco products should be evaluated for their synergistic action on enhancing the inflammatory response to virus infection or viral PAMP (poly[I:C]) in the lungs of rodent models.
Cancer Disease Rodent Models for Combusted Tobacco Products Some studies have shown that inhaled tobacco smoke can induce tumors and cancers in animal models, but the data are inconsistent. Studies in hamsters have produced convincing evidence that exposure to cigarette smoke
induced an increased incidence of larynx alterations and cancers. In these experiments, the severity of the alterations correlated to dose and duration, while control hamsters did not develop any alterations (Dontenwill et al., 1973). Studies in mice and rats have produced less consistent results, but two relatively recent studies demonstrated significant incidences of respiratory tract tumors. In a study by Mauderly et al. (2004), rats exposed to cigarette smoke had a convincing, although moderate, increase in tumors of the lung and nasal mucosa. In a study by Hutt et al. (2005) with mice, exposure to cigarette smoke significantly increased incidence of lung adenoma (28.2 percent in treated, 6.7 percent in control), adenocarcinoma (20.3 percent versus 2.8 percent), total benign pulmonary neoplasms (30.9 percent versus 6.7 percent), and other changes. Both the Mauderly and Hutt studies were characterized by lengthy exposures to high concentrations of cigarette smoke. It would be important to replicate these results and to determine whether either of these protocols could become a standard model for cigarette smoke evaluation (Hutt et al., 2005; Mauderly et al., 2004).
The A/J mouse is highly susceptible to lung tumor induction and has been widely used as a screening test system in carcinogenicity evaluations. K-ras oncogene activation is associated with enhanced risk for lung tumor susceptibility, illustrated by presentation of pulmonary adenoma. In one replicated exposure protocol, benign lung tumors are reproducibly induced in this strain by a mixture of 89 percent cigarette sidestream smoke and 11 percent mainstream smoke, using an exposure period of 5 months followed by a 4-month recovery period. The response was due to the gas phase of cigarette smoke, and can be used to investigate the effect of secondhand smoke on lung tumorigenesis (Witschi, 2004). Whole-body exposure to diluted cigarette mainstream smoke for 5 months followed by a 4-month postinhalation period gave a concentration dependent tumorigenic response, mainly as pulmonary adenomas in A/J mice as well as in Swiss SWR/J mice. Using this protocol, Stinn et al. (2010) demonstrated that the particulate phase presented the major tumorigenic potency. Further exploration of these models for routine evaluation of combusted products would be desirable.
Experiments exposing animals to fractions of tobacco smoke and its condensate have been conducted to evaluate the carcinogenicity of tobacco smoke constituents. Mouse skin testing of smoke condensate and its subfractions has consistently demonstrated induction of both benign and malignant tumors. Mouse skin testing is particularly sensitive to polycyclic aromatic hydrocarbons, tumor promoters, and cocarcinogens, and should be part of any battery of evaluative assays. Skin application studies have also been conducted in rats, Syrian hamsters, and rabbits (IARC, 2004). Table 3-4 summarizes selected studies of carcinogenicity
|Strain||Sex||No. of Treated Animals/Group||Type of Exposure||Exposure Dosage and Duration||Tumor Incidence||Reference|
|CAF1||M + F,
|44-112||Skin painting (dorsal) of CSC||CSC/acetone solution (40 mg CSC/application), x 3/wk, lifetime||36/81 (skin epidermoid carcinoma), 0/30 (acetone controls)||(Wynder et al., 1953)|
|ICR Swiss||F||5,200||Skin painting (dorsal) of CSC||CSC/acetone solution (150 mg or 300 mg CSC/week), x 6/wk, 78 wks||482/5,200 (skin carcinoma), 3/800 (acetone controls)a||(Gargus et al., 1976)|
|ICR Swiss||F||4,900||Skin painting (dorsal) of CSC||CSC/acetone solution (25 mg or 50 mg CSC/application), x 6/wk, 78 wks||1,157/4,900 (skin carcinoma), 0/800 (acetone controls)||(Gori, 1976)|
|ICR/Ha Swiss||F||100||Topical application with CSC to oral mucosa (lips and oral area)||CSC/acetone (26 mg CSC/application), x 5/wk, 15 months||52/81 (lung tumorsb (P < .0001), 20/89b (acetone controls)||(DiPaolo and Levin, 1965)|
|ICR/Ha Swiss||F||30||Skin painting (dorsal) with CSC active fraction with or without subsequent painting of the skin with croton oil||CSC active fraction/acetone (2.5 mg of 0.6% SC/application), 10 x on alternate days croton oil (2.5%), x 3/wk, up to 15 months, 10 days after last CSC active fraction application||After 12 and 15 months: 4/30 (skin carcinoma), 0/65 (croton oil controls)||(Hoffmann and Wynder, 1971)|
|Strain||Sex||No. of Treated Animals/Group||Type of Exposure||Exposure Dosage and Duration||Tumor Incidence||Reference|
|Swiss||F||30-50||Skin painting (dorsal) of CSC with or without initiation by DMBA application||DMBA (75 μg); CSC/acetone (75 mg CSC/application, start: 1 wk after DMBA application), x 2-3/wk, 12 months—animals observed 3 months later||DMBA: 2/30 (skin carcinoma) (7%) 2 x CSC: 1/40 (skin carcinoma) (3%) DMBA + 2 x CSC: 8/30 (skin carcinoma) (27%) 3 x CSC: 11/50 (skin carcinoma) (22%) DMBA + 3 x CSC: 11/30 (skin carcinoma) (37%)||(Wynder and Hoffmann, 1961)|
|Swiss albino||M||15||Oral gavage of Indian bidi smoke condensate||1 mg bidi smoke condensate/0.1 mg DMSO, 5 days/wk, 55 wks, termination 90 wks||4 hepatic hemangiomas, 1 stomach papilloma and carcinoma, and 1 esophageal carcinoma/15 mice; 0/15 (untreated or DMSO-treated controls)||(Pakhale et al., 1988)|
|Osborne Mendel||F||NG||Intrapulmonary administration of CSC pellet||CSC/beeswax:tricaprylin (24 mg CSC/injection), up to 107 wks after implantation||14/40c (lung squamous-cell carcinoma), 0/63c (beeswax:tricaprylin controls)||(Stanton et al., 1972)|
|OM/NCR||F||120d||Intrapulmonary administration of CSC pellet||CSC/beeswax:tricaprylin (5, 10, 20 or 67 mg CSC/injection), 120 wks after implantation||4%, 10%, 20%, and 42% pulmonary carcinoma prevalence; 0% carcinoma prevalence for three control groups of about 190 rats each||(Dagle et al., 1978)|
|Albino New Zealand||M + F||38||Skin painting of CSC (both ears)||SC/acetone solution (100 mg CSC/application/ear), x 5/wk, lifetime (4-6 yrs)||4/38 (2 skin carcinoma + 1 skin liposarcoma + 1 skin fibrosarcoma), 0/7 (acetone controls)||(Graham et al., 1957)|
in response to different methods of tobacco smoke condensate administration.
Detection of mutagens in the urine of smokers has been shown to be an effective and reliable method of quantifying human exposure to mutagens created by combusted tobacco (Kriebel et al., 1985; Putzrath et al., 1981; Yamasaki and Ames, 1977). These methods involve concentrating organic compounds from urine and evaluating the mutagenicity of the resulting mixture with the Ames test. Aromatic amines and heterocyclic aromatic amines have particularly high activities in these assays, so the results obtained from studies of smokers’ urine may mainly reflect the concentrations of these compounds. Studies have shown that urinary mutagenicity increases with the number of cigarettes smoked (Kuenemann-Migeot et al., 1996; Tuomisto et al., 1986), and that mutagenicity of urine from individuals who used products that heat rather than burn tobacco is similar to that of nonsmokers (DeBethizy et al., 1990; Doolittle et al., 1989; Smith et al., 1996).
Cytogenetic damage, including micronuclei (Bonassi et al., 2003), sister chromatid exchange, and other chromosomal aberrations can also be detected in the cells of smokers. Sister chromatid exchange in peripheral lymphocytes of smokers has been shown to be consistently higher in smokers than in nonsmokers (Rowland and Harding, 1999; Sarto et al., 1985).
Summary of Preclinical Studies
Although preclinical assays for toxicity and carcinogenicity can provide relevant and meaningful data about tobacco products, these assays are limited in their usefulness in this regulatory context. For the purposes of evaluating MRTPs, scientific evidence should be able to support the inference that a particular MRTP will reduce the rates of tobacco-related disease compared to another conventional product.
Preclinical assays alone are fundamentally incapable of supporting such a claim. The majority of the technologies used to test in vitro toxicology were not generated for testing tobacco products and their toxicity (Johnson et al., 2009). These methods “are not reliably quantitative to allow valid comparisons of substantially different tobacco products with differing yields of complex chemical mixtures” and “provide data that cannot reliably be extrapolated to infer human cancer risk” (Johnson et al., 2009). As such, evidence produced by these methods cannot by itself
support the inference that an MRTP will produce less harm than another product.
Nevertheless, preclinical assays of toxicity still play an integral role in the evaluation of MRTPs. These toxicology methods are primarily intended to be used as screening methods to identify potential human carcinogens. These assays are essential in identifying particularly risky or toxic products that should not be tested in humans and for identifying products that have reasonable potential for success and should therefore proceed to clinical evaluation. The role of these tests is to ensure that products that proceed to clinical evaluation in people are not unnecessarily risky and have a reasonable potential to ultimately reduce harm. No one assay can do this alone, because each assay is limited in its scope. A complete battery of preclinical assays should be required prior to committing a product to clinical evaluation. At a minimum, the battery should include assays with consistent and reproducible results and that reach across a wide spectrum of mechanisms and types of toxicity, such as: (1) in vitro toxicity and genetic toxicology tests; (2) appropriate animal studies; and (3) urinary mutagenicity and sister chromatid exchange in smokers. The proper role of these assays is as gatekeepers to long-term studies in humans, which include studies of not only health effects in individuals but also population effects and behavioral effects.
Going forward, it should be anticipated that new assays that specifically focus on tobacco products, that are intended to produce evidence upon which reliable comparisons can be made between products, and for which inferences about human effects can be reliably made, will be developed and should be added to the evaluation process. Over time, the assays discussed in this section may become outdated as technology advances and develops. In the future, it is possible and indeed likely that new assays will be developed that may specifically focus on tobacco products. These assays could be designed to produce evidence intended for comparisons between products, or evidence intended for inference about human effects. These assays should be added to the evaluation process.
Clinical Trial Methods
The use of appropriately designed clinical trials will be important to establish whether use of the MRTP reduces exposure to toxicants or induces positive changes in surrogate markers as claimed by the manufacturer. Although people who have never used tobacco products cannot be randomized to begin using tobacco products (including MRTPs) in the longer term, there may be advantages from certain trial designs involving
substitution of conventional tobacco products with the MRTPs. This design has similarities with a clinical trial evaluation of a smoking cessation intervention. This topic has recently been reviewed (Hatsukami et al., 2009), and so the committee will not reiterate this material here. Short-and intermediate-term clinical trials—where the research participants use the product regularly throughout the day rather than in the confines of a laboratory setting—are thought to provide a better approximation to real-world use. This is particularly true in regard to the question of an MRTP’s ability to be a substitute for cigarettes. Typically, there are “forced switching” studies, where the participant ceases using traditional cigarettes and uses the MRTP for a fixed period of time. Use patterns of the MRTP can be prescribed (controlled use) or can be left to the participant (ad libitum). Such studies can be conducted in the field (i.e., research participant brings the MRTP home) or in a residential setting (i.e., research participant is confined for the duration of the study). Residential settings offer the advantage of stricter control over exogenous factors that could affect biomarkers of exposure or risk (e.g., diet, environmental exposures, etc.) and facilitate compliance with product use. However, these are necessarily contrived and so represent a best-case scenario for product use. Nonresidential studies are more difficult to control and compliance is more difficult to assure, but they are more accurate representations of user behavior. Studies in this idiom have consisted of 12-120 participants, typically containing 10-20 participants per experimental arm. Intermediate-term trials have typically been conducted in the field, but designs have been more varied, ranging from relatively tight prescription of product use to more observational designs. Intermediate-term trials have an advantage of stabilization of use of the MRTP with time.
As one can readily appreciate, demonstrating that an MRTP can achieve measurable changes on clinical endpoints may require large, long-term trials. These designs are sometimes questionable from a perspective of feasibility, are undoubtedly costly, and can only provide useful data after years of investigation. In most studies described in the literature, biomarkers of exposure (e.g., NNAL, cotinine, 1-HOP) and/or risk (8-epi-prostaglandin F2a, forced expiratory volume in one second [FEV1], CRP) have been assessed as the main outcomes. The ability to demonstrate in a randomized trial the significant reduction of a range of biomarkers of exposure and/or risk, in the absence of significant elevation of others, will be critical to the consideration of an MRTP application.
There may be other situations where randomized trials can be employed for the evaluation of specific health effects of MRTPs. While the pathogenesis of the primary tobacco-related chronic diseases (e.g., various cancers, heart disease, and stroke) is thought to take place over many years, there are a number of conditions where MRTP effects could
1. Short-term vascular phenomena, such as intermittent claudication or Raynaud’s disease, which may be responsive over a short term Ankle-Brachial index
2. Mitigation of tobacco-related skin conditions, such as psoriasis or hyperhidrosis
3. Alterations in surgical wound healing, which are known to be tobacco sensitive
4. Variation in the progression and impact of periodontal disease, which is sensitive to tobacco use
5. Alteration in the progression or regression of precancerous mucosal lesions in the oral cavity, where frequent evaluation is feasible
6. Time required for a fracture to heal, also related to tobacco exposure
7. Alteration in the rates of tobacco-related outcomes of pregnancy associated with MRTP use, including fetal death, premature labor and delivery, and low birth weight infants, could be assessed in a relatively short period of time
8. Lung function, pulmonary function testing
9. Blood pressure
be evaluated over a relatively short (< 2 years) time frame. An emphasis on shorter-term clinical outcomes might be one important way to achieve relevant information about the potential health impact of an MRTP. Although not an exhaustive list, Box 3-2 presents a list of examples of health outcomes that MRTPs might be evaluated for relative to smoking and smoking cessation.
In clinical trial design, the use of at least one control arm is crucial. Previous trials have employed various control groups, including arms involving those that continued smoking, those that undertook smoking cessation, and those that switched to medicinal nicotine. Use of a continued smoking arm is necessary to compare exposure and risk reduction while using novel products with levels associated with traditional product use. A cessation arm (where participants may quit with or without pharmaceutical aids) provides researchers with a comparison of the MRTP with the greatest possible exposure or risk reduction. Broadly speaking, a desirable outcome for MRTPs would be a pattern of exposure and risk biomarkers closer to the cessation level than the smoking level. Standard analytical techniques, such as the intention-to-treat principle, would generally be applied to this fundamental design. It is important to recognize that no single randomized controlled trial can address all of the health effects caused by tobacco use. Replication of clinical trial results is an almost universal requirement in the regulation of drugs. Although replication frequently is interpreted as the replication of results using an
identical protocol design, replication requirements can be met by the confirmatory evidence standard. In fact, from a psychometric point of view, stronger conclusions are possible if congruent results are obtained using different measures and methods.
Participant selection and recruitment are important considerations for the generalizability of clinical trial findings. Typically, pregnant and breastfeeding women, children, and those with unstable physical or mental illness have been excluded from MRTP studies. Typically, minimum daily cigarette consumption values are specified for smokers (often > 10 cigarettes per day), and concurrent use of other forms of nicotine or prior experience with the MRTP is proscribed. Research participants have typically been recruited through community advertising (e.g., flyers, newspapers) seeking smokers willing to test new and potentially less risky products.
The Role of Clinical Trials in the Evaluation of MRTPs for Health Effects
Overall, despite the limitations of clinical trials for product evaluation, the committee recognizes the critical role for clinical trials in evaluating the effects of MRTPs on human health. The committee suggests that clinical trial designs consider the following key points, adapted from recommendations provided by Hatsukami et al. (2009):
• Trial designs where biomarkers are used as primary or secondary endpoints should be informed by the half-life of the biomarker(s) examined and the time needed to stabilize use behavior of the MRTP. Determining the stabilization of product use behavior may require a longitudinal trial.
• Clinical trial designs should use both a controlled use approach and an ad libitum approach in complementary studies.
• Any clinical trial should include at least two control conditions—usual brand use and cessation—to allow examination of the relative effects of the MRTP on biomarkers of exposure or risk.
• Short-term residential and nonresidential studies and intermediate-term clinical studies have different strengths and limitations, and proper evaluation of MRTP effects may require several or even all of these study designs. Use of these different study designs will assist for cross-validation.
• Participants in trials should be drawn from a broad cross-section of the population, considering sex, race or ethnicity, smoking or tobacco use history, degree of dependence, stage of change, socioeconomic status, and genetic makeup (e.g., rate of nicotine metabolism).
Observational epidemiologic studies play a critical and central role in the evaluation of MRTPs. Although they will rarely, if ever, have the compelling scientific credibility of experimental designs, these methods form the basis for most evaluation studies of regulated products in the community. This is true particularly in the postlicensure/postcertification period, but also during the initial regulatory evaluation.
Given the great diversity of health consequences of tobacco use (see Table 1-1 in Chapter 1), determining the contrasting potential effects of MRTPs on disease outcomes and population health is a difficult matter. Long, intensive, and robust studies of actual health outcomes would be required to fully evaluate the net effects of MRTPs relative to conventional tobacco products.
An exhaustive, multidisciplinary approach to plan and execute epidemiologic studies to evaluate the relative impact of various MRTPs on health status and outcomes—behavioral, biochemical, genetic, and patho-physiological—are all necessary at some level. In some cases, full answers may not be possible; however, in many cases, rigorously designed studies are likely to be extremely useful in making important policy decisions.
This section provides an overview of the types of epidemiologic and related studies that can address the issues noted above. It is divided into four sections relevant to regulatory and related policy decisions:
1. Considerations on studying disease-exposure associations
2. General design issues for epidemiologic and related studies
3. Evaluating outcomes for various conditions, including the selection of research conditions and the contingencies for each disease category
4. Types of feasible study designs
Preliminary Considerations in Studying the
Disease Outcomes Associated with MRTPs
There May Be Many Potential Types of MRTPs and Many Patterns of Usage
One of the critical general issues in exploring the health impact of MRTPs is the multiplicity of products that may become available and the potential variation in their characteristics. Product type and the purported mechanisms by which it is expected to reduce disease risk by necessity inform the type of epidemiologic studies that can be effective in evaluating its health effects. If there are many products with potentially diverse pharmacological and biological effects, evaluating each separately could be a great logistical challenge. In observational studies, it is axiomatic that
the product(s) involved should be unambiguously identified so the effects of MRTP exposure can be disentangled from other tobacco products.
Based on available information, it may be necessary to combine various products into a manageable number of analytical categories in order to conduct statistically robust studies. The construction of these categories should, however, be scientifically credible. A related issue is that over time, individuals using MRTPs may switch products at irregular intervals, use them at varying rates, use them interchangeably, or even use them simultaneously with conventional tobacco products, making it very difficult to credibly document use patterns that can be related to health outcomes in observational studies. A similar issue arises if many products are not widely used in the general population; in this case, there may be insufficient population exposure to confidently assess particular health outcomes. It is likely that only products with substantial and long-term general market sales in the general population will be suitable for epidemiologic assessment of MRTP-related disease occurrence, that is, largely for postcertification activities. It is also possible that MRTPs that have been consumed in the community over a long period may have changed in content and exposure yields, thus complicating exposure assessment.
The Diversity of Diseases and Conditions Caused by
Conventional Tobacco Products
Tobacco use, particularly cigarette smoking, causes a large number of diseases and conditions (HHS, 2004a, 2006). Diseases and conditions caused by active cigarette smoking are summarized in Table 1-1. Thus, an important conceptual issue is which conditions should be evaluated for alteration when evaluating MRTPs. It is obvious that not all tobacco-related conditions can be assessed, and policy decisions on evaluative strategies need to be made. Also, it is possible that different MRTPs will have different effects on different disease processes. For example, there is no necessary a priori reason to believe that an MRTP that reduces the risk of atherosclerotic disease may yield the same effects on risk of various cancers, bone fracture, premature delivery, or Alzheimer’s disease. The multiplicity of potentially available health outcomes requires careful consideration when selecting epidemiologic study designs. Epidemiologic assessments will be much more efficient if targeted to specific diseases and conditions based on hypotheses grounded in previous literature reviews of the product—disease associations, the known chemical constituents of the MRTP, the constituents to which product users are exposed, and other suggestions from professionals or the public. It is conceivable or even likely that different types of study designs may be needed for different disease outcomes. For example, a study evaluating the effects of an
MRTP on lung cancer may be structured differently from one assessing atherosclerotic outcomes. In fact, it is entirely possible that an MRTP may decrease the risk of some conditions while increasing the risk of others, even those that are not necessarily caused by tobacco smoking. The design of studies to assess offsetting risk can be extremely complex, and policy decisions will have to be made as to how this issue should be regarded. All of this reinforces the need to mandate epidemiologic precertification studies that are directed by the best exposure and toxicological data available.
General Design Issues for Epidemiologic and Related Studies of MRTP-Disease Associations
The Importance of Determining “Acceptable” Effect Size Differences
As population studies are developed for evaluating the health impact of MRTPs, there may be value in establishing in advance the policy for interpreting various study effect sizes as the differences between outcomes of MRTPs versus conventional tobacco products emerge. That is, how much of a decrease in disease rates is important to individuals trying to change their smoking habits, and what differences should lead to certain regulatory decisions? And how much difference should occur before a product can be called an MRTP? In general, such policies should be determined aside from statistical significance, although the latter is important. For example, if an MRTP, ceteris paribus, yielded a hypothetical 2 percent reduction in lung or bladder cancer rates over a defined time period, would that be a suitable basis for regulatory decisions or compelling enough for a smoker to change products? And what if the effect is different over a longer time period? Such regulatory decisions may be more problematic given that the impact of the MRTP on other conditions may not be well understood. Considering acceptable “effect sizes” early on may help define the sample sizes and other design features of proposed studies.
The structure of studies that contrast risk of disease among MRTPs and conventional tobacco products is of paramount interest, and, speculatively, many potential MRTPs with substantial reduction in toxic exposures may show reductions in disease risks. This is likely given the high toxic exposures that occur due to use of conventional cigarettes. However, studies that contrast disease risks conferred by competing MRTPs may be more challenging because exposures are lower and confounding factors may become more important. This problem should be considered in structuring such studies.
Strategies to Increase the Efficiency of Study Designs in Exploring
MRTP-Associated Disease Risks
In conventional cohort studies, as discussed below in more detail, health outcomes among those persons using MRTPs are prospectively compared to those using conventional tobacco products, and it may take many years for answers to appear. This is because the incubation period of many smoking-related conditions may extend to decades, and there may be no basis for understanding how long it may take to show differences among those using MRTPs, even with lesser cumulative exposures to certain tobacco constituents. This is especially true because many persons in these study cohorts would be former smokers, and disease pathogenesis is already under way. However, some strategies exist that may be explored to enable acquisition of earlier answers:
1. Some diseases emerge earlier than others after tobacco initiation, or decrease more rapidly when conventional tobacco products are withdrawn, and in these situations it may be possible to acquire earlier answers regarding MRTP health effects. An important example is coronary heart disease, where withdrawal of cigarette smoking is associated with a clear reduction in disease risk within a few years of smoking cessation. Another important example could be evaluating the relative effects of MRTPs and conventional tobacco products on pregnancy outcomes. While all pregnant women should be strongly discouraged from all tobacco use, those who cannot or will not quit may be approached to use alternative products, and answers to problems such as fetal loss or premature delivery may be available relatively quickly. Such study designs will require substantial consideration and thorough ethical review.
2. An additional approach is to focus on various population groups that are at particularly high risk of disease outcomes of interest. One obvious approach is to enrich disease outcome studies with individuals possessing high-risk factor levels (other than tobacco use) for those conditions. This may allow smaller sample sizes and possibly shorter study intervals. For example, rates of cardiovascular disease outcomes would be increased by enrolling those with elevated blood pressure and cholesterol levels, or familial hypercholesterolemia and diabetics. Selected occupational groups where smoking levels are high in addition to job-related exposures may be at special risk of lung tumors, such as in uranium or asbestos miners, and textile workers. Smokers who have had one tumor that is “cured” are at greater risk of a second tumor (e.g., those with head and neck cancers) and may be important research participants. All of these groups may be suitable for clinical trials or observational studies of MRTP health effects.
3. A related general approach to possibly accelerate informative studies on the role of MRTPs in altering risks and rates for important diseases and conditions is to focus on population groups with higher prevalence rates of conventional tobacco product use. Examples of groups with higher cigarette smoking rates include certain minority groups, persons with lower socioeconomic status, sexual minorities, individuals with psychiatric conditions and substance abuse other than tobacco, and disabled individuals. Focusing on such populations may lead to efficiencies in study recruitment, and because of higher rates of smoking, such populations should be given special consideration for emphasis in reducing smoking-related morbidity and mortality. As above, these high-risk populations may be important candidates for trials or observational studies.
4. The use of composite health outcomes could also increase the efficiency of MRTP evaluation studies. Conventional cohort studies (see below) can yield data on all disease outcomes for which information is sought. However, understandably, these studies examine single disease outcomes separately, following specified hypotheses and exploring biologically and toxicologically plausible causal pathways. But because these studies yield many tobacco-related outcomes of importance, such as major diseases and causes of death associated with cigarette smoking, there may be scientifically credible value in pre-specifying and exploring composite outcomes, possibly increasing their efficiency and decreasing their duration. One of the most obvious strategies would be to create an outcome consisting of any of these major diseases, whichever comes first. In fact, this is an approach used in some cohort studies and clinical trials. The issue has been perhaps best evaluated with respect to varying causes of death, where smoking leads to any number of important illnesses, most precluding the occurrence of the others, a phenomenon called “competing mortality.” Because it is an important goal for MRTPs to prevent and alleviate suffering from a variety of diseases, composite outcomes may more accurately reflect general health outcomes, in the same way that self-reported health status and disability-adjusted life years summarize health status across individual disease states. It might even be worthwhile to weight disease outcomes in terms of clinical importance or likely relation to product use (e.g., with lung cancer receiving a higher weighting than chronic bronchitis). The use of a composite category in no way precludes evaluating individual disease outcomes.
Consideration of Confounding Factors in Epidemiologic Studies of Tobacco,
MRTPs, and Altered Disease Risk
Almost all epidemiologic studies can be subverted if confounding factors—factors associated with both the likelihood of exposure and disease
outcomes—are not taken into account. Often, these are the very risk factors that explain why some persons are at greater risk for various diseases, such as hypertension, hypercholesterolemia and diabetes, and atherosclerotic diseases. One would not want to falsely attribute an altered disease risk associated with an MRTP when the contrast groups actually differ in other risk factors that can explain the observed differences.
Another example of a confounding factor is the common situation where epidemiologic studies contrast continuing cigarette smokers with those who change to MRTPs. The latter group may be less addicted to tobacco and nicotine and thus may have quantitative tobacco exposure differences that need to be considered in assessing disease risk. Determining patterns of MRTP use and levels of exposure will be very important in assessing product-disease associations. Comparative studies of these groups should attempt to adjust for these exposure differences among the contrast groups.
Also, as suggested above, an important example of a confounding factor associated with tobacco addiction is the fact that cigarette smoking is associated with increased prevalence rates of a variety of psychiatric illnesses, including various substance use and abuse syndromes including alcohol and illicit drugs. Thus, for maximum analytical specificity in evaluating MRTPs, it would generally be important to try to acquire a history of psychiatric illnesses and related substance abuse activities that may in themselves lead to adverse health outcomes. If the illnesses and substance use rates are lower among those able to switch to MRTPs, this could confound study findings.
Other situations exist where confounding factors may be important when considering studies that contrast disease outcomes of MRTPs versus conventional tobacco products:
a. Cigarette smokers often try to stop smoking, as documented in this report, but it may be important to understand some of the motivations. Some smokers stop smoking or change tobacco products because of overt incident diseases or the self-perception of abnormal symptoms or related clinical problems. It is important to obtain a history of these events when conducting epidemiologic studies; otherwise, product use may appear to be associated with increased disease risk when in fact the product was initiated because of the advent of clinical problems.
b. Another related issue that occurs with such studies is that smokers often use aids to assist in smoking cessation, such as nicotine-containing products or other medications. It is documented that many of these smoking cessation aids have their own set of adverse clinical events (Singh et al., 2011), and to the extent possible their use should be carefully monitored and not be confused with MRTP-associated effects.
c. Comorbid conditions in smokers and MRTP users are also potential confounders that will need attention. It is not surprising that current and former smokers may have higher rates of various medical conditions than nonsmoking populations. The presence of such conditions and their treatments prior to study onset can confound the evaluation and interpretation of MRTP or conventional tobacco disease outcomes and should be scrupulously documented in all epidemiologic and related studies of product contrasts. It should be emphasized that assessment of disease treatments is also extremely important. Extensive treatments of important diseases may be indicative of more severe disease processes, and they may be associated with higher rates of secondary complications, such as from percutaneous coronary stents or adjuvant chemotherapy.
d. In the genetics/genomics era, gene variants have been discovered that may affect the pharmacological and pathogenetic effects of both conventional cigarettes and MRTPs, as well as various disease outcomes (NCI, 2009). In a sense, for the purposes of product evaluation, genes may become confounders of product-outcome assessments, because they may relate both to product use behavior and to the clinical outcomes. The relevant genetic literature should be monitored so genetic studies can be made if a gene variant becomes an important part of causal pathways. The committee recognizes that people may be increasingly likely to have genome scans or other genetic tests, and availability of such information should be monitored.
e. For the past several years in the United States, cigarette smoking rates have been higher among persons with lower socioeconomic status (CDC, 2011a); that is, those with lower educational attainment, those with lower personal and family income, and “blue-collar” workers are more likely to encumber higher rates of adverse occupational or environmental exposures. Epidemiologic studies that compare conventional cigarette smokers with nonsmokers or MRTP users thus need to scrupulously adjust for socioeconomic differences among these groups in order to avoid confounding by this potent factor, which is related to both rates of tobacco use and adverse health outcomes.
f. Because of social and regulatory pressure on smoking behaviors, cigarette smokers tend to congregate with each other, or find themselves together in designated smoking venues. Thus, in epidemiologic studies of MRTPs that include biomarkers or more concrete health outcomes, the role of secondhand smoke exposure can be an important determinant. This problem should lead to routine data collection of secondhand smoke exposure as part of observational study methodology.
g. In this era of rapid changes in tobacco-related public health policies, legislation (e.g., increased tobacco taxes), and health information, it is possible that changes in secular events could significantly influence such
outcomes as tobacco use and cessation, likelihood of adoption of MRTP use, and engagement in other health-relevant behaviors (exercise, use of statin drugs). This calls for careful evaluation of not only such secular events but also the possible consequences of such events, so that these can be used as covariates or time-varying covariates, depending on the nature of the research design.
h. In certain observational studies, ascertainment and detection bias may be an issue. For example, ex-smokers switching to an MRTP might be under more surveillance than other populations, or higher-risk subjects may undergo additional diagnostic tests or screening, which may skew the results. Consideration of detection and ascertainment bias is particularly important in the design and evaluation of longer-term observational studies.
Benchmarking the Health Effects of MRTPs
A generally useful but sometimes tacit presumption in evaluation studies of MRTPs is that conventional tobacco product use is the health benchmark against which MRTPs are evaluated. However, this could be difficult to execute in the common situation where a credible lifetime history of cigarette smoking is difficult to obtain. Researchers may also benchmark MRTPs with each other and with the health outcomes of nonsmokers, and this may also be of value in making policy decisions. An explicit approach to benchmarking health outcome levels is extremely useful and could encompass a range of tobacco products or MRTPs. These should be declared in advance of proposed health studies.
Evaluating Health and Disease Outcomes in the Study of MRTPs
There are several potential types of response variables (outcomes) to MRTPs and tobacco products in observational studies and other clinical and population research. The advantages and limitations of using biomarkers and surrogate endpoints were discussed earlier in this chapter. It should be noted that with respect to reflecting true disease outcomes, biomarkers have been controversial. In general, because there have been many documented instances where pharmacological alteration of biomarker levels has not led to disease progress in the predicted direction, biomarkers have received limited credibility as disease endpoints (Hatsukami et al., 2006; Hecht et al., 2010). In general, they are not acceptable alternatives to true disease endpoints.
Furthermore, for many years there has been substantial concern about adopting surrogate endpoints as the sole measure of therapeutic efficacy in clinical trials, particularly because there are very important counter
examples in the history of drug regulation where surrogate endpoint control did not lead to disease prevention or amelioration; such intermediate endpoints included blood pressure control, antiarrhythmic treatments, and cholesterol-lowering agents.
General Epidemiologic and Related Study Designs for Assessing Altered Disease Risk or Mitigation Associated with MRTP Use
In general, except for short-term pharmacological or toxicological studies and some behavioral interventions, disease risks associated with MRTPs will be assessed with observational studies, although there is certainly room for clinical trial methodology, as noted previously, because of their growing importance to translational science. A panoply of observational research designs is available, and only a few of the most basic and central will be discussed here.
Cohort Studies in MRTP Assessment
Cohort studies are obvious candidates for the evaluation of MRTPs, and over the years they have been an important instrument of tobacco product evaluation (FDA, 2011b). In this type of study design, persons with various product use habits are followed into the future to assess variation in clinical outcomes. These studies have several important strengths:
• Biochemical tobacco and MRTP exposure assessments can be made at baseline, offering “unbiased” exposure assessment before health outcomes occur.
• There is less of a problem with retrospective recall of product use, because this information is summarized at the start of the study and followed prospectively.
• Changing product use habits can be monitored concurrently as the study progresses.
• Outcomes are documented as they occur, and verification becomes more efficient.
• A wide variety of outcomes can be evaluated in the same study, particularly those that are more common.
Indeed, cohort studies allow assessment of overall health status and outcomes. However, there are also prominent or at least potential limitations to this design:
• Important and severe chronic illnesses may be uncommon and take many years to occur, even in a population of cigarette smokers, and thus
the studies may require many years, large sample sizes, and substantial resources to complete.
• If exposure to MRTPs is limited in the community, these studies may be underpowered and inefficient.
• MRTP or conventional tobacco product use habits may change over time, and thus determining and analyzing differential exposures may be complex.
Efficiencies could be obtained in part by enrolling only persons who use certain tobacco products or MRTPs in a cohort study. Depending on whether the contrast group for MRTP evaluation is nonsmokers or never smokers, one variation of a general cohort study approach is to create “inception cohorts” of those beginning MRTP use for the first time. This is similar to a “new-user” cohort when evaluating drug use outcomes. However, the logistics of this type of study and maintaining the cohort for many years would always be challenging.
One additional, possibly more efficient approach would be a retrospective cohort study, where the data have already been collected. This might occur in the situation, for example, where the product (conventional tobacco and MRTP) purchasing behavior of a large group of persons has been previously recorded, and the population had been monitored for relevant health outcomes. However, this would not apply to MRTPs never on the market, and there would be a problem of ascertaining important confounding variables to conduct a credible analysis. It is not likely that many such retrospective cohorts would be available, and in such situations, important confounding factors may not have been collected.
Finally, an additional strategy to increase efficiency of prospective cohort studies is to include additional questions and measures of biomarkers to concurrent studies.
The Important Role of Case-Control Studies
Another important instrument of observational epidemiology is the case-control study, where persons with a particular health outcome or disease are the cases and a healthy control group is used to contrast the history of exposure to whatever exposure is being evaluated. Case-control studies are commonly used because of their efficiency in assembling study participants, including the circumstance where the outcomes are not common in general populations. Further, this method has been used widely to evaluate the impact of preventive interventions (Weiss, 1994). It is possible to contrast product exposures among those with varying levels of biomarkers or disease outcomes—intermediate or clinically
overt—with “controls” who have no evidence of disease and have normal disease-related biomarker levels.
As in the situation of cohort studies, however, case-control studies encumber many important methodological issues that require attention, in addition to the confounding problem noted above. These include
• cases and controls should have the same source population to allow more credible contrast;
• exposures to MRTPs should have had sufficient time to occur and be generally available so usage can be evaluated;
• diseases (cases) can only be assessed one at a time; no overall health impact is usually possible; and
• exposures can only be assessed retrospectively, which can be a problem because of lapses in recall or memory (the so-called recall bias).
As in all case-control studies, the accuracy of retrospective recall of exposure can decrease scientific credibility and usefulness. Nonetheless, for evaluating MRTPs that have had community usage, the case-control study will remain an important tool.
When the outcomes are short term and/or recurrent, particularly when using biomarkers or intermediate endpoints, an observational crossover (or “case-crossover”) design becomes feasible and informative. In its most simple form, research participants serve as their own control, and the outcome of interest is assessed during each of the exposures of interest. Then, for example, the effects of cigarette smoking can be compared with exposure to an MRTP. This “self-control” approach eliminates many potential “within-person” confounders, but it assumes that the effect of the first exposure does not carry over when the switch to the other exposure occurs. Crossover designs could be used to evaluate participants who switch from one MRTP to another, or who switch from an MRTP back to conventional cigarettes or other tobacco products.
Applying the Methodology of Comparative Effectiveness Research (CER)
The methodology of CER is basically designed to more critically inform health care and policy decisions by comparing health outcomes associated with different clinical interventions (usually therapies) for a particular disease or other clinical situations. Although CER methods include clinical trials, most approaches have been developed for observational study application, particularly in the analysis of clinical cohorts.
CER can sharpen or extend observational methodology that could provide additional approaches for comparing smokers and nonusers of tobacco with those using certain MRTPs. Methods such as propensity scoring (the likelihood of switching options, e.g., MRTP or conventional products) and instrumental variable analysis (to adjust for unmeasured confounders) are routinely used in non-CER research, but offer additional techniques for exploring causation. CER also offers techniques to review and synthesize the medical literature and identify important gaps, promote new analytical tools, and translate research findings to diverse stakeholders (IOM, 2009). Although CER does not claim to provide the same level of causal inference that might be derived from a randomized experimental design, its promise is to provide more credible answers when only observational data are available.
Summary of Observational Studies
There is no overriding conflict between observational and experimental methods and designs; rather, the contributions of both study designs are complementary and will be necessary for thorough evaluation of MRTPs. It is important to note that even when randomized clinical trials for health and behavioral outcomes are feasible and performed, subgroup analyses of these data are essentially observational in nature. Each general approach to scientific inquiry is really a large suite of study designs, to be chosen and exploited in the combinations that yield the best possible answers to the health and safety questions of interest. Consideration of study designs in general depend on them having suitable feasibility in execution, scientific credibility, responsiveness to informing policy decisions, and efficient use of available resources. The scientific and regulatory reality is that most of the population outcome studies can only be satisfied with the best observational studies possible. There are several reasons for the centrality of observational methods:
a. There are substantial ethical limitations on the application of MRTPs or contrasting conventional tobacco products in planned intervention studies, although some situations do allow for such interventions. A discussion on the ethical considerations of tobacco research is found in Chapter 2.
b. Research participants in randomized trials can rarely be expected to adhere to a particular intervention or product for long periods of time, as is true of drug or other intervention trials, so that summarization of usage patterns will require detailed and complex observational techniques. This is crucial to measuring personal exposure to MRTPs and conventional
tobacco products, so the exposure “dose” can be assessed as accurately as possible and related to health or behavioral outcomes of interest.
c. Behavioral patterns of MRTP use and myriad health outcomes that may be anticipated yield a level of complexity that can often not be captured in experimental designs, no matter how desirable. This complexity limits the nature and execution of experimental designs.
d. Data on MRTP marketing and distribution are in essence observational. These data play an increasingly important role in the evaluation of population exposure to various MRTPs or conventional tobacco products, and they help set boundaries to better understand potential rates of potential adverse events.
e. MRTP manufacturers and marketers will change product formulations, designs, and advertising modes and presentations in order to maximize sales. Such activities will likely work to subvert experimental studies where adherence to a particular MRTP is a fundamental part of the study design.
f. Community surveillance for adverse effects of MRTPs is for the most part an observational activity, including such sources as (a) citizen reports to health agencies; (b) individual case reports or case series reported by health professionals, which are uncontrolled clinical observations; (c) ad hoc institutional or multicenter disease registers, which in general do not have a geographic-based catchment area; or (d) monitoring of electronic medical records for health events not otherwise anticipated.
g. Most adverse health and behavioral effects of MRTPs will be detected and validated in the longer term, to the extent possible, using observational methods such as cohort, case-control, or other related study designs.
h. As with most other consumer products released in the community, MRTPs are subject to personal misuse and abuse, accidental contamination, conscious adulteration, faulty manufacturing, the release of imitation (and thus unregulated) products, and long-term unanticipated alterations in product content and potential health outcomes. The development of protocols for inspection and other problem-control protocols for these problems, as is true for other consumer products in general, are essentially observational in nature and rarely subject to experimental designs.
While evaluating the empirical evidence emerging from studies on the health effects of MRTPs, researchers and regulators should anticipate how an intended exposure reduction affects disease risks. For this, models based on scientific data, rather than on speculation, can provide
relevant insight. Mathematical modeling for estimating health effects of tobacco products is one method to improve the quantification of exposure response data from product development. Modeling can generate data on complex issues of product and constituent interaction and can provide insight for trials in specific subpopulations.
Risk assessment models, developed to represent the mechanistic pathways leading to clinical endpoints, can be used to study disease endpoints. There is a history of using models to understand the health impacts of tobacco use. For example, following the emergence of evidence linking smoking to lung cancer in the 1950s, Levin proposed a model linking smoking to lung cancer; this method is still in use today (Levin, 1953).
A model linking a reduced carcinogen exposure to a reduced risk of cancer should include a causality assessment, which details how the targeted carcinogen affects an individual’s health and risk for cancer. The model should also include knowledge of the dose-response relationship for the carcinogen, as well as an individual susceptibility assessment. Additionally, it should include an understanding of the targeted carcinogen in context of the other carcinogens present in the product (IOM, 2001). Exposure can be measured using a validated biomarker, rather than by individual constituents present in the tobacco product or its smoke.
While designing a model, researchers should take into account the potential limitations of its inputs. For example, a dose-response curve may change for individuals with different histories of tobacco use. An integrated mathematical model for tobacco harm reduction should consider dose-response relationships for multiple disease outcomes. That is, a model with a dose-response relationship for only a single disease outcome will limit the relevance of the data, as tobacco product use leads to multiple health outcomes.
Failures in modeling design can lead to unsuccessful future studies and other product safety issues (FDA, 2009). It is likely that discussion of quantitative tobacco product development methods between the FDA and product sponsors will improve these results.
Adam, T., S. Mitschke, T. Streibel, R. R. Baker, and R. Zimmermann. 2006. Quantitative puff-by-puff-resolved characterization of selected toxic compounds in cigarette mainstream smoke. Chemical Research in Toxicology 19(4):511-520.
Andrei, G. 2006. Three-dimensional culture models for human viral diseases and antiviral drug development. Antiviral Research 71(2-3):96-107.
Andrei, G., S. Duraffour, J. Van den Oord, and R. Snoeck. 2010. Epithelial raft cultures for investigations of virus growth, pathogenesis and efficacy of antiviral agents. Antiviral Research 85(3):431-449.
Anzueto, A., S. Sethi, and F. J. Martinez. 2007. Exacerbations of chronic obstructive pulmonary disease. Proceedings of the American Thoracic Society 4(7):554-564.
Ashrafi, S., A. Das, R. Worowongvasu, B. Mehdinejad, and J. Waterhouse. 1992. A light, transmission and scanning electron microscope study of snuff-treated hamster cheek pouch epithelium. Scanning Microscopy 6(1):183.
Baarsma, H. A., H. Meurs, A. J. Halayko, M. H. Menzen, M. Schmidt, H. A. Kerstjens, and R. Gosens. 2011. Glycogen synthase kinase-3 regulates cigarette smoke extract-and IL-10-induced cytokine secretion by airway smooth muscle. American Journal of Physiology—Lung Cellular and Molecular Physiology 300(6):L910-L919.
Bagchi, D., M. Bagchi, E. A. Hassoun, and S. J. Stohs. 1996. Cadmium-induced excretion of urinary lipid metabolites, DNA damage, glutathione depletion, and hepatic lipid peroxidation in Sprague-Dawley rats. Biological Trace Element Research 52(2):143-154.
Bagchi, M., J. Balmoori, D. Bagchi, S. D. Ray, C. Kuszynski, and S. J. Stohs. 1999. Smokeless tobacco, oxidative stress, apoptosis, and antioxidants in human oral keratinocytes. Free Radical Biology and Medicine 26(7-8):992-1000.
Bagchi, M., C. A. Kuszynski, J. Balmoori, S. S. Joshi, S. J. Stohs, and D. Bagchi. 2001. Protective effects of antioxidants against smokeless tobacco-induced oxidative stress and modulation of Bcl-2 and p53 genes in human oral keratinocytes. Free Radical Research 35(2):181-194.
Bagchi, M., J. Balmoori, D. Bagchi, S. J. Stohs, J. Chakrabarti, and D. K. Das. 2002. Role of reactive oxygen species in the development of cytotoxicity with various forms of chewing tobacco and pan masala. Toxicology 179(3):247-255.
Banerjee, A., V. Gopalakrishnan, and J. Vishwanatha. 2007. Inhibition of nitric oxide-induced apoptosis by nicotine in oral epithelial cells. Molecular and Cellular Biochemistry 305(1):113-121.
Basbaum, C. B., A. Chow, B. A. Macher, W. E. Finkbeiner, D. Veissiere, and L. S. Forsberg. 1986. Tracheal carbohydrate antigens identified by monoclonal antibodies. Archives of Biochemistry and Biophysics 249(2):363-373.
Batariova, A., V. Spevackova, B. Benes, M. Cejchanova, J. Smid, and M. Cerna. 2006. Blood and urine levels of Pb, Cd and Hg in the general population of the Czech Republic and proposed reference values. International Journal of Hygiene and Environmental Health 209(4):359-366.
Bauer, C. M., C. C. Zavitz, F. M. Botelho, K. N. Lambert, E. G. Brown, K. L. Mossman, J. D. Taylor, and M. R. Stampfli. 2010. Treating viral exacerbations of chronic obstructive pulmonary disease: Insights from a mouse model of cigarette smoke and H1N1 influenza infection. PLoS One 5(10):e13251.
Bernzweig, E., J. B. Payne, R. A. Reinhardt, J. K. Dyer, and K. D. Patil. 1998. Nicotine and smokeless tobacco effects on gingival and peripheral blood mononuclear cells. Journal of Clinical Periodontology 25(3):246-252.
Bertram, K. M., C. J. Baglole, R. P. Phipps, and R. T. Libby. 2009. Molecular regulation of cigarette smoke induced-oxidative stress in human retinal pigment epithelial cells: Implications for age-related macular degeneration. American Journal of Physiology—Cell Physiology 297(5):C1200-1210.
Boffetta, P., S. Clark, M. Shen, R. Gislefoss, R. Peto, and A. Andersen. 2006. Serum cotinine level as predictor of lung cancer risk. Cancer Epidemiology, Biomarkers & Prevention 15(6):1184-1188.
Bonassi, S., M. Neri, C. Lando, M. Ceppi, Y. Lin, W. P. Chang, N. Holland, M. Kirsch-Volders, E. Zeiger, and M. Fenech. 2003. Effect of smoking habit on the frequency of micronuclei in human lymphocytes: Results from the human micronucleus project. Mutation Research/Reviews in Mutation Research 543(2):155-166.
Bond, J. A., B. T. Chen, W. C. Griffith, and J. L. Mauderly. 1989. Cigarette smoke induces DNA adducts in lungs of rats after inhalation. Experimental Pathology 37(1-4):190-193.
Borgerding, M., and H. Klus. 2005. Analysis of complex mixtures—cigarette smoke. Experimental and Toxicologic Pathology 57(Suppl. 1):43-73.
Brunnemann, K. D., J. Qi, and D. Hoffmann. 2002. Chemical profile of two types of oral snuff tobacco. Food and Chemical Toxicology 40(11):1699-1703.
Brusselle, G. G., G. F. Joos, and K. R. Bracke. 2011. New insights into the immunology of chronic obstructive pulmonary disease. Lancet 378(9795):1015-1026.
Burns, D. M., E. Dybing, N. Gray, S. Hecht, C. Anderson, T. Sanner, R. O’Connor, M. Djordjevic, C. Dresler, P. Hainaut, M. Jarvis, A. Opperhuizen, and K. Straif. 2008. Mandated lowering of toxicants in cigarette smoke: A description of the World Health Organization TobReg proposal. Tobacco Control 17(2):132-141.
Carmella, S. G., M. Chen, S. Han, A. Briggs, J. Jensen, D. K. Hatsukami, and S. S. Hecht. 2009. Effects of smoking cessation on eight urinary tobacco carcinogen and toxicant biomarkers. Chemical Research in Toxicology 22(4):734-741.
CAST II Investigators. 1992. Effect of the antiarrhythmic agent moricizine on survival after myocardial infarction. NEJM 327(4):227-233.
CDC (Centers for Disease Control and Prevention). 1999. Tobacco use—United States, 19901999. Morbidity and Mortality Weekly Report 48(43):986-993.
CDC. 2010. Vital signs: Current cigarette smoking among adults aged ≥ 18 years—United States, 2009. Morbidity and Mortality Weekly Report 59(35):1135-1140.
CDC. 2011a. Cigarette smoking—United States, 1965-2008. Morbidity and Mortality Weekly Report 60(1):109-113.
CDC. 2011b. Ten great public health achievements—United States, 2001-2010. Morbidity and Mortality Weekly Report 60(19):619-623.
Chen, C., and J. F. Pankow. 2009. Gas/particle partitioning of two acid-base active compounds in mainstream tobacco smoke: Nicotine and ammonia. Journal of Agricultural and Food Chemistry 57(7):2678-2690.
Chen, H. W., M. L. Chien, Y. H. Chaung, C. K. Lii, and T. S. Wang. 2004. Extracts from cigarette smoke induce DNA damage and cell adhesion molecule expression through different pathways. Chemico-Biological Interactions 150(3):233-241.
Chen, H. W., C. K. Lii, H. J. Ku, and T. S. Wang. 2009. Cigarette smoke extract induces expression of cell adhesion molecules in HUVEC via actin filament reorganization. Environmental and Molecular Mutagenesis 50(2):96-104.
Chen, P., and S. Molduveanu. 2003. Mainstream smoke chemical analyses for 2R4F Kentucky reference cigarette. Beitrage zur Tabak forschung International 20(7):448-458.
Cheng, S. E., S. F. Luo, M. J. Jou, C. C. Lin, Y. R. Kou, I. T. Lee, H. L. Hsieh, and C. M. Yang. 2009. Cigarette smoke extract induces cytosolic phospholipase A2 expression via NADPH oxidase, MAPKs, AP-1, and NF-kappaB in human tracheal smooth muscle cells. Free Radical Biology and Medicine 46(7):948-960.
Chung, F. L., L. Zhang, J. E. Ocando, and R. G. Nath. 1999. Role of 1, N-2-propanodeoxyguanosine adducts as endogenous DNA lesions in rodents and humans. Exocyclic DNA Adducts in Mutagenesis and Carcinogenesis(150):45-54.
Church, T. R., K. E. Anderson, N. E. Caporaso, M. S. Geisser, C. T. Le, Y. Zhang, A. R. Benoit, S. G. Carmella, and S. S. Hecht. 2009. A prospectively measured serum biomarker for a tobacco-specific carcinogen and lung cancer in smokers. Cancer Epidemiology, Biomarkers & Prevention 18(1):260-266.
Church, T. R., K. E. Anderson, C. Le, Y. Zhang, D. M. Kampa, A. R. Benoit, A. R. Yoder, S. G. Carmella, and S. S. Hecht. 2010. Temporal stability of urinary and plasma biomarkers of tobacco smoke exposure among cigarette smokers. Biomarkers 15:345-352.
Churg, A., S. Zhou, X. Wang, R. Wang, and J. L. Wright. 2009. The role of interleukin-1 P in murine cigarette smoke-induced emphysema and small airway remodeling. American Journal of Respiratory Cell and Molecular Biology 40(4):482-490.
Clauss, M., R. Voswinckel, G. Rajashekhar, N. L. Sigua, H. Fehrenbach, N. I. Rush, K. S. Schweitzer, A. O. Yildirim, K. Kamocki, A. J. Fisher, Y. Gu, B. Safadi, S. Nikam, W. C. Hubbard, R. M. Tuder, H. L. Twigg, 3rd, R. G. Presson, S. Sethi, and I. Petrache. 2011. Lung endothelial monocyte-activating protein 2 is a mediator of cigarette smoke-induced emphysema in mice. Journal of Clinical Investigation 121(6):2470-2479.
Copp, J., G. Manning, and T. Hunter. 2009. TORC-specific phosphorylation of mammalian target of rapamycin (mTOR): Phospho-Ser2481 is a marker for intact mTOR signaling complex 2. Cancer Research 69(5):1821-1827.
Correa, E., P. A. Joshi, A. Castonguay, and H. M. Schuller. 1990. The tobacco-specific nitrosamine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone is an active transplacental carcinogen in Syrian golden hamsters. Cancer Research 50(11):3435-3438.
Counts, M. E., F. S. Hsu, S. W. Laffoon, R. W. Dwyer, and R. H. Cox. 2004. Mainstream smoke constituent yields and predicting relationships from a worldwide market sample of cigarette brands: ISO smoking conditions. Regulatory Toxicology and Pharmacology 39(2):111-134.
Dagle, G., K. McDonald, L. Smith, and D. Stevens Jr. 1978. Pulmonary carcinogenesis in rats given implants of cigarette smoke condensate in beeswax pellets. Journal of the National Cancer Institute 61(3):905.
DeBethizy, J., M. Borgerding, D. Doolittle, J. Robinson, K. McManus, C. Rahn, R. Davis, and G. Burger. 1990. Chemical and biological studies of a cigarette that heats rather than burns tobacco. The Journal of Clinical Pharmacology 30(8):755.
Ding, Y. S., X. J. Yan, R. B. Jain, E. Lopp, A. Tavakoli, G. M. Polzin, S. B. Stanfill, D. L. Ashley, and C. H. Watson. 2006. Determination of 14 polycyclic aromatic hydrocarbons in mainstream smoke from U.S. Brand and non-U.S. Brand cigarettes. Environmental Science and Technology 40(4):1133-1138.
Ding, Y. S., D. L. Ashley, and C. H. Watson. 2007. Determination of 10 carcinogenic polycyclic aromatic hydrocarbons in mainstream cigarette smoke. Journal of Agricultural and Food Chemistry 55(15):5966-5973.
Ding, Y. S., B. C. Blount, L. Valentin-Blasini, H. AS. Applewhite, Y. CIA, C. AH. Watson, and D. L. Ashley. 2009. Simultaneous determination of six mercapturic acid metabolites of volatile organic compounds in human urine. Chemical Research in Toxicology 22(6):1018-1025.
Diablo, J. A. 1962. Effect of tobacco diets on rodents. Nature 195:1316.
Diablo, J. A., and M. L. Levin. 1965. Tumor incidence in mice after oral painting with cigarette smoke condensate. Journal of the National Cancer Institute 34:595.
Djordjevic, M. V., J. Fan, L. P. Bush, K. D. Brunnemann, and D. Hoffann. 1993. Effects of storage conditions on levels of tobacco-specific N-nitrosamines and N-nitrosamino acids in U.S. moist snuff. Journal of Agricultural and Food Chemistry 41(10):1790-1794.
Dontenwill, W., H. Chevalier, H. Harke, U. Lafrenz, G. Reckzeh, and B. Schneider. 1973. Investigations on the effects of chronic cigarette-smoke inhalation in Syrian golden hamsters. Journal of the National Cancer Institute 51(6):1781.
Doolittle, D., C. Rahn, G. Burger, R. Davis, J. deBethizy, G. Howard, C. Lee, S. McKarns, E. Riccio, and J. Robinson. 1989. Human urine mutagenicity study comparing cigarettes which burn or only heat tobacco. Mutation Research/Genetic Toxicology 223(2):221-232.
Dunham, L. J., C. S. Muir, and J. E. Hamner, 3rd. 1966. Epithelial atypia in hamster cheek pouches treated repeatedly with calcium hydroxide. British Journal of Cancer 20(3):588-593.
Dunham, L. J., K. C. Snell, and H. L. Stewart. 1975. Argyrophilic carcinoids in two Syrian hamsters (mesocricetus auratus). Journal of the National Cancer Institute 54(2):507-513.
FDA (Food and Drug Administration). 2009. Guidance for industry: End-of-phase 2a meetings. http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm079690.pdf (accessed November 28, 2011).
FDA. 2011a. Harmful and potentially harmful constituents in tobacco products and tobacco smoke. Federal Register 76(156).
FDA. 2011b. FDA and NIH announce joint study on tobacco use and risk perceptions. http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm274626.htm (accessed November 28, 2011).
FDA. 2011c. Draft initial list of harmful/potentially harmful constituents in tobacco smoke or smokeless tobacco products. http://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/TobaccoProductsScientificAdvisoryCommittee/UCM221804.pdf (accessed October 26, 2011).
Federal Trade Commission. 2007. Smokeless tobacco reports for the years 2002-2005. Washington, DC: Federal Trade Commission.
Feng, S., H. J. Roethig, Q. Liang, R. Kinser, Y. Jin, G. Scherer, M. Urban, J. Engl, and K. Riedel. 2006a. Evaluation of urinary 1-hydroxypyrene, S-phenylmercapturic acid, trans,transmuconic acid, 3-methyladenine, 3-ethyladenine, 8-hydroxy-2’-deoxyguanosine and thioethers as biomarkers of exposure to cigarette smoke. Biomarkers 11(1):28-52.
Feng, Z., W. Hu, Y. Hu, and M. S. Tang. 2006b. Acrolein is a major cigarette-related lung cancer agent. Preferential binding at p53 mutational hotspots and inhibition of DNA repair. Proceedings of the National Academy of Sciences 103:15404-15409.
Fisher, M. A., G. W. Taylor, and K. R. Tilashalski. 2005. Smokeless tobacco and severe active periodontal disease, NHANES III. Journal of Dental Research 84(8):705-710.
Fleming, T. R., and D. L. DeMets. 1996. Surrogate end points in clinical trials: Are we being misled? Annals of Internal Medicine 125(7):605-613.
Forster, K., R. Preuss, B. Rossbach, T. Bruning, J. Angerer, and P. Simon. 2008. 3-hydroxybenzo[a]pyrene in the urine of workers with occupational exposure to polycyclic aromatic hydrocarbons in different industries. Occupational and Environmental Medicine 65(4):224-229.
Frost-Pineda, K., Q. Liang, J. Liu, L. Rimmer, Y. Jin, S. Feng, S. Kapur, P. Mendes, H. Roethig, and M. Sarkar. 2011. Biomarkers of potential harm among adult smokers and non-smokers in the total exposure study. Nicotine & Tobacco Research 13(3):182-193.
Furie, M. B., J. A. Raffanello, E. I. Gergel, T. J. Lisinski, and L. D. Horb. 2000. Extracts of smokeless tobacco induce pro-inflammatory changes in cultured human vascular endothelial cells. Immunopharmacology 47(1):13-23.
Gargus, J. L., M. B. H. Powers, R.T., and J. R. Everly. 1976. Mouse dermal bioassays of cigarette smoke condensates. In Report no. 1. Toward less hazardous cigarettes. The first set of experimental cigarettes. Vol. DHEW Publ. No. (NIH) 76-905, edited by G. B. Gori. Washington, DC: Department of Health, Education, and Welfare, Public Health Service, National Institutes of Health, National Cancer Institute. Pp. 85-94.
Gijare, P. S., K. V. Rao, and S. V. Bhide. 1989. Effects of tobacco-specific nitrosamines and snuff extract on cell proliferation and activities of ornithine decarboxylase and aryl hydrocarbon hydroxylase in mouse tongue primary epithelial cell cultures. Journal of Cancer Research and Clinical Oncology 115(6):558-563.
Goniewicz, M. L., M. D. Eisner, E. Lazcano-Ponce, W. Zielinska-Danch, B. Koszowski, A. Sobczak, C. Havel, P. Jacob, and N. L. Benowitz. 2011. Comparison of urine cotinine and the tobacco-specific nitrosamine metabolite 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) and their ratio to discriminate active from passive smoking. Nicotine & Tobacco Research 13(3):202-208.
Gori, G. B. 1976. Report on the second set of experimental cigarettes. In Report no. 2. Toward less hazardous cigarettes. The second set of experimental cigarettes. Vol. DHEW Publ. No. (NIH) 76-1111, edited by G. B. Gori. Washington, DC: Department of Health, Education, and Welfare, Public Health Service, National Institutes of Health, National Cancer Institute. Pp. 4-15.
Graham, E. A., A. B. Croninger, and E. L. Wynder. 1957. Experimental production of carcinoma with cigarette tar. Cancer Research 17(11):1058-1066.
Grasso, P., and A. H. Mann. 1998. Smokeless tobacco and oral cancer: An assessment of evidence derived from laboratory animals. Food and Chemical Toxicology 36(11):1015-1029.
Gregg, E., C. Hill, M. Hollywood, M. Kearney, K. McAdam, D. McLaughlin, S. Purkis, and M. Williams. 2004. The UK smoke constituents testing study. Summary of results and comparison with other studies. Beitrage zur Tabakforschung International 21:117-138.
Gregory, R. L., and L. E. Gfell. 1996. Effect of nicotine on secretory component synthesis by secretory epithelial cells. Clinical and Diagnostic Laboratory Immunology 3(5):578-583.
Guarino, F., G. Cantarella, M. Caruso, C. Russo, S. Mancuso, G. Arcidiacono, R. R. Cacciola, R. Bernardini, and R. Polosa. 2011. Endothelial activation and injury by cigarette smoke exposure. Journal of Biological Regulators and Homeostatic Agents 25(2):259-268.
Hakura, A., H. Shimada, M. Nakajima, H. Sui, S. Kitamoto, S. Suzuki, and T. Satoh. 2005. Salmonella/human S9 mutagenicity test: A collaborative study with 58 compounds. Mutagenesis 20(3):217-228.
Hammond, D., and R. J. O’Connor. 2008. Constituents in tobacco and smoke emissions from Canadian cigarettes. Tobacco Control 17(Suppl. 1):i24-i31.
Harvey, C. J., R. K. Thimmulappa, S. Sethi, X. Kong, L. Yarmus, R. H. Brown, D. Feller-Kopman, R. Wise, and S. Biswal. 2011. Targeting Nrf2 signaling improves bacterial clearance by alveolar macrophages in patients with COPD and in a mouse model. Science Translational Medicine 3(78):78ra32.
Hasseus, B., M. Wallstrom, B.-G. Osterdahl, J.-M. Hirsch, and M. Jontell. 1997. Immunotoxic effects of smokeless tobacco on the accessory cell function of rat oral epithelium. European Journal of Oral Sciences 105(1):45-51.
Hatsukami, D. K., C. Lemmonds, Y. Zhang, S. E. Murphy, C. Le, S. G. Carmella, and S. S. Hecht. 2004. Evaluation of carcinogen exposure in people who used “reduced exposure” tobacco products. Journal of the National Cancer Institute 96(11):844-852.
Hatsukami, D. K., N. L. Benowitz, S. I. Rennard, C. Oncken, and S. S. Hecht. 2006. Biomarkers to assess the utility of potential reduced exposure tobacco products. Nicotine & Tobacco Research 8(2):169-191.
Hatsukami, D. K., R. M. Feuer, J. O. Ebbert, I. Stepanov, and S. S. Hecht. 2007a. Changing smokeless tobacco products: New tobacco delivery systems. American Journal of Preventive Medicine 33:S368-S378.
Hatsukami, D. K., A. M. Joseph, M. LeSage, J. Jensen, S. E. Murphy, P. Pentel, M. Kotlyar, E. Borgida, C. Le, and S. S. Hecht. 2007b. The science base for reducing tobacco harm. Nicotine & Tobacco Research 9:S537-S553.
Hatsukami, D. K., K. Hanson, A. Briggs, M. Parascandola, J. M. Genkinger, R. O’Connor, and P. G. Shields. 2009. Clinical trials methods for evaluation of potential reduced exposure products. Cancer Epidemiology, Biomarkers & Prevention 18(12):3143-3195.
Hatsukami, D. K., M. Kotlyar, L. A. Hertsgaard, Y. Zhang, S. G. Carmella, J. A. Jensen, P. G. Shields, S. E. Murphy, I. Stepanov, and S. S. Hecht. 2010. Reduced nicotine content cigarettes: Effects on toxicant exposure, dependence and cessation. Addiction 105:343-355.
Hautamaki, R. D., D. K. Kobayashi, R. M. Senior, and S. D. Shapiro. 1997. Requirement for macrophage elastase for cigarette smoke-induced emphysema in mice. Science 277(5334):2002-2004.
Hecht, S. S. 1999. Tobacco smoke carcinogens and lung cancer. Journal of the National Cancer Institute 91:1194-1210.
Hecht, S. S. 2002. Human urinary carcinogen metabolites: Biomarkers for investigating tobacco and cancer. Carcinogenesis 23:907-922.
Hecht, S. S. 2003a. Tobacco carcinogens, their biomarkers, and tobacco-induced cancer. Nature Reviews. Cancer 3:733-744.
Hecht, S. S. 2003b. Carcinogen derived biomarkers: Applications in studies of human exposure to secondhand tobacco smoke. Tobacco Control 13(Suppl. 1):i48-i56.
Hecht, S. S. 2008. Progress and challenges in selected areas of tobacco carcinogenesis. Chemical Research in Toxicology 21:160-171.
Hecht, S. S. 2010. Tobacco smoke carcinogens and lung cancer. In Chemical carcinogenesis, edited by T. M. Penning. New York: Springer.
Hecht, S. S., A. Rivenson, J. Braley, J. DiBello, J. D. Adams, and D. Hoffmann. 1986. Induction of oral cavity tumors in F344 rats by tobacco-specific nitrosamines and snuff. Cancer Research 46(8):4162-4166.
Hecht, S. S., M. Chen, A. Yoder, J. Jensen, D. Hatsukami, C. Le, and S. G. Carmella. 2005. Longitudinal study of urinary phenanthrene metabolite ratios: Effect of smoking on the diol epoxide pathway. Cancer Epidemiology, Biomarkers & Prevention 14(12):2969-2974.
Hecht, S. S., J. M. Yuan, and D. Hatsukami. 2010. Applying tobacco carcinogen and toxicant biomarkers in product regulation and cancer prevention. Chemical Research in Toxicology 23(6):1001-1008.
Herrold, K. M., and L. J. Dunham. 1962. Induction of carcinoma and papilloma of the Syrian hamster by intratracheal instillation of benzo[a]-pyrene. Journal of the National Cancer Institute 28:467-491.
HHS (U.S. Department of Health and Human Services). 2004a. The health consequences of smoking: A report of the Surgeon General. Rockville, MD: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, Center for Health Promotion and Education, Office on Smoking and Health.
HHS. 2004b. Report on carcinogens, 11th edition. Research Triangle Park, NC: HHS, National Toxicology Program.
HHS. 2006. The health consequences of involuntary exposure to tobacco smoke: A report of the Surgeon General. Washington, DC: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health.
HHS. 2010. How tobacco smoke causes disease: The biology and behavioral basis for smoking-attributable disease: A report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health.
Hirsch, J. M., and S. L. Johansson. 1983. Effect of long-term application of snuff on the oral mucosa: An experimental study in the rat. Journal of Oral Pathology 12(3):187-198.
Hirsch, J. M., and H. Thilander. 1981. Snuff-induced lesions of the oral mucosa—an experimental model in the rat. Journal of Oral Pathology 10(5):342-353.
Hirsch, J. M., S. L. Johansson, and A. Vahlne. 1984. Effect of snuff and herpes simplex virus-1 on rat oral mucosa: Possible associations with the development of squamous cell carcinoma. Journal of Oral Pathology 13(1):52-62.
Hoffmann, D., and J. D. Adams. 1981. Carcinogenic tobacco-specific N-nitrosamines in snuff and in the saliva of snuff dippers. Cancer Research 41(11 Pt 1):4305-4308.
Hoffmann, D., and E. L. Wynder. 1971. A study of tobacco carcinogenesis. XI. Tumor initiators, tumor accelerators, and tumor promoting activity of condensate fractions. Cancer 27(4):848-864.
Hoffmann, D., A. Castonguay, A. Rivenson, and S. S. Hecht. 1981. Comparative carcinogenicity and metabolism of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone and N’-nitrosonornicotine in Syrian golden hamsters. Cancer Research 41(6):2386-2393.
Hoffmann, D., A. Rivenson, and S. S. Hecht. 1992. Carcinogenesis of smokeless tobacco. In NCI smoking and tobacco control monograph. Vol. 92-3461. Bethesda, MD: National Cancer Institute, National Institutes of Health.
Hoffmann, K., K. Becker, C. Friedrich, D. Helm, C. Krause, and B. Seifert. 2000. The German environmental survey 1990/1992 (GerES II): Cadmium in blood, urine and hair of adults and children. Journal of Exposure Analysis and Environmental Epidemiology 10:126-135.
Homburger, F. 1971. Mechanical irritation, polycyclic hydrocarbons, and snuff. Effects on facial skin, cheek pouch, and oral mucosa in Syrian hamsters. Archives of Pathology 91(5):411-417.
Hukkanen, J., P. Jacob, III, and N. L. Benowitz. 2005. Metabolism and disposition kinetics of nicotine. Pharmacological Reviews 57(1):79-115.
Hutt, J. A., B. R. Vuillemenot, E. B. Barr, M. J. Grimes, F. F. Hahn, C. H. Hobbs, T. H. March, A. P. Gigliotti, S. K. Seilkop, and G. L. Finch. 2005. Life-span inhalation exposure to mainstream cigarette smoke induces lung cancer in B6C3F1 mice through genetic and epigenetic pathways. Carcinogenesis 26(11):1999.
Huvenne, W., E. A. Lanckacker, O. Krysko, K. R. Bracke, T. Demoor, P. W. Hellings, G. G. Brusselle, G. F. Joos, C. Bachert, and T. Maes. 2011. Exacerbation of cigarette smoke-induced pulmonary inflammation by Staphylococcus aureus enterotoxin B in mice. Respiratory Research 12:69.
IARC (International Agency for Research on Cancer). 1985. Tobacco habits other than smoking; betel-quid & areca-nut chewing; and some related nitrosamines. In IARC monographs on the evaluation of carcinogenic risks to humans. Vol. 37. Lyon, France: IARC.
IARC. 1987. Overall evaluations of carcinogenicity: An updating of IARC monographs volumes 1 to 42. In IARC monographs on the evaluation of the carcinogenic risk of chemicals to humans, suppl. 7. Lyon, France: IARC. Pp. 91-122.
IARC. 1995. Acrolein. In IARC monographs on the evaluation of carcinogenic risks to humans. Vol. 63. Lyon, France: IARC. Pp. 337-391.
IARC. 1999a. Beryllium, cadmium, mercury, and exposures in the glass manufacturing industry. In IARC monographs on the evaluation of carcinogenic risks to humans. Vol. 58. Lyon, France: IARC. Pp. 119-237.
IARC. 1999b. Re-evaluation of some organic chemicals, hydrazine and hydrogen peroxide (part one). In IARC monographs on the evaluation of carcinogenic risks to humans. Vol. 71. Lyon, France: IARC. Pp. 43-108.
IARC. 2004. IARC monographs on the evaluation of carcinogenic risks to humans: Volume 83. Tobacco smoke and involuntary smoking. Lyon, France: IARC.
IARC. 2006. Formaldehyde, 2-butoxyethanol and 1-tert -butoxypropan-2-ol. In IARC monographs on the evaluation of carcinogenic risks to humans, vol. 88. Lyon, France: IARC. Pp. 37-325.
IARC. 2007. Smokeless tobacco and tobacco-specific nitrosamines. Lyon, France: IARC.
IARC. 2008. 1,3-butadiene, ethylene oxide and vinyl halides (vinyl fluoride, vinyl chloride and vinyl bromide). In IARC monographs on the evaluation of carcinogenic risks to humans, vol. 97. Lyon, France: IARC. Pp. 45-309.
IOM (Institute of Medicine). 2001. Clearing the smoke: Assessing the science base for tobacco harm reduction. Washington, DC: National Academy Press.
IOM. 2009. Initial national priorities for comparative effectiveness research. Washington, DC: The National Academies Press.
IOM. 2010. Evaluation of biomarkers and surrogate endpoints in chronic disease. Washington, DC: The National Academies Press.
Ismail, A. I., B. A. Burt, and S. A. Eklund. 1983. Epidemiologic patterns of smoking and periodontal disease in the United States. Journal of the American Dental Association 106(5):617-621.
Jacob, P., III, M. Wilson, and N. L. Benowitz. 2007. Determination of phenolic metabolites of polycyclic aromatic hydrocarbons in human urine as their pentafluorobenzyl ether derivatives using liquid chromatography-tandem mass spectrometry. Analytical Chemistry 79(2):587-598.
Jaju, R., R. Patel, S. Bakshi, A. Trivedi, B. Dave, and S. Adhvaryu. 1992. Chromosome damaging effects of pan masala. Cancer Letters 65(3):221-226.
Jansson, T., L. Romert, J. Magnusson, and D. Jenssen. 1991. Genotoxicity testing of extracts of a Swedish moist oral snuff. Mutation Research/Genetic Toxicology 261(2):101-115.
Johansson, S. L., J. M. Hirsch, P. A. Larsson, J. Saidi, and B. G. Osterdahl. 1989. Snuff-induced carcinogenesis: Effect of snuff in rats initiated with 4-nitroquinoline N-oxide. Cancer Research 49(11):3063.
Johansson, S. L., J. M. Hirsch, P. A. Larsson, J. Saidi, and B. G. osterdahl. 1991a. Lack of promoting ability of snuff in rats initiated with 4-nitroquinoline-N-oxide. IARC Scientific Publications (105):507-509.
Johansson, S. L., J. Saidi, B. G. Österdahl, and R. A. Smith. 1991b. Promoting effect of snuff in rats initiated by 4-nitroquinoline-N-oxide or 7, 12-dimethylbenz (a) anthracene. Cancer Research 51(16):4388.
Johnson, M. D., J. Schilz, M. V. Djordjevic, J. R. Rice, and P. G. Shields. 2009. Evaluation of in vitro assays for assessing the toxicity of cigarette smoke and smokeless tobacco. Cancer Epidemiology, Biomarkers & Prevention 18(12):3263-3304.
Jorquera, R., A. Castonguay, and H. M. Schuller. 1992. Placental transfer of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone instilled intratracheally in Syrian golden hamsters. Cancer Research 52(12):3273-3280.
Kang, M. J., C. G. Lee, J. Y. Lee, C. S. Dela Cruz, Z. J. Chen, R. Enelow, and J. A. Elias. 2008. Cigarette smoke selectively enhances viral pamp- and virus-induced pulmonary innate immune and remodeling responses in mice. Journal of Clinical Investigation 118(8):2771-2784.
Kavvadias, D., G. Scherer, F. Cheung, G. Errington, J. Shepperd, and M. McEwan. 2009a. Determination of tobacco-specific N-nitrosamines in urine of smokers and non-smokers. Biomarkers 14(8):547-553.
Kavvadias, D., G. Scherer, M. Urban, F. Cheung, G. Errington, J. Shepperd, and M. McEwan. 2009b. Simultaneous determination of four tobacco-specific N-nitrosamines (TSNA) in human urine. Journal of Chromatography B 877(11-12):1185-1192.
Kim, K. C., J. I. Rearick, P. Nettesheim, and A. M. Jetten. 1985. Biochemical characterization of mucous glycoproteins synthesized and secreted by hamster tracheal epithelial cells in primary culture. Journal of Biological Chemistry 260(7):4021-4027.
Klaunig, J. E. 2008. Acrylamide carcinogenicity. Journal of Agricultural and Food Chemistry 56(15):5984-5988.
Kotlyar, M., L. A. Hertsgaard, B. R. Lindgren, J. A. Jensen, S. G. Carmella, I. Stepanov, S. E. Murphy, S. S. Hecht, and D. K. Hatsukami. 2011. Effect of oral snus and medicinal nicotine in smokers on toxicant exposure and withdrawal symptoms: A feasibility study. Cancer Epidemiology, Biomarkers & Prevention 20(1):91-100.
Kriebel, D., J. Henry, J. Gold, A. Bronsdon, and B. Commoner. 1985. The mutagenicity of cigarette smokers’ urine. Journal of Environmental Pathology, Toxicology and Oncology 6(2):157.
Kuenemann-Migeot, C., F. Callais, I. Momas, and B. Festy. 1996. Urinary promutagens of smokers: Comparison of concentration methods and relation to cigarette consumption. Mutation Research/Genetic Toxicology 368(2):141-147.
Larsson, P. A., S. L. Johansson, A. Vahlne, and J. M. Hirsch. 1989. Snuff tumorigenesis: Effects of long-term snuff administration after initiation with 4-nitroquinoline-N-oxide and herpes simplex virus type 1. Journal of Oral Pathology and Medicine 18(4):187-192.
Leslie, E. M., G. Ghibellini, K. Nezasa, and K. L. Brouwer. 2007. Biotransformation and transport of the tobacco-specific carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) in bile duct-cannulated wild-type and Mrp2/Abcc2-deficient (TR-) wistar rats. Carcinogenesis 28(12):2650-2656.
Levin, M. 1953. The occurrence of lung cancer in man. Acta-Unio Internationalis Contra Cancrum 9(3):531.
Lin, H., D. M. Carlson, J. A. St. George, C. G. Plopper, and R. Wu. 1989. An ELISA method for the quantitation of tracheal mucins from human and nonhuman primates. American Journal of Respiratory Cell and Molecular Biology 1(1):41-48.
Lindemann, R. A., and N. H. Park. 1988. Inhibition of human lymphokine-activated killer activity by smokeless tobacco (snuff) extract. Archives of Oral Biology 33(5):317-321.
Liu, J., Q. Liang, K. Frost-Pineda, R. Muhammad-Kah, L. Rimmer, H. Roethig, P. Mendes, and M. Sarkar. 2011. Relationship between biomarkers of cigarette smoke exposure and biomarkers of inflammation, oxidative stress, and platelet activation in adult cigarette smokers. Cancer Epidemiology, Biomarkers & Prevention 20(8):1760-1769.
Lowe, F. J., E. O. Gregg, and M. McEwan. 2009. Evaluation of biomarkers of exposure and potential harm in smokers, former smokers and never-smokers. Clinical Chemistry and Laboratory Medicine 47(3):311-320.
Ma, B., M. J. Kang, C. G. Lee, S. Chapoval, W. Liu, Q. Chen, A. J. Coyle, J. M. Lora, D. Picarella, R. J. Homer, and J. A. Elias. 2005. Role of CCR5 in IFN-^-induced and cigarette smoke-induced emphysema. Journal of Clinical Investigation 115(12):3460-3472.
Malhotra, D., R. Thimmulappa, A. Navas-Acien, A. Sandford, M. Elliott, A. Singh, L. Chen, X. Zhuang, J. Hogg, P. Pare, R. M. Tuder, and S. Biswal. 2008. Decline in NRF2-regulated antioxidants in chronic obstructive pulmonary disease lungs due to loss of its positive regulator, DJ-1. American Journal of Respiratory and Critical Care Medicine 178(6):592-604.
Mallia, P., S. D. Message, V. Gielen, M. Contoli, K. Gray, T. Kebadze, J. Aniscenko, V. Laza-Stanca, M. R. Edwards, L. Slater, A. Papi, L. A. Stanciu, O. M. Kon, M. Johnson, and S. L. Johnston. 2011. Experimental rhinovirus infection as a human model of chronic obstructive pulmonary disease exacerbation. American Journal of Respiratory and Critical Care Medicine 183(6):734-742.
Mangipudy, R. S., and J. K. Vishwanatha. 1999. Role of nitric oxide in the induction of apoptosis by smokeless tobacco extract. Molecular and Cellular Biochemistry 200(1-2):51-57.
Mariner, D. C., M. Ashley, C. J. Shepperd, G. Mullard, and M. Dixon. 2010. Mouth level smoke exposure using analysis of filters from smoked cigarettes: A study of eight countries. Regulatory Toxicology and Pharmacology.
Mauderly, J. L., W. E. Bechtold, J. A. Bond, A. L. Brooks, B. T. Chen, R. G. Cuddihy, J. R. Harkema, R. F. Henderson, N. F. Johnson, K. Rithidech, et al. 1989. Comparison of 3 methods of exposing rats to cigarette smoke. Experimental Pathology 37(1-4):194-197.
Mauderly, J. L., A. P. Gigliotti, E. B. Barr, W. E. Bechtold, S. A. Belinsky, F. F. Hahn, C. A. Hobbs, T. H. March, S. K. Seilkop, and G. L. Finch. 2004. Chronic inhalation exposure to mainstream cigarette smoke increases lung and nasal tumor incidence in rats. Toxicological Sciences 81(2):280.
McElroy, J. A., M. M. Shafer, A. Trentham-Dietz, J. M. Hampton, and P. A. Newcomb. 2007. Urinary cadmium levels and tobacco smoke exposure in women age 20-69 years in the United States. Journal of Toxicology and Environmental Health. Part A 70(20):1779-1782.
Melikian, A. A., M. V. Djordjevic, S. Chen, J. Richie, and S. D. Stellman. 2007. Effect of delivered dosage of cigarette smoke toxins on the levels of urinary biomarkers of exposure. Cancer Epidemiology, Biomarkers & Prevention 16(7):1408-1415.
Mendes, P., S. Kapur, J. Wang, S. Feng, and H. Roethig. 2008. A randomized, controlled exposure study in adult smokers of full flavor Marlboro cigarettes switching to Marlboro lights or Marlboro ultra lights cigarettes. Regulatory Toxicology and Pharmacology 51(3):295-305.
Merne, M., K. Heikinheimo, I. Saloniemi, and S. Syrjanen. 2004. Effects of snuff extract on epithelial growth and differentiation in vitro. Oral Oncology 40(1):6-12.
Morin, A., C. J. Shepperd, A. C. Eldridge, N. Poirier, and R. Voisine. 2011. Estimation and correlation of cigarette smoke exposure in Canadian smokers as determined by filter analysis and biomarkers of exposure. Regulatory Toxicology and Pharmacology 61(3, Supplement):S3-S12.
Mortaz, E., A. D. Kraneveld, J. J. Smit, M. Kool, B. N. Lambrecht, S. L. Kunkel, N. W. Lukacs, F. P. Nijkamp, and G. Folkerts. 2009. Effect of cigarette smoke extract on dendritic cells and their impact on T-cell proliferation. PLoS One 4(3):e4946.
Muns, G., J. K. Vishwanatha, and I. Rubinstein. 1994. Effects of smokeless tobacco on chemically transformed hamster oral keratinocytes: Role of angiotensin I-converting enzyme. Carcinogenesis 15(7):1325-1327.
Naufal, Z. S., K. M. Marano, S. J. Kathman, and C. L. Wilson. 2011. Differential exposure biomarker levels among cigarette smokers and smokeless tobacco consumers in the National Health and Nutrition Examination Survey 1999-2008. Biomarkers 16(3):222-235.
NCI (National Cancer Institute). 1998. Cigars: Health effects and trends, Smoking and tobacco control monograph no. 9. Bethesda, MD: U.S. Department of Health and Human Services, National Institutes of Health. NIH Pub. No. 98-4645.
NCI. 2001. Risks associated with smoking cigarettes with low machine-measured yields of tar and nicotine, Smoking and tobacco control monograph no. 13. Bethesda, MD: U.S. Department of Health and Human Services, National Institutes of Health, National Cancer Institute. NIH Pub. No. 99-4645.
NCI. 2009. Phenotypes and endophenotypes: Foundations for genetic studies of nicotine use and dependence. Bethesda, MD: U.S. Department of Health and Human Services, National Institutes of Health, National Cancer Institute.
Nelson, D. E., P. Mowery, S. Tomar, S. Marcus, G. Giovino, and L. Zhao. 2006. Trends in smokeless tobacco use among adults and adolescents in the United States. American Journal of Public Health 96(5):897-905.
Niphadkar, M. P., A. N. Bagwe, and R. A. Bhisey. 1996. Mutagenic potential of Indian tobacco products. Mutagenesis 11(2):151.
Pakhale, S., S. Sarkar, K. Jayant, and S. Bhide. 1988. Carcinogenicity of Indian bidi and cigarette smoke condensate in Swiss albino mice. Journal of Cancer Research and Clinical Oncology 114(6):647-649.
Palladino, G., J. D. Adams, K. D. Brunnemann, N. J. Haley, and D. Hoffmann. 1986. Snuffdipping in college students: A clinical profile. Military Medicine 151(6):342-346.
Pankow, J. F., A. D. Tavakoli, W. Luo, and L. M. Isabelle. 2003. Percent free base nicotine in the tobacco smoke particulate matter of selected commercial and reference cigarettes. Chemical Research in Toxicology 16(8):1014-1018.
Papageorge, M. B., E. Cataldo, and E. G. Jahngen. 1996. The effect of N-nitrosonornicotine on the buccal mucosa of Syrian hamsters. Journal of Oral and Maxillofacial Surgery 54(2):187-190.
Park, N. H., J. P. Sapp, and E. G. Herbosa. 1986. Oral cancer induced in hamsters with herpes simplex infection and simulated snuff dipping. Oral Surgery, Oral Medicine, Oral Pathology 62(2):164-168.
Paschal, D. C., V. Burt, S. P. Caudill, E. W. Gunter, J. L. Pirkle, E. J. Sampson, D. T. Miller, and R. J. Jackson. 2000. Exposure of the U.S. population aged 6 years and older to cadmium: 1988-1994. Archives of Environmental Contamination and Toxicology 38:377-383.
Patel, R. K., R. J. Jaju, S. R. Bakshi, A. H. Trivedi, B. J. Dave, and S. G. Adhvaryu. 1994. Pan masala—a genotoxic menace. Mutation Research/Genetic Toxicology 320(3):245-249.
Pauly, J. L., R. J. O’Connor, G. M. Paszkiewicz, K. M. Cummings, M. V. Djordjevic, and P. G. Shields. 2009. Cigarette filter-based assays as proxies for toxicant exposure and smoking behavior—a literature review. Cancer Epidemiology, Biomarkers & Prevention 18(12):3321-3333.
Peacock, E. E., Jr., and B. W. Brawley. 1959. An evaluation of snuff and tobacco in the production of mouth cancer. Plastic and Reconstructive Surgery and the Transplantation Bulletin 23(6):628-635.
Peacock, E. E., Jr., B. G. Greenberg, and B. W. Brawley. 1960. The effect of snuff and tobacco on the production of oral carcinoma: An experimental and epidemiological study. Annals of Surgery 151:542-550.
Peluffo, G., P. Calcerrada, L. Piacenza, N. Pizzano, and R. Radi. 2009. Superoxide-mediated inactivation of nitric oxide and peroxynitrite formation by tobacco smoke in vascular endothelium: Studies in cultured cells and smokers. American Journal of Physiology—Heart and Circulatory Physiology 296(6):H1781-H1792.
Phillips, D. H. 2002. Smoking-related DNA and protein adducts in human tissues. Carcinogenesis 23(12):1979-2004.
Prentice, R. L. 1989. Surrogate endpoints in clinical trials: Definition and operational criteria. Statistics in Medicine 8(4):431-440.
Putzrath, R. M., D. Langley, and E. Eisenstadt. 1981. Analysis of mutagenic activity in cigarette smokers’ urine by high performance liquid chromatography. Mutation Research/Environmental Mutagenesis and Related Subjects 85(3):97-108.
Rainey, C. L., P. A. Conder, and J. V. Goodpaster. 2011. Chemical characterization of dissolvable tobacco products promoted to reduce harm. Journal of Agricultural and Food Chemistry 59(6):2745-2751.
Rangasamy, T., C. Y. Cho, R. K. Thimmulappa, L. Zhen, S. S. Srisuma, T. W. Kensler, M. Yamamoto, I. Petrache, R. M. Tuder, and S. Biswal. 2004. Genetic ablation of Nrf2 enhances susceptibility to cigarette smoke-induced emphysema in mice. Journal of Clinical Investigation 114(9):1248-1259.
Reilly, K. B., S. Srinivasan, M. E. Hatley, M. K. Patricia, J. Lannigan, D. T. Bolick, G. Vandenhoff, H. Pei, R. Natarajan, J. L. Nadler, and C. C. Hedrick. 2004. 12/15-lipoxygenase activity mediates inflammatory monocyte/endothelial interactions and atherosclerosis in vivo. Journal of Biological Chemistry 279(10):9440-9450.
Richter, P., and F. W. Spierto. 2003. Surveillance of smokeless tobacco nicotine, pH, moisture, and unprotonated nicotine content. Nicotine & Tobacco Research 5(6):885-889.
Richter, P., K. Hodge, S. Stanfill, L. Zhang, and C. Watson. 2008. Surveillance of moist snuff: Total nicotine, moisture, pH, un-ionized nicotine, and tobacco-specific nitrosamines. Nicotine & Tobacco Research 10(11):1645-1652.
Rickert, W. S., P. J. Joza, M. Sharifi, J. Wu, and J. H. Lauterbach. 2008. Reductions in the tobacco specific nitrosamine (TSNA) content of tobaccos taken from commercial Canadian cigarettes and corresponding reductions in TSNA deliveries in mainstream smoke from such cigarettes. Regulatory Toxicology and Pharmacology 51(3):306-310.
Rodgman, A., and T. Perfetti. 2009. The chemical components of tobacco and tobacco smoke. Boca Raton, FL: CRC Press.
Roemer, E., R. Stabbert, K. Rustemeier, D. J. Veltel, T. J. Meisgen, W. Reininghaus, R. A. Carchman, C. L. Gaworski, and K. F. Podraza. 2004. Chemical composition, cytotoxicity and mutagenicity of smoke from U.S. commercial and reference cigarettes smoked under two sets of machine smoking conditions. Toxicology 195(1):31-52.
Roethig, H. J., B. K. Zedler, R. D. Kinser, S. Feng, B. L. Nelson, and Q. Liang. 2007. Short-term clinical exposure evaluation of a second-generation electrically heated cigarette smoking system. Journal of Clinical Pharmacology 47(4):518-530.
Roethig, H. J., S. Munjal, S. Feng, Q. Liang, M. Sarkar, R. A. Walk, and P. E. Mendes. 2009. Population estimates for biomarkers of exposure to cigarette smoke in adult U.S. cigarette smokers. Nicotine & Tobacco Research 11(10):1216-1225.
Rohatgi, N., J. Kaur, A. Srivastava, and R. Ralhan. 2005. Smokeless tobacco (khaini) extracts modulate gene expression in epithelial cell culture from an oral hyperplasia. Oral Oncology 41(8):806-820.
Rowland, R., and K. Harding. 1999. Increased sister chromatid exchange in the peripheral blood lymphocytes of young women who smoke cigarettes. Hereditas 131(2):143-146.
Rundle, A., and H. Ahsan. 2008. Molecular epidemiological studies that can be nested within cohorts. In Molecular epidemiology of chronic diseases, edited by C. Wild, P. Vineis, and S. Garte. West Sussex, England: J.W. Wiley. Pp. 23-37.
SAMHSA (Substance Abuse and Mental Health Services Administration). 2007. Results from the 2006 national survey on drug use and health: National findings. DHHS Publication No. SMA 07-4293. Rockville, MD: Substance Abuse and Mental Health Services Administration.
SAMHSA. 2011. Results from the 2010 national survey on drug use and health: National findings. NSDUH Series H-41, HHS Publication No. (SMA) 11-4658. Rockville, MD: Substance Abuse and Mental Health Services Administration.
Sarkar, M., S. Kapur, K. Frost-Pineda, S. Feng, J. Wang, Q. Liang, and H. Roethig. 2008. Evaluation of biomarkers of exposure to selected cigarette smoke constituents in adult smokers switched to carbon-filtered cigarettes in short-term and long-term clinical studies. Nicotine & Tobacco Research 10(12):1761-1772.
Sarto, F., M. Faccioli, I. Cominato, and A. Levis. 1985. Aging and smoking increase the frequency of sister-chromatid exchanges (SCE) in man. Mutation Research Letters 144(3):183-187.
Scherer, G. 2005. Biomonitoring of inhaled complex mixtures—ambient air, diesel exhaust and cigarette smoke. Experimental and Toxicologic Pathology 57(Suppl. 1):75-110.
Scherer, G. 2006. Carboxyhemoglobin and thiocyanate as biomarkers of exposure to carbon monoxide and hydrogen cyanide in tobacco smoke. Experimental and Toxicologic Pathology 58(2-3):101-124.
Scherer, G., M. Urban, J. Engl, H. W. Hagedorn, and K. Riedel. 2006. Influence of smoking charcoal filter tipped cigarettes on various biomarkers of exposure. Inhalation Toxicology 18(10):821-829.
Scherer, G., J. Engl, M. Urban, G. Gilch, D. Janket, and K. Riedel. 2007a. Relationship between machine-derived smoke yields and biomarkers in cigarette smokers in Germany. Regulatory Toxicology and Pharmacology 47(2):171-183.
Scherer, G., M. Urban, H. W. Hagedorn, S. Feng, R. D. Kinser, M. Sarkar, Q. Liang, and H. J. Roethig. 2007b. Determination of two mercapturic acids related to crotonaldehyde in human urine: Influence of smoking. Human & Experimental Toxicology 26(1):37-47.
Scherer, G., M. Urban, H. W. Hagedorn, R. Serafin, S. Feng, S. Kapur, R. Muhammad, Y. Jin, M. Sarkar, and H. J. Roethig. 2010. Determination of methyl-, 2-hydroxyethyl- and 2-cyanoethylmercapturic acids as biomarkers of exposure to alkylating agents in cigarette smoke. Journal of Chromatography B 878(27):2520-2528.
Schuller, H. M., R. Jorquera, A. Reichert, and A. Castonguay. 1993. Transplacental induction of pancreas tumors in hamsters by ethanol and the tobacco-specific nitrosamine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone. Cancer Research 53(11):2498-2501.
Schuller, H. M., R. Jorquera, X. Lu, A. Riechert, and A. Castonguay. 1994. Transplacental carcinogenicity of low doses of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone administered subcutaneously or intratracheally to hamsters. Journal of Cancer Research and Clinical Oncology 120(4):200-203.
Schwartz, J., and X. Gu. 2002. Chapter 8: Hamster oral cancer model. In Tumor models in cancer research. Vol. 10, edited by B. A. Teicher. Totowa, NJ: Humana Press, Inc.
Schwartz, J. L., K. D. Brunnemann, A. J. Adami, S. Panda, S. C. Gordon, D. Hoffmann, and G. R. Adami. 2010. Brand specific responses to smokeless tobacco in a rat lip canal model. Journal of Oral Pathology and Medicine 39(6):453-459.
Secretan, B., K. Straif, R. Baan, Y. Grosse, F. El Ghissassi, V. Bouvard, L. Benbrahim-Tallaa, N. Guha, C. Freeman, L. Galichet, and V. Cogliano. 2009. A review of human carcinogens—Part E: Tobacco, areca nut, alcohol, coal smoke, and salted fish. Lancet Oncology 10(11):1033-1034.
Shirname-More, L. 1991. Smokeless tobacco extracts mutate human cells. Carcinogenesis 12(5):927-930.
Shklar, G., K. Niukian, M. Hassan, and E. G. Herbosa. 1985. Effects of smokeless tobacco and snuff on oral mucosa of experimental animals. Journal of Oral and Maxillofacial Surgery 43(2):80-86.
Singh, R., and H. S. Nalwa. 2011. Medical applications of nanoparticles in biological imaging, cell labeling, antimicrobial agents, and anticancer nanodrugs. Journal of Biomedical Nanotechnology 7(4):489-503.
Singh, S., Y. K. Loke, J. G. Spangler, and C. D. Furberg. 2011. Risk of serious adverse cardiovascular events associated with varenicline: A systematic review and meta-analysis. Canadian Medical Association Journal 183(12):1359-1366.
Smith, C., S. McKarns, R. Davis, S. Livingston, B. Bombick, J. Avalos, W. Morgan, and D. Doolittle. 1996. Human urine mutagenicity study comparing cigarettes which burn or primarily heat tobacco. Mutation Research/Environmental Mutagenesis and Related Subjects 361(1):1-9.
Song, S. Q., and D. L. Ashley. 1999. Supercritical fluid extraction and gas chromatography mass spectrometry for the analysis of tobacco-specific nitrosamines in cigarettes. Analytical Chemistry 71(7):1303-1308.
St. George, J. A., D. L. Cranz, S. C. Zicker, J. R. Etchison, D. L. Dungworth, and C. G. Plopper. 1985. An immunohistochemical characterization of rhesus monkey respiratory secretions using monoclonal antibodies. American Review of Respiratory Disease 132(3):556-563.
Stamm, S. C., B. Z. Zhong, W. Z. Whong, and T. Ong. 1994. Mutagenicity of coal-dust and smokeless-tobacco extracts in salmonella typhimurium strains with differing levels of O-acetyltransferase activities. Mutation Research 321(4):253-264.
Stanton, M., E. Miller, C. Wrench, and R. Blackwell. 1972. Experimental induction of epidermoid carcinoma in the lungs of rats by cigarette smoke condensate. Journal of the National Cancer Institute 49(3):867.
Starrett, W., and D. J. Blake. 2011. Sulforaphane inhibits de novo synthesis of IL-8 and MCP-1 in human epithelial cells generated by cigarette smoke extract. Journal of Immunotoxicology 8(2):150-158.
Stenstrom, B., C. M. Zhao, A. B. Rogers, H. O. Nilsson, E. Sturegard, S. Lundgren, J. G. Fox, T. C. Wang, T. M. Wadstrom, and D. Chen. 2007. Swedish moist snuff accelerates gastric cancer development in Helicobacter pylori-infected wild-type and gastrin transgenic mice. Carcinogenesis 28(9):2041-2046.
Stepanov, I., and S. S. Hecht. 2005. Tobacco-specific nitrosamines and their N-glucuronides in the urine of smokers and smokeless tobacco users. Cancer Epidemiology, Biomarkers & Prevention 14:885-891.
Stepanov, I., J. Jensen, D. Hatsukami, and S. S. Hecht. 2008. New and traditional smokeless tobacco: Comparison of toxicant and carcinogen levels. Nicotine & Tobacco Research 10:1773-1782.
Stepanov, I., S. G. Carmella, A. Briggs, L. Hertsgaard, B. Lindgren, D. K. Hatsukami, and S. S. Hecht. 2009. Presence of the carcinogen N’-nitrosonornicotine in the urine of some users of oral nicotine replacement therapy products. Cancer Research 69:8236-8240.
Stepanov, I., P. W. Villalta, A. Knezevich, J. Jensen, D. Hatsukami, and S. S. Hecht. 2010. Analysis of 23 polycyclic aromatic hydrocarbons in smokeless tobacco by gas chromatography-mass spectrometry. Chemical Research in Toxicology 23:66-73.
Stinn, W., J. H. E. Arts, A. Buettner, E. Duistermaat, K. Janssens, C. F. Kuper, and H. J. Haussmann. 2010. Murine lung tumor response after exposure to cigarette mainstream smoke or its particulate and gas/vapor phase fractions. Toxicology 275(1-3):10-20.
Straif, K., R. Baan, Y. Grosse, B. Secretan, F. El Ghissassi, and V. Cogliano. 2005. Carcinogenicity of polycyclic aromatic hydrocarbons. Lancet Oncology 6(12):931-932.
Sussan, T. E., T. Rangasamy, D. J. Blake, D. Malhotra, H. El-Haddad, D. Bedja, M. S. Yates, P. Kombairaju, M. Yamamoto, K. T. Liby, M. B. Sporn, K. L. Gabrielson, H. C. Champion, R. M. Tuder, T. W. Kensler, and S. Biswal. 2009. Targeting Nrf2 with the triterpenoid CDDO- imidazolide attenuates cigarette smoke-induced emphysema and cardiac dysfunction in mice. Proceedings of the National Academy of Sciences 106(1):250-255.
Suwan-ampai, P., A. Navas-Acien, P. T. Strickland, and J. Agnew. 2009. Involuntary tobacco smoke exposure and urinary levels of polycyclic aromatic hydrocarbons in the United States, 1999 to 2002. Cancer Epidemiology, Biomarkers & Prevention 18(3):884-893.
Thompson, C. A., and P. C. Burcham. 2008. Genome-wide transcriptional responses to acrolein. Chemical Research in Toxicology 21:2245-2256.
Trivedi, A. H., B. J. Dave, and S. G. Adhvaryu. 1993. Genotoxic effects of tobacco extract on Chinese hamster ovary cells. Cancer Letters 70(1-2):107-112.
Tuomisto, J., S. Kolonen, M. Sorsa, and P. Einisto. 1986. No difference between urinary mutagenicity in smokers of low-tar and medium-tar cigarettes: A double-blind crossover study. Archives of Toxicology. Supplement 9:115.
Wang, Y., E. Rotem, F. Andriani, and J. A. Garlick. 2001. Smokeless tobacco extracts modulate keratinocyte and fibroblast growth in organotypic culture. Journal of Dental Research 80(9):1862-1866.
Wang, D., Y. Wang, and Y. N. Liu. 2010. Experimental pulmonary infection and colonization of Haemophilus influenzae in emphysematous hamsters. Pulmonary Pharmacology and Therapeutics 23(4):292-299.
Weed, D. L. 2000. Epidemiologic evidence and causal inference. Hematology/Oncology Clinics of North America 14(4):797-807.
Weiss, N. S. 1994. Application of the case-control method in the evaluation of screening. Epidemiologic Reviews 16(1):102-108.
Whincup, P. H., J. A. Gilg, J. R. Emberson, M. J. Jarvis, C. Feyerabend, A. Bryant, M. Walker, and D. G. Cook. 2004. Passive smoking and risk of coronary heart disease and stroke: Prospective study with cotinine measurement. BMJ 329(7459):200-205.
Whitcutt, M. J., K. B. Adler, and R. Wu. 1988. A biphasic chamber system for maintaining polarity of differentiation of cultured respiratory tract epithelial cells. In Vitro Cellular and Developmental Biology 24(5):420-428.
Whong, W. Z., R. G. Ames, and T. M. Ong. 1984. Mutagenicity of tobacco snuff: Possible health implications for coal miners. Journal of Toxicology and Environmental Health 14(4):491-496.
Whong, W. Z., J. D. Stewart, and T. Ong. 1985. Formation of bacterial mutagens from the reaction of chewing tobacco with nitrite. Mutation Research/Genetic Toxicology 158(3):105-110.
Witschi, H. 2004. A/J mouse as a model for lung tumorigenesis caused by tobacco smoke: Strengths and weaknesses. Experimental Lung Research 31(1):3-18.
Worawongvasu, R., S. Ashrafi, A. Das, J. Waterhouse, and H. Medak. 1991. A light and scanning electron microscopic study of snuff induced early changes in hamster cheek pouch mucosa. Biomedical Research (India) 2:240-250.
Wu, R., E. Nolan, and C. Turner. 1985. Expression of tracheal differentiated functions in serum-free hormone-supplemented medium. Journal of Cellular Physiology 125(2):167-181.
Wu, R., C. G. Plopper, and P. W. Cheng. 1991. Mucin-like glycoprotein secreted by cultured hamster tracheal epithelial cells. Biochemical and immunological characterization. Biochemical Journal 277(3):713-718.
Wu, Z., P. Upadhyaya, S. G. Carmella, S. S. Hecht, and C. L. Zimmerman. 2002. Disposition of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) in bile duct-cannulated rats: Stereoselective metabolism and tissue distribution. Carcinogenesis 23(1):171-179.
Wynder, E. L., and D. Hoffmann. 1961. A study of tobacco carcinogenesis. VIII. The role of the acidic fractions as promoters. Cancer 14(6):1306-1315.
Wynder, E. L., E. A. Graham, and A. B. Croninger. 1953. Experimental production of carcinoma with cigarette tar. Cancer Research 13(12):855-864.
Yamasaki, E., and B. N. Ames. 1977. Concentration of mutagens from urine by absorption with the nonpolar resin XAD-2: Cigarette smokers have mutagenic urine. Proceedings of the National Academy of Sciences 74(8):3555.
Yildiz, D., Y. S. Liu, N. Ercal, and D. W. Armstrong. 1999. Comparison of pure nicotine- and smokeless tobacco extract-induced toxicities and oxidative stress. Archives of Environmental Contamination and Toxicology 37(4):434-439.
Yoshida, T., I. Mett, A. K. Bhunia, J. Bowman, M. Perez, L. Zhang, A. Gandjeva, L. Zhen, U. Chukwueke, T. Mao, A. Richter, E. Brown, H. Ashush, N. Notkin, A. Gelfand, R. K. Thimmulappa, T. Rangasamy, T. Sussan, G. Cosgrove, M. Mouded, S. D. Shapiro, I. Petrache, S. Biswal, E. Feinstein, and R. M. Tuder. 2010. Rtp801, a suppressor of mTOR signaling, is an essential mediator of cigarette smoke-induced pulmonary injury and emphysema. Nature Medicine 16(7):767-773.
Yuan, J. M., W. P. Koh, S. E. Murphy, Y. Fan, R. Wang, S. G. Carmella, S. Han, K. Wickham, Y. T. Gao, M. C. Yu, and S. S. Hecht. 2009. Urinary levels of tobacco-specific nitrosamine metabolites in relation to lung cancer development in two prospective cohorts of cigarette smokers. Cancer Research 69:2990-2995.
Zedler, B. K., R. Kinser, J. Oey, B. Nelson, H. J. Roethig, R. A. Walk, P. Kuhl, K. Rustemeier, G. Schepers, K. Von Holt, and A. R. Tricker. 2006. Biomarkers of exposure and potential harm in adult smokers of 3-7 mg tar yield (Federal Trade Commission) cigarettes and in adult non-smokers. Biomarkers 11(3):201-220.
Zhong, Y., S. G. Carmella, J. B. Hochalter, S. Balbo, and S. S. Hecht. 2010. Analysis of r -7, t -8,9, c-10-tetrahydroxy-7,8,9,10-tetrahydrobenzo[a]pyrene in human urine: A biomarker for directly assessing carcinogenic polycyclic aromatic hydrocarbon exposure plus metabolic activation. Chemical Research in Toxicology 24:73-80.
Zhou, J., E. A. Eksioglu, N. R. Fortenbery, X. Chen, H. Wang, P. K. Epling-Burnette, J. Y. Djeu, and S. Wei. 2011. Bone marrow mononuclear cells up-regulate toll-like receptor expression and produce inflammatory mediators in response to cigarette smoke extract. PLoS One 6(6):e21173.