Diagnosis and Management of Lung Cancer, 3rd ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines Online Only Articles |

Epidemiology of Lung CancerEpidemiology of Lung Cancer: Diagnosis and Management of Lung Cancer, 3rd ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines FREE TO VIEW

Anthony J. Alberg, PhD, MPH; Malcolm V. Brock, MD; Jean G. Ford, MD, MPH, FCCP; Jonathan M. Samet, MD, FCCP; Simon D. Spivack, MD, MPH
Author and Funding Information

From the Hollings Cancer Center (Dr Alberg) and the Department of Public Health Sciences, Medical University of South Carolina, Charleston, SC; Department of Surgery (Dr Brock), School of Medicine, Johns Hopkins University, Baltimore, MD; Department of Epidemiology (Dr Ford), Johns Hopkins Bloomberg School of Public Health, Baltimore, MD; Department of Preventive Medicine (Dr Samet), Keck School of Medicine, University of Southern California, Los Angeles, CA; and Division of Pulmonary Medicine (Dr Spivack), Department of Medicine, Albert Einstein College of Medicine, Bronx, NY.

Correspondence to: Anthony J. Alberg, PhD, MPH, Hollings Cancer Center, Medical University of South Carolina, 68 President St, MSC 955, Charleston, SC 29425; e-mail: alberg@musc.edu

Dr Ford is currently at the Department of Medicine, Brooklyn Hospital Center (Brooklyn, NY).

Funding/Sponsors: The overall process for the development of these guidelines, including matters pertaining to funding and conflicts of interest, are described in the methodology article.1 The development of this guideline was supported primarily by the American College of Chest Physicians. The lung cancer guidelines conference was supported in part by a grant from the Lung Cancer Research Foundation. The publication and dissemination of the guidelines was supported in part by a 2009 independent educational grant from Boehringer Ingelheim Pharmaceuticals, Inc.

COI grids reflecting the conflicts of interest that were current as of the date of the conference and voting are posted in the online supplementary materials.

Disclaimer: American College of Chest Physician guidelines are intended for general information only, are not medical advice, and do not replace professional medical care and physician advice, which always should be sought for any medical condition. The complete disclaimer for this guideline can be accessed at http://dx.doi.org/10.1378/chest.1435S1.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.

Chest. 2013;143(5_suppl):e1S-e29S. doi:10.1378/chest.12-2345
Text Size: A A A
Published online

Background:  Ever since a lung cancer epidemic emerged in the mid-1900s, the epidemiology of lung cancer has been intensively investigated to characterize its causes and patterns of occurrence. This report summarizes the key findings of this research.

Methods:  A detailed literature search provided the basis for a narrative review, identifying and summarizing key reports on population patterns and factors that affect lung cancer risk.

Results:  Established environmental risk factors for lung cancer include smoking cigarettes and other tobacco products and exposure to secondhand tobacco smoke, occupational lung carcinogens, radiation, and indoor and outdoor air pollution. Cigarette smoking is the predominant cause of lung cancer and the leading worldwide cause of cancer death. Smoking prevalence in developing nations has increased, starting new lung cancer epidemics in these nations. A positive family history and acquired lung disease are examples of host factors that are clinically useful risk indicators. Risk prediction models based on lung cancer risk factors have been developed, but further refinement is needed to provide clinically useful risk stratification. Promising biomarkers of lung cancer risk and early detection have been identified, but none are ready for broad clinical application.

Conclusions:  Almost all lung cancer deaths are caused by cigarette smoking, underscoring the need for ongoing efforts at tobacco control throughout the world. Further research is needed into the reasons underlying lung cancer disparities, the causes of lung cancer in never smokers, the potential role of HIV in lung carcinogenesis, and the development of biomarkers.

Lung cancer is the leading cause of cancer death in the world. In 2008, > 1.6 million people received a new diagnosis of lung cancer, comprising 13% of all new cancer diagnoses, and 1.4 million died of lung cancer, which was 18% of all cancer deaths.2

Many causes of lung cancer have been identified, including active cigarette smoking3; exposure to secondhand cigarette smoke (passive smoking)4; pipe and cigar smoking5; occupational exposure to agents such as asbestos, nickel, chromium, and arsenic6; exposure to radiation, including radon gas in homes and mines7; and exposure to indoor and outdoor air pollution.8 Despite the identification of this constellation of well-established causal risk factors, the global epidemic of lung cancer is primarily caused by a single factor: cigarette smoking. This dominance of cigarette smoking reflects effective marketing and promotion of an addicting and deadly product by multinational corporations.9

Refining the understanding of the etiology and pathogenesis of lung cancer remains a vibrant area of research. As described in this chapter, foci of current research include understanding the root causes of racial and socioeconomic disparities, elucidation of the role of lifestyle factors other than cigarette smoking (eg, diet, physical activity), the risk of indoor and outdoor pollutants, genetic determinants of risk, biomarkers of risk and early detection, and the potential role of infections such as HIV.

An understanding of the epidemiology of lung cancer provides background and contextual information regarding lung cancer that is important for management of guidelines. This article includes no recommendations for individual patients but is included in the American College of Chest Physicians (ACCP) Lung Cancer Guidelines to provide a foundation.

A narrative review of published evidence on the epidemiology of lung cancer was carried out. Key reports that described the occurrence of lung cancer in populations and factors that affect lung cancer risk were identified. This review was accomplished through a combination of approaches that included cataloging reports from the authors’ files and augmenting this with Medline searches that included the term ”lung cancer” and terms for various exposures that have been studied in relation to lung cancer (eg, “smoking,” “asbestos,” “radiation”). Emphasis was placed on systematic reviews, if available.

The objective was to provide a summary of the epidemiologic evidence on lung cancer, emphasizing issues that are currently relevant to prevention. The literature is extraordinarily large, and we did not attempt to conduct a comprehensive review and systematic synthesis. Such syntheses have been carried out by expert review groups, including the committees assembled to prepare the US Surgeon General’s reports on smoking and health and committees of other governments and organizations, including the UK Royal College of Physicians and Scientific Committee on Tobacco, the World Health Organization (WHO) International Agency for Research on Cancer (IARC), and the World Cancer Research Fund (WCRF).

The topics covered were agreed on by consensus of the writing committee with initial input from the ACCP Lung Cancer Guidelines Panel. Topics were added as recommended by external reviewers from the ACCP Lung Cancer Guidelines Panel, the Thoracic Oncology NetWork, Guidelines Oversight Committee (formerly known as the Health and Science Policy Committee), and the Board of Regents of the ACCP. All parties agreed to make no attempt to grade the evidence or generate formal guidelines.

The patterns of occurrence of lung cancer with respect to survival, incidence, and mortality rates are reviewed in this section, using the United States as a specific example before going on to consider global variation in rates.

2.1 Survival

The 5-year relative survival rate for lung cancer in the United States for the period of 2001 to 2007 is 16.3%, which is up from 12.3% in 1975 to 1977.10 The 5-year relative survival rate varies markedly depending on the stage at diagnosis, from 52% to 24% to 4% for local, regional, and distant stage disease, respectively.10 Stage at diagnosis accounts for the most marked variation in prognosis, but patient characteristics associated with poorer survival also include being older, male, and African American.10

2.2 Temporal Trends

Because of the high case fatality rate of lung cancer, incidence and mortality rates are nearly equivalent, and consequently, routinely collected vital statistics provide a long record of the occurrence of lung cancer. We are presently amid an epidemic of lung cancer that began in the 1930s in the United States.

2.3 Sex

As the leading cause of cancer death among women, lung cancer is a major women’s health issue. Historical trends indicate that cigarette smoking prevalence peaked about 2 decades earlier in men than in women; thus, the epidemic of lung cancer started later in women. In contrast to men, lung cancer incidence rates in women have not yet begun to decrease consistently,10 but a recent analysis of 2003 to 2007 data for the first time detected a significant downturn in incidence and mortality rates in US women.11 Far more men than women still die of lung cancer each year, but the gender gap in lung cancer mortality is steadily narrowing and is expected to close.

2.4 Race and Ethnicity
2.4.1 African Americans

The patterns of occurrence of lung cancer by race and ethnicity make lung cancer a relevant disease for cancer disparities research. Lung cancer incidence rates are similar among African American and white women, but rates are about 47% higher among African American men than among white men.10 African American men have also experienced a greater mortality from lung cancer, with the largest disparity in rates being 42% greater than for European American men in 1990; the excess decreased to 25% in 2008. Clues to birth cohort changes in disease occurrence may be found in changes in rates in young people. An analysis of trends in lung cancer incidence and mortality rates (per 100,000) from 1992 to 2006 among 20- to 39-year-olds revealed a similar narrowing of the racial gap in this age group, leading to the inference that the drop among African Americans resulted from the striking decrease in smoking prevalence among African American youth since the 1970s. If this inference is correct, continued narrowing of the racial disparity can be anticipated in the coming decades.12

This racial disparity may be partially due to a greater susceptibility of African American smokers to smoking-induced lung carcinogenesis,13 but historical differences in smoking prevalence do not explain all of the higher risks seen in African Americans compared with European Americans.14 The racial disparity in mortality reflects not only the differences in incidence but also 20% poorer 5-year relative survival among African Americans compared with whites.10 The poorer lung cancer survival in African Americans remains to be explained, but in multivariate analyses, the racial disparity is diminished by adjustment for receipt of evidence-based therapy15 and pretreatment health status.16 Thus, potential contributors to the racial disparity in survival could be treatment-related factors such as later stage at diagnosis and lack of access to, or uptake of, evidence-based stage-specific lung cancer treatment.17

Compared with African Americans (72.7) and whites (63.3), age-adjusted incidence rates per 100,000 for the years 2004 to 2008 were significantly lower among American Indians/Alaskan Natives (44.5), Asians/Pacific Islanders (39.0), and Hispanics (32.5).10 Similar patterns are observed for lung cancer mortality rates.

2.4.2 Asians

Among patients with lung cancer, those of Asian ancestry have consistently been observed to have better survival than whites.18 The reasons for the more favorable prognosis in Asians are incompletely understood, but one contributory factor is differences in tumor characteristics. For example, in Asians, the prevalence of epidermal growth factor receptor mutations in lung tumors is much higher than in whites, and epidermal growth factor receptor-positive tumors are responsive to treatment with gefitinib.19 Further delineating the distinct features of the etiology and prognosis of lung cancer in Asians compared with other ethnic groups may lead to novel insights into lung cancer pathogenesis.

2.5 Socioeconomic Status

Increasingly, lung cancer is more likely to occur in poorer and less-educated populations, primarily reflecting the increasing gradient of smoking with socioeconomic indicators that include income, education, and occupation. This pattern, noted decades ago in the United States,20 has now been observed in many countries worldwide. For example, in Canada, the risk of lung cancer was inversely associated with income, education, and social class,21 and despite universal health care, lower socioeconomic status was significantly associated with poorer lung cancer survival.22 In China, a sixfold variation in lung cancer risk was observed between the lowest and highest income categories.23 In The Netherlands, lung cancer risk was inversely associated with attained education.24 Lower socioeconomic status has also been associated with later stage at diagnosis for lung cancer as for other cancers.25 In the United States, studies of lung cancer prognosis that have examined both race/ethnicity and socioeconomic status have shown lower socioeconomic status to be a strong determinant of worse prognosis, whereas racial differences in prognosis tend to diminish when adjusted for socioeconomic status.26,27 Socioeconomic status is associated with an unfavorable profile of interacting determinants of lung cancer risk, such as smoking, diet, and exposure to inhaled carcinogens in the workplace and general environment.

2.6 Geographic Patterns

Internationally, lung cancer rates vary markedly across countries: Age-standardized incidence rates vary > 60-fold in both men and women.28 The geographic distribution is predominantly driven by historical patterns in cigarette smoking prevalence, with an approximately 20-year lag period from change in smoking pattern to change in incidence, reflecting the slow and multistep process of cancer initiation and progression.29

In men, the highest annual lung cancer incidence rates are in central and eastern Europe and North America (65.7 and 61.2 per 100,000, respectively). In women, the lung cancer incidence rates are highest in North America and northern Europe (35.6 and 21.3 per 100,000, respectively).30 For both sexes, the lowest incidence rates are in Africa. These patterns are fluid because lung cancer rates will change commensurate with changes in smoking prevalence.

The situation in China is both unique and of particular concern. Contrary to elsewhere, the high lung cancer mortality rates among Chinese women are not due to a high prevalence of cigarette smoking. Rather, the high rates appear to be a result of exposure to other risk factors that include indoor air pollution from cooking fumes.2 Chinese men are a high-risk population of particular concern because of a striking increase in their smoking rates. Per capita cigarette consumption in Chinese men increased from one cigarette per day in 1952, to four in 1972, to 10 in 1992.31 As a consequence, the lung cancer incidence rates have already increased and will continue to rise substantially. The increase in lung cancer among Chinese men will have a major impact on the global burden of lung cancer in the 21st century, given the size of this group of smokers.

The tobacco addiction epidemic in China exemplifies a shift in the global burden of lung cancer from high-income western countries to low- and middle-income countries, particularly in Asia. In 2008, newly diagnosed lung cancers in developing countries (884,500) exceeded the number in developed countries (724,300) by 22%.2 The trend of the lung cancer burden becoming increasingly concentrated in the developing world is expected to continue for the foreseeable future.

Substantial geographic variation in lung cancer mortality rates is also present within countries. For example, in the United States from 2004 to 2008, the age-adjusted lung cancer incidence rates varied 3.6-fold between the states with the highest (Kentucky, 101 per 100,000) and the lowest (Utah, 28 per 100,000) rates.10

The etiology of lung cancer can be conceptualized as reflecting the joint consequences of the interrelationship between (1) exposure to etiologic agents and (2) individual susceptibility to these agents. Synergistic interactions among risk factors can have substantial consequences for lung cancer risk. Well-known examples include the synergistic effect of cigarette smoking on the lung cancer risk associated with asbestos exposure and radon.32

Given the many known risk factors for lung cancer, a practical question for guiding prevention is the relative contribution of these factors to the overall burden of lung cancer. The population attributable risk approach takes into account the magnitude of the relative risk associated with an exposure along with the likelihood of exposure in the general population.33 These attributable risk estimates include joint contributions of risk factors that sometimes have synergistic relationships. For example, the attributable risk estimate for cigarette smoking includes the lung cancer risk attributed to the independent effects of cigarette smoking and further includes the risk of lung cancer from smoking because of its synergistic interactions with factors such as asbestos and radon. For this reason, the total percentage can exceed 100%. Lung cancer has a well-characterized set of important risk factors and established synergistic interactions between risk factors, and these reasons contribute to the attributable risks summing to considerably > 100%. For example, population attributable risk estimates for lung cancer indicate that in the United States, active smoking is responsible for 90% of lung cancer,3 and radon is responsible for 15%.7

3.1 Environmental and Occupational Agents
3.1.1 Tobacco Smoking

A single etiologic agent—cigarette smoking—is by far the leading cause of lung cancer, accounting for about 80% to 90% of lung cancer cases in the United States and other countries where cigarette smoking is common.34 Compared with never smokers, US smokers who have not quit successfully have about a 20-fold increase in lung cancer risk. Few exposures to environmental agents convey such risks for any disease. In general, spatial and temporal trends of lung cancer occurrence closely reflect patterns of smoking, but rates of occurrence lag smoking rates by about 20 years. Prior versions of this review35,36 covered smoking and lung cancer extensively, so only a summary of this voluminous literature is provided here. Lung cancer occurs in four major types as classified by light microscopy: adenocarcinoma, squamous cell carcinoma, large cell carcinoma, and small cell carcinoma. All four types are caused by cigarette smoking.37 The histologic characteristics of lung cancer in developed countries have changed during the past 50 years. Adenocarcinoma has become more common, whereas squamous cell carcinoma has declined. This shift is notable because adenocarcinoma tends to arise more peripherally and squamous cell carcinoma more centrally.38 The most likely explanation for the rise in adenocarcinoma is the changing cigarette, leading to changes in smoking topography that has included greater depth of inhalation.

Cigar smoking is also an established cause of lung cancer.5 The lung cancer risks associated with cigar smoking are substantial but less than the risks observed for cigarette smoking because of differences in smoking frequency and depth of inhalation. The same pattern holds true for pipe smoking.39 Bidi, loose tobacco rolled in a leaf, is the most commonly smoked tobacco product in India; bidi smoking also causes lung cancer. With respect to smoking of nontobacco products, despite the plausibility of marijuana as a risk factor for lung cancer, the evidence to date has not documented an association after adjusting for tobacco smoking.40 Smoking Cessation—

Cigarette smokers can benefit at any age by quitting smoking. The likelihood of developing lung cancer decreases among those who quit smoking compared with those who continue to smoke.41 As the period of abstinence from smoking cigarettes increases, the risk of lung cancer decreases.42 However, even for periods of abstinence of > 40 years, the risk of lung cancer among former smokers remains elevated compared with never smokers.42,43 The benefits derived from smoking cessation also depend on the duration of smoking; for a given period of abstinence, the decrease in risk increases as the duration of smoking decreases.42 In general, studies have shown comparable reductions in risk following cessation, regardless of sex, type of tobacco smoked, and histologic type of lung cancer.44 Tobacco dependence treatments are reviewed in the ACCP Lung Cancer Guidelines article “Treatment of Tobacco Use in Lung Cancer.”45 The Changing Cigarette—

The composition of cigarettes has evolved considerably since the 1950s. The marketplace has shifted from mainly unfiltered cigarettes to predominantly filtered cigarettes. In the mid-1960s, ventilation holes were added to the filter, which dilute the smoke with air drawn through them. However, smokers can easily block the holes with their fingers, which are left unblocked by the machines used to test cigarettes. Reconstituted tobacco has been used increasingly since the 1960s, there have been changes to the cigarette paper and additives used, and most cigarettes are more ammoniated in the United States.46 A concomitant shift toward lowered levels of tar and nicotine, as measured by a smoking machine, has occurred.47 Cigarette tar refers to the condensable residue of cigarette smoke, that is, the total particulate matter of cigarette smoke deposited on the machine’s filter less the moisture and nicotine. Tar is a complex mixture that includes many carcinogens.47

Studies show little relation between biomarkers of cigarette smoke and cigarette tar or nicotine yield as measured by Federal Trade Commission (FTC) protocol.48 These studies have been conducted in both the population context and the controlled laboratory setting. The lack of association of tar and nicotine yields with biomarker levels partially reflects compensatory changes in smoking patterns for smokers switching from higher to lower yield products. The compensation includes blocking the ventilation holes, more frequent and deeper puffs, and an increase in the number of cigarettes smoked.49

The gradual reduction in machine-measured tar yield over recent decades would be expected to have reduced smokers’ exposures to carcinogens if the FTC test protocol were predictive of carcinogen doses delivered to the lung.47 However, substantial evidence indicates that the FTC test method is not informative with regard to lung cancer risk or risks of smoking-caused diseases more generally.49,50 For lung cancer and other diseases, three lines of epidemiologic data have been available on changes in products. The first comes from case-control studies that compared the smoking history profiles of persons with lung cancer with those of control subjects. The second comes from cohort studies that tracked the risk of lung cancer over time as the products smoked changed. The third comes from assessment of the temporal changes in age-specific patterns of lung cancer mortality rates compared with changes in cigarette characteristics. These lines of evidence are convergent, and national as well as international groups that have evaluated the evidence concluded that changes in yield over time have not reduced lung cancer risk in smokers; conversely, the cigarette changes may have actually increased lung cancer risk.2,49,5153 Under the 2009 Family Smoking Prevention and Tobacco Control Act, cigarette packages will no longer provide machine-measured yields. Menthol—

Menthol is a flavoring agent that can be either derived naturally or synthesized in the laboratory. Menthol cigarettes were invented in the 1920s and currently comprise approximately one-third of the US cigarette market.54,55 The prevalence of menthol cigarette smoking is by far the highest among African Americans, reflecting patterns of aggressive and targeted marketing that began in the 1960s.9,56,57 Use of menthol cigarettes is increasing among adolescents,58,59 an increase that is greater among minorities.55,60

Menthol acts on receptors expressed primarily on sensory nerves in the nose, mouth, and airway to produce a minty taste and aroma.61 It has cooling,62 counterirritant,63 and analgesic properties.64 These sensory actions of menthol in cigarette smoke have raised concern that it facilitates experimentation and initiation of regular smoking and alters smoking topography in ways that increase doses of exposure to tobacco smoke toxins. These effects of menthol, combined with the fact that African Americans have disproportionately high lung cancer rates and a very high prevalence of menthol cigarette use, are compatible with the hypothesis that menthol cigarettes may be even more strongly associated with lung cancer risk than nonmenthol cigarettes.

However, two relevant lines of evidence involve comparing smokers of menthol and nonmenthol cigarettes with respect to (1) biomarkers and (2) lung cancer risk. Comparisons of menthol and nonmenthol cigarette smokers with respect to biomarkers of tobacco smoke exposure and dose have not revealed consistent differences, including concentration of nicotine and tobacco-specific nitrosamines.6569 The results of numerous case-control7074 and cohort75,76 studies have consistently reinforced the conclusion that menthol and nonmenthol cigarettes are associated with nearly equivalent risks of lung cancer. These lines of evidence were recently summarized by the Tobacco Products Scientific Advisory Committee (TPSAC) of the Food and Drug Administration. Under the 2009 Family Smoking Prevention and Tobacco Control Act, TPSAC was required to develop a report and recommendations to address the impact of menthol in cigarettes on public health, including specifically considering use among children, African Americans, Hispanics, and other racial and ethnic minorities. The TPSAC report concluded that the totality of the evidence did not support the hypothesis that smoking menthol cigarettes was associated with a greater risk of lung cancer than smoking nonmenthol cigarettes.77 Studies subsequent to the TPSAC report provided evidence to support TPSAC’s conclusion; notable among these was a study nested with a large cohort of racially diverse adults in which a lower lung cancer incidence was noted in menthol vs nonmenthol smokers.78

However, menthol cigarettes may still have adverse public health consequences. Several studies suggested that menthol cigarettes are a starter product that may be associated with smoking initiation.58,79 The evidence is conflicting with regard to the effects of menthol on dependence in adult smokers but more clearly points to menthol cigarettes being associated with greater nicotine dependence among adolescents.77 Furthermore, among African Americans, smokers of menthol cigarettes may have a more difficult time successfully quitting smoking than smokers of nonmenthol cigarettes.8082 The TPSAC concluded on the basis of this evidence that menthol cigarettes are detrimental to public health by increasing the number of smokers and the duration of smoking, resulting in increased smoking prevalence. Secondhand Smoke Exposure—

Passive smokers inhale a complex mixture of smoke widely referred to as secondhand smoke. The 2006 US Surgeon General’s report reinforced earlier conclusions that secondhand smoke exposure is a cause of lung cancer among nonsmokers.4 This association holds true regardless of the source of exposure to secondhand smoke, but the most abundant evidence is for nonsmokers who live with a smoker, which is associated with a 20% to 30% increased risk of lung cancer.4 IARC has classified secondhand tobacco smoke exposure as a known human (class A) carcinogen.52 Secondhand smoking is estimated to cause 3,000 lung cancer deaths per year in the United States83 and 21,400 deaths per year globally.84 Are Women More Susceptible to Smoking-Induced Lung Cancer? —

Results of some studies have suggested a potentially higher risk of smoking-associated lung cancer in women compared with men,8587 but methodologic issues cloud the interpretation of these studies, particularly due to a lack of focus on the most informative comparisons.88 Furthermore, the evidence from prospective cohort studies fails to support the notion of a sex differential in susceptibility to lung cancer from smoking.89 The equal rates of lung cancer mortality in younger US men and women corresponding to a time of equal smoking prevalence also provides evidence against an important gender difference in susceptibility to smoking-induced lung cancer.90 At present, the evidence does not favor the hypothesis because for a specific degree of smoking history, the relative risk estimates for men and women are very similar.89

Suspicion is naturally cast on a potential hormonal role for any hypothesized gender difference in disease susceptibility. A meta-analysis of two large-scale randomized controlled trials of hormonal therapy with an estrogen plus progestin formulation found a significantly increased risk of lung cancer (relative risk, 1.4; 95% CI, 1.03-1.8), pointing to a potential hormonal role in the etiology of lung cancer.91 Reports published since the meta-analysis further suggested that the lung cancer risk associated with hormonal therapy may be specific to estrogen plus progestin formulations. In the Women’s Health Initiative trial, results were null for estrogen-only formulations.92 In a prospective cohort study, increased lung cancer risk was observed for estrogen plus progestin formulations, with null results for estrogen-only formulations.93 Risk Prediction—

Models for predicting risk will be needed to guide lung cancer screening with CT scanning or other interventions targeting high-risk smokers. Given the dominance of cigarette smoking as a risk factor for lung cancer in the general population, many analyses have explored the relationships between quantitative measures of smoking and lung cancer risk. These analyses have shown the importance of smoking duration; number of cigarettes smoked; and, for former smokers, the time since quitting. A landmark analysis of the data from a cohort study of British physicians showed that duration of smoking and number of cigarettes smoked have quantitatively distinct effects on lung cancer risk and that they should not be combined for estimating lung cancer risk.94 Further information on the quantitative relationships of measures of smoking with lung cancer risk can be found in the comprehensive review presented in the 2004 IARC monograph and the 2004 report of the US Surgeon General.3,52

More recently, prediction models have been developed to estimate the probability of lung cancer occurring during specified time intervals. Such models are of particular interest for risk stratification to identify candidates for screening, given the recent demonstration of reduction in lung cancer mortality by low-dose CT scanning in the National Lung Cancer Screening Trial (NLST).95 Prediction models parallel the approach taken for breast cancer, wherein a risk factor-based model, the Gail model, has long been available.96 This model’s predictions are used for a variety of purposes, including informing patients of their potential risk, guiding screening, and selecting women for clinical trials.

Three prediction models for lung cancer are now available: the Bach model, which is based on data from the CARET (β-Carotene and Retinol Efficacy Trial); the Spitz model, which is based on data from an ongoing case-control study; and the Liverpool Lung Project, which is based on a case-control study in Liverpool, England.9799 The general approach used to develop the models is similar. The epidemiologic data are analyzed to identify risk factors and estimate the associated relative risk. The relative risk estimates are then used to project risk over time. Various statistical approaches are used to identify the most informative variables and to validate the final model. The measure of prediction used generally is the area under the curve (AUC) from receiver operating characteristic (ROC) analysis; the AUC varies from 0 to 1.0, that is, from no predictive value to perfect prediction.

Etzel and Bach100 provided a useful comparison of the three models. The AUC values for the models range from about 0.60 to 0.70. D’Amelio and colleagues101 compared the performance of the three models in an independent data set, a case-control study carried out in Boston, Massachusetts. The AUC values for the Spitz and Liverpool Lung Project models were 0.69, whereas that for the Bach model was 0.66. Spitz and colleagues98 refined the original model in two ways: (1) by adding two markers of DNA repair capacity with a slight gain in AUC98 and (2) by developing a model specifically for African Americans, which performed better than the model based on whites, in this racial group.102

Undoubtedly, these models will continue to be refined. Their predictive power may be greatly enhanced if additional genetic markers are identified to determine susceptibility to lung cancer in smokers. For now, they provide a tool for risk stratification but lack the sensitivity and specificity needed for managing individual patients.

3.1.2 Never Smokers

Tobacco smoking causes such a large proportion of all lung cancer cases that there have been few data on the occurrence of lung cancer among never smokers. Global estimates indicate that about 300,000 lung cancer deaths annually are not due to tobacco use.103 Even though this estimate represents a minority of the lung cancer burden, the incidence of lung cancer in never smokers ranges from 4.8 to 20.8 per 100,000 among individuals aged 40 to 79 years,104 comparable to the rates of myeloma in men and cervical cancer in women.

An analysis of data from 35 cohort and cancer registry studies from around the world revealed the absence of temporal trends of lung cancer in never smokers but indicated that among never smokers, lung cancer death rates are greater in men than in women and greater in African Americans and in Asians living in Asia compared with those of European ancestry.105 Among patients with lung cancer who never smoked, the known causes of lung cancer other than cigarette smoking, such as exposure to secondhand smoke, radon, and occupational carcinogens, are present in a substantial105 proportion. Conversely, a large fraction of cases were not attributable to these factors.106

3.1.3 Diet and Physical Activity

Lifestyle factors other than cigarette smoking, such as diet and exercise, have been extensively investigated for a potential role in influencing lung cancer risk. Diet—

The most thoroughly investigated dietary factors are also those that appear to have the greatest implications for prevention: fruits, vegetables, and specific antioxidant micronutrients that are commonly found in fruits and vegetables. This section relies primarily on evidence ratings from a systematic review of the world’s evidence summarized in the 2007 report of the WCRF.107 This rating scale included categories of convincing, probable, and limited-suggestive evidence of an association between a dietary factor and lung cancer.

With a WCRF evidence rating of probable, the evidence points toward greater levels of fruit consumption being inversely associated with lung cancer risk. Similar to fruit, the evidence suggests that vegetable consumption is inversely associated with lung cancer risk, but the results have been less consistent and weaker than for fruit. Hence, the overall evidence for vegetables was rated as limited-suggestive in the WCRF report.

To better understand the basis of these inverse associations, fruits and vegetables have been grouped into classes and studied in relation to lung cancer risk. For example, cruciferous vegetables have been associated with a reduced risk of lung cancer in a number of studies.108 The association with cruciferous vegetable intake has persisted even after careful control for cigarette smoking in the design of some studies.109 The evidence of an inverse association between cruciferous vegetable intake and lung cancer risk has bolstered interest in isothiocyanates as a promising chemopreventive agent. Isothiocyanates, metabolites of the class of phytochemicals known as glucosinolates, could exert anticancer effects by blocking carcinogens through induction of phase 2 detoxification enzymes, such as glutathione S-transferase. Lung cancer risk has also been consistently inversely associated with higher dietary intakes or urinary levels of isothiocyanates,110112 a constituent of cruciferous vegetables.

Fruits and vegetables are the major dietary source of antioxidant micronutrients, such as carotenoids. The example of carotenoids exemplifies the complexities involved in attempting to determine the role of diet in the etiology of lung cancer. Prospective studies of both dietary intake and prediagnostic blood concentrations suggest an inverse association between carotenoids and lung cancer.113 For example, both dietary intake and circulating concentrations of total carotenoids were associated with a 20% to 30% lower risk of lung cancer in the highest vs lowest exposure categories. On the basis of these data, the WCRF rates foods containing carotenoids as probable protective factors for lung cancer. However, it cannot be determined with certainty whether the inverse association between carotenoids and lung cancer is directly due to carotenoid intake or whether carotenoid intake merely serves as a marker of the intake of other protective substances or healthier dietary habits in general. This uncertainty is heightened by the results of large-scale randomized controlled trials that conclusively demonstrated that high-dose β-carotene consumption is associated with an increased risk of lung cancer in smokers.113

Studies of fruits, vegetables, and micronutrients have been the centerpiece of studies of diet and lung cancer, but a wide range of dietary and anthropometric factors have been investigated. For example, the results of meta-analyses suggested that alcohol drinking in the highest consumption categories is associated with an increased risk of lung cancer,114,115 but a study of alcohol drinking in never smokers was null.116 Anthropometric measures have also been studied, indicating a tendency for persons with a lower BMI to have increased lung cancer risk relative to heavier people.117,118 However, effects of both alcohol drinking and low BMI may be difficult to separate from the concomitant effects of smoking because those who smoke more cigarettes per day tend to be leaner and to drink more. At present, when considering the possible relationships between lung cancer and factors such as alcohol drinking and lower BMI, uncontrolled confounding by cigarette smoking cannot be dismissed as a possible explanation.

The overwhelming contribution of cigarette smoking as a cause of lung cancer poses a challenge to detecting the role that lifestyle factors, such as diet, may play in the etiology of lung cancer. Cigarette smoking is now so closely associated with unhealthy lifestyles in the United States and some other countries119 that it often is difficult to disentangle the dietary factors of interest from the effects of smoking. Cigarette smoke can directly affect circulating concentrations of dietary factors; for example, smokers tend to have lower circulating concentrations of antioxidant micronutrients even after accounting for differences in dietary intake.119 Additionally, associations between dietary factors and lung cancer risk are likely to be far weaker than the association with active smoking, and in general, diet is measured with much greater error than smoking. Even for a dietary factor such as fruit consumption that is inversely associated with lower lung cancer risk, the highest exposure category is typically associated with at most a halving in the risk of lung cancer. Thus, in interpreting evidence on diet and lung cancer risk, residual confounding cannot be readily set aside as an explanation for the observed associations between dietary factors and lung cancer.