Tobacco Research

Home | Research | Investigators | Harm Reduction | Resources | Publications

Ongoing Research

Following are brief descriptions of the ongoing tobacco research studies at Lombardi.

 

ADHD and Smoking

Dr. Kenneth Tercyak's translational tobacco research is supported by an NCI Research Career Development Award and examines smoking and other risky behaviors among children and adolescents seeking primary medical care, medical care for chronic illness, and medical care for attention-deficit/hyperactivity disorder (ADHD); a genetic component of this work explores genes thought to relate to tobacco use. The premise of this line of inquiry is that general and illness-specific stressors contribute to smoking among affected youth (despite smoking's well-known negative side effects on physical health), and that both smoking and ADHD symptoms are over-expressed among youth with dopamine transporter/receptor gene polymorphisms. Dr. Tercyak has found that, indeed, youth with ADHD symptoms are at high risk for smoking and that chronically ill youth are just as likely to smoke as their unaffected peers. The results of this work are expected to lead to targeted prevention and intervention programs to decrease tobacco use among high-risk populations.

Behavioral Research in the National Lung Screening Trial

The Lombardi Comprehensive Cancer Center is a national study site for both the Prostate, Lung, Colon, and Ovarian Cancer Screening Trial (PLCO) and the National Lung Screening Trial (NLST). Both trials address major research questions about screening for lung cancer in current and former smokers. The principle investigator for the PLCO and NLST trials is Edward Gelmann, MD. Matthew Freedman, MD, MBA, is lead radiology investigator for PLCO and principal co-investigator of NLST at Georgetown University.

The PLCO Trial is closed to accrual, but has randomized 155,000 men and women to screening versus observation. Included in the screening arm is an annual chest X-ray for lung cancer detection. NLST addresses the risks and efficacy of a new approach to lung cancer screening, use of spiral computerized tomography (CT). Participants in this trial were randomized to annual chest X-ray or spiral CT examinations for 3 years. This trial is also closed to accrual and screening of 50,000 participants is ongoing at Lombardi and other centers a cross the country.

Carcinogens in Human Lung Tissue

Building upon Lombardi researchers' expertise in tobacco-related carcinogenesis and nicotine addiction, studies in the area of lung cancer have been increasing. These studies range from the detection of carcinogen and DNA damage in lung tissues to studies of radiation-induced lung cancer. While at the NCI, Drs. Radoslav Goldman and Peter Shields investigated the formation of DNA adducts in human lung tissue. This work is continuing but also is being expanded to improve both the exposure assessment and the cellular reactions in these autopsy donors.

In collaboration with Dr. Edo Pellizzari of Research Triangle Institute, they have measured 11 polycyclic aromatic hydrocarbons (PAHs), known human carcinogens, in 70 lung tissue samples from cancer-free autopsy donors by gas chromatography-mass spectrometry. There were 37 smokers and 33 nonsmokers as estimated by serum cotinine concentration. The sum of PAH concentrations was higher in smokers (p = 0.01), and there was a dose-response relationship for greater smoking (p < 0.01). Smoking increased the concentration of five PAHs including benzo(a)pyrene, which increased approximately two-fold. The risk for increasing carcinogenic PAHs (odds ratio, 8.20; 95% confidence interval, 2.39-28.09) was three-fold compared with noncarcinogenic PAHs (odds ratio, 2.61; 95% confidence interval, 0.75-9.12). A higher concentration of PAHs was detected in the lung tissue of males, although the estimated smoking was similar in males and females. Race was not associated with PAH concentrations overall, but PAH concentrations appeared to be higher in African-American males than in any other group. Age was weakly correlated with an increase in fluoranthene and pyrene. The measurement of PAHs in human lung tissue can be used to estimate the actual dose to the target organ. It has more recently been determined that the level of cytochrome P450 Ia1 in the lung tissue correlates with the PAH levels. This work will now allow the investigators to relate the formation of DNA adducts to the actual carcinogen level in the target organ, as well as other cellular responses (e.g., p53 induction, estrogen receptor expression).

Computer-Aided Detection

In collaboration with the ISIS Imaging Science and Information Systems Research Center at Georgetown, the Lombardi Comprehensive Cancer Center is working to develop methods for the computer-aided analysis of chest radiographs and lung CTs for cancer. The studies have both corporate and National Cancer Institute funding. Three different types of computer tasks are under development: programs that will assist radiologists in the detection of lung nodules that could represent cancer, programs that will assist radiologists in determining whether a detected nodule is benign or malignant, and tools to assist radiologists as they interpret lung CT images by automating some of the tasks that they currently perform. This work is under the direction of Matthew Freedman, MD, from the Lombardi Comprehensive Cancer Center, and Shih-Chung Benedict Lo, PhD, from the ISIS Research Center.

DNA Repair and Lung Cancer

The primary focus of this study is to assess the risk associated with mutagen sensitivity. Drs. Christopher Loffredo and Peter Shields, in conjunction with Drs. Curtis Harris of the NCI and Anthony Alberg of Johns Hopkins University, are conducting a five-hospital case-control study in Baltimore, Maryland. Controls are recruited using Department of Motor Vehicles records. There will be 200 Caucasian and 200 African American cases, with the latter oversampled so that the investigators can assess racial differences in cancer risk. Thus far, individual sensitivity to bleomycin-induced chromosome breaks has been compared in 152 newly diagnosed and previously untreated lung cancer patients with 94 population controls and 85 hospital controls with no history of cancer. The mean number of bleomycin-induced breaks per cell was 1.01 for the cases compared with 0.86 for hospital controls (P<0.01) and 0.89 for population controls (P<0.01). The mean number of breaks per cell was 1.01 for those >65 years old and 0.81 for those <65 years old (P<0.01) among population controls. Defining "bleomycin sensitive" as greater than 0.84 break/cell (the median level in population controls), 66% of the cases were bleomycin sensitive compared with 46% of the hospital controls (adjusted odds ratio [aOR] = 2.69, 95% confidence interval [CI] = 1.33, 5.04), and 49% of the population controls (aOR = 2.18, 95% CI = 1.13, 4.21). When analyzed as a continuous variable, there was an approximately 14% increase in lung cancer risk with each unit (one unit = 0.1 breaks/cell) of increase in bleomycin-induced chromosome breaks, adjusted for age, gender, race and smoking history. These data indicate that the increased number of bleomycin-induced chromosome breaks is significantly associated with an increased risk of lung cancer in the first 331 subjects.

Evaluation of Harm Reduction through Use of Modified Cigarette Products or Nicotine Replacement Therapy

In 2000 and 2001, Dr. Peter Shields devoted an important effort to the activities of the Institute of Medicine's Committee for Assessing the Science Base for Harm Reduction. There, he conducted an extensive study of biomarkers and lung carcinogenesis in the context of establishing a quantitative dose-response relationship for smoking and cancer, and how one could assess the effectiveness of exposure reduction through the use of newly manufactured cigarette-like products or the use of supplemental nicotine replacement therapy in smokers. Integral to understanding harm reduction is the assessment of risk of low tar and full flavor cigarettes, such as the light cigarette, erroneously considered a harm reduction method in the 1960s.

Genetics of Nicotine Addiction and Smoking Behavior: Transdisciplinary Tobacco Use Research Center (TTURC)

The Transdisciplinary Tobacco Use Research Center (TTURC), awarded to Lombardi in 1999, spawned much of the center's tobacco-related research. The overall goals of the TTURC are to: 1) obtain a more complete understanding of the role of specific genetic factors and bio-behavioral mechanisms that promote tobacco use; and 2) apply this knowledge to the prevention, treatment and reductions of harm from tobacco exposure. In 2001, Dr. Lerman, PI of the TTURC, moved to the University of Pennsylvania. Several components of the TTURC remained here, including a project to investigate smoking genotypes (Dr. Peter Shields), a genetics laboratory (Dr. Peter Shields), and an investigation of the cost-effectiveness of alternative smoking cessation approaches (Dr. William Lawrence).

The unifying theme for the TTURC is that there is a genetic basis to tobacco addition, cessation and smoking behavior, and that the genes governing such then affect carcinogenic outcomes. The initial basis of the TTURC was based upon a case-control study of smokers and non-smokers, and work on this study set continues today. Drs. Shields and Lerman initially reported in 1999 that genetic polymorphisms in the dopamine transporter gene and the dopamine D2 receptor gene predicted tobacco use and number of quitting attempts. They also reported that there was an interaction with depression and a genetic polymorphism in the dopamine D4 receptor, while a different allele was a risk factor in smoking and smoking relapse in African Americans. Also, a genetic polymorphism in the serotonin transporter gene affects people with higher neuroticism scores to increase smoking risk. Several other genes have been investigated, but no relationship to addiction was found.

The current three TTURC projects examine the role of the above genes and others. The projects are designed to span the spectrum, from smoking initiation in adolescents to determinants of individuals' smoking amounts and how this affects DNA damage in the blood. The project based at Lombardi is carried out by Drs. Peter Shields, Christopher Loffredo and Radoslav Goldman. The aim is to determine if an individual's need to have a specific nicotine level is genetically controlled. It also will determine if genetic variability affects how people smoke (e.g., puffs per cigarette, depth of inhalation) and how this is related to the downstream formation of carcinogen-DNA adducts. The study is now in the field, where 200 Caucasian and 200 African American regular smokers are being recruited from the general population. Oversampling for African Americans will allow for the study of more typical smoking patterns within that group (i.e., menthol cigarette use).

Dr. Shields also provides the molecular analyses for the off-site TTURC projects.

  • Dr. Janet Audrain (University of Pennsylvania) is carrying out a prospective study of high school students to determine if genetic risk factors predict which adolescents progress to becoming regular smokers after experimentation (1,136 subjects accrued).
  • Dr. Lerman has completed a placebo controlled study for the genetic determinants of smoking cessation relating to Zyban, which is the only non-nicotine FDA approved drug for smoking cessation. Genetic polymorphism in CYP2B6 predicted successful quitting in a bupropion cessation trial, and that there was a greater effect in women then men. The polymorphism did not predict the success of bupropion, but it did identify successful quitting in both treated and untreated subjects, suggesting that it affects nicotine metabolism.
  • Dr. Lerman also directs a clinical trial comparing nicotine patches to nicotine nasal spray to see if genetic polymorphisms predict the optimal type of replacement therapy in an individual.

Molecular Epidemiology of Secondary Lung Cancer Risk

Several studies indicate that women with breast cancer who undergo radiotherapy are susceptible to secondary lung cancer, whether or not they are smokers. However, all studies to date have had methodological limitations, have used crude estimates of radiation doses, and have included small numbers of individuals. Moreover, none have used molecular markers, which can improve exposure assessments and elucidate mediating mechanisms. In collaboration with Dr. Per Hall and coworkers at the Karolinska Institute, along with Dr. Timothy Jorgensen and Baljit Singh, Drs. Peter Shields and Bassem Haddad are studying this risk in detail by conducting a population-based study using the Swedish Cancer Registry (SCR). In this study, the investigators can provide accurate radiation dose estimates for the whole lung and also for each side of the lung (which incorporates an estimate of scatter) and will have reliable smoking data.

Using the SCR and the unparalleled ability to obtain tissue blocks dating from the 1950s, Dr. Shields and colleagues have the unique opportunity to understand risk in the context of molecular markers, namely those that reflect p53 inactivation pathways (i.e., p53 mutations or loss of heterozygosity). Also, given that tumor blocks will be available for both the breast and lung tumors from the same women (proven to be useful even though dating back to 1958, see below), they also have the unique opportunity to study women who might be considered to have a phenotype of susceptibility to multiple primary cancers.

The team obtained data from the SCR to estimate the risks of developing lung cancer in Sweden. From the SCR, 141,053 women had histologically confirmed invasive breast cancer as the first cancer reported between 1958 and 1997, and 613 of these had a subsequent primary lung cancer. A statistically significant increased standardized incidence ratio (SIR) of 1.32 (95% CI 1.22-1.43) for lung cancer risk was seen. During the first five years there was a significantly decreased risk, followed by significantly increased risks in the subsequent periods, reaching its maximum >20 years after breast diagnosis (SIR = 2.53; 95% CI 2.10 - 3.04). To study the potential impact of radiation therapy on the lung cancer risk, the predictive value of treatment to one breast was considered in relation to the side of the lung where the lung cancer subsequently occurred. There was a statistically significant concordance after more than 10 years after breast cancer diagnosis. Also, the team has been reviewing medical records and obtaining tumor blocks to evaluate p53 mutations, and has shown that they can successfully amplify p53 from 90% of the blocks, even those dating back to 1958.

An extension of the p53 analysis is also underway with Dr. Frederica Perera of Columbia University to evaluate men in the Physician's Health Study. Dr. Perera already demonstrated that carcinogen-DNA adduct levels in the blood were prospectively predictive of lung cancer. The results from this study indicated that the men with G®T transversions, thought to be related to smoking, also had the highest adduct levels (unpublished data). However, the number of study subjects examined was small and the investigators will replicate these activities when tumor blocks are collected for an updated cohort.

Neuronal Nicotinic Acetylcholine Receptors

Dr. Ken Kellar is studying nicotine receptors and how they relate to nicotine addiction. Dr. Kellar provided the first characterization of the brain receptor's pharmacological properties, brain region distribution and regulation. Subsequently, he has shown that these receptors also are increased in the brains of humans who smoke, indicating that the nicotine-induced increase in these receptors is an important consequence of smoking and suggesting it may be related to nicotine's addictive actions. He also published the first pictures (autoradiographs) demonstrating the effects of smoking on nicotinic receptors in human brain. In recent years, Dr. Kellar has developed new tools and methods, including radiolabeled epibatidine and antibody methods, which allow for the study of the characteristics and regulation of different subtypes of nicotinic acetylcholine receptors in the nervous system. In addition, he has created a library of stably transfected cell lines that express six different neuronal nicotinic receptor subtypes that are thought to serve normal physiological functions and that may be potential targets for nicotine's actions.

Tobacco Research in Egypt

Drs. Christopher Loffredo and Radoslav Goldman, in collaboration with researchers at the University of Maryland, were recently funded to develop a tobacco research infrastructure in Egypt. They will conduct two studies within a combined NIH/WHO tobacco research center for Egypt and the Middle East, a large program of research, community education and prevention, and policy studies. The research projects are: 1) an assessment of the attributable risk of cigarette and water-pipe smoking behaviors to lung, head and neck, and bladder cancers and; 2) a study of dopamine pathway genetic polymorphisms as mediators for tobacco addition, withdrawal, behavior and quitting success.

Using the Internet to Disseminate Tobacco Dependence Treatment

Reducing smoking prevalence requires innovative approaches to reach and treat current smokers. The Public Health Service (PHS) Guideline [3] recommends key components (behavioral coping, social support, pharmacological), greater intensity and combined delivery modes (telephone, self-help) as part of effective treatment. The Internet is an untapped resource that can potentially deliver proven cessation interventions to smokers. Little is known about what type of smoker uses the Internet, the components of Internet interventions that are most used, and what the mediators and moderators of effective use are. This study, being one of the first to evaluate the Internet, is focused on initial efficacy with an eye toward eventual dissemination. The Internet could have public health benefit as a cost-effective way to make an overall population impact.

This study aims to extend existing theory and application by comparing the efficacy and cost-effectiveness of a full service, tailored Internet intervention (Premium Internet) alone or in conjunction with proactive telephone counseling (Premium Internet plus Telephone) against a standard Internet comparison condition (Basic Internet). Since increased intensity and combined treatments are associated with better outcomes, adding proactive telephone contacts to Internet treatment will further improve population impact.

This study will recruit smokers who use an Internet search engine to find smoking cessation programs. A random subset will be directed to a Web page that will describe the study and procedures for enrollment and consent them to participate. Using a randomized, controlled design with repeated measures at baseline, and 3-, 6-, 12-, and 18-months post randomization (6-, 12-, and 18-month follow-ups corresponding to 0-, 6-, and 12-months post-treatment), consented smokers will be assigned to one of three treatment conditions: 1) Basic Internet comparison group; 2) Premium Internet; and 3) Premium Internet plus Proactive Telephone Counseling.